diff --git "a/train_job_output.txt" "b/train_job_output.txt"
--- "a/train_job_output.txt"
+++ "b/train_job_output.txt"
@@ -1,4 +1,4 @@
-slurm submission log: 2024-05-23 14:58:55.959945
+slurm submission log: 2024-05-24 11:42:13.175719
 created following sbatch script: 
 
 ###############################
@@ -7,24 +7,24 @@ created following sbatch script:
 
 #SBATCH --account=nlp
 #SBATCH --cpus-per-task=16
-#SBATCH --dependency=afterok:7645749
+#SBATCH --dependency=afterok:7648457
 #SBATCH --gres=gpu:2
-#SBATCH --job-name=tthrush-job-2504325
+#SBATCH --job-name=tthrush-job-3227437
 #SBATCH --mem=100G
 #SBATCH --nodelist=sphinx2
 #SBATCH --open-mode=append
-#SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_initial_init/llms/pythia-70m_arc_easy_1/train_job_output.txt
+#SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_initial_init_min_threshold/llms/pythia-70m_arc_easy_1/train_job_output.txt
 #SBATCH --partition=sphinx
 #SBATCH --time=14-0
 
 # activate your desired anaconda environment
-. /nlp/scr/tthrush/miniconda3/etc/profile.d/conda.sh ; conda activate pretraining-coreset-selection
+. /nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/etc/profile.d/conda.sh ; conda activate pretraining-coreset-selection
 
 # cd to working directory
 cd .
 
 # launch commands
-srun --unbuffered run_as_child_processes 'torchrun --master_port 29527 --nproc_per_node=2 train_llm.py --dataset_id /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_initial_init/data/arc_easy --output_dir /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_initial_init/llms/pythia-70m_arc_easy_1 --output_hub_id pythia-70m_arc_easy --model_id EleutherAI/pythia-70m --learning_rate 1e-3 --warmup_ratio=0.1 --gradient_accumulation_steps 2 --per_device_train_batch_size 256 --seed 1 --num_train_epochs 1'
+srun --unbuffered run_as_child_processes 'torchrun --master_port 29527 --nproc_per_node=2 train_llm.py --dataset_id /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_initial_init_min_threshold/data/arc_easy --output_dir /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_initial_init_min_threshold/llms/pythia-70m_arc_easy_1 --output_hub_id pythia-70m_arc_easy --model_id EleutherAI/pythia-70m --learning_rate 1e-3 --warmup_ratio=0.1 --gradient_accumulation_steps 2 --per_device_train_batch_size 256 --seed 1 --num_train_epochs 14'
 
 ###############################
 
@@ -34,462 +34,221 @@ submission to slurm complete!
 ###############################
 slurm submission output
 
-Submitted batch job 7645750
+Submitted batch job 7648458
 
 
 
 ###############################
 
+/var/lib/slurm/slurmd/job7648458/slurm_script: line 16: /nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/etc/profile.d/conda.sh: No such file or directory
+
+CommandNotFoundError: Your shell has not been properly configured to use 'conda activate'.
+To initialize your shell, run
+
+    $ conda init <SHELL_NAME>
+
+Currently supported shells are:
+  - bash
+  - fish
+  - tcsh
+  - xonsh
+  - zsh
+  - powershell
+
+See 'conda init --help' for more information and options.
+
+IMPORTANT: You may need to close and restart your shell after running 'conda init'.
+
+
+###############################
+start time: 2024-05-24 11:43:09.843114
+machine: sphinx2
+conda env: pretraining-coreset-selection
+###############################
+running following processes
+
+	torchrun --master_port 29527 --nproc_per_node=2 train_llm.py --dataset_id /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_initial_init_min_threshold/data/arc_easy --output_dir /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_initial_init_min_threshold/llms/pythia-70m_arc_easy_1 --output_hub_id pythia-70m_arc_easy --model_id EleutherAI/pythia-70m --learning_rate 1e-3 --warmup_ratio=0.1 --gradient_accumulation_steps 2 --per_device_train_batch_size 256 --seed 1 --num_train_epochs 14
+
+
+###############################
+command outputs: 
+
+
+[2024-05-24 11:43:11,843] torch.distributed.run: [WARNING] 
+[2024-05-24 11:43:11,843] torch.distributed.run: [WARNING] *****************************************
+[2024-05-24 11:43:11,843] torch.distributed.run: [WARNING] Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. 
+[2024-05-24 11:43:11,843] torch.distributed.run: [WARNING] *****************************************
+05/24/2024 11:43:15 - INFO - __main__ - Script parameters ScriptArguments(seed=1, dataset_id='/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_initial_init_min_threshold/data/arc_easy', output_dir='/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_initial_init_min_threshold/llms/pythia-70m_arc_easy_1', output_hub_id='pythia-70m_arc_easy', hf_hub_token=True, model_id='EleutherAI/pythia-70m', per_device_train_batch_size=256, num_train_epochs=14.0, learning_rate=0.001, gradient_accumulation_steps=2, from_scratch=True, warmup_ratio=0.1, adam_beta1=0.9, adam_beta2=0.95, adam_epsilon=1e-08, weight_decay=0.01, lr_scheduler_type='cosine', local_rank=0, resume_from_checkpoint=False, deepspeed=None, peft=False)
+05/24/2024 11:43:16 - INFO - __main__ - Script parameters ScriptArguments(seed=1, dataset_id='/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_initial_init_min_threshold/data/arc_easy', output_dir='/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_initial_init_min_threshold/llms/pythia-70m_arc_easy_1', output_hub_id='pythia-70m_arc_easy', hf_hub_token=True, model_id='EleutherAI/pythia-70m', per_device_train_batch_size=256, num_train_epochs=14.0, learning_rate=0.001, gradient_accumulation_steps=2, from_scratch=True, warmup_ratio=0.1, adam_beta1=0.9, adam_beta2=0.95, adam_epsilon=1e-08, weight_decay=0.01, lr_scheduler_type='cosine', local_rank=0, resume_from_checkpoint=False, deepspeed=None, peft=False)
+Traceback (most recent call last):
+  File "/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/train_llm.py", line 204, in <module>
+    train_model()
+  File "/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/train_llm.py", line 164, in train_model
+    train_dataset = load_from_disk(script_args.dataset_id)
+  File "/nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/lib/python3.10/site-packages/datasets/load.py", line 2638, in load_from_disk
+    raise FileNotFoundError(
+FileNotFoundError: Directory /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_initial_init_min_threshold/data/arc_easy is neither a `Dataset` directory nor a `DatasetDict` directory.
+Traceback (most recent call last):
+  File "/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/train_llm.py", line 204, in <module>
+    train_model()
+  File "/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/train_llm.py", line 164, in train_model
+    train_dataset = load_from_disk(script_args.dataset_id)
+  File "/nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/lib/python3.10/site-packages/datasets/load.py", line 2638, in load_from_disk
+    raise FileNotFoundError(
+FileNotFoundError: Directory /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_initial_init_min_threshold/data/arc_easy is neither a `Dataset` directory nor a `DatasetDict` directory.
+[2024-05-24 11:43:21,865] torch.distributed.elastic.multiprocessing.api: [ERROR] failed (exitcode: 1) local_rank: 0 (pid: 2635429) of binary: /nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/bin/python
+Traceback (most recent call last):
+  File "/nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/bin/torchrun", line 8, in <module>
+    sys.exit(main())
+  File "/nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/lib/python3.10/site-packages/torch/distributed/elastic/multiprocessing/errors/__init__.py", line 347, in wrapper
+    return f(*args, **kwargs)
+  File "/nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/lib/python3.10/site-packages/torch/distributed/run.py", line 812, in main
+    run(args)
+  File "/nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/lib/python3.10/site-packages/torch/distributed/run.py", line 803, in run
+    elastic_launch(
+  File "/nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/lib/python3.10/site-packages/torch/distributed/launcher/api.py", line 135, in __call__
+    return launch_agent(self._config, self._entrypoint, list(args))
+  File "/nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/lib/python3.10/site-packages/torch/distributed/launcher/api.py", line 268, in launch_agent
+    raise ChildFailedError(
+torch.distributed.elastic.multiprocessing.errors.ChildFailedError: 
+============================================================
+train_llm.py FAILED
+------------------------------------------------------------
+Failures:
+[1]:
+  time      : 2024-05-24_11:43:21
+  host      : sphinx2.stanford.edu
+  rank      : 1 (local_rank: 1)
+  exitcode  : 1 (pid: 2635430)
+  error_file: <N/A>
+  traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html
+------------------------------------------------------------
+Root Cause (first observed failure):
+[0]:
+  time      : 2024-05-24_11:43:21
+  host      : sphinx2.stanford.edu
+  rank      : 0 (local_rank: 0)
+  exitcode  : 1 (pid: 2635429)
+  error_file: <N/A>
+  traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html
+============================================================
+###############################
+end time: 2024-05-24 11:43:29.871511
+elapsed time: 0:00:20.028397
+slurm submission log: 2024-05-24 11:46:18.773501
+created following sbatch script: 
+
+###############################
+
+#!/bin/bash
+
+#SBATCH --account=nlp
+#SBATCH --cpus-per-task=16
+#SBATCH --dependency=afterok:7648489
+#SBATCH --gres=gpu:2
+#SBATCH --job-name=tthrush-job-1194216
+#SBATCH --mem=100G
+#SBATCH --nodelist=sphinx2
+#SBATCH --open-mode=append
+#SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_initial_init_min_threshold/llms/pythia-70m_arc_easy_1/train_job_output.txt
+#SBATCH --partition=sphinx
+#SBATCH --time=14-0
+
+# activate your desired anaconda environment
+. /nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/etc/profile.d/conda.sh ; conda activate pretraining-coreset-selection
+
+# cd to working directory
+cd .
+
+# launch commands
+srun --unbuffered run_as_child_processes 'torchrun --master_port 29527 --nproc_per_node=2 train_llm.py --dataset_id /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_initial_init_min_threshold/data/arc_easy --output_dir /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_initial_init_min_threshold/llms/pythia-70m_arc_easy_1 --output_hub_id pythia-70m_arc_easy --model_id EleutherAI/pythia-70m --learning_rate 1e-3 --warmup_ratio=0.1 --gradient_accumulation_steps 2 --per_device_train_batch_size 256 --seed 1 --num_train_epochs 14'
+
+###############################
+
+submission to slurm complete!
+
+
+###############################
+slurm submission output
+
+Submitted batch job 7648490
+
+
+
+###############################
+
+/var/lib/slurm/slurmd/job7648490/slurm_script: line 16: /nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/etc/profile.d/conda.sh: No such file or directory
+
+CommandNotFoundError: Your shell has not been properly configured to use 'conda activate'.
+To initialize your shell, run
+
+    $ conda init <SHELL_NAME>
+
+Currently supported shells are:
+  - bash
+  - fish
+  - tcsh
+  - xonsh
+  - zsh
+  - powershell
+
+See 'conda init --help' for more information and options.
+
+IMPORTANT: You may need to close and restart your shell after running 'conda init'.
+
+
 ###############################
-start time: 2024-05-23 23:44:14.982041
+start time: 2024-05-24 13:44:10.592875
 machine: sphinx2
 conda env: pretraining-coreset-selection
 ###############################
 running following processes
 
-	torchrun --master_port 29527 --nproc_per_node=2 train_llm.py --dataset_id /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_initial_init/data/arc_easy --output_dir /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_initial_init/llms/pythia-70m_arc_easy_1 --output_hub_id pythia-70m_arc_easy --model_id EleutherAI/pythia-70m --learning_rate 1e-3 --warmup_ratio=0.1 --gradient_accumulation_steps 2 --per_device_train_batch_size 256 --seed 1 --num_train_epochs 1
+	torchrun --master_port 29527 --nproc_per_node=2 train_llm.py --dataset_id /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_initial_init_min_threshold/data/arc_easy --output_dir /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_initial_init_min_threshold/llms/pythia-70m_arc_easy_1 --output_hub_id pythia-70m_arc_easy --model_id EleutherAI/pythia-70m --learning_rate 1e-3 --warmup_ratio=0.1 --gradient_accumulation_steps 2 --per_device_train_batch_size 256 --seed 1 --num_train_epochs 14
 
 
 ###############################
 command outputs: 
 
 
-[2024-05-23 23:44:17,343] torch.distributed.run: [WARNING] 
-[2024-05-23 23:44:17,343] torch.distributed.run: [WARNING] *****************************************
-[2024-05-23 23:44:17,343] torch.distributed.run: [WARNING] Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. 
-[2024-05-23 23:44:17,343] torch.distributed.run: [WARNING] *****************************************
-05/23/2024 23:44:24 - INFO - __main__ - Script parameters ScriptArguments(seed=1, dataset_id='/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_initial_init/data/arc_easy', output_dir='/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_initial_init/llms/pythia-70m_arc_easy_1', output_hub_id='pythia-70m_arc_easy', hf_hub_token=True, model_id='EleutherAI/pythia-70m', per_device_train_batch_size=256, num_train_epochs=1.0, learning_rate=0.001, gradient_accumulation_steps=2, from_scratch=True, warmup_ratio=0.1, adam_beta1=0.9, adam_beta2=0.95, adam_epsilon=1e-08, weight_decay=0.01, lr_scheduler_type='cosine', local_rank=0, resume_from_checkpoint=False, deepspeed=None, peft=False)
-05/23/2024 23:44:24 - INFO - __main__ - Script parameters ScriptArguments(seed=1, dataset_id='/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_initial_init/data/arc_easy', output_dir='/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_initial_init/llms/pythia-70m_arc_easy_1', output_hub_id='pythia-70m_arc_easy', hf_hub_token=True, model_id='EleutherAI/pythia-70m', per_device_train_batch_size=256, num_train_epochs=1.0, learning_rate=0.001, gradient_accumulation_steps=2, from_scratch=True, warmup_ratio=0.1, adam_beta1=0.9, adam_beta2=0.95, adam_epsilon=1e-08, weight_decay=0.01, lr_scheduler_type='cosine', local_rank=0, resume_from_checkpoint=False, deepspeed=None, peft=False)
-
  0%|          | 0/10707 [00:00<?, ?it/s][rank1]:[W reducer.cpp:1360] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters in the forward pass. This flag results in an extra traversal of the autograd graph every iteration,  which can adversely affect performance. If your model indeed never has any unused parameters in the forward pass, consider turning this flag off. Note that this warning may be a false positive if your model has flow control causing later iterations to have unused parameters. (function operator())
-[rank0]:[W reducer.cpp:1360] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters in the forward pass. This flag results in an extra traversal of the autograd graph every iteration,  which can adversely affect performance. If your model indeed never has any unused parameters in the forward pass, consider turning this flag off. Note that this warning may be a false positive if your model has flow control causing later iterations to have unused parameters. (function operator())
-
  0%|          | 1/10707 [00:14<42:20:00, 14.24s/it]
  0%|          | 2/10707 [00:22<31:34:53, 10.62s/it]
  0%|          | 3/10707 [00:29<27:23:26,  9.21s/it]
  0%|          | 4/10707 [00:36<24:35:26,  8.27s/it]
  0%|          | 5/10707 [00:43<22:44:55,  7.65s/it]
  0%|          | 6/10707 [00:49<21:20:09,  7.18s/it]
  0%|          | 7/10707 [00:55<20:07:06,  6.77s/it]
  0%|          | 8/10707 [01:01<19:05:20,  6.42s/it]
  0%|          | 9/10707 [01:06<18:16:50,  6.15s/it]
  0%|          | 10/10707 [01:11<17:28:06,  5.88s/it]
  0%|          | 11/10707 [01:16<16:43:18,  5.63s/it]
  0%|          | 12/10707 [01:21<15:51:04,  5.34s/it]
  0%|          | 13/10707 [01:26<15:12:17,  5.12s/it]
  0%|          | 14/10707 [01:30<14:22:46,  4.84s/it]
  0%|          | 15/10707 [01:34<13:27:18,  4.53s/it]
  0%|          | 16/10707 [01:38<12:47:17,  4.31s/it]
  0%|          | 17/10707 [01:41<12:00:56,  4.05s/it]
  0%|          | 18/10707 [01:45<11:42:21,  3.94s/it]
  0%|          | 19/10707 [01:48<11:16:09,  3.80s/it]
  0%|          | 20/10707 [01:52<10:55:44,  3.68s/it]
  0%|          | 21/10707 [01:55<10:31:29,  3.55s/it]
  0%|          | 22/10707 [01:58<10:15:25,  3.46s/it]
  0%|          | 23/10707 [02:01<9:59:36,  3.37s/it] 
  0%|          | 24/10707 [02:04<9:42:50,  3.27s/it]
  0%|          | 25/10707 [02:07<9:11:46,  3.10s/it]
                                                    {'loss': 10.6692, 'grad_norm': 1.3864414691925049, 'learning_rate': 2.3342670401493933e-05, 'epoch': 0.0}
-

  0%|          | 25/10707 [02:07<9:11:46,  3.10s/it]
  0%|          | 26/10707 [02:10<8:53:12,  3.00s/it]
  0%|          | 27/10707 [02:12<8:38:32,  2.91s/it]
  0%|          | 28/10707 [02:15<8:26:45,  2.85s/it]
  0%|          | 29/10707 [02:18<8:07:14,  2.74s/it]
  0%|          | 30/10707 [02:20<7:52:25,  2.65s/it]
  0%|          | 31/10707 [02:22<7:31:49,  2.54s/it]
  0%|          | 32/10707 [02:25<7:18:49,  2.47s/it]
  0%|          | 33/10707 [02:27<7:04:58,  2.39s/it]
  0%|          | 34/10707 [02:29<6:54:16,  2.33s/it]
  0%|          | 35/10707 [02:31<6:44:04,  2.27s/it]
  0%|          | 36/10707 [02:33<6:40:55,  2.25s/it]
  0%|          | 37/10707 [02:35<6:28:40,  2.19s/it]
  0%|          | 38/10707 [02:38<6:26:18,  2.17s/it]
  0%|          | 39/10707 [02:40<6:17:14,  2.12s/it]
  0%|          | 40/10707 [02:41<6:05:10,  2.05s/it]
  0%|          | 41/10707 [02:43<5:59:26,  2.02s/it]
  0%|          | 42/10707 [02:45<5:50:12,  1.97s/it]
  0%|          | 43/10707 [02:47<5:38:14,  1.90s/it]
  0%|          | 44/10707 [02:49<5:33:57,  1.88s/it]
  0%|          | 45/10707 [02:51<6:00:05,  2.03s/it]
  0%|          | 46/10707 [02:53<6:10:50,  2.09s/it]
  0%|          | 47/10707 [02:56<6:13:45,  2.10s/it]
  0%|          | 48/10707 [02:58<6:25:05,  2.17s/it]
  0%|          | 49/10707 [03:00<6:37:44,  2.24s/it]
  0%|          | 50/10707 [03:03<6:49:43,  2.31s/it]
                                                    {'loss': 9.959, 'grad_norm': 1.237808108329773, 'learning_rate': 4.6685340802987865e-05, 'epoch': 0.0}
-
  0%|          | 50/10707 [03:03<6:49:43,  2.31s/it]
  0%|          | 51/10707 [03:05<6:37:09,  2.24s/it]
  0%|          | 52/10707 [03:07<6:50:24,  2.31s/it]
  0%|          | 53/10707 [03:10<6:52:18,  2.32s/it]
  1%|          | 54/10707 [03:12<7:05:29,  2.40s/it]
  1%|          | 55/10707 [03:15<7:08:37,  2.41s/it]
  1%|          | 56/10707 [03:17<7:22:52,  2.49s/it]
  1%|          | 57/10707 [03:20<7:40:24,  2.59s/it]
  1%|          | 58/10707 [03:22<6:39:27,  2.25s/it]
  1%|          | 59/10707 [03:23<5:53:43,  1.99s/it]
  1%|          | 60/10707 [03:24<5:24:23,  1.83s/it]
  1%|          | 61/10707 [03:26<4:56:50,  1.67s/it]
  1%|          | 62/10707 [03:27<4:37:30,  1.56s/it]
  1%|          | 63/10707 [03:28<4:23:11,  1.48s/it]
  1%|          | 64/10707 [03:30<4:18:18,  1.46s/it]
  1%|          | 65/10707 [03:31<4:11:50,  1.42s/it]
  1%|          | 66/10707 [03:32<4:06:04,  1.39s/it]
  1%|          | 67/10707 [03:34<3:58:31,  1.35s/it]
  1%|          | 68/10707 [03:35<3:53:32,  1.32s/it]
  1%|          | 69/10707 [03:36<3:55:23,  1.33s/it]
  1%|          | 70/10707 [03:38<3:49:16,  1.29s/it]
  1%|          | 71/10707 [03:39<3:42:57,  1.26s/it]
  1%|          | 72/10707 [03:40<3:36:57,  1.22s/it]
  1%|          | 73/10707 [03:41<3:37:42,  1.23s/it]
  1%|          | 74/10707 [03:42<3:38:32,  1.23s/it]
  1%|          | 75/10707 [03:43<3:34:42,  1.21s/it]
                                                    
{'loss': 9.2536, 'grad_norm': 1.0226235389709473, 'learning_rate': 7.00280112044818e-05, 'epoch': 0.01}
-
  1%|          | 75/10707 [03:43<3:34:42,  1.21s/it]
  1%|          | 76/10707 [03:45<3:33:45,  1.21s/it]
  1%|          | 77/10707 [03:46<3:32:56,  1.20s/it]
  1%|          | 78/10707 [03:47<3:33:27,  1.20s/it]
  1%|          | 79/10707 [03:48<3:32:02,  1.20s/it]
  1%|          | 80/10707 [03:49<3:24:58,  1.16s/it]
  1%|          | 81/10707 [03:51<3:27:20,  1.17s/it]
  1%|          | 82/10707 [03:52<3:25:59,  1.16s/it]
  1%|          | 83/10707 [03:53<3:25:20,  1.16s/it]
  1%|          | 84/10707 [03:54<3:18:07,  1.12s/it]
  1%|          | 85/10707 [03:55<3:21:19,  1.14s/it]
  1%|          | 86/10707 [03:56<3:17:25,  1.12s/it]
  1%|          | 87/10707 [03:57<3:17:58,  1.12s/it]
  1%|          | 88/10707 [03:58<3:14:30,  1.10s/it]
  1%|          | 89/10707 [03:59<3:15:54,  1.11s/it]
  1%|          | 90/10707 [04:00<3:13:52,  1.10s/it]
  1%|          | 91/10707 [04:01<3:09:10,  1.07s/it]
  1%|          | 92/10707 [04:03<3:07:48,  1.06s/it]
  1%|          | 93/10707 [04:03<3:01:36,  1.03s/it]
  1%|          | 94/10707 [04:04<3:02:13,  1.03s/it]
  1%|          | 95/10707 [04:05<2:58:56,  1.01s/it]
  1%|          | 96/10707 [04:06<2:52:32,  1.02it/s]
  1%|          | 97/10707 [04:07<2:56:30,  1.00it/s]
  1%|          | 98/10707 [04:08<2:50:11,  1.04it/s]
  1%|          | 99/10707 [04:09<2:53:16,  1.02it/s]
  1%|          | 100/10707 [04:10<2:54:13,  1.01it/s]{'loss': 8.4913, 'grad_norm': 0.7243988513946533, 'learning_rate': 9.337068160597573e-05, 'epoch': 0.01}
                                                     
-
  1%|          | 100/10707 [04:10<2:54:13,  1.01it/s]
  1%|          | 101/10707 [04:11<2:50:49,  1.03it/s]
  1%|          | 102/10707 [04:12<2:51:32,  1.03it/s]
  1%|          | 103/10707 [04:13<2:43:53,  1.08it/s]
  1%|          | 104/10707 [04:14<2:41:48,  1.09it/s]
  1%|          | 105/10707 [04:15<2:45:41,  1.07it/s]
  1%|          | 106/10707 [04:16<2:40:20,  1.10it/s]
  1%|          | 107/10707 [04:17<2:42:37,  1.09it/s]
  1%|          | 108/10707 [04:18<2:41:16,  1.10it/s]
  1%|          | 109/10707 [04:18<2:35:33,  1.14it/s]
  1%|          | 110/10707 [04:19<2:40:03,  1.10it/s]
  1%|          | 111/10707 [04:20<2:40:02,  1.10it/s]
  1%|          | 112/10707 [04:21<2:36:02,  1.13it/s]
  1%|          | 113/10707 [04:22<2:37:33,  1.12it/s]
  1%|          | 114/10707 [04:23<2:36:47,  1.13it/s]
  1%|          | 115/10707 [04:24<2:37:03,  1.12it/s]
  1%|          | 116/10707 [04:25<2:36:37,  1.13it/s]
  1%|          | 117/10707 [04:25<2:32:59,  1.15it/s]
  1%|          | 118/10707 [04:26<2:34:54,  1.14it/s]
  1%|          | 119/10707 [04:27<2:34:53,  1.14it/s]
  1%|          | 120/10707 [04:28<2:29:48,  1.18it/s]
  1%|          | 121/10707 [04:29<2:30:54,  1.17it/s]
  1%|          | 122/10707 [04:30<2:29:50,  1.18it/s]
  1%|          | 123/10707 [04:30<2:23:59,  1.23it/s]
  1%|          | 124/10707 [04:31<2:26:58,  1.20it/s]
  1%|          | 125/10707 [04:32<2:24:38,  1.22it/s]{'loss': 7.8692, 'grad_norm': 0.45226630568504333, 'learning_rate': 0.00011671335200746966, 'epoch': 0.01}
-
                                                     

  1%|          | 125/10707 [04:32<2:24:38,  1.22it/s]
  1%|          | 126/10707 [04:33<2:20:55,  1.25it/s]
  1%|          | 127/10707 [04:34<2:23:54,  1.23it/s]
  1%|          | 128/10707 [04:35<2:24:50,  1.22it/s]
  1%|          | 129/10707 [04:35<2:18:39,  1.27it/s]
  1%|          | 130/10707 [04:36<2:20:36,  1.25it/s]
  1%|          | 131/10707 [04:37<2:22:06,  1.24it/s]
  1%|          | 132/10707 [04:38<2:17:01,  1.29it/s]
  1%|          | 133/10707 [04:38<2:14:52,  1.31it/s]
  1%|▏         | 134/10707 [04:39<2:19:10,  1.27it/s]
  1%|▏         | 135/10707 [04:40<2:16:55,  1.29it/s]
  1%|▏         | 136/10707 [04:41<2:15:45,  1.30it/s]
  1%|▏         | 137/10707 [04:42<2:21:15,  1.25it/s]
  1%|▏         | 138/10707 [04:42<2:19:05,  1.27it/s]
  1%|▏         | 139/10707 [04:43<2:19:41,  1.26it/s]
  1%|▏         | 140/10707 [04:44<2:24:02,  1.22it/s]
  1%|▏         | 141/10707 [04:45<2:23:29,  1.23it/s]
  1%|▏         | 142/10707 [04:46<2:21:04,  1.25it/s]
  1%|▏         | 143/10707 [04:46<2:23:19,  1.23it/s]
  1%|▏         | 144/10707 [04:47<2:19:52,  1.26it/s]
  1%|▏         | 145/10707 [04:48<2:18:54,  1.27it/s]
  1%|▏         | 146/10707 [04:49<2:21:13,  1.25it/s]
  1%|▏         | 147/10707 [04:50<2:17:24,  1.28it/s]
  1%|▏         | 148/10707 [04:50<2:12:16,  1.33it/s]
  1%|▏         | 149/10707 [04:51<2:11:12,  1.34it/s]
  1%|▏         | 150/10707 [04:52<2:15:01,  1.30it/s]
                                                     
{'loss': 7.408, 'grad_norm': 0.40946343541145325, 'learning_rate': 0.0001400560224089636, 'epoch': 0.01}
-
  1%|▏         | 150/10707 [04:52<2:15:01,  1.30it/s]
  1%|▏         | 151/10707 [04:53<2:15:05,  1.30it/s]
  1%|▏         | 152/10707 [04:53<2:12:01,  1.33it/s]
  1%|▏         | 153/10707 [04:54<2:11:25,  1.34it/s]
  1%|▏         | 154/10707 [04:55<2:14:33,  1.31it/s]
  1%|▏         | 155/10707 [04:56<2:14:57,  1.30it/s]
  1%|▏         | 156/10707 [04:56<2:12:12,  1.33it/s]
  1%|▏         | 157/10707 [04:57<2:14:36,  1.31it/s]
  1%|▏         | 158/10707 [04:58<2:13:16,  1.32it/s]
  1%|▏         | 159/10707 [04:59<2:10:52,  1.34it/s]
  1%|▏         | 160/10707 [04:59<2:12:26,  1.33it/s]
  2%|▏         | 161/10707 [05:00<2:11:15,  1.34it/s]
  2%|▏         | 162/10707 [05:01<2:07:34,  1.38it/s]
  2%|▏         | 163/10707 [05:01<2:06:39,  1.39it/s]
  2%|▏         | 164/10707 [05:02<2:08:17,  1.37it/s]
  2%|▏         | 165/10707 [05:03<2:07:00,  1.38it/s]
  2%|▏         | 166/10707 [05:04<2:02:41,  1.43it/s]
  2%|▏         | 167/10707 [05:04<2:00:36,  1.46it/s]
  2%|▏         | 168/10707 [05:05<2:01:51,  1.44it/s]
  2%|▏         | 169/10707 [05:06<2:02:09,  1.44it/s]
  2%|▏         | 170/10707 [05:06<2:00:19,  1.46it/s]
  2%|▏         | 171/10707 [05:07<1:59:15,  1.47it/s]
  2%|▏         | 172/10707 [05:08<1:59:32,  1.47it/s]
  2%|▏         | 173/10707 [05:08<2:01:34,  1.44it/s]
  2%|▏         | 174/10707 [05:09<1:58:56,  1.48it/s]
  2%|▏         | 175/10707 [05:10<1:56:40,  1.50it/s]{'loss': 7.0268, 'grad_norm': 0.32415705919265747, 'learning_rate': 0.00016339869281045753, 'epoch': 0.02}
-
                                                     

  2%|▏         | 175/10707 [05:10<1:56:40,  1.50it/s]
  2%|▏         | 176/10707 [05:10<1:55:25,  1.52it/s]
  2%|▏         | 177/10707 [05:11<1:58:38,  1.48it/s]
  2%|▏         | 178/10707 [05:12<1:56:23,  1.51it/s]
  2%|▏         | 179/10707 [05:12<1:54:58,  1.53it/s]
  2%|▏         | 180/10707 [05:13<1:54:36,  1.53it/s]
  2%|▏         | 181/10707 [05:14<1:56:58,  1.50it/s]
  2%|▏         | 182/10707 [05:14<1:55:40,  1.52it/s]
  2%|▏         | 183/10707 [05:15<1:54:38,  1.53it/s]
  2%|▏         | 184/10707 [05:16<1:53:28,  1.55it/s]
  2%|▏         | 185/10707 [05:16<1:52:18,  1.56it/s]
  2%|▏         | 186/10707 [05:17<1:55:20,  1.52it/s]
  2%|▏         | 187/10707 [05:17<1:54:55,  1.53it/s]
  2%|▏         | 188/10707 [05:18<1:53:08,  1.55it/s]
  2%|▏         | 189/10707 [05:19<1:52:58,  1.55it/s]
  2%|▏         | 190/10707 [05:19<1:51:46,  1.57it/s]
  2%|▏         | 191/10707 [05:20<1:52:14,  1.56it/s]
  2%|▏         | 192/10707 [05:21<1:52:53,  1.55it/s]
  2%|▏         | 193/10707 [05:21<1:52:16,  1.56it/s]
  2%|▏         | 194/10707 [05:22<1:51:03,  1.58it/s]
  2%|▏         | 195/10707 [05:23<1:49:55,  1.59it/s]
  2%|▏         | 196/10707 [05:23<1:49:34,  1.60it/s]
  2%|▏         | 197/10707 [05:24<1:51:15,  1.57it/s]
  2%|▏         | 198/10707 [05:24<1:52:57,  1.55it/s]
  2%|▏         | 199/10707 [05:25<1:50:38,  1.58it/s]
  2%|▏         | 200/10707 [05:26<1:48:46,  1.61it/s]{'loss': 6.6841, 'grad_norm': 0.42098233103752136, 'learning_rate': 0.00018674136321195146, 'epoch': 0.02}
-
                                                     

  2%|▏         | 200/10707 [05:26<1:48:46,  1.61it/s]
  2%|▏         | 201/10707 [05:26<1:48:04,  1.62it/s]
  2%|▏         | 202/10707 [05:27<1:48:37,  1.61it/s]
  2%|▏         | 203/10707 [05:28<1:49:25,  1.60it/s]
  2%|▏         | 204/10707 [05:28<1:50:58,  1.58it/s]
  2%|▏         | 205/10707 [05:29<1:49:08,  1.60it/s]
  2%|▏         | 206/10707 [05:29<1:48:00,  1.62it/s]
  2%|▏         | 207/10707 [05:30<1:48:10,  1.62it/s]
  2%|▏         | 208/10707 [05:31<1:47:26,  1.63it/s]
  2%|▏         | 209/10707 [05:31<1:46:38,  1.64it/s]
  2%|▏         | 210/10707 [05:32<1:47:34,  1.63it/s]
  2%|▏         | 211/10707 [05:32<1:48:01,  1.62it/s]
  2%|▏         | 212/10707 [05:33<1:46:34,  1.64it/s]
  2%|▏         | 213/10707 [05:34<1:45:17,  1.66it/s]
  2%|▏         | 214/10707 [05:34<1:45:41,  1.65it/s]
  2%|▏         | 215/10707 [05:35<1:44:10,  1.68it/s]
  2%|▏         | 216/10707 [05:35<1:44:37,  1.67it/s]
  2%|▏         | 217/10707 [05:36<1:46:11,  1.65it/s]
  2%|▏         | 218/10707 [05:37<1:46:54,  1.64it/s]
  2%|▏         | 219/10707 [05:37<1:46:17,  1.64it/s]
  2%|▏         | 220/10707 [05:38<1:45:40,  1.65it/s]
  2%|▏         | 221/10707 [05:38<1:44:47,  1.67it/s]
  2%|▏         | 222/10707 [05:39<1:44:29,  1.67it/s]
  2%|▏         | 223/10707 [05:40<1:43:39,  1.69it/s]
  2%|▏         | 224/10707 [05:40<1:43:30,  1.69it/s]
  2%|▏         | 225/10707 [05:41<1:44:29,  1.67it/s]{'loss': 6.4355, 'grad_norm': 0.5124174356460571, 'learning_rate': 0.0002100840336134454, 'epoch': 0.02}
                                                     
-
  2%|▏         | 225/10707 [05:41<1:44:29,  1.67it/s]
  2%|▏         | 226/10707 [05:41<1:45:32,  1.66it/s]
  2%|▏         | 227/10707 [05:42<1:44:23,  1.67it/s]
  2%|▏         | 228/10707 [05:43<1:43:27,  1.69it/s]
  2%|▏         | 229/10707 [05:43<1:42:54,  1.70it/s]
  2%|▏         | 230/10707 [05:44<1:41:52,  1.71it/s]
  2%|▏         | 231/10707 [05:44<1:43:34,  1.69it/s]
  2%|▏         | 232/10707 [05:45<1:43:33,  1.69it/s]
  2%|▏         | 233/10707 [05:46<1:43:30,  1.69it/s]
  2%|▏         | 234/10707 [05:46<1:42:52,  1.70it/s]
  2%|▏         | 235/10707 [05:47<1:42:06,  1.71it/s]
  2%|▏         | 236/10707 [05:47<1:41:25,  1.72it/s]
  2%|▏         | 237/10707 [05:48<1:40:41,  1.73it/s]
  2%|▏         | 238/10707 [05:48<1:40:24,  1.74it/s]
  2%|▏         | 239/10707 [05:49<1:41:08,  1.73it/s]
  2%|▏         | 240/10707 [05:50<1:42:30,  1.70it/s]
  2%|▏         | 241/10707 [05:50<1:42:37,  1.70it/s]
  2%|▏         | 242/10707 [05:51<1:42:03,  1.71it/s]
  2%|▏         | 243/10707 [05:51<1:40:28,  1.74it/s]
  2%|▏         | 244/10707 [05:52<1:40:36,  1.73it/s]
  2%|▏         | 245/10707 [05:53<1:39:48,  1.75it/s]
  2%|▏         | 246/10707 [05:53<1:39:19,  1.76it/s]
  2%|▏         | 247/10707 [05:54<1:39:25,  1.75it/s]
  2%|▏         | 248/10707 [05:54<1:39:32,  1.75it/s]
  2%|▏         | 249/10707 [05:55<1:40:12,  1.74it/s]
  2%|▏         | 250/10707 [05:55<1:40:13,  1.74it/s]{'loss': 6.2296, 'grad_norm': 0.7482662796974182, 'learning_rate': 0.00023342670401493932, 'epoch': 0.02}
-
                                                     

  2%|▏         | 250/10707 [05:55<1:40:13,  1.74it/s]
  2%|▏         | 251/10707 [05:56<1:40:01,  1.74it/s]
  2%|▏         | 252/10707 [05:57<1:39:17,  1.76it/s]
  2%|▏         | 253/10707 [05:57<1:39:13,  1.76it/s]
  2%|▏         | 254/10707 [05:58<1:38:46,  1.76it/s]
  2%|▏         | 255/10707 [05:58<1:38:36,  1.77it/s]
  2%|▏         | 256/10707 [05:59<1:38:30,  1.77it/s]
  2%|▏         | 257/10707 [05:59<1:37:04,  1.79it/s]
  2%|▏         | 258/10707 [06:00<1:38:07,  1.77it/s]
  2%|▏         | 259/10707 [06:00<1:38:22,  1.77it/s]
  2%|▏         | 260/10707 [06:01<1:38:08,  1.77it/s]
  2%|▏         | 261/10707 [06:02<1:38:18,  1.77it/s]
  2%|▏         | 262/10707 [06:02<1:38:08,  1.77it/s]
  2%|▏         | 263/10707 [06:03<1:38:29,  1.77it/s]
  2%|▏         | 264/10707 [06:03<1:37:58,  1.78it/s]
  2%|▏         | 265/10707 [06:04<1:37:49,  1.78it/s]
  2%|▏         | 266/10707 [06:04<1:36:57,  1.79it/s]
  2%|▏         | 267/10707 [06:05<1:37:48,  1.78it/s]
  3%|▎         | 268/10707 [06:06<1:38:18,  1.77it/s]
  3%|▎         | 269/10707 [06:06<1:38:41,  1.76it/s]
  3%|▎         | 270/10707 [06:07<1:38:06,  1.77it/s]
  3%|▎         | 271/10707 [06:07<1:38:41,  1.76it/s]
  3%|▎         | 272/10707 [06:08<1:37:54,  1.78it/s]
  3%|▎         | 273/10707 [06:08<1:37:30,  1.78it/s]
  3%|▎         | 274/10707 [06:09<1:36:51,  1.80it/s]
  3%|▎         | 275/10707 [06:09<1:36:46,  1.80it/s]
                                                     {'loss': 6.0507, 'grad_norm': 0.6746402382850647, 'learning_rate': 0.0002567693744164332, 'epoch': 0.03}
-
  3%|▎         | 275/10707 [06:09<1:36:46,  1.80it/s]
  3%|▎         | 276/10707 [06:10<1:38:12,  1.77it/s]
  3%|▎         | 277/10707 [06:11<1:37:19,  1.79it/s]
  3%|▎         | 278/10707 [06:11<1:36:52,  1.79it/s]
  3%|▎         | 279/10707 [06:12<1:36:25,  1.80it/s]
  3%|▎         | 280/10707 [06:12<1:36:17,  1.80it/s]
  3%|▎         | 281/10707 [06:13<1:35:26,  1.82it/s]
  3%|▎         | 282/10707 [06:13<1:36:09,  1.81it/s]
  3%|▎         | 283/10707 [06:14<1:37:10,  1.79it/s]
  3%|▎         | 284/10707 [06:14<1:37:15,  1.79it/s]
  3%|▎         | 285/10707 [06:15<1:36:51,  1.79it/s]
  3%|▎         | 286/10707 [06:16<1:37:25,  1.78it/s]
  3%|▎         | 287/10707 [06:16<1:37:03,  1.79it/s]
  3%|▎         | 288/10707 [06:17<1:37:26,  1.78it/s]
  3%|▎         | 289/10707 [06:17<1:37:57,  1.77it/s]
  3%|▎         | 290/10707 [06:18<1:37:36,  1.78it/s]
  3%|▎         | 291/10707 [06:18<1:38:11,  1.77it/s]
  3%|▎         | 292/10707 [06:19<1:37:37,  1.78it/s]
  3%|▎         | 293/10707 [06:20<1:37:22,  1.78it/s]
  3%|▎         | 294/10707 [06:20<1:38:13,  1.77it/s]
  3%|▎         | 295/10707 [06:21<1:36:55,  1.79it/s]
  3%|▎         | 296/10707 [06:21<1:35:57,  1.81it/s]
  3%|▎         | 297/10707 [06:22<1:35:16,  1.82it/s]
  3%|▎         | 298/10707 [06:22<1:36:25,  1.80it/s]
  3%|▎         | 299/10707 [06:23<1:37:51,  1.77it/s]
  3%|▎         | 300/10707 [06:23<1:37:33,  1.78it/s]{'loss': 5.9103, 'grad_norm': 0.500686526298523, 'learning_rate': 0.0002801120448179272, 'epoch': 0.03}
                                                     
-
  3%|▎         | 300/10707 [06:23<1:37:33,  1.78it/s]
  3%|▎         | 301/10707 [06:24<1:37:37,  1.78it/s]
  3%|▎         | 302/10707 [06:25<1:37:49,  1.77it/s]
  3%|▎         | 303/10707 [06:25<1:37:34,  1.78it/s]
  3%|▎         | 304/10707 [06:26<1:36:43,  1.79it/s]
  3%|▎         | 305/10707 [06:26<1:36:49,  1.79it/s]
  3%|▎         | 306/10707 [06:27<1:36:45,  1.79it/s]
  3%|▎         | 307/10707 [06:27<1:36:21,  1.80it/s]
  3%|▎         | 308/10707 [06:28<1:36:17,  1.80it/s]
  3%|▎         | 309/10707 [06:28<1:37:46,  1.77it/s]
  3%|▎         | 310/10707 [06:29<1:37:41,  1.77it/s]
  3%|▎         | 311/10707 [06:30<1:37:51,  1.77it/s]
  3%|▎         | 312/10707 [06:30<1:36:27,  1.80it/s]
  3%|▎         | 313/10707 [06:31<1:36:27,  1.80it/s]
  3%|▎         | 314/10707 [06:31<1:36:10,  1.80it/s]
  3%|▎         | 315/10707 [06:32<1:35:30,  1.81it/s]
  3%|▎         | 316/10707 [06:32<1:35:35,  1.81it/s]
  3%|▎         | 317/10707 [06:33<1:35:40,  1.81it/s]
  3%|▎         | 318/10707 [06:33<1:35:55,  1.81it/s]
  3%|▎         | 319/10707 [06:34<1:35:06,  1.82it/s]
  3%|▎         | 320/10707 [06:35<1:36:17,  1.80it/s]
  3%|▎         | 321/10707 [06:35<1:36:25,  1.80it/s]
  3%|▎         | 322/10707 [06:36<1:35:39,  1.81it/s]
  3%|▎         | 323/10707 [06:36<1:35:22,  1.81it/s]
  3%|▎         | 324/10707 [06:37<1:35:15,  1.82it/s]
  3%|▎         | 325/10707 [06:37<1:35:17,  1.82it/s]{'loss': 5.7868, 'grad_norm': 0.654743492603302, 'learning_rate': 0.0003034547152194211, 'epoch': 0.03}
-
                                                     

  3%|▎         | 325/10707 [06:37<1:35:17,  1.82it/s]
  3%|▎         | 326/10707 [06:38<1:34:19,  1.83it/s]
  3%|▎         | 327/10707 [06:39<1:48:31,  1.59it/s]
  3%|▎         | 328/10707 [06:39<1:57:26,  1.47it/s]
  3%|▎         | 329/10707 [06:40<1:49:58,  1.57it/s]
  3%|▎         | 330/10707 [06:41<1:44:31,  1.65it/s]
  3%|▎         | 331/10707 [06:41<1:41:23,  1.71it/s]
  3%|▎         | 332/10707 [06:42<1:40:04,  1.73it/s]
  3%|▎         | 333/10707 [06:42<1:38:17,  1.76it/s]
  3%|▎         | 334/10707 [06:43<1:36:44,  1.79it/s]
  3%|▎         | 335/10707 [06:43<1:36:40,  1.79it/s]
  3%|▎         | 336/10707 [06:44<1:35:47,  1.80it/s]
  3%|▎         | 337/10707 [06:44<1:35:38,  1.81it/s]
  3%|▎         | 338/10707 [06:45<1:35:13,  1.81it/s]
  3%|▎         | 339/10707 [06:45<1:35:24,  1.81it/s]
  3%|▎         | 340/10707 [06:46<1:35:51,  1.80it/s]
  3%|▎         | 341/10707 [06:47<1:36:20,  1.79it/s]
  3%|▎         | 342/10707 [06:47<1:35:41,  1.81it/s]
  3%|▎         | 343/10707 [06:48<1:34:59,  1.82it/s]
  3%|▎         | 344/10707 [06:48<1:35:18,  1.81it/s]
  3%|▎         | 345/10707 [06:49<1:34:58,  1.82it/s]
  3%|▎         | 346/10707 [06:49<1:35:19,  1.81it/s]
  3%|▎         | 347/10707 [06:50<1:35:20,  1.81it/s]
  3%|▎         | 348/10707 [06:50<1:33:58,  1.84it/s]
  3%|▎         | 349/10707 [06:51<1:33:09,  1.85it/s]
  3%|▎         | 350/10707 [06:51<1:32:36,  1.86it/s]{'loss': 5.673, 'grad_norm': 0.5446093082427979, 'learning_rate': 0.00032679738562091506, 'epoch': 0.03}
-
                                                     

  3%|▎         | 350/10707 [06:51<1:32:36,  1.86it/s]
  3%|▎         | 351/10707 [06:52<1:32:00,  1.88it/s]
  3%|▎         | 352/10707 [06:53<1:31:57,  1.88it/s]
  3%|▎         | 353/10707 [06:53<1:32:43,  1.86it/s]
  3%|▎         | 354/10707 [06:54<1:32:43,  1.86it/s]
  3%|▎         | 355/10707 [06:54<1:32:11,  1.87it/s]
  3%|▎         | 356/10707 [06:55<1:31:41,  1.88it/s]
  3%|▎         | 357/10707 [06:55<1:31:13,  1.89it/s]
  3%|▎         | 358/10707 [06:56<1:31:16,  1.89it/s]
  3%|▎         | 359/10707 [06:56<1:31:20,  1.89it/s]
  3%|▎         | 360/10707 [06:57<1:30:48,  1.90it/s]
  3%|▎         | 361/10707 [06:57<1:30:21,  1.91it/s]
  3%|▎         | 362/10707 [06:58<1:29:56,  1.92it/s]
  3%|▎         | 363/10707 [06:58<1:30:10,  1.91it/s]
  3%|▎         | 364/10707 [06:59<1:30:12,  1.91it/s]
  3%|▎         | 365/10707 [06:59<1:30:40,  1.90it/s]
  3%|▎         | 366/10707 [07:00<1:30:54,  1.90it/s]
  3%|▎         | 367/10707 [07:00<1:30:57,  1.89it/s]
  3%|▎         | 368/10707 [07:01<1:30:46,  1.90it/s]
  3%|▎         | 369/10707 [07:02<1:30:54,  1.90it/s]
  3%|▎         | 370/10707 [07:02<1:30:39,  1.90it/s]
  3%|▎         | 371/10707 [07:03<1:30:31,  1.90it/s]
  3%|▎         | 372/10707 [07:03<1:30:29,  1.90it/s]
  3%|▎         | 373/10707 [07:04<1:30:29,  1.90it/s]
  3%|▎         | 374/10707 [07:04<1:30:33,  1.90it/s]
  4%|▎         | 375/10707 [07:05<1:29:56,  1.91it/s]
                                                     {'loss': 5.5826, 'grad_norm': 0.739492654800415, 'learning_rate': 0.00035014005602240897, 'epoch': 0.04}
-

  4%|▎         | 375/10707 [07:05<1:29:56,  1.91it/s]
  4%|▎         | 376/10707 [07:05<1:29:54,  1.92it/s]
  4%|▎         | 377/10707 [07:06<1:29:22,  1.93it/s]
  4%|▎         | 378/10707 [07:06<1:29:21,  1.93it/s]
  4%|▎         | 379/10707 [07:07<1:29:18,  1.93it/s]
  4%|▎         | 380/10707 [07:07<1:29:20,  1.93it/s]
  4%|▎         | 381/10707 [07:08<1:29:32,  1.92it/s]
  4%|▎         | 382/10707 [07:08<1:29:34,  1.92it/s]
  4%|▎         | 383/10707 [07:09<1:29:47,  1.92it/s]
  4%|▎         | 384/10707 [07:09<1:29:57,  1.91it/s]
  4%|▎         | 385/10707 [07:10<1:29:24,  1.92it/s]
  4%|▎         | 386/10707 [07:10<1:29:02,  1.93it/s]
  4%|▎         | 387/10707 [07:11<1:29:18,  1.93it/s]
  4%|▎         | 388/10707 [07:11<1:29:15,  1.93it/s]
  4%|▎         | 389/10707 [07:12<1:29:45,  1.92it/s]
  4%|▎         | 390/10707 [07:12<1:29:39,  1.92it/s]
  4%|▎         | 391/10707 [07:13<1:29:19,  1.92it/s]
  4%|▎         | 392/10707 [07:13<1:29:37,  1.92it/s]
  4%|▎         | 393/10707 [07:14<1:29:15,  1.93it/s]
  4%|▎         | 394/10707 [07:15<1:29:24,  1.92it/s]
  4%|▎         | 395/10707 [07:15<1:29:46,  1.91it/s]
  4%|▎         | 396/10707 [07:16<1:29:17,  1.92it/s]
  4%|▎         | 397/10707 [07:16<1:29:43,  1.92it/s]
  4%|▎         | 398/10707 [07:17<1:29:27,  1.92it/s]
  4%|▎         | 399/10707 [07:17<1:28:58,  1.93it/s]
  4%|▎         | 400/10707 [07:18<1:28:50,  1.93it/s]{'loss': 5.4916, 'grad_norm': 0.7197967171669006, 'learning_rate': 0.0003734827264239029, 'epoch': 0.04}
-
                                                     

  4%|▎         | 400/10707 [07:18<1:28:50,  1.93it/s]
  4%|▎         | 401/10707 [07:18<1:28:40,  1.94it/s]
  4%|▍         | 402/10707 [07:19<1:28:53,  1.93it/s]
  4%|▍         | 403/10707 [07:19<1:28:49,  1.93it/s]
  4%|▍         | 404/10707 [07:20<1:29:38,  1.92it/s]
  4%|▍         | 405/10707 [07:20<1:29:22,  1.92it/s]
  4%|▍         | 406/10707 [07:21<1:29:15,  1.92it/s]
  4%|▍         | 407/10707 [07:21<1:29:12,  1.92it/s]
  4%|▍         | 408/10707 [07:22<1:29:04,  1.93it/s]
  4%|▍         | 409/10707 [07:22<1:28:37,  1.94it/s]
  4%|▍         | 410/10707 [07:23<1:28:36,  1.94it/s]
  4%|▍         | 411/10707 [07:23<1:28:56,  1.93it/s]
  4%|▍         | 412/10707 [07:24<1:28:38,  1.94it/s]
  4%|▍         | 413/10707 [07:24<1:28:09,  1.95it/s]
  4%|▍         | 414/10707 [07:25<1:28:24,  1.94it/s]
  4%|▍         | 415/10707 [07:25<1:28:05,  1.95it/s]
  4%|▍         | 416/10707 [07:26<1:28:06,  1.95it/s]
  4%|▍         | 417/10707 [07:26<1:28:03,  1.95it/s]
  4%|▍         | 418/10707 [07:27<1:28:09,  1.95it/s]
  4%|▍         | 419/10707 [07:27<1:28:09,  1.95it/s]
  4%|▍         | 420/10707 [07:28<1:28:04,  1.95it/s]
  4%|▍         | 421/10707 [07:28<1:28:32,  1.94it/s]
  4%|▍         | 422/10707 [07:29<1:28:37,  1.93it/s]
  4%|▍         | 423/10707 [07:30<1:28:35,  1.93it/s]
  4%|▍         | 424/10707 [07:30<1:28:43,  1.93it/s]
  4%|▍         | 425/10707 [07:31<1:28:26,  1.94it/s]{'loss': 5.4065, 'grad_norm': 0.863642156124115, 'learning_rate': 0.0003968253968253968, 'epoch': 0.04}
-
                                                     

  4%|▍         | 425/10707 [07:31<1:28:26,  1.94it/s]
  4%|▍         | 426/10707 [07:31<1:28:29,  1.94it/s]
  4%|▍         | 427/10707 [07:32<1:28:09,  1.94it/s]
  4%|▍         | 428/10707 [07:32<1:28:34,  1.93it/s]
  4%|▍         | 429/10707 [07:33<1:28:24,  1.94it/s]
  4%|▍         | 430/10707 [07:33<1:28:19,  1.94it/s]
  4%|▍         | 431/10707 [07:34<1:27:59,  1.95it/s]
  4%|▍         | 432/10707 [07:34<1:27:49,  1.95it/s]
  4%|▍         | 433/10707 [07:35<1:27:48,  1.95it/s]
  4%|▍         | 434/10707 [07:35<1:27:32,  1.96it/s]
  4%|▍         | 435/10707 [07:36<1:27:47,  1.95it/s]
  4%|▍         | 436/10707 [07:36<1:27:50,  1.95it/s]
  4%|▍         | 437/10707 [07:37<1:27:44,  1.95it/s]
  4%|▍         | 438/10707 [07:37<1:27:47,  1.95it/s]
  4%|▍         | 439/10707 [07:38<1:27:57,  1.95it/s]
  4%|▍         | 440/10707 [07:38<1:27:40,  1.95it/s]
  4%|▍         | 441/10707 [07:39<1:27:25,  1.96it/s]
  4%|▍         | 442/10707 [07:39<1:27:29,  1.96it/s]
  4%|▍         | 443/10707 [07:40<1:27:29,  1.96it/s]
  4%|▍         | 444/10707 [07:40<1:27:33,  1.95it/s]
  4%|▍         | 445/10707 [07:41<1:27:31,  1.95it/s]
  4%|▍         | 446/10707 [07:41<1:27:30,  1.95it/s]
  4%|▍         | 447/10707 [07:42<1:27:42,  1.95it/s]
  4%|▍         | 448/10707 [07:42<1:27:44,  1.95it/s]
  4%|▍         | 449/10707 [07:43<1:27:29,  1.95it/s]
  4%|▍         | 450/10707 [07:43<1:27:22,  1.96it/s]{'loss': 5.3304, 'grad_norm': 0.5300434231758118, 'learning_rate': 0.0004201680672268908, 'epoch': 0.04}
                                                     
-
  4%|▍         | 450/10707 [07:43<1:27:22,  1.96it/s]
  4%|▍         | 451/10707 [07:44<1:27:30,  1.95it/s]
  4%|▍         | 452/10707 [07:44<1:27:24,  1.96it/s]
  4%|▍         | 453/10707 [07:45<1:27:44,  1.95it/s]
  4%|▍         | 454/10707 [07:45<1:27:41,  1.95it/s]
  4%|▍         | 455/10707 [07:46<1:28:08,  1.94it/s]
  4%|▍         | 456/10707 [07:46<1:27:40,  1.95it/s]
  4%|▍         | 457/10707 [07:47<1:27:53,  1.94it/s]
  4%|▍         | 458/10707 [07:47<1:27:45,  1.95it/s]
  4%|▍         | 459/10707 [07:48<1:27:30,  1.95it/s]
  4%|▍         | 460/10707 [07:49<1:27:18,  1.96it/s]
  4%|▍         | 461/10707 [07:49<1:27:08,  1.96it/s]
  4%|▍         | 462/10707 [07:50<1:27:03,  1.96it/s]
  4%|▍         | 463/10707 [07:50<1:27:14,  1.96it/s]
  4%|▍         | 464/10707 [07:51<1:27:03,  1.96it/s]
  4%|▍         | 465/10707 [07:51<1:26:55,  1.96it/s]
  4%|▍         | 466/10707 [07:52<1:27:06,  1.96it/s]
  4%|▍         | 467/10707 [07:52<1:26:54,  1.96it/s]
  4%|▍         | 468/10707 [07:53<1:26:43,  1.97it/s]
  4%|▍         | 469/10707 [07:53<1:27:03,  1.96it/s]
  4%|▍         | 470/10707 [07:54<1:26:57,  1.96it/s]
  4%|▍         | 471/10707 [07:54<1:27:24,  1.95it/s]
  4%|▍         | 472/10707 [07:55<1:27:18,  1.95it/s]
  4%|▍         | 473/10707 [07:55<1:27:04,  1.96it/s]
  4%|▍         | 474/10707 [07:56<1:27:00,  1.96it/s]
  4%|▍         | 475/10707 [07:56<1:27:03,  1.96it/s]
                                                     
{'loss': 5.2672, 'grad_norm': 0.5380837321281433, 'learning_rate': 0.0004435107376283847, 'epoch': 0.04}
-
  4%|▍         | 475/10707 [07:56<1:27:03,  1.96it/s]
  4%|▍         | 476/10707 [07:57<1:27:16,  1.95it/s]
  4%|▍         | 477/10707 [07:57<1:27:25,  1.95it/s]
  4%|▍         | 478/10707 [07:58<1:27:12,  1.95it/s]
  4%|▍         | 479/10707 [07:58<1:26:57,  1.96it/s]
  4%|▍         | 480/10707 [07:59<1:26:50,  1.96it/s]
  4%|▍         | 481/10707 [07:59<1:26:49,  1.96it/s]
  5%|▍         | 482/10707 [08:00<1:26:45,  1.96it/s]
  5%|▍         | 483/10707 [08:00<1:26:51,  1.96it/s]
  5%|▍         | 484/10707 [08:01<1:26:52,  1.96it/s]
  5%|▍         | 485/10707 [08:01<1:27:09,  1.95it/s]
  5%|▍         | 486/10707 [08:02<1:27:03,  1.96it/s]
  5%|▍         | 487/10707 [08:02<1:26:44,  1.96it/s]
  5%|▍         | 488/10707 [08:03<1:26:36,  1.97it/s]
  5%|▍         | 489/10707 [08:03<1:26:52,  1.96it/s]
  5%|▍         | 490/10707 [08:04<1:26:44,  1.96it/s]
  5%|▍         | 491/10707 [08:04<1:26:45,  1.96it/s]
  5%|▍         | 492/10707 [08:05<1:26:30,  1.97it/s]
  5%|▍         | 493/10707 [08:05<1:26:31,  1.97it/s]
  5%|▍         | 494/10707 [08:06<1:26:23,  1.97it/s]
  5%|▍         | 495/10707 [08:06<1:26:14,  1.97it/s]
  5%|▍         | 496/10707 [08:07<1:26:15,  1.97it/s]
  5%|▍         | 497/10707 [08:07<1:26:24,  1.97it/s]
  5%|▍         | 498/10707 [08:08<1:26:38,  1.96it/s]
  5%|▍         | 499/10707 [08:08<1:26:21,  1.97it/s]
  5%|▍         | 500/10707 [08:09<1:26:15,  1.97it/s]{'loss': 5.2, 'grad_norm': 0.5683251619338989, 'learning_rate': 0.00046685340802987864, 'epoch': 0.05}
-
                                                     

  5%|▍         | 500/10707 [08:09<1:26:15,  1.97it/s]
  5%|▍         | 501/10707 [08:09<1:26:16,  1.97it/s]
  5%|▍         | 502/10707 [08:10<1:25:58,  1.98it/s]
  5%|▍         | 503/10707 [08:10<1:26:01,  1.98it/s]
  5%|▍         | 504/10707 [08:11<1:25:53,  1.98it/s]
  5%|▍         | 505/10707 [08:11<1:26:03,  1.98it/s]
  5%|▍         | 506/10707 [08:12<1:26:26,  1.97it/s]
  5%|▍         | 507/10707 [08:12<1:26:22,  1.97it/s]
  5%|▍         | 508/10707 [08:13<1:26:07,  1.97it/s]
  5%|▍         | 509/10707 [08:13<1:25:58,  1.98it/s]
  5%|▍         | 510/10707 [08:14<1:25:58,  1.98it/s]
  5%|▍         | 511/10707 [08:14<1:25:50,  1.98it/s]
  5%|▍         | 512/10707 [08:15<1:25:46,  1.98it/s]
  5%|▍         | 513/10707 [08:15<1:25:44,  1.98it/s]
  5%|▍         | 514/10707 [08:16<1:25:46,  1.98it/s]
  5%|▍         | 515/10707 [08:16<1:25:46,  1.98it/s]
  5%|▍         | 516/10707 [08:17<1:25:43,  1.98it/s]
  5%|▍         | 517/10707 [08:17<1:25:39,  1.98it/s]
  5%|▍         | 518/10707 [08:18<1:25:42,  1.98it/s]
  5%|▍         | 519/10707 [08:18<1:25:38,  1.98it/s]
  5%|▍         | 520/10707 [08:19<1:25:34,  1.98it/s]
  5%|▍         | 521/10707 [08:19<1:25:38,  1.98it/s]
  5%|▍         | 522/10707 [08:20<1:25:46,  1.98it/s]
  5%|▍         | 523/10707 [08:21<1:25:47,  1.98it/s]
  5%|▍         | 524/10707 [08:21<1:25:59,  1.97it/s]
  5%|▍         | 525/10707 [08:22<1:25:53,  1.98it/s]
                                                     {'loss': 5.1367, 'grad_norm': 0.5235716700553894, 'learning_rate': 0.0004901960784313725, 'epoch': 0.05}
-

  5%|▍         | 525/10707 [08:22<1:25:53,  1.98it/s]
  5%|▍         | 526/10707 [08:22<1:25:48,  1.98it/s]
  5%|▍         | 527/10707 [08:23<1:25:53,  1.98it/s]
  5%|▍         | 528/10707 [08:23<1:25:47,  1.98it/s]
  5%|▍         | 529/10707 [08:24<1:25:35,  1.98it/s]
  5%|▍         | 530/10707 [08:24<1:25:31,  1.98it/s]
  5%|▍         | 531/10707 [08:25<1:25:23,  1.99it/s]
  5%|▍         | 532/10707 [08:25<1:25:18,  1.99it/s]
  5%|▍         | 533/10707 [08:26<1:25:22,  1.99it/s]
  5%|▍         | 534/10707 [08:26<1:25:21,  1.99it/s]
  5%|▍         | 535/10707 [08:27<1:25:43,  1.98it/s]
  5%|▌         | 536/10707 [08:27<1:25:41,  1.98it/s]
  5%|▌         | 537/10707 [08:28<1:25:37,  1.98it/s]
  5%|▌         | 538/10707 [08:28<1:25:26,  1.98it/s]
  5%|▌         | 539/10707 [08:29<1:25:21,  1.99it/s]
  5%|▌         | 540/10707 [08:29<1:25:47,  1.98it/s]
  5%|▌         | 541/10707 [08:30<1:25:54,  1.97it/s]
  5%|▌         | 542/10707 [08:30<1:25:48,  1.97it/s]
  5%|▌         | 543/10707 [08:31<1:25:38,  1.98it/s]
  5%|▌         | 544/10707 [08:31<1:25:29,  1.98it/s]
  5%|▌         | 545/10707 [08:32<1:25:26,  1.98it/s]
  5%|▌         | 546/10707 [08:32<1:25:30,  1.98it/s]
  5%|▌         | 547/10707 [08:33<1:25:30,  1.98it/s]
  5%|▌         | 548/10707 [08:33<1:25:26,  1.98it/s]
  5%|▌         | 549/10707 [08:34<1:25:20,  1.98it/s]
  5%|▌         | 550/10707 [08:34<1:25:15,  1.99it/s]
                                                     
{'loss': 5.0866, 'grad_norm': 0.45653992891311646, 'learning_rate': 0.0005135387488328664, 'epoch': 0.05}
-
  5%|▌         | 550/10707 [08:34<1:25:15,  1.99it/s]
  5%|▌         | 551/10707 [08:35<1:25:19,  1.98it/s]
  5%|▌         | 552/10707 [08:35<1:25:16,  1.98it/s]
  5%|▌         | 553/10707 [08:36<1:25:14,  1.99it/s]
  5%|▌         | 554/10707 [08:36<1:25:11,  1.99it/s]
  5%|▌         | 555/10707 [08:37<1:25:06,  1.99it/s]
  5%|▌         | 556/10707 [08:37<1:25:14,  1.98it/s]
  5%|▌         | 557/10707 [08:38<1:25:10,  1.99it/s]
  5%|▌         | 558/10707 [08:38<1:24:59,  1.99it/s]
  5%|▌         | 559/10707 [08:39<1:25:00,  1.99it/s]
  5%|▌         | 560/10707 [08:39<1:25:06,  1.99it/s]
  5%|▌         | 561/10707 [08:40<1:25:01,  1.99it/s]
  5%|▌         | 562/10707 [08:40<1:24:58,  1.99it/s]
  5%|▌         | 563/10707 [08:41<1:25:13,  1.98it/s]
  5%|▌         | 564/10707 [08:41<1:25:05,  1.99it/s]
  5%|▌         | 565/10707 [08:42<1:25:41,  1.97it/s]
  5%|▌         | 566/10707 [08:42<1:25:20,  1.98it/s]
  5%|▌         | 567/10707 [08:43<1:25:13,  1.98it/s]
  5%|▌         | 568/10707 [08:43<1:25:13,  1.98it/s]
  5%|▌         | 569/10707 [08:44<1:25:03,  1.99it/s]
  5%|▌         | 570/10707 [08:44<1:24:57,  1.99it/s]
  5%|▌         | 571/10707 [08:45<1:25:00,  1.99it/s]
  5%|▌         | 572/10707 [08:45<1:25:06,  1.98it/s]
  5%|▌         | 573/10707 [08:46<1:25:23,  1.98it/s]
  5%|▌         | 574/10707 [08:46<1:25:46,  1.97it/s]
  5%|▌         | 575/10707 [08:47<1:25:46,  1.97it/s]{'loss': 5.0322, 'grad_norm': 0.5589630007743835, 'learning_rate': 0.0005368814192343605, 'epoch': 0.05}
-
                                                     

  5%|▌         | 575/10707 [08:47<1:25:46,  1.97it/s]
  5%|▌         | 576/10707 [08:47<1:25:51,  1.97it/s]
  5%|▌         | 577/10707 [08:48<1:25:52,  1.97it/s]
  5%|▌         | 578/10707 [08:48<1:26:10,  1.96it/s]
  5%|▌         | 579/10707 [08:49<1:26:18,  1.96it/s]
  5%|▌         | 580/10707 [08:49<1:26:00,  1.96it/s]
  5%|▌         | 581/10707 [08:50<1:25:55,  1.96it/s]
  5%|▌         | 582/10707 [08:50<1:25:46,  1.97it/s]
  5%|▌         | 583/10707 [08:51<1:36:34,  1.75it/s]
  5%|▌         | 584/10707 [08:52<1:34:03,  1.79it/s]
  5%|▌         | 585/10707 [08:52<1:31:10,  1.85it/s]
  5%|▌         | 586/10707 [08:53<1:29:06,  1.89it/s]
  5%|▌         | 587/10707 [08:53<1:28:14,  1.91it/s]
  5%|▌         | 588/10707 [08:54<1:27:35,  1.93it/s]
  6%|▌         | 589/10707 [08:54<1:27:05,  1.94it/s]
  6%|▌         | 590/10707 [08:55<1:26:13,  1.96it/s]
  6%|▌         | 591/10707 [08:55<1:25:51,  1.96it/s]
  6%|▌         | 592/10707 [08:56<1:25:31,  1.97it/s]
  6%|▌         | 593/10707 [08:56<1:26:03,  1.96it/s]
  6%|▌         | 594/10707 [08:57<1:25:52,  1.96it/s]
  6%|▌         | 595/10707 [08:57<1:25:38,  1.97it/s]
  6%|▌         | 596/10707 [08:58<1:25:23,  1.97it/s]
  6%|▌         | 597/10707 [08:58<1:25:23,  1.97it/s]
  6%|▌         | 598/10707 [08:59<1:25:50,  1.96it/s]
  6%|▌         | 599/10707 [08:59<1:25:53,  1.96it/s]
  6%|▌         | 600/10707 [09:00<1:26:05,  1.96it/s]{'loss': 4.9878, 'grad_norm': 0.582036018371582, 'learning_rate': 0.0005602240896358544, 'epoch': 0.06}
-
                                                     

  6%|▌         | 600/10707 [09:00<1:26:05,  1.96it/s]
  6%|▌         | 601/10707 [09:00<1:26:12,  1.95it/s]
  6%|▌         | 602/10707 [09:01<1:26:01,  1.96it/s]
  6%|▌         | 603/10707 [09:01<1:25:46,  1.96it/s]
  6%|▌         | 604/10707 [09:02<1:26:18,  1.95it/s]
  6%|▌         | 605/10707 [09:02<1:26:43,  1.94it/s]
  6%|▌         | 606/10707 [09:03<1:26:32,  1.95it/s]
  6%|▌         | 607/10707 [09:03<1:26:31,  1.95it/s]
  6%|▌         | 608/10707 [09:04<1:27:12,  1.93it/s]
  6%|▌         | 609/10707 [09:04<1:26:46,  1.94it/s]
  6%|▌         | 610/10707 [09:05<1:26:49,  1.94it/s]
  6%|▌         | 611/10707 [09:05<1:27:15,  1.93it/s]
  6%|▌         | 612/10707 [09:06<1:27:13,  1.93it/s]
  6%|▌         | 613/10707 [09:06<1:26:18,  1.95it/s]
  6%|▌         | 614/10707 [09:07<1:25:40,  1.96it/s]
  6%|▌         | 615/10707 [09:07<1:25:23,  1.97it/s]
  6%|▌         | 616/10707 [09:08<1:25:58,  1.96it/s]
  6%|▌         | 617/10707 [09:08<1:25:43,  1.96it/s]
  6%|▌         | 618/10707 [09:09<1:25:23,  1.97it/s]
  6%|▌         | 619/10707 [09:09<1:25:20,  1.97it/s]
  6%|▌         | 620/10707 [09:10<1:25:09,  1.97it/s]
  6%|▌         | 621/10707 [09:10<1:25:08,  1.97it/s]
  6%|▌         | 622/10707 [09:11<1:24:58,  1.98it/s]
  6%|▌         | 623/10707 [09:11<1:24:59,  1.98it/s]
  6%|▌         | 624/10707 [09:12<1:24:50,  1.98it/s]
  6%|▌         | 625/10707 [09:12<1:24:39,  1.98it/s]
                                                     
{'loss': 4.9491, 'grad_norm': 0.6276235580444336, 'learning_rate': 0.0005835667600373484, 'epoch': 0.06}
-
  6%|▌         | 625/10707 [09:12<1:24:39,  1.98it/s]
  6%|▌         | 626/10707 [09:13<1:25:13,  1.97it/s]
  6%|▌         | 627/10707 [09:13<1:25:17,  1.97it/s]
  6%|▌         | 628/10707 [09:14<1:25:33,  1.96it/s]
  6%|▌         | 629/10707 [09:14<1:25:15,  1.97it/s]
  6%|▌         | 630/10707 [09:15<1:25:47,  1.96it/s]
  6%|▌         | 631/10707 [09:16<1:25:31,  1.96it/s]
  6%|▌         | 632/10707 [09:16<1:25:36,  1.96it/s]
  6%|▌         | 633/10707 [09:17<1:25:18,  1.97it/s]
  6%|▌         | 634/10707 [09:17<1:24:55,  1.98it/s]
  6%|▌         | 635/10707 [09:18<1:24:59,  1.97it/s]
  6%|▌         | 636/10707 [09:18<1:25:27,  1.96it/s]
  6%|▌         | 637/10707 [09:19<1:25:08,  1.97it/s]
  6%|▌         | 638/10707 [09:19<1:25:22,  1.97it/s]
  6%|▌         | 639/10707 [09:20<1:25:10,  1.97it/s]
  6%|▌         | 640/10707 [09:20<1:24:58,  1.97it/s]
  6%|▌         | 641/10707 [09:21<1:24:38,  1.98it/s]
  6%|▌         | 642/10707 [09:21<1:24:34,  1.98it/s]
  6%|▌         | 643/10707 [09:22<1:24:21,  1.99it/s]
  6%|▌         | 644/10707 [09:22<1:24:10,  1.99it/s]
  6%|▌         | 645/10707 [09:23<1:24:32,  1.98it/s]
  6%|▌         | 646/10707 [09:23<1:24:25,  1.99it/s]
  6%|▌         | 647/10707 [09:24<1:24:16,  1.99it/s]
  6%|▌         | 648/10707 [09:24<1:24:06,  1.99it/s]
  6%|▌         | 649/10707 [09:25<1:24:02,  1.99it/s]
  6%|▌         | 650/10707 [09:25<1:24:07,  1.99it/s]
                                                     {'loss': 4.9062, 'grad_norm': 0.4354894161224365, 'learning_rate': 0.0006069094304388422, 'epoch': 0.06}
-

  6%|▌         | 650/10707 [09:25<1:24:07,  1.99it/s]
  6%|▌         | 651/10707 [09:26<1:24:17,  1.99it/s]
  6%|▌         | 652/10707 [09:26<1:24:08,  1.99it/s]
  6%|▌         | 653/10707 [09:27<1:24:19,  1.99it/s]
  6%|▌         | 654/10707 [09:27<1:24:05,  1.99it/s]
  6%|▌         | 655/10707 [09:28<1:24:01,  1.99it/s]
  6%|▌         | 656/10707 [09:28<1:23:51,  2.00it/s]
  6%|▌         | 657/10707 [09:29<1:23:47,  2.00it/s]
  6%|▌         | 658/10707 [09:29<1:23:38,  2.00it/s]
  6%|▌         | 659/10707 [09:30<1:23:41,  2.00it/s]
  6%|▌         | 660/10707 [09:30<1:23:45,  2.00it/s]
  6%|▌         | 661/10707 [09:31<1:23:45,  2.00it/s]
  6%|▌         | 662/10707 [09:31<1:23:38,  2.00it/s]
  6%|▌         | 663/10707 [09:32<1:23:42,  2.00it/s]
  6%|▌         | 664/10707 [09:32<1:23:47,  2.00it/s]
  6%|▌         | 665/10707 [09:33<1:23:41,  2.00it/s]
  6%|▌         | 666/10707 [09:33<1:23:33,  2.00it/s]
  6%|▌         | 667/10707 [09:34<1:23:31,  2.00it/s]
  6%|▌         | 668/10707 [09:34<1:23:37,  2.00it/s]
  6%|▌         | 669/10707 [09:35<1:23:42,  2.00it/s]
  6%|▋         | 670/10707 [09:35<1:23:42,  2.00it/s]
  6%|▋         | 671/10707 [09:36<1:23:46,  2.00it/s]
  6%|▋         | 672/10707 [09:36<1:23:38,  2.00it/s]
  6%|▋         | 673/10707 [09:37<1:23:47,  2.00it/s]
  6%|▋         | 674/10707 [09:37<1:23:46,  2.00it/s]
  6%|▋         | 675/10707 [09:38<1:23:46,  2.00it/s]{'loss': 4.8631, 'grad_norm': 0.4510965645313263, 'learning_rate': 0.0006302521008403362, 'epoch': 0.06}
-
                                                     

  6%|▋         | 675/10707 [09:38<1:23:46,  2.00it/s]
  6%|▋         | 676/10707 [09:38<1:23:44,  2.00it/s]
  6%|▋         | 677/10707 [09:39<1:23:30,  2.00it/s]
  6%|▋         | 678/10707 [09:39<1:23:35,  2.00it/s]
  6%|▋         | 679/10707 [09:40<1:23:35,  2.00it/s]
  6%|▋         | 680/10707 [09:40<1:23:40,  2.00it/s]
  6%|▋         | 681/10707 [09:41<1:23:40,  2.00it/s]
  6%|▋         | 682/10707 [09:41<1:23:40,  2.00it/s]
  6%|▋         | 683/10707 [09:42<1:23:38,  2.00it/s]
  6%|▋         | 684/10707 [09:42<1:23:36,  2.00it/s]
  6%|▋         | 685/10707 [09:43<1:23:43,  1.99it/s]
  6%|▋         | 686/10707 [09:43<1:23:45,  1.99it/s]
  6%|▋         | 687/10707 [09:44<1:23:38,  2.00it/s]
  6%|▋         | 688/10707 [09:44<1:23:30,  2.00it/s]
  6%|▋         | 689/10707 [09:45<1:23:25,  2.00it/s]
  6%|▋         | 690/10707 [09:45<1:23:38,  2.00it/s]
  6%|▋         | 691/10707 [09:46<1:23:30,  2.00it/s]
  6%|▋         | 692/10707 [09:46<1:23:25,  2.00it/s]
  6%|▋         | 693/10707 [09:47<1:23:23,  2.00it/s]
  6%|▋         | 694/10707 [09:47<1:23:23,  2.00it/s]
  6%|▋         | 695/10707 [09:48<1:23:30,  2.00it/s]
  7%|▋         | 696/10707 [09:48<1:23:32,  2.00it/s]
  7%|▋         | 697/10707 [09:49<1:23:23,  2.00it/s]
  7%|▋         | 698/10707 [09:49<1:23:22,  2.00it/s]
  7%|▋         | 699/10707 [09:50<1:23:22,  2.00it/s]
  7%|▋         | 700/10707 [09:50<1:23:38,  1.99it/s]{'loss': 4.8222, 'grad_norm': 0.4342808425426483, 'learning_rate': 0.0006535947712418301, 'epoch': 0.07}
-
                                                     

  7%|▋         | 700/10707 [09:50<1:23:38,  1.99it/s]
  7%|▋         | 701/10707 [09:51<1:23:37,  1.99it/s]
  7%|▋         | 702/10707 [09:51<1:23:29,  2.00it/s]
  7%|▋         | 703/10707 [09:52<1:23:22,  2.00it/s]
  7%|▋         | 704/10707 [09:52<1:23:25,  2.00it/s]
  7%|▋         | 705/10707 [09:53<1:23:24,  2.00it/s]
  7%|▋         | 706/10707 [09:53<1:23:33,  1.99it/s]
  7%|▋         | 707/10707 [09:54<1:23:32,  1.99it/s]
  7%|▋         | 708/10707 [09:54<1:23:24,  2.00it/s]
  7%|▋         | 709/10707 [09:55<1:23:28,  2.00it/s]
  7%|▋         | 710/10707 [09:55<1:23:22,  2.00it/s]
  7%|▋         | 711/10707 [09:56<1:23:26,  2.00it/s]
  7%|▋         | 712/10707 [09:56<1:23:24,  2.00it/s]
  7%|▋         | 713/10707 [09:57<1:23:20,  2.00it/s]
  7%|▋         | 714/10707 [09:57<1:23:23,  2.00it/s]
  7%|▋         | 715/10707 [09:58<1:23:16,  2.00it/s]
  7%|▋         | 716/10707 [09:58<1:23:11,  2.00it/s]
  7%|▋         | 717/10707 [09:59<1:23:16,  2.00it/s]
  7%|▋         | 718/10707 [09:59<1:23:14,  2.00it/s]
  7%|▋         | 719/10707 [10:00<1:23:16,  2.00it/s]
  7%|▋         | 720/10707 [10:00<1:23:11,  2.00it/s]
  7%|▋         | 721/10707 [10:01<1:23:10,  2.00it/s]
  7%|▋         | 722/10707 [10:01<1:23:02,  2.00it/s]
  7%|▋         | 723/10707 [10:02<1:23:11,  2.00it/s]
  7%|▋         | 724/10707 [10:02<1:23:09,  2.00it/s]
  7%|▋         | 725/10707 [10:03<1:23:00,  2.00it/s]{'loss': 4.8001, 'grad_norm': 0.4419580101966858, 'learning_rate': 0.0006769374416433241, 'epoch': 0.07}
                                                     
-
  7%|▋         | 725/10707 [10:03<1:23:00,  2.00it/s]
  7%|▋         | 726/10707 [10:03<1:23:03,  2.00it/s]
  7%|▋         | 727/10707 [10:04<1:23:00,  2.00it/s]
  7%|▋         | 728/10707 [10:04<1:22:56,  2.01it/s]
  7%|▋         | 729/10707 [10:05<1:23:03,  2.00it/s]
  7%|▋         | 730/10707 [10:05<1:23:02,  2.00it/s]
  7%|▋         | 731/10707 [10:06<1:22:54,  2.01it/s]
  7%|▋         | 732/10707 [10:06<1:22:55,  2.01it/s]
  7%|▋         | 733/10707 [10:07<1:23:04,  2.00it/s]
  7%|▋         | 734/10707 [10:07<1:22:57,  2.00it/s]
  7%|▋         | 735/10707 [10:08<1:23:00,  2.00it/s]
  7%|▋         | 736/10707 [10:08<1:22:52,  2.01it/s]
  7%|▋         | 737/10707 [10:09<1:23:07,  2.00it/s]
  7%|▋         | 738/10707 [10:09<1:23:11,  2.00it/s]
  7%|▋         | 739/10707 [10:10<1:23:07,  2.00it/s]
  7%|▋         | 740/10707 [10:10<1:22:57,  2.00it/s]
  7%|▋         | 741/10707 [10:11<1:22:55,  2.00it/s]
  7%|▋         | 742/10707 [10:11<1:22:54,  2.00it/s]
  7%|▋         | 743/10707 [10:12<1:22:55,  2.00it/s]
  7%|▋         | 744/10707 [10:12<1:22:55,  2.00it/s]
  7%|▋         | 745/10707 [10:13<1:23:09,  2.00it/s]
  7%|▋         | 746/10707 [10:13<1:23:07,  2.00it/s]
  7%|▋         | 747/10707 [10:14<1:23:11,  2.00it/s]
  7%|▋         | 748/10707 [10:14<1:23:10,  2.00it/s]
  7%|▋         | 749/10707 [10:15<1:23:17,  1.99it/s]
  7%|▋         | 750/10707 [10:15<1:23:05,  2.00it/s]
                                                     {'loss': 4.7654, 'grad_norm': 0.3877808451652527, 'learning_rate': 0.0007002801120448179, 'epoch': 0.07}
-

  7%|▋         | 750/10707 [10:15<1:23:05,  2.00it/s]
  7%|▋         | 751/10707 [10:16<1:23:01,  2.00it/s]
  7%|▋         | 752/10707 [10:16<1:22:59,  2.00it/s]
  7%|▋         | 753/10707 [10:17<1:22:47,  2.00it/s]
  7%|▋         | 754/10707 [10:17<1:22:41,  2.01it/s]
  7%|▋         | 755/10707 [10:18<1:22:45,  2.00it/s]
  7%|▋         | 756/10707 [10:18<1:22:47,  2.00it/s]
  7%|▋         | 757/10707 [10:19<1:22:44,  2.00it/s]
  7%|▋         | 758/10707 [10:19<1:22:43,  2.00it/s]
  7%|▋         | 759/10707 [10:20<1:22:47,  2.00it/s]
  7%|▋         | 760/10707 [10:20<1:22:45,  2.00it/s]
  7%|▋         | 761/10707 [10:21<1:22:42,  2.00it/s]
  7%|▋         | 762/10707 [10:21<1:22:41,  2.00it/s]
  7%|▋         | 763/10707 [10:22<1:22:40,  2.00it/s]
  7%|▋         | 764/10707 [10:22<1:22:39,  2.00it/s]
  7%|▋         | 765/10707 [10:23<1:22:34,  2.01it/s]
  7%|▋         | 766/10707 [10:23<1:22:30,  2.01it/s]
  7%|▋         | 767/10707 [10:24<1:22:28,  2.01it/s]
  7%|▋         | 768/10707 [10:24<1:22:29,  2.01it/s]
  7%|▋         | 769/10707 [10:25<1:22:35,  2.01it/s]
  7%|▋         | 770/10707 [10:25<1:22:26,  2.01it/s]
  7%|▋         | 771/10707 [10:26<1:22:30,  2.01it/s]
  7%|▋         | 772/10707 [10:26<1:22:50,  2.00it/s]
  7%|▋         | 773/10707 [10:27<1:22:57,  2.00it/s]
  7%|▋         | 774/10707 [10:27<1:22:42,  2.00it/s]
  7%|▋         | 775/10707 [10:28<1:22:40,  2.00it/s]{'loss': 4.7439, 'grad_norm': 0.3865131735801697, 'learning_rate': 0.0007236227824463119, 'epoch': 0.07}
-
                                                     

  7%|▋         | 775/10707 [10:28<1:22:40,  2.00it/s]
  7%|▋         | 776/10707 [10:28<1:22:44,  2.00it/s]
  7%|▋         | 777/10707 [10:29<1:22:35,  2.00it/s]
  7%|▋         | 778/10707 [10:29<1:22:28,  2.01it/s]
  7%|▋         | 779/10707 [10:30<1:22:29,  2.01it/s]
  7%|▋         | 780/10707 [10:30<1:22:26,  2.01it/s]
  7%|▋         | 781/10707 [10:31<1:22:26,  2.01it/s]
  7%|▋         | 782/10707 [10:31<1:22:26,  2.01it/s]
  7%|▋         | 783/10707 [10:32<1:22:26,  2.01it/s]
  7%|▋         | 784/10707 [10:32<1:22:21,  2.01it/s]
  7%|▋         | 785/10707 [10:33<1:22:23,  2.01it/s]
  7%|▋         | 786/10707 [10:33<1:22:24,  2.01it/s]
  7%|▋         | 787/10707 [10:34<1:22:21,  2.01it/s]
  7%|▋         | 788/10707 [10:34<1:22:20,  2.01it/s]
  7%|▋         | 789/10707 [10:35<1:22:29,  2.00it/s]
  7%|▋         | 790/10707 [10:35<1:22:21,  2.01it/s]
  7%|▋         | 791/10707 [10:36<1:22:18,  2.01it/s]
  7%|▋         | 792/10707 [10:36<1:22:25,  2.00it/s]
  7%|▋         | 793/10707 [10:37<1:22:21,  2.01it/s]
  7%|▋         | 794/10707 [10:37<1:22:21,  2.01it/s]
  7%|▋         | 795/10707 [10:38<1:22:22,  2.01it/s]
  7%|▋         | 796/10707 [10:38<1:22:28,  2.00it/s]
  7%|▋         | 797/10707 [10:39<1:22:21,  2.01it/s]
  7%|▋         | 798/10707 [10:39<1:22:19,  2.01it/s]
  7%|▋         | 799/10707 [10:40<1:22:22,  2.00it/s]
  7%|▋         | 800/10707 [10:40<1:22:13,  2.01it/s]
                                                     
{'loss': 4.7023, 'grad_norm': 0.36365842819213867, 'learning_rate': 0.0007469654528478058, 'epoch': 0.07}
-
  7%|▋         | 800/10707 [10:40<1:22:13,  2.01it/s]
  7%|▋         | 801/10707 [10:41<1:22:17,  2.01it/s]
  7%|▋         | 802/10707 [10:41<1:22:26,  2.00it/s]
  7%|▋         | 803/10707 [10:42<1:22:23,  2.00it/s]
  8%|▊         | 804/10707 [10:42<1:22:21,  2.00it/s]
  8%|▊         | 805/10707 [10:43<1:22:27,  2.00it/s]
  8%|▊         | 806/10707 [10:43<1:22:26,  2.00it/s]
  8%|▊         | 807/10707 [10:44<1:22:36,  2.00it/s]
  8%|▊         | 808/10707 [10:44<1:22:29,  2.00it/s]
  8%|▊         | 809/10707 [10:45<1:22:27,  2.00it/s]
  8%|▊         | 810/10707 [10:45<1:22:36,  2.00it/s]
  8%|▊         | 811/10707 [10:46<1:22:30,  2.00it/s]
  8%|▊         | 812/10707 [10:46<1:22:18,  2.00it/s]
  8%|▊         | 813/10707 [10:47<1:22:20,  2.00it/s]
  8%|▊         | 814/10707 [10:47<1:22:25,  2.00it/s]
  8%|▊         | 815/10707 [10:48<1:22:32,  2.00it/s]
  8%|▊         | 816/10707 [10:48<1:22:28,  2.00it/s]
  8%|▊         | 817/10707 [10:49<1:22:24,  2.00it/s]
  8%|▊         | 818/10707 [10:49<1:22:24,  2.00it/s]
  8%|▊         | 819/10707 [10:50<1:22:18,  2.00it/s]
  8%|▊         | 820/10707 [10:50<1:22:20,  2.00it/s]
  8%|▊         | 821/10707 [10:51<1:22:15,  2.00it/s]
  8%|▊         | 822/10707 [10:51<1:22:13,  2.00it/s]
  8%|▊         | 823/10707 [10:52<1:22:16,  2.00it/s]
  8%|▊         | 824/10707 [10:52<1:22:16,  2.00it/s]
  8%|▊         | 825/10707 [10:53<1:22:14,  2.00it/s]{'loss': 4.6907, 'grad_norm': 0.3555740714073181, 'learning_rate': 0.0007703081232492998, 'epoch': 0.08}
-
                                                     

  8%|▊         | 825/10707 [10:53<1:22:14,  2.00it/s]
  8%|▊         | 826/10707 [10:53<1:22:20,  2.00it/s]
  8%|▊         | 827/10707 [10:54<1:22:18,  2.00it/s]
  8%|▊         | 828/10707 [10:54<1:22:15,  2.00it/s]
  8%|▊         | 829/10707 [10:55<1:22:11,  2.00it/s]
  8%|▊         | 830/10707 [10:55<1:22:06,  2.00it/s]
  8%|▊         | 831/10707 [10:56<1:22:04,  2.01it/s]
  8%|▊         | 832/10707 [10:56<1:22:01,  2.01it/s]
  8%|▊         | 833/10707 [10:57<1:22:03,  2.01it/s]
  8%|▊         | 834/10707 [10:57<1:22:01,  2.01it/s]
  8%|▊         | 835/10707 [10:58<1:22:11,  2.00it/s]
  8%|▊         | 836/10707 [10:58<1:22:09,  2.00it/s]
  8%|▊         | 837/10707 [10:59<1:22:01,  2.01it/s]
  8%|▊         | 838/10707 [10:59<1:22:03,  2.00it/s]
  8%|▊         | 839/10707 [11:00<1:22:01,  2.01it/s]
  8%|▊         | 840/10707 [11:00<1:21:57,  2.01it/s]
  8%|▊         | 841/10707 [11:01<1:21:58,  2.01it/s]
  8%|▊         | 842/10707 [11:01<1:22:00,  2.00it/s]
  8%|▊         | 843/10707 [11:02<1:21:51,  2.01it/s]
  8%|▊         | 844/10707 [11:02<1:21:51,  2.01it/s]
  8%|▊         | 845/10707 [11:03<1:21:55,  2.01it/s]
  8%|▊         | 846/10707 [11:03<1:21:49,  2.01it/s]
  8%|▊         | 847/10707 [11:04<1:21:46,  2.01it/s]
  8%|▊         | 848/10707 [11:04<1:21:45,  2.01it/s]
  8%|▊         | 849/10707 [11:05<1:21:56,  2.00it/s]
  8%|▊         | 850/10707 [11:05<1:21:52,  2.01it/s]{'loss': 4.6621, 'grad_norm': 0.3475082814693451, 'learning_rate': 0.0007936507936507937, 'epoch': 0.08}
-
                                                     

  8%|▊         | 850/10707 [11:05<1:21:52,  2.01it/s]
  8%|▊         | 851/10707 [11:06<1:22:03,  2.00it/s]
  8%|▊         | 852/10707 [11:06<1:21:49,  2.01it/s]
  8%|▊         | 853/10707 [11:06<1:22:07,  2.00it/s]
  8%|▊         | 854/10707 [11:07<1:22:00,  2.00it/s]
  8%|▊         | 855/10707 [11:07<1:21:54,  2.00it/s]
  8%|▊         | 856/10707 [11:08<1:21:43,  2.01it/s]
  8%|▊         | 857/10707 [11:08<1:21:52,  2.01it/s]
  8%|▊         | 858/10707 [11:09<1:21:48,  2.01it/s]
  8%|▊         | 859/10707 [11:09<1:21:41,  2.01it/s]
  8%|▊         | 860/10707 [11:10<1:21:43,  2.01it/s]
  8%|▊         | 861/10707 [11:10<1:21:33,  2.01it/s]
  8%|▊         | 862/10707 [11:11<1:21:35,  2.01it/s]
  8%|▊         | 863/10707 [11:11<1:21:31,  2.01it/s]
  8%|▊         | 864/10707 [11:12<1:21:37,  2.01it/s]
  8%|▊         | 865/10707 [11:12<1:21:46,  2.01it/s]
  8%|▊         | 866/10707 [11:13<1:21:43,  2.01it/s]
  8%|▊         | 867/10707 [11:13<1:21:39,  2.01it/s]
  8%|▊         | 868/10707 [11:14<1:21:37,  2.01it/s]
  8%|▊         | 869/10707 [11:14<1:21:34,  2.01it/s]
  8%|▊         | 870/10707 [11:15<1:21:26,  2.01it/s]
  8%|▊         | 871/10707 [11:15<1:21:29,  2.01it/s]
  8%|▊         | 872/10707 [11:16<1:21:32,  2.01it/s]
  8%|▊         | 873/10707 [11:16<1:21:32,  2.01it/s]
  8%|▊         | 874/10707 [11:17<1:21:31,  2.01it/s]
  8%|▊         | 875/10707 [11:17<1:21:34,  2.01it/s]{'loss': 4.6319, 'grad_norm': 0.41164302825927734, 'learning_rate': 0.0008169934640522876, 'epoch': 0.08}
                                                     
-
  8%|▊         | 875/10707 [11:17<1:21:34,  2.01it/s]
  8%|▊         | 876/10707 [11:18<1:21:36,  2.01it/s]
  8%|▊         | 877/10707 [11:18<1:21:41,  2.01it/s]
  8%|▊         | 878/10707 [11:19<1:21:47,  2.00it/s]
  8%|▊         | 879/10707 [11:19<1:21:36,  2.01it/s]
  8%|▊         | 880/10707 [11:20<1:21:30,  2.01it/s]
  8%|▊         | 881/10707 [11:20<1:21:26,  2.01it/s]
  8%|▊         | 882/10707 [11:21<1:21:26,  2.01it/s]
  8%|▊         | 883/10707 [11:21<1:21:25,  2.01it/s]
  8%|▊         | 884/10707 [11:22<1:21:23,  2.01it/s]
  8%|▊         | 885/10707 [11:22<1:21:27,  2.01it/s]
  8%|▊         | 886/10707 [11:23<1:21:33,  2.01it/s]
  8%|▊         | 887/10707 [11:23<1:21:23,  2.01it/s]
  8%|▊         | 888/10707 [11:24<1:21:23,  2.01it/s]
  8%|▊         | 889/10707 [11:24<1:21:25,  2.01it/s]
  8%|▊         | 890/10707 [11:25<1:21:19,  2.01it/s]
  8%|▊         | 891/10707 [11:25<1:21:27,  2.01it/s]
  8%|▊         | 892/10707 [11:26<1:21:23,  2.01it/s]
  8%|▊         | 893/10707 [11:26<1:21:30,  2.01it/s]
  8%|▊         | 894/10707 [11:27<1:21:30,  2.01it/s]
  8%|▊         | 895/10707 [11:27<1:21:28,  2.01it/s]
  8%|▊         | 896/10707 [11:28<1:21:29,  2.01it/s]
  8%|▊         | 897/10707 [11:28<1:21:29,  2.01it/s]
  8%|▊         | 898/10707 [11:29<1:21:29,  2.01it/s]
  8%|▊         | 899/10707 [11:29<1:21:26,  2.01it/s]
  8%|▊         | 900/10707 [11:30<1:21:24,  2.01it/s]{'loss': 4.613, 'grad_norm': 0.3965432345867157, 'learning_rate': 0.0008403361344537816, 'epoch': 0.08}
-
                                                     

  8%|▊         | 900/10707 [11:30<1:21:24,  2.01it/s]
  8%|▊         | 901/10707 [11:30<1:21:40,  2.00it/s]
  8%|▊         | 902/10707 [11:31<1:21:39,  2.00it/s]
  8%|▊         | 903/10707 [11:31<1:21:36,  2.00it/s]
  8%|▊         | 904/10707 [11:32<1:21:29,  2.00it/s]
  8%|▊         | 905/10707 [11:32<1:21:23,  2.01it/s]
  8%|▊         | 906/10707 [11:33<1:21:17,  2.01it/s]
  8%|▊         | 907/10707 [11:33<1:21:16,  2.01it/s]
  8%|▊         | 908/10707 [11:34<1:21:17,  2.01it/s]
  8%|▊         | 909/10707 [11:34<1:21:13,  2.01it/s]
  8%|▊         | 910/10707 [11:35<1:21:16,  2.01it/s]
  9%|▊         | 911/10707 [11:35<1:21:46,  2.00it/s]
  9%|▊         | 912/10707 [11:36<1:21:40,  2.00it/s]
  9%|▊         | 913/10707 [11:36<1:21:28,  2.00it/s]
  9%|▊         | 914/10707 [11:37<1:21:29,  2.00it/s]
  9%|▊         | 915/10707 [11:37<1:21:30,  2.00it/s]
  9%|▊         | 916/10707 [11:38<1:21:19,  2.01it/s]
  9%|▊         | 917/10707 [11:38<1:21:22,  2.01it/s]
  9%|▊         | 918/10707 [11:39<1:21:15,  2.01it/s]
  9%|▊         | 919/10707 [11:39<1:21:07,  2.01it/s]
  9%|▊         | 920/10707 [11:40<1:21:12,  2.01it/s]
  9%|▊         | 921/10707 [11:40<1:21:11,  2.01it/s]
  9%|▊         | 922/10707 [11:41<1:21:07,  2.01it/s]
  9%|▊         | 923/10707 [11:41<1:21:09,  2.01it/s]
  9%|▊         | 924/10707 [11:42<1:21:11,  2.01it/s]
  9%|▊         | 925/10707 [11:42<1:21:08,  2.01it/s]
                                                     
{'loss': 4.5928, 'grad_norm': 0.35606470704078674, 'learning_rate': 0.0008636788048552755, 'epoch': 0.09}
-
  9%|▊         | 925/10707 [11:42<1:21:08,  2.01it/s]
  9%|▊         | 926/10707 [11:43<1:21:12,  2.01it/s]
  9%|▊         | 927/10707 [11:43<1:21:07,  2.01it/s]
  9%|▊         | 928/10707 [11:44<1:21:04,  2.01it/s]
  9%|▊         | 929/10707 [11:44<1:21:03,  2.01it/s]
  9%|▊         | 930/10707 [11:45<1:20:59,  2.01it/s]
  9%|▊         | 931/10707 [11:45<1:21:02,  2.01it/s]
  9%|▊         | 932/10707 [11:46<1:21:06,  2.01it/s]
  9%|▊         | 933/10707 [11:46<1:20:58,  2.01it/s]
  9%|▊         | 934/10707 [11:47<1:21:02,  2.01it/s]
  9%|▊         | 935/10707 [11:47<1:21:01,  2.01it/s]
  9%|▊         | 936/10707 [11:48<1:21:00,  2.01it/s]
  9%|▉         | 937/10707 [11:48<1:21:05,  2.01it/s]
  9%|▉         | 938/10707 [11:49<1:20:57,  2.01it/s]
  9%|▉         | 939/10707 [11:49<1:20:58,  2.01it/s]
  9%|▉         | 940/10707 [11:50<1:21:05,  2.01it/s]
  9%|▉         | 941/10707 [11:50<1:20:57,  2.01it/s]
  9%|▉         | 942/10707 [11:51<1:20:59,  2.01it/s]
  9%|▉         | 943/10707 [11:51<1:21:01,  2.01it/s]
  9%|▉         | 944/10707 [11:52<1:20:59,  2.01it/s]
  9%|▉         | 945/10707 [11:52<1:20:59,  2.01it/s]
  9%|▉         | 946/10707 [11:53<1:21:01,  2.01it/s]
  9%|▉         | 947/10707 [11:53<1:20:51,  2.01it/s]
  9%|▉         | 948/10707 [11:54<1:20:56,  2.01it/s]
  9%|▉         | 949/10707 [11:54<1:20:57,  2.01it/s]
  9%|▉         | 950/10707 [11:55<1:20:56,  2.01it/s]
                                                     {'loss': 4.5603, 'grad_norm': 0.3743130564689636, 'learning_rate': 0.0008870214752567694, 'epoch': 0.09}
-

  9%|▉         | 950/10707 [11:55<1:20:56,  2.01it/s]
  9%|▉         | 951/10707 [11:55<1:21:01,  2.01it/s]
  9%|▉         | 952/10707 [11:56<1:21:03,  2.01it/s]
  9%|▉         | 953/10707 [11:56<1:21:00,  2.01it/s]
  9%|▉         | 954/10707 [11:57<1:20:58,  2.01it/s]
  9%|▉         | 955/10707 [11:57<1:20:57,  2.01it/s]
  9%|▉         | 956/10707 [11:58<1:20:51,  2.01it/s]
  9%|▉         | 957/10707 [11:58<1:20:47,  2.01it/s]
  9%|▉         | 958/10707 [11:59<1:20:45,  2.01it/s]
  9%|▉         | 959/10707 [11:59<1:20:54,  2.01it/s]
  9%|▉         | 960/10707 [12:00<1:20:49,  2.01it/s]
  9%|▉         | 961/10707 [12:00<1:21:02,  2.00it/s]
  9%|▉         | 962/10707 [12:01<1:21:02,  2.00it/s]
  9%|▉         | 963/10707 [12:01<1:20:55,  2.01it/s]
  9%|▉         | 964/10707 [12:02<1:20:53,  2.01it/s]
  9%|▉         | 965/10707 [12:02<1:20:53,  2.01it/s]
  9%|▉         | 966/10707 [12:03<1:20:50,  2.01it/s]
  9%|▉         | 967/10707 [12:03<1:20:47,  2.01it/s]
  9%|▉         | 968/10707 [12:04<1:20:40,  2.01it/s]
  9%|▉         | 969/10707 [12:04<1:20:38,  2.01it/s]
  9%|▉         | 970/10707 [12:05<1:20:40,  2.01it/s]
  9%|▉         | 971/10707 [12:05<1:20:34,  2.01it/s]
  9%|▉         | 972/10707 [12:06<1:20:36,  2.01it/s]
  9%|▉         | 973/10707 [12:06<1:20:41,  2.01it/s]
  9%|▉         | 974/10707 [12:07<1:20:38,  2.01it/s]
  9%|▉         | 975/10707 [12:07<1:20:38,  2.01it/s]{'loss': 4.536, 'grad_norm': 0.34302517771720886, 'learning_rate': 0.0009103641456582633, 'epoch': 0.09}
-
                                                     

  9%|▉         | 975/10707 [12:07<1:20:38,  2.01it/s]
  9%|▉         | 976/10707 [12:08<1:20:40,  2.01it/s]
  9%|▉         | 977/10707 [12:08<1:20:38,  2.01it/s]
  9%|▉         | 978/10707 [12:09<1:20:38,  2.01it/s]
  9%|▉         | 979/10707 [12:09<1:20:35,  2.01it/s]
  9%|▉         | 980/10707 [12:10<1:20:41,  2.01it/s]
  9%|▉         | 981/10707 [12:10<1:20:40,  2.01it/s]
  9%|▉         | 982/10707 [12:11<1:20:39,  2.01it/s]
  9%|▉         | 983/10707 [12:11<1:20:39,  2.01it/s]
  9%|▉         | 984/10707 [12:12<1:20:49,  2.01it/s]
  9%|▉         | 985/10707 [12:12<1:20:40,  2.01it/s]
  9%|▉         | 986/10707 [12:13<1:20:38,  2.01it/s]
  9%|▉         | 987/10707 [12:13<1:20:34,  2.01it/s]
  9%|▉         | 988/10707 [12:14<1:20:32,  2.01it/s]
  9%|▉         | 989/10707 [12:14<1:20:40,  2.01it/s]
  9%|▉         | 990/10707 [12:15<1:20:33,  2.01it/s]
  9%|▉         | 991/10707 [12:15<1:20:33,  2.01it/s]
  9%|▉         | 992/10707 [12:16<1:20:34,  2.01it/s]
  9%|▉         | 993/10707 [12:16<1:20:31,  2.01it/s]
  9%|▉         | 994/10707 [12:17<1:20:29,  2.01it/s]
  9%|▉         | 995/10707 [12:17<1:20:27,  2.01it/s]
  9%|▉         | 996/10707 [12:18<1:20:24,  2.01it/s]
  9%|▉         | 997/10707 [12:18<1:20:28,  2.01it/s]
  9%|▉         | 998/10707 [12:19<1:20:23,  2.01it/s]
  9%|▉         | 999/10707 [12:19<1:20:22,  2.01it/s]
  9%|▉         | 1000/10707 [12:20<1:20:42,  2.00it/s]
                                                      
{'loss': 4.5232, 'grad_norm': 0.33623871207237244, 'learning_rate': 0.0009337068160597573, 'epoch': 0.09}
-
  9%|▉         | 1000/10707 [12:20<1:20:42,  2.00it/s]
  9%|▉         | 1001/10707 [12:20<1:20:37,  2.01it/s]
  9%|▉         | 1002/10707 [12:21<1:20:34,  2.01it/s]
  9%|▉         | 1003/10707 [12:21<1:20:31,  2.01it/s]
  9%|▉         | 1004/10707 [12:22<1:20:29,  2.01it/s]
  9%|▉         | 1005/10707 [12:22<1:20:25,  2.01it/s]
  9%|▉         | 1006/10707 [12:23<1:20:22,  2.01it/s]
  9%|▉         | 1007/10707 [12:23<1:20:32,  2.01it/s]
  9%|▉         | 1008/10707 [12:24<1:20:22,  2.01it/s]
  9%|▉         | 1009/10707 [12:24<1:20:25,  2.01it/s]
  9%|▉         | 1010/10707 [12:25<1:20:32,  2.01it/s]
  9%|▉         | 1011/10707 [12:25<1:20:28,  2.01it/s]
  9%|▉         | 1012/10707 [12:26<1:20:30,  2.01it/s]
  9%|▉         | 1013/10707 [12:26<1:20:27,  2.01it/s]
  9%|▉         | 1014/10707 [12:27<1:20:27,  2.01it/s]
  9%|▉         | 1015/10707 [12:27<1:20:27,  2.01it/s]
  9%|▉         | 1016/10707 [12:28<1:20:27,  2.01it/s]
  9%|▉         | 1017/10707 [12:28<1:20:20,  2.01it/s]
 10%|▉         | 1018/10707 [12:29<1:20:18,  2.01it/s]
 10%|▉         | 1019/10707 [12:29<1:20:13,  2.01it/s]
 10%|▉         | 1020/10707 [12:30<1:20:13,  2.01it/s]
 10%|▉         | 1021/10707 [12:30<1:20:17,  2.01it/s]
 10%|▉         | 1022/10707 [12:31<1:20:18,  2.01it/s]
 10%|▉         | 1023/10707 [12:31<1:20:21,  2.01it/s]
 10%|▉         | 1024/10707 [12:32<1:20:20,  2.01it/s]
 10%|▉         | 1025/10707 [12:32<1:20:20,  2.01it/s]
                                                      
{'loss': 4.4932, 'grad_norm': 0.38535675406455994, 'learning_rate': 0.0009570494864612512, 'epoch': 0.1}
-
 10%|▉         | 1025/10707 [12:32<1:20:20,  2.01it/s]
 10%|▉         | 1026/10707 [12:33<1:20:23,  2.01it/s]
 10%|▉         | 1027/10707 [12:33<1:20:19,  2.01it/s]
 10%|▉         | 1028/10707 [12:34<1:20:19,  2.01it/s]
 10%|▉         | 1029/10707 [12:34<1:20:16,  2.01it/s]
 10%|▉         | 1030/10707 [12:35<1:20:12,  2.01it/s]
 10%|▉         | 1031/10707 [12:35<1:20:08,  2.01it/s]
 10%|▉         | 1032/10707 [12:36<1:20:16,  2.01it/s]
 10%|▉         | 1033/10707 [12:36<1:20:12,  2.01it/s]
 10%|▉         | 1034/10707 [12:37<1:20:10,  2.01it/s]
 10%|▉         | 1035/10707 [12:37<1:20:03,  2.01it/s]
 10%|▉         | 1036/10707 [12:38<1:20:05,  2.01it/s]
 10%|▉         | 1037/10707 [12:38<1:19:56,  2.02it/s]
 10%|▉         | 1038/10707 [12:39<1:19:56,  2.02it/s]
 10%|▉         | 1039/10707 [12:39<1:20:01,  2.01it/s]
 10%|▉         | 1040/10707 [12:40<1:20:00,  2.01it/s]
 10%|▉         | 1041/10707 [12:40<1:20:09,  2.01it/s]
 10%|▉         | 1042/10707 [12:41<1:32:51,  1.73it/s]
 10%|▉         | 1043/10707 [12:41<1:29:02,  1.81it/s]
 10%|▉         | 1044/10707 [12:42<1:26:16,  1.87it/s]
 10%|▉         | 1045/10707 [12:42<1:24:20,  1.91it/s]
 10%|▉         | 1046/10707 [12:43<1:23:00,  1.94it/s]
 10%|▉         | 1047/10707 [12:43<1:22:04,  1.96it/s]
 10%|▉         | 1048/10707 [12:44<1:34:06,  1.71it/s]
 10%|▉         | 1049/10707 [12:45<1:29:50,  1.79it/s]
 10%|▉         | 1050/10707 [12:45<1:26:54,  1.85it/s]{'loss': 4.4756, 'grad_norm': 0.3656974732875824, 'learning_rate': 0.000980392156862745, 'epoch': 0.1}
-
                                                      

 10%|▉         | 1050/10707 [12:45<1:26:54,  1.85it/s]
 10%|▉         | 1051/10707 [12:46<1:24:49,  1.90it/s]
 10%|▉         | 1052/10707 [12:46<1:23:29,  1.93it/s]
 10%|▉         | 1053/10707 [12:47<1:22:25,  1.95it/s]
 10%|▉         | 1054/10707 [12:47<1:21:35,  1.97it/s]
 10%|▉         | 1055/10707 [12:48<1:21:03,  1.98it/s]
 10%|▉         | 1056/10707 [12:48<1:20:37,  2.00it/s]
 10%|▉         | 1057/10707 [12:49<1:20:25,  2.00it/s]
 10%|▉         | 1058/10707 [12:49<1:20:07,  2.01it/s]
 10%|▉         | 1059/10707 [12:50<1:20:08,  2.01it/s]
 10%|▉         | 1060/10707 [12:50<1:20:00,  2.01it/s]
 10%|▉         | 1061/10707 [12:51<1:20:01,  2.01it/s]
 10%|▉         | 1062/10707 [12:51<1:20:03,  2.01it/s]
 10%|▉         | 1063/10707 [12:52<1:20:00,  2.01it/s]
 10%|▉         | 1064/10707 [12:52<1:19:54,  2.01it/s]
 10%|▉         | 1065/10707 [12:53<1:19:56,  2.01it/s]
 10%|▉         | 1066/10707 [12:53<1:19:56,  2.01it/s]
 10%|▉         | 1067/10707 [12:54<1:19:57,  2.01it/s]
 10%|▉         | 1068/10707 [12:54<1:19:56,  2.01it/s]
 10%|▉         | 1069/10707 [12:55<1:19:51,  2.01it/s]
 10%|▉         | 1070/10707 [12:55<1:19:50,  2.01it/s]
 10%|█         | 1071/10707 [12:56<1:19:52,  2.01it/s]
 10%|█         | 1072/10707 [12:56<1:19:48,  2.01it/s]
 10%|█         | 1073/10707 [12:57<1:19:53,  2.01it/s]
 10%|█         | 1074/10707 [12:57<1:19:48,  2.01it/s]
 10%|█         | 1075/10707 [12:57<1:19:50,  2.01it/s]
                                                      {'loss': 4.4589, 'grad_norm': 0.33178970217704773, 'learning_rate': 0.0009999995748265938, 'epoch': 0.1}
-
 10%|█         | 1075/10707 [12:58<1:19:50,  2.01it/s]
 10%|█         | 1076/10707 [12:58<1:19:54,  2.01it/s]
 10%|█         | 1077/10707 [12:58<1:19:58,  2.01it/s]
 10%|█         | 1078/10707 [12:59<1:19:56,  2.01it/s]
 10%|█         | 1079/10707 [12:59<1:19:51,  2.01it/s]
 10%|█         | 1080/10707 [13:00<1:19:46,  2.01it/s]
 10%|█         | 1081/10707 [13:00<1:19:46,  2.01it/s]
 10%|█         | 1082/10707 [13:01<1:19:51,  2.01it/s]
 10%|█         | 1083/10707 [13:01<1:19:44,  2.01it/s]
 10%|█         | 1084/10707 [13:02<1:19:38,  2.01it/s]
 10%|█         | 1085/10707 [13:02<1:19:36,  2.01it/s]
 10%|█         | 1086/10707 [13:03<1:19:30,  2.02it/s]
 10%|█         | 1087/10707 [13:03<1:19:32,  2.02it/s]
 10%|█         | 1088/10707 [13:04<1:19:34,  2.01it/s]
 10%|█         | 1089/10707 [13:04<1:19:34,  2.01it/s]
 10%|█         | 1090/10707 [13:05<1:19:56,  2.00it/s]
 10%|█         | 1091/10707 [13:05<1:19:48,  2.01it/s]
 10%|█         | 1092/10707 [13:06<1:19:42,  2.01it/s]
 10%|█         | 1093/10707 [13:06<1:19:43,  2.01it/s]
 10%|█         | 1094/10707 [13:07<1:19:41,  2.01it/s]
 10%|█         | 1095/10707 [13:07<1:19:45,  2.01it/s]
 10%|█         | 1096/10707 [13:08<1:19:50,  2.01it/s]
 10%|█         | 1097/10707 [13:08<1:19:37,  2.01it/s]
 10%|█         | 1098/10707 [13:09<1:19:39,  2.01it/s]
 10%|█         | 1099/10707 [13:09<1:19:39,  2.01it/s]
 10%|█         | 1100/10707 [13:10<1:19:32,  2.01it/s]{'loss': 4.4539, 'grad_norm': 0.3041117787361145, 'learning_rate': 0.000999977651986143, 'epoch': 0.1}
-
                                                      

 10%|█         | 1100/10707 [13:10<1:19:32,  2.01it/s]
 10%|█         | 1101/10707 [13:10<1:19:39,  2.01it/s]
 10%|█         | 1102/10707 [13:11<1:19:34,  2.01it/s]
 10%|█         | 1103/10707 [13:11<1:19:34,  2.01it/s]
 10%|█         | 1104/10707 [13:12<1:19:34,  2.01it/s]
 10%|█         | 1105/10707 [13:12<1:19:27,  2.01it/s]
 10%|█         | 1106/10707 [13:13<1:19:34,  2.01it/s]
 10%|█         | 1107/10707 [13:13<1:19:34,  2.01it/s]
 10%|█         | 1108/10707 [13:14<1:19:27,  2.01it/s]
 10%|█         | 1109/10707 [13:14<1:19:27,  2.01it/s]
 10%|█         | 1110/10707 [13:15<1:19:21,  2.02it/s]
 10%|█         | 1111/10707 [13:15<1:19:23,  2.01it/s]
 10%|█         | 1112/10707 [13:16<1:19:20,  2.02it/s]
 10%|█         | 1113/10707 [13:16<1:19:25,  2.01it/s]
 10%|█         | 1114/10707 [13:17<1:19:20,  2.02it/s]
 10%|█         | 1115/10707 [13:17<1:19:19,  2.02it/s]
 10%|█         | 1116/10707 [13:18<1:19:16,  2.02it/s]
 10%|█         | 1117/10707 [13:18<1:19:17,  2.02it/s]
 10%|█         | 1118/10707 [13:19<1:19:21,  2.01it/s]
 10%|█         | 1119/10707 [13:19<1:19:16,  2.02it/s]
 10%|█         | 1120/10707 [13:20<1:19:19,  2.01it/s]
 10%|█         | 1121/10707 [13:20<1:19:24,  2.01it/s]
 10%|█         | 1122/10707 [13:21<1:19:23,  2.01it/s]
 10%|█         | 1123/10707 [13:21<1:19:29,  2.01it/s]
 10%|█         | 1124/10707 [13:22<1:19:22,  2.01it/s]
 11%|█         | 1125/10707 [13:22<1:19:27,  2.01it/s]{'loss': 4.4126, 'grad_norm': 0.30170753598213196, 'learning_rate': 0.0009999225141371527, 'epoch': 0.11}
-
                                                      

 11%|█         | 1125/10707 [13:22<1:19:27,  2.01it/s]
 11%|█         | 1126/10707 [13:23<1:19:33,  2.01it/s]
 11%|█         | 1127/10707 [13:23<1:19:22,  2.01it/s]
 11%|█         | 1128/10707 [13:24<1:19:22,  2.01it/s]
 11%|█         | 1129/10707 [13:24<1:19:20,  2.01it/s]
 11%|█         | 1130/10707 [13:25<1:19:19,  2.01it/s]
 11%|█         | 1131/10707 [13:25<1:19:32,  2.01it/s]
 11%|█         | 1132/10707 [13:26<1:19:20,  2.01it/s]
 11%|█         | 1133/10707 [13:26<1:19:19,  2.01it/s]
 11%|█         | 1134/10707 [13:27<1:19:15,  2.01it/s]
 11%|█         | 1135/10707 [13:27<1:19:13,  2.01it/s]
 11%|█         | 1136/10707 [13:28<1:19:06,  2.02it/s]
 11%|█         | 1137/10707 [13:28<1:19:09,  2.01it/s]
 11%|█         | 1138/10707 [13:29<1:19:07,  2.02it/s]
 11%|█         | 1139/10707 [13:29<1:19:13,  2.01it/s]
 11%|█         | 1140/10707 [13:30<1:19:14,  2.01it/s]
 11%|█         | 1141/10707 [13:30<1:19:08,  2.01it/s]
 11%|█         | 1142/10707 [13:31<1:19:05,  2.02it/s]
 11%|█         | 1143/10707 [13:31<1:19:08,  2.01it/s]
 11%|█         | 1144/10707 [13:32<1:19:10,  2.01it/s]
 11%|█         | 1145/10707 [13:32<1:19:09,  2.01it/s]
 11%|█         | 1146/10707 [13:33<1:19:08,  2.01it/s]
 11%|█         | 1147/10707 [13:33<1:19:08,  2.01it/s]
 11%|█         | 1148/10707 [13:34<1:19:06,  2.01it/s]
 11%|█         | 1149/10707 [13:34<1:19:06,  2.01it/s]
 11%|█         | 1150/10707 [13:35<1:19:07,  2.01it/s]{'loss': 4.398, 'grad_norm': 0.287154883146286, 'learning_rate': 0.000999834164942594, 'epoch': 0.11}
                                                      
-
 11%|█         | 1150/10707 [13:35<1:19:07,  2.01it/s]
 11%|█         | 1151/10707 [13:35<1:19:15,  2.01it/s]
 11%|█         | 1152/10707 [13:36<1:19:29,  2.00it/s]
 11%|█         | 1153/10707 [13:36<1:19:23,  2.01it/s]
 11%|█         | 1154/10707 [13:37<1:19:16,  2.01it/s]
 11%|█         | 1155/10707 [13:37<1:19:06,  2.01it/s]
 11%|█         | 1156/10707 [13:38<1:19:12,  2.01it/s]
 11%|█         | 1157/10707 [13:38<1:19:00,  2.01it/s]
 11%|█         | 1158/10707 [13:39<1:19:03,  2.01it/s]
 11%|█         | 1159/10707 [13:39<1:19:00,  2.01it/s]
 11%|█         | 1160/10707 [13:40<1:19:00,  2.01it/s]
 11%|█         | 1161/10707 [13:40<1:18:55,  2.02it/s]
 11%|█         | 1162/10707 [13:41<1:18:51,  2.02it/s]
 11%|█         | 1163/10707 [13:41<1:18:58,  2.01it/s]
 11%|█         | 1164/10707 [13:42<1:18:51,  2.02it/s]
 11%|█         | 1165/10707 [13:42<1:18:56,  2.01it/s]
 11%|█         | 1166/10707 [13:43<1:18:55,  2.01it/s]
 11%|█         | 1167/10707 [13:43<1:19:01,  2.01it/s]
 11%|█         | 1168/10707 [13:44<1:19:02,  2.01it/s]
 11%|█         | 1169/10707 [13:44<1:18:59,  2.01it/s]
 11%|█         | 1170/10707 [13:45<1:18:56,  2.01it/s]
 11%|█         | 1171/10707 [13:45<1:18:56,  2.01it/s]
 11%|█         | 1172/10707 [13:46<1:18:53,  2.01it/s]
 11%|█         | 1173/10707 [13:46<1:18:59,  2.01it/s]
 11%|█         | 1174/10707 [13:47<1:18:56,  2.01it/s]
 11%|█         | 1175/10707 [13:47<1:18:59,  2.01it/s]{'loss': 4.3852, 'grad_norm': 0.285279244184494, 'learning_rate': 0.0009997126102717687, 'epoch': 0.11}
-
                                                      

 11%|█         | 1175/10707 [13:47<1:18:59,  2.01it/s]
 11%|█         | 1176/10707 [13:48<1:19:06,  2.01it/s]
 11%|█         | 1177/10707 [13:48<1:18:55,  2.01it/s]
 11%|█         | 1178/10707 [13:49<1:18:51,  2.01it/s]
 11%|█         | 1179/10707 [13:49<1:18:55,  2.01it/s]
 11%|█         | 1180/10707 [13:50<1:18:52,  2.01it/s]
 11%|█         | 1181/10707 [13:50<1:18:52,  2.01it/s]
 11%|█         | 1182/10707 [13:51<1:18:46,  2.02it/s]
 11%|█         | 1183/10707 [13:51<1:18:51,  2.01it/s]
 11%|█         | 1184/10707 [13:52<1:18:46,  2.01it/s]
 11%|█         | 1185/10707 [13:52<1:18:51,  2.01it/s]
 11%|█         | 1186/10707 [13:53<1:18:54,  2.01it/s]
 11%|█         | 1187/10707 [13:53<1:18:48,  2.01it/s]
 11%|█         | 1188/10707 [13:54<1:18:49,  2.01it/s]
 11%|█         | 1189/10707 [13:54<1:18:44,  2.01it/s]
 11%|█         | 1190/10707 [13:55<1:18:47,  2.01it/s]
 11%|█         | 1191/10707 [13:55<1:18:52,  2.01it/s]
 11%|█         | 1192/10707 [13:56<1:18:44,  2.01it/s]
 11%|█         | 1193/10707 [13:56<1:18:45,  2.01it/s]
 11%|█         | 1194/10707 [13:57<1:18:41,  2.01it/s]
 11%|█         | 1195/10707 [13:57<1:18:40,  2.01it/s]
 11%|█         | 1196/10707 [13:58<1:18:42,  2.01it/s]
 11%|█         | 1197/10707 [13:58<1:18:37,  2.02it/s]
 11%|█         | 1198/10707 [13:59<1:18:37,  2.02it/s]
 11%|█         | 1199/10707 [13:59<1:18:32,  2.02it/s]
 11%|█         | 1200/10707 [14:00<1:18:41,  2.01it/s]
                                                      
{'loss': 4.3595, 'grad_norm': 0.2913043797016144, 'learning_rate': 0.0009995578581999157, 'epoch': 0.11}
-
 11%|█         | 1200/10707 [14:00<1:18:41,  2.01it/s]
 11%|█         | 1201/10707 [14:00<1:18:40,  2.01it/s]
 11%|█         | 1202/10707 [14:01<1:18:38,  2.01it/s]
 11%|█         | 1203/10707 [14:01<1:18:36,  2.01it/s]
 11%|█         | 1204/10707 [14:02<1:18:33,  2.02it/s]
 11%|█▏        | 1205/10707 [14:02<1:18:38,  2.01it/s]
 11%|█▏        | 1206/10707 [14:03<1:18:36,  2.01it/s]
 11%|█▏        | 1207/10707 [14:03<1:18:38,  2.01it/s]
 11%|█▏        | 1208/10707 [14:04<1:18:34,  2.01it/s]
 11%|█▏        | 1209/10707 [14:04<1:18:36,  2.01it/s]
 11%|█▏        | 1210/10707 [14:05<1:18:38,  2.01it/s]
 11%|█▏        | 1211/10707 [14:05<1:18:32,  2.02it/s]
 11%|█▏        | 1212/10707 [14:06<1:18:31,  2.02it/s]
 11%|█▏        | 1213/10707 [14:06<1:18:38,  2.01it/s]
 11%|█▏        | 1214/10707 [14:07<1:18:38,  2.01it/s]
 11%|█▏        | 1215/10707 [14:07<1:18:36,  2.01it/s]
 11%|█▏        | 1216/10707 [14:08<1:18:32,  2.01it/s]
 11%|█▏        | 1217/10707 [14:08<1:18:32,  2.01it/s]
 11%|█▏        | 1218/10707 [14:09<1:18:26,  2.02it/s]
 11%|█▏        | 1219/10707 [14:09<1:18:26,  2.02it/s]
 11%|█▏        | 1220/10707 [14:10<1:18:24,  2.02it/s]
 11%|█▏        | 1221/10707 [14:10<1:18:29,  2.01it/s]
 11%|█▏        | 1222/10707 [14:11<1:18:29,  2.01it/s]
 11%|█▏        | 1223/10707 [14:11<1:18:27,  2.01it/s]
 11%|█▏        | 1224/10707 [14:12<1:18:25,  2.02it/s]
 11%|█▏        | 1225/10707 [14:12<1:18:27,  2.01it/s]{'loss': 4.3367, 'grad_norm': 0.2734946310520172, 'learning_rate': 0.0009993699190076779, 'epoch': 0.11}
-
                                                      

 11%|█▏        | 1225/10707 [14:12<1:18:27,  2.01it/s]
 11%|█▏        | 1226/10707 [14:13<1:18:29,  2.01it/s]
 11%|█▏        | 1227/10707 [14:13<1:18:25,  2.01it/s]
 11%|█▏        | 1228/10707 [14:14<1:18:25,  2.01it/s]
 11%|█▏        | 1229/10707 [14:14<1:18:25,  2.01it/s]
 11%|█▏        | 1230/10707 [14:14<1:18:22,  2.02it/s]
 11%|█▏        | 1231/10707 [14:15<1:18:22,  2.02it/s]
 12%|█▏        | 1232/10707 [14:15<1:18:39,  2.01it/s]
 12%|█▏        | 1233/10707 [14:16<1:18:33,  2.01it/s]
 12%|█▏        | 1234/10707 [14:16<1:18:26,  2.01it/s]
 12%|█▏        | 1235/10707 [14:17<1:18:23,  2.01it/s]
 12%|█▏        | 1236/10707 [14:17<1:18:17,  2.02it/s]
 12%|█▏        | 1237/10707 [14:18<1:18:17,  2.02it/s]
 12%|█▏        | 1238/10707 [14:18<1:18:25,  2.01it/s]
 12%|█▏        | 1239/10707 [14:19<1:18:21,  2.01it/s]
 12%|█▏        | 1240/10707 [14:19<1:18:25,  2.01it/s]
 12%|█▏        | 1241/10707 [14:20<1:18:20,  2.01it/s]
 12%|█▏        | 1242/10707 [14:20<1:18:19,  2.01it/s]
 12%|█▏        | 1243/10707 [14:21<1:18:17,  2.01it/s]
 12%|█▏        | 1244/10707 [14:21<1:18:22,  2.01it/s]
 12%|█▏        | 1245/10707 [14:22<1:18:23,  2.01it/s]
 12%|█▏        | 1246/10707 [14:22<1:18:21,  2.01it/s]
 12%|█▏        | 1247/10707 [14:23<1:18:23,  2.01it/s]
 12%|█▏        | 1248/10707 [14:23<1:18:18,  2.01it/s]
 12%|█▏        | 1249/10707 [14:24<1:18:19,  2.01it/s]
 12%|█▏        | 1250/10707 [14:24<1:18:20,  2.01it/s]{'loss': 4.3267, 'grad_norm': 0.25569283962249756, 'learning_rate': 0.0009991488051804167, 'epoch': 0.12}
-
                                                      

 12%|█▏        | 1250/10707 [14:24<1:18:20,  2.01it/s]
 12%|█▏        | 1251/10707 [14:25<1:18:29,  2.01it/s]
 12%|█▏        | 1252/10707 [14:25<1:18:28,  2.01it/s]
 12%|█▏        | 1253/10707 [14:26<1:18:20,  2.01it/s]
 12%|█▏        | 1254/10707 [14:26<1:18:13,  2.01it/s]
 12%|█▏        | 1255/10707 [14:27<1:18:12,  2.01it/s]
 12%|█▏        | 1256/10707 [14:27<1:18:18,  2.01it/s]
 12%|█▏        | 1257/10707 [14:28<1:18:13,  2.01it/s]
 12%|█▏        | 1258/10707 [14:28<1:18:14,  2.01it/s]
 12%|█▏        | 1259/10707 [14:29<1:18:10,  2.01it/s]
 12%|█▏        | 1260/10707 [14:29<1:18:10,  2.01it/s]
 12%|█▏        | 1261/10707 [14:30<1:18:09,  2.01it/s]
 12%|█▏        | 1262/10707 [14:30<1:18:07,  2.01it/s]
 12%|█▏        | 1263/10707 [14:31<1:18:12,  2.01it/s]
 12%|█▏        | 1264/10707 [14:31<1:18:16,  2.01it/s]
 12%|█▏        | 1265/10707 [14:32<1:18:12,  2.01it/s]
 12%|█▏        | 1266/10707 [14:32<1:18:10,  2.01it/s]
 12%|█▏        | 1267/10707 [14:33<1:18:03,  2.02it/s]
 12%|█▏        | 1268/10707 [14:33<1:18:06,  2.01it/s]
 12%|█▏        | 1269/10707 [14:34<1:18:08,  2.01it/s]
 12%|█▏        | 1270/10707 [14:34<1:18:07,  2.01it/s]
 12%|█▏        | 1271/10707 [14:35<1:18:05,  2.01it/s]
 12%|█▏        | 1272/10707 [14:35<1:18:07,  2.01it/s]
 12%|█▏        | 1273/10707 [14:36<1:18:03,  2.01it/s]
 12%|█▏        | 1274/10707 [14:36<1:18:04,  2.01it/s]
 12%|█▏        | 1275/10707 [14:37<1:18:07,  2.01it/s]
                                                      
{'loss': 4.312, 'grad_norm': 0.28991401195526123, 'learning_rate': 0.000998894531407384, 'epoch': 0.12}
-
 12%|█▏        | 1275/10707 [14:37<1:18:07,  2.01it/s]
 12%|█▏        | 1276/10707 [14:37<1:18:12,  2.01it/s]
 12%|█▏        | 1277/10707 [14:38<1:18:10,  2.01it/s]
 12%|█▏        | 1278/10707 [14:38<1:18:05,  2.01it/s]
 12%|█▏        | 1279/10707 [14:39<1:18:02,  2.01it/s]
 12%|█▏        | 1280/10707 [14:39<1:17:59,  2.01it/s]
 12%|█▏        | 1281/10707 [14:40<1:17:58,  2.01it/s]
 12%|█▏        | 1282/10707 [14:40<1:18:04,  2.01it/s]
 12%|█▏        | 1283/10707 [14:41<1:18:04,  2.01it/s]
 12%|█▏        | 1284/10707 [14:41<1:18:01,  2.01it/s]
 12%|█▏        | 1285/10707 [14:42<1:17:59,  2.01it/s]
 12%|█▏        | 1286/10707 [14:42<1:18:05,  2.01it/s]
 12%|█▏        | 1287/10707 [14:43<1:17:54,  2.02it/s]
 12%|█▏        | 1288/10707 [14:43<1:17:57,  2.01it/s]
 12%|█▏        | 1289/10707 [14:44<1:17:55,  2.01it/s]
 12%|█▏        | 1290/10707 [14:44<1:17:54,  2.01it/s]
 12%|█▏        | 1291/10707 [14:45<1:17:49,  2.02it/s]
 12%|█▏        | 1292/10707 [14:45<1:17:47,  2.02it/s]
 12%|█▏        | 1293/10707 [14:46<1:17:51,  2.02it/s]
 12%|█▏        | 1294/10707 [14:46<1:17:48,  2.02it/s]
 12%|█▏        | 1295/10707 [14:47<1:17:50,  2.02it/s]
 12%|█▏        | 1296/10707 [14:47<1:17:50,  2.01it/s]
 12%|█▏        | 1297/10707 [14:48<1:17:48,  2.02it/s]
 12%|█▏        | 1298/10707 [14:48<1:17:49,  2.02it/s]
 12%|█▏        | 1299/10707 [14:49<1:17:48,  2.02it/s]
 12%|█▏        | 1300/10707 [14:49<1:17:45,  2.02it/s]{'loss': 4.3013, 'grad_norm': 0.27162185311317444, 'learning_rate': 0.0009986071145807462, 'epoch': 0.12}
-
                                                      

 12%|█▏        | 1300/10707 [14:49<1:17:45,  2.02it/s]
 12%|█▏        | 1301/10707 [14:50<1:17:47,  2.02it/s]
 12%|█▏        | 1302/10707 [14:50<1:17:49,  2.01it/s]
 12%|█▏        | 1303/10707 [14:51<1:17:45,  2.02it/s]
 12%|█▏        | 1304/10707 [14:51<1:17:47,  2.01it/s]
 12%|█▏        | 1305/10707 [14:52<1:17:44,  2.02it/s]
 12%|█▏        | 1306/10707 [14:52<1:17:45,  2.02it/s]
 12%|█▏        | 1307/10707 [14:53<1:17:45,  2.01it/s]
 12%|█▏        | 1308/10707 [14:53<1:17:43,  2.02it/s]
 12%|█▏        | 1309/10707 [14:54<1:17:46,  2.01it/s]
 12%|█▏        | 1310/10707 [14:54<1:17:40,  2.02it/s]
 12%|█▏        | 1311/10707 [14:55<1:17:41,  2.02it/s]
 12%|█▏        | 1312/10707 [14:55<1:17:38,  2.02it/s]
 12%|█▏        | 1313/10707 [14:56<1:17:36,  2.02it/s]
 12%|█▏        | 1314/10707 [14:56<1:17:36,  2.02it/s]
 12%|█▏        | 1315/10707 [14:57<1:17:35,  2.02it/s]
 12%|█▏        | 1316/10707 [14:57<1:17:33,  2.02it/s]
 12%|█▏        | 1317/10707 [14:58<1:17:29,  2.02it/s]
 12%|█▏        | 1318/10707 [14:58<1:17:32,  2.02it/s]
 12%|█▏        | 1319/10707 [14:59<1:17:32,  2.02it/s]
 12%|█▏        | 1320/10707 [14:59<1:17:32,  2.02it/s]
 12%|█▏        | 1321/10707 [15:00<1:17:29,  2.02it/s]
 12%|█▏        | 1322/10707 [15:00<1:17:35,  2.02it/s]
 12%|█▏        | 1323/10707 [15:01<1:17:36,  2.02it/s]
 12%|█▏        | 1324/10707 [15:01<1:17:34,  2.02it/s]
 12%|█▏        | 1325/10707 [15:02<1:17:34,  2.02it/s]
                                                      {'loss': 4.2797, 'grad_norm': 0.2555503845214844, 'learning_rate': 0.0009982865737944613, 'epoch': 0.12}
-

 12%|█▏        | 1325/10707 [15:02<1:17:34,  2.02it/s]
 12%|█▏        | 1326/10707 [15:02<1:17:34,  2.02it/s]
 12%|█▏        | 1327/10707 [15:03<1:17:35,  2.01it/s]
 12%|█▏        | 1328/10707 [15:03<1:17:38,  2.01it/s]
 12%|█▏        | 1329/10707 [15:04<1:17:33,  2.02it/s]
 12%|█▏        | 1330/10707 [15:04<1:17:39,  2.01it/s]
 12%|█▏        | 1331/10707 [15:05<1:17:33,  2.01it/s]
 12%|█▏        | 1332/10707 [15:05<1:17:29,  2.02it/s]
 12%|█▏        | 1333/10707 [15:06<1:17:23,  2.02it/s]
 12%|█▏        | 1334/10707 [15:06<1:17:29,  2.02it/s]
 12%|█▏        | 1335/10707 [15:07<1:17:31,  2.01it/s]
 12%|█▏        | 1336/10707 [15:07<1:17:25,  2.02it/s]
 12%|█▏        | 1337/10707 [15:08<1:17:32,  2.01it/s]
 12%|█▏        | 1338/10707 [15:08<1:17:29,  2.02it/s]
 13%|█▎        | 1339/10707 [15:09<1:17:29,  2.02it/s]
 13%|█▎        | 1340/10707 [15:09<1:17:29,  2.01it/s]
 13%|█▎        | 1341/10707 [15:10<1:17:25,  2.02it/s]
 13%|█▎        | 1342/10707 [15:10<1:17:24,  2.02it/s]
 13%|█▎        | 1343/10707 [15:11<1:17:18,  2.02it/s]
 13%|█▎        | 1344/10707 [15:11<1:17:24,  2.02it/s]
 13%|█▎        | 1345/10707 [15:12<1:17:20,  2.02it/s]
 13%|█▎        | 1346/10707 [15:12<1:17:25,  2.02it/s]
 13%|█▎        | 1347/10707 [15:13<1:17:18,  2.02it/s]
 13%|█▎        | 1348/10707 [15:13<1:17:22,  2.02it/s]
 13%|█▎        | 1349/10707 [15:14<1:17:20,  2.02it/s]
 13%|█▎        | 1350/10707 [15:14<1:17:17,  2.02it/s]{'loss': 4.2794, 'grad_norm': 0.27959683537483215, 'learning_rate': 0.000997932930343011, 'epoch': 0.13}
-
                                                      

 13%|█▎        | 1350/10707 [15:14<1:17:17,  2.02it/s]
 13%|█▎        | 1351/10707 [15:15<1:17:25,  2.01it/s]
 13%|█▎        | 1352/10707 [15:15<1:17:22,  2.02it/s]
 13%|█▎        | 1353/10707 [15:16<1:17:19,  2.02it/s]
 13%|█▎        | 1354/10707 [15:16<1:17:16,  2.02it/s]
 13%|█▎        | 1355/10707 [15:17<1:17:21,  2.01it/s]
 13%|█▎        | 1356/10707 [15:17<1:17:17,  2.02it/s]
 13%|█▎        | 1357/10707 [15:18<1:17:23,  2.01it/s]
 13%|█▎        | 1358/10707 [15:18<1:17:22,  2.01it/s]
 13%|█▎        | 1359/10707 [15:19<1:17:17,  2.02it/s]
 13%|█▎        | 1360/10707 [15:19<1:17:19,  2.01it/s]
 13%|█▎        | 1361/10707 [15:20<1:17:13,  2.02it/s]
 13%|█▎        | 1362/10707 [15:20<1:17:14,  2.02it/s]
 13%|█▎        | 1363/10707 [15:21<1:17:13,  2.02it/s]
 13%|█▎        | 1364/10707 [15:21<1:17:14,  2.02it/s]
 13%|█▎        | 1365/10707 [15:22<1:17:16,  2.01it/s]
 13%|█▎        | 1366/10707 [15:22<1:17:20,  2.01it/s]
 13%|█▎        | 1367/10707 [15:23<1:17:17,  2.01it/s]
 13%|█▎        | 1368/10707 [15:23<1:17:18,  2.01it/s]
 13%|█▎        | 1369/10707 [15:24<1:17:23,  2.01it/s]
 13%|█▎        | 1370/10707 [15:24<1:17:22,  2.01it/s]
 13%|█▎        | 1371/10707 [15:24<1:17:20,  2.01it/s]
 13%|█▎        | 1372/10707 [15:25<1:17:14,  2.01it/s]
 13%|█▎        | 1373/10707 [15:25<1:17:16,  2.01it/s]
 13%|█▎        | 1374/10707 [15:26<1:17:13,  2.01it/s]
 13%|█▎        | 1375/10707 [15:26<1:17:15,  2.01it/s]{'loss': 4.2582, 'grad_norm': 0.2557609975337982, 'learning_rate': 0.0009975462077199857, 'epoch': 0.13}
-
                                                      

 13%|█▎        | 1375/10707 [15:26<1:17:15,  2.01it/s]
 13%|█▎        | 1376/10707 [15:27<1:17:24,  2.01it/s]
 13%|█▎        | 1377/10707 [15:27<1:17:23,  2.01it/s]
 13%|█▎        | 1378/10707 [15:28<1:17:20,  2.01it/s]
 13%|█▎        | 1379/10707 [15:28<1:17:21,  2.01it/s]
 13%|█▎        | 1380/10707 [15:29<1:17:14,  2.01it/s]
 13%|█▎        | 1381/10707 [15:29<1:17:13,  2.01it/s]
 13%|█▎        | 1382/10707 [15:30<1:17:10,  2.01it/s]
 13%|█▎        | 1383/10707 [15:30<1:17:07,  2.01it/s]
 13%|█▎        | 1384/10707 [15:31<1:17:02,  2.02it/s]
 13%|█▎        | 1385/10707 [15:31<1:17:04,  2.02it/s]
 13%|█▎        | 1386/10707 [15:32<1:17:05,  2.02it/s]
 13%|█▎        | 1387/10707 [15:32<1:17:03,  2.02it/s]
 13%|█▎        | 1388/10707 [15:33<1:17:06,  2.01it/s]
 13%|█▎        | 1389/10707 [15:33<1:17:03,  2.02it/s]
 13%|█▎        | 1390/10707 [15:34<1:17:01,  2.02it/s]
 13%|█▎        | 1391/10707 [15:34<1:17:06,  2.01it/s]
 13%|█▎        | 1392/10707 [15:35<1:17:09,  2.01it/s]
 13%|█▎        | 1393/10707 [15:35<1:17:08,  2.01it/s]
 13%|█▎        | 1394/10707 [15:36<1:17:05,  2.01it/s]
 13%|█▎        | 1395/10707 [15:36<1:17:03,  2.01it/s]
 13%|█▎        | 1396/10707 [15:37<1:17:10,  2.01it/s]
 13%|█▎        | 1397/10707 [15:37<1:17:07,  2.01it/s]
 13%|█▎        | 1398/10707 [15:38<1:17:07,  2.01it/s]
 13%|█▎        | 1399/10707 [15:38<1:17:03,  2.01it/s]
 13%|█▎        | 1400/10707 [15:39<1:16:58,  2.02it/s]{'loss': 4.2493, 'grad_norm': 0.2693474590778351, 'learning_rate': 0.0009971264316165243, 'epoch': 0.13}
-
                                                      

 13%|█▎        | 1400/10707 [15:39<1:16:58,  2.02it/s]
 13%|█▎        | 1401/10707 [15:39<1:17:07,  2.01it/s]
 13%|█▎        | 1402/10707 [15:40<1:17:03,  2.01it/s]
 13%|█▎        | 1403/10707 [15:40<1:17:02,  2.01it/s]
 13%|█▎        | 1404/10707 [15:41<1:17:01,  2.01it/s]
 13%|█▎        | 1405/10707 [15:41<1:17:00,  2.01it/s]
 13%|█▎        | 1406/10707 [15:42<1:16:53,  2.02it/s]
 13%|█▎        | 1407/10707 [15:42<1:16:54,  2.02it/s]
 13%|█▎        | 1408/10707 [15:43<1:16:50,  2.02it/s]
 13%|█▎        | 1409/10707 [15:43<1:16:54,  2.02it/s]
 13%|█▎        | 1410/10707 [15:44<1:16:52,  2.02it/s]
 13%|█▎        | 1411/10707 [15:44<1:16:58,  2.01it/s]
 13%|█▎        | 1412/10707 [15:45<1:16:50,  2.02it/s]
 13%|█▎        | 1413/10707 [15:45<1:16:52,  2.02it/s]
 13%|█▎        | 1414/10707 [15:46<1:16:48,  2.02it/s]
 13%|█▎        | 1415/10707 [15:46<1:16:51,  2.01it/s]
 13%|█▎        | 1416/10707 [15:47<1:16:47,  2.02it/s]
 13%|█▎        | 1417/10707 [15:47<1:16:51,  2.01it/s]
 13%|█▎        | 1418/10707 [15:48<1:16:56,  2.01it/s]
 13%|█▎        | 1419/10707 [15:48<1:16:49,  2.01it/s]
 13%|█▎        | 1420/10707 [15:49<1:16:52,  2.01it/s]
 13%|█▎        | 1421/10707 [15:49<1:16:48,  2.01it/s]
 13%|█▎        | 1422/10707 [15:50<1:16:49,  2.01it/s]
 13%|█▎        | 1423/10707 [15:50<1:16:52,  2.01it/s]
 13%|█▎        | 1424/10707 [15:51<1:16:51,  2.01it/s]
 13%|█▎        | 1425/10707 [15:51<1:16:53,  2.01it/s]{'loss': 4.2238, 'grad_norm': 0.2845945656299591, 'learning_rate': 0.000996673629919607, 'epoch': 0.13}
-
                                                      

 13%|█▎        | 1425/10707 [15:51<1:16:53,  2.01it/s]
 13%|█▎        | 1426/10707 [15:52<1:17:00,  2.01it/s]
 13%|█▎        | 1427/10707 [15:52<1:16:53,  2.01it/s]
 13%|█▎        | 1428/10707 [15:53<1:16:52,  2.01it/s]
 13%|█▎        | 1429/10707 [15:53<1:16:48,  2.01it/s]
 13%|█▎        | 1430/10707 [15:54<1:16:47,  2.01it/s]
 13%|█▎        | 1431/10707 [15:54<1:16:44,  2.01it/s]
 13%|█▎        | 1432/10707 [15:55<1:16:45,  2.01it/s]
 13%|█▎        | 1433/10707 [15:55<1:16:45,  2.01it/s]
 13%|█▎        | 1434/10707 [15:56<1:16:39,  2.02it/s]
 13%|█▎        | 1435/10707 [15:56<1:16:47,  2.01it/s]
 13%|█▎        | 1436/10707 [15:57<1:16:39,  2.02it/s]
 13%|█▎        | 1437/10707 [15:57<1:16:39,  2.02it/s]
 13%|█▎        | 1438/10707 [15:58<1:16:34,  2.02it/s]
 13%|█▎        | 1439/10707 [15:58<1:16:39,  2.01it/s]
 13%|█▎        | 1440/10707 [15:59<1:16:34,  2.02it/s]
 13%|█▎        | 1441/10707 [15:59<1:16:39,  2.01it/s]
 13%|█▎        | 1442/10707 [16:00<1:16:42,  2.01it/s]
 13%|█▎        | 1443/10707 [16:00<1:16:36,  2.02it/s]
 13%|█▎        | 1444/10707 [16:01<1:16:40,  2.01it/s]
 13%|█▎        | 1445/10707 [16:01<1:16:40,  2.01it/s]
 14%|█▎        | 1446/10707 [16:02<1:16:39,  2.01it/s]
 14%|█▎        | 1447/10707 [16:02<1:16:40,  2.01it/s]
 14%|█▎        | 1448/10707 [16:03<1:16:32,  2.02it/s]
 14%|█▎        | 1449/10707 [16:03<1:16:31,  2.02it/s]
 14%|█▎        | 1450/10707 [16:04<1:16:29,  2.02it/s]
                                                      
{'loss': 4.2288, 'grad_norm': 0.28249067068099976, 'learning_rate': 0.0009961878327102026, 'epoch': 0.14}
-
 14%|█▎        | 1450/10707 [16:04<1:16:29,  2.02it/s]
 14%|█▎        | 1451/10707 [16:04<1:16:37,  2.01it/s]
 14%|█▎        | 1452/10707 [16:05<1:16:42,  2.01it/s]
 14%|█▎        | 1453/10707 [16:05<1:16:33,  2.01it/s]
 14%|█▎        | 1454/10707 [16:06<1:16:34,  2.01it/s]
 14%|█▎        | 1455/10707 [16:06<1:16:37,  2.01it/s]
 14%|█▎        | 1456/10707 [16:07<1:16:30,  2.02it/s]
 14%|█▎        | 1457/10707 [16:07<1:16:38,  2.01it/s]
 14%|█▎        | 1458/10707 [16:08<1:16:31,  2.01it/s]
 14%|█▎        | 1459/10707 [16:08<1:16:34,  2.01it/s]
 14%|█▎        | 1460/10707 [16:09<1:16:26,  2.02it/s]
 14%|█▎        | 1461/10707 [16:09<1:16:28,  2.01it/s]
 14%|█▎        | 1462/10707 [16:10<1:16:26,  2.02it/s]
 14%|█▎        | 1463/10707 [16:10<1:16:20,  2.02it/s]
 14%|█▎        | 1464/10707 [16:11<1:16:22,  2.02it/s]
 14%|█▎        | 1465/10707 [16:11<1:16:17,  2.02it/s]
 14%|█▎        | 1466/10707 [16:12<1:16:17,  2.02it/s]
 14%|█▎        | 1467/10707 [16:12<1:16:14,  2.02it/s]
 14%|█▎        | 1468/10707 [16:13<1:16:20,  2.02it/s]
 14%|█▎        | 1469/10707 [16:13<1:16:18,  2.02it/s]
 14%|█▎        | 1470/10707 [16:14<1:16:20,  2.02it/s]
 14%|█▎        | 1471/10707 [16:14<1:16:19,  2.02it/s]
 14%|█▎        | 1472/10707 [16:15<1:16:17,  2.02it/s]
 14%|█▍        | 1473/10707 [16:15<1:16:19,  2.02it/s]
 14%|█▍        | 1474/10707 [16:16<1:16:19,  2.02it/s]
 14%|█▍        | 1475/10707 [16:16<1:16:22,  2.01it/s]{'loss': 4.2226, 'grad_norm': 0.25083908438682556, 'learning_rate': 0.0009956690722612706, 'epoch': 0.14}
-
                                                      

 14%|█▍        | 1475/10707 [16:16<1:16:22,  2.01it/s]
 14%|█▍        | 1476/10707 [16:17<1:16:26,  2.01it/s]
 14%|█▍        | 1477/10707 [16:17<1:16:20,  2.02it/s]
 14%|█▍        | 1478/10707 [16:18<1:16:20,  2.01it/s]
 14%|█▍        | 1479/10707 [16:18<1:16:15,  2.02it/s]
 14%|█▍        | 1480/10707 [16:19<1:16:14,  2.02it/s]
 14%|█▍        | 1481/10707 [16:19<1:16:12,  2.02it/s]
 14%|█▍        | 1482/10707 [16:20<1:16:15,  2.02it/s]
 14%|█▍        | 1483/10707 [16:20<1:16:17,  2.02it/s]
 14%|█▍        | 1484/10707 [16:21<1:16:21,  2.01it/s]
 14%|█▍        | 1485/10707 [16:21<1:16:18,  2.01it/s]
 14%|█▍        | 1486/10707 [16:22<1:16:18,  2.01it/s]
 14%|█▍        | 1487/10707 [16:22<1:16:19,  2.01it/s]
 14%|█▍        | 1488/10707 [16:23<1:16:25,  2.01it/s]
 14%|█▍        | 1489/10707 [16:23<1:16:22,  2.01it/s]
 14%|█▍        | 1490/10707 [16:24<1:16:23,  2.01it/s]
 14%|█▍        | 1491/10707 [16:24<1:16:19,  2.01it/s]
 14%|█▍        | 1492/10707 [16:25<1:16:25,  2.01it/s]
 14%|█▍        | 1493/10707 [16:25<1:16:27,  2.01it/s]
 14%|█▍        | 1494/10707 [16:26<1:16:24,  2.01it/s]
 14%|█▍        | 1495/10707 [16:26<1:16:18,  2.01it/s]
 14%|█▍        | 1496/10707 [16:27<1:16:20,  2.01it/s]
 14%|█▍        | 1497/10707 [16:27<1:16:25,  2.01it/s]
 14%|█▍        | 1498/10707 [16:28<1:16:22,  2.01it/s]
 14%|█▍        | 1499/10707 [16:28<1:16:17,  2.01it/s]
 14%|█▍        | 1500/10707 [16:29<1:16:13,  2.01it/s]{'loss': 4.2044, 'grad_norm': 0.24389806389808655, 'learning_rate': 0.0009951173830356167, 'epoch': 0.14}
-
                                                      

 14%|█▍        | 1500/10707 [16:29<1:16:13,  2.01it/s]
 14%|█▍        | 1501/10707 [16:29<1:16:21,  2.01it/s]
 14%|█▍        | 1502/10707 [16:30<1:16:21,  2.01it/s]
 14%|█▍        | 1503/10707 [16:30<1:16:12,  2.01it/s]
 14%|█▍        | 1504/10707 [16:31<1:16:14,  2.01it/s]
 14%|█▍        | 1505/10707 [16:31<1:16:09,  2.01it/s]
 14%|█▍        | 1506/10707 [16:32<1:16:11,  2.01it/s]
 14%|█▍        | 1507/10707 [16:32<1:16:14,  2.01it/s]
 14%|█▍        | 1508/10707 [16:33<1:16:09,  2.01it/s]
 14%|█▍        | 1509/10707 [16:33<1:16:05,  2.01it/s]
 14%|█▍        | 1510/10707 [16:34<1:16:02,  2.02it/s]
 14%|█▍        | 1511/10707 [16:34<1:16:08,  2.01it/s]
 14%|█▍        | 1512/10707 [16:35<1:16:12,  2.01it/s]
 14%|█▍        | 1513/10707 [16:35<1:16:08,  2.01it/s]
 14%|█▍        | 1514/10707 [16:36<1:16:07,  2.01it/s]
 14%|█▍        | 1515/10707 [16:36<1:16:09,  2.01it/s]
 14%|█▍        | 1516/10707 [16:36<1:16:07,  2.01it/s]
 14%|█▍        | 1517/10707 [16:37<1:16:06,  2.01it/s]
 14%|█▍        | 1518/10707 [16:37<1:16:01,  2.01it/s]
 14%|█▍        | 1519/10707 [16:38<1:15:59,  2.02it/s]
 14%|█▍        | 1520/10707 [16:38<1:15:57,  2.02it/s]
 14%|█▍        | 1521/10707 [16:39<1:15:58,  2.01it/s]
 14%|█▍        | 1522/10707 [16:39<1:15:56,  2.02it/s]
 14%|█▍        | 1523/10707 [16:40<1:16:00,  2.01it/s]
 14%|█▍        | 1524/10707 [16:40<1:15:55,  2.02it/s]
 14%|█▍        | 1525/10707 [16:41<1:15:55,  2.02it/s]{'loss': 4.1997, 'grad_norm': 0.2523472309112549, 'learning_rate': 0.0009945328016836038, 'epoch': 0.14}
                                                      
-
 14%|█▍        | 1525/10707 [16:41<1:15:55,  2.02it/s]
 14%|█▍        | 1526/10707 [16:41<1:16:06,  2.01it/s]
 14%|█▍        | 1527/10707 [16:42<1:16:01,  2.01it/s]
 14%|█▍        | 1528/10707 [16:42<1:16:03,  2.01it/s]
 14%|█▍        | 1529/10707 [16:43<1:16:05,  2.01it/s]
 14%|█▍        | 1530/10707 [16:43<1:16:05,  2.01it/s]
 14%|█▍        | 1531/10707 [16:44<1:16:03,  2.01it/s]
 14%|█▍        | 1532/10707 [16:44<1:16:00,  2.01it/s]
 14%|█▍        | 1533/10707 [16:45<1:15:58,  2.01it/s]
 14%|█▍        | 1534/10707 [16:45<1:15:54,  2.01it/s]
 14%|█▍        | 1535/10707 [16:46<1:15:54,  2.01it/s]
 14%|█▍        | 1536/10707 [16:46<1:15:48,  2.02it/s]
 14%|█▍        | 1537/10707 [16:47<1:15:53,  2.01it/s]
 14%|█▍        | 1538/10707 [16:47<1:15:50,  2.02it/s]
 14%|█▍        | 1539/10707 [16:48<1:15:51,  2.01it/s]
 14%|█▍        | 1540/10707 [16:48<1:15:52,  2.01it/s]
 14%|█▍        | 1541/10707 [16:49<1:15:49,  2.01it/s]
 14%|█▍        | 1542/10707 [16:49<1:15:51,  2.01it/s]
 14%|█▍        | 1543/10707 [16:50<1:15:47,  2.02it/s]
 14%|█▍        | 1544/10707 [16:50<1:15:48,  2.01it/s]
 14%|█▍        | 1545/10707 [16:51<1:15:51,  2.01it/s]
 14%|█▍        | 1546/10707 [16:51<1:15:45,  2.02it/s]
 14%|█▍        | 1547/10707 [16:52<1:15:48,  2.01it/s]
 14%|█▍        | 1548/10707 [16:52<1:15:41,  2.02it/s]
 14%|█▍        | 1549/10707 [16:53<1:15:42,  2.02it/s]
 14%|█▍        | 1550/10707 [16:53<1:15:36,  2.02it/s]{'loss': 4.1861, 'grad_norm': 0.24572913348674774, 'learning_rate': 0.0009939153670407172, 'epoch': 0.14}
-
                                                      

 14%|█▍        | 1550/10707 [16:53<1:15:36,  2.02it/s]
 14%|█▍        | 1551/10707 [16:54<1:15:43,  2.02it/s]
 14%|█▍        | 1552/10707 [16:54<1:15:46,  2.01it/s]
 15%|█▍        | 1553/10707 [16:55<1:15:43,  2.01it/s]
 15%|█▍        | 1554/10707 [16:55<1:15:42,  2.02it/s]
 15%|█▍        | 1555/10707 [16:56<1:15:40,  2.02it/s]
 15%|█▍        | 1556/10707 [16:56<1:15:42,  2.01it/s]
 15%|█▍        | 1557/10707 [16:57<1:15:42,  2.01it/s]
 15%|█▍        | 1558/10707 [16:57<1:15:39,  2.02it/s]
 15%|█▍        | 1559/10707 [16:58<1:15:38,  2.02it/s]
 15%|█▍        | 1560/10707 [16:58<1:15:41,  2.01it/s]
 15%|█▍        | 1561/10707 [16:59<1:15:38,  2.02it/s]
 15%|█▍        | 1562/10707 [16:59<1:15:39,  2.01it/s]
 15%|█▍        | 1563/10707 [17:00<1:15:44,  2.01it/s]
 15%|█▍        | 1564/10707 [17:00<1:15:48,  2.01it/s]
 15%|█▍        | 1565/10707 [17:01<1:15:41,  2.01it/s]
 15%|█▍        | 1566/10707 [17:01<1:15:38,  2.01it/s]
 15%|█▍        | 1567/10707 [17:02<1:15:35,  2.02it/s]
 15%|█▍        | 1568/10707 [17:02<1:15:37,  2.01it/s]
 15%|█▍        | 1569/10707 [17:03<1:15:32,  2.02it/s]
 15%|█▍        | 1570/10707 [17:03<1:15:32,  2.02it/s]
 15%|█▍        | 1571/10707 [17:04<1:15:34,  2.01it/s]
 15%|█▍        | 1572/10707 [17:04<1:15:35,  2.01it/s]
 15%|█▍        | 1573/10707 [17:05<1:15:25,  2.02it/s]
 15%|█▍        | 1574/10707 [17:05<1:15:30,  2.02it/s]
 15%|█▍        | 1575/10707 [17:06<1:15:35,  2.01it/s]{'loss': 4.1825, 'grad_norm': 0.232056125998497, 'learning_rate': 0.0009932651201249839, 'epoch': 0.15}
-
                                                      

 15%|█▍        | 1575/10707 [17:06<1:15:35,  2.01it/s]
 15%|█▍        | 1576/10707 [17:06<1:15:35,  2.01it/s]
 15%|█▍        | 1577/10707 [17:07<1:15:34,  2.01it/s]
 15%|█▍        | 1578/10707 [17:07<1:15:32,  2.01it/s]
 15%|█▍        | 1579/10707 [17:08<1:15:27,  2.02it/s]
 15%|█▍        | 1580/10707 [17:08<1:15:29,  2.01it/s]
 15%|█▍        | 1581/10707 [17:09<1:15:27,  2.02it/s]
 15%|█▍        | 1582/10707 [17:09<1:15:25,  2.02it/s]
 15%|█▍        | 1583/10707 [17:10<1:15:16,  2.02it/s]
 15%|█▍        | 1584/10707 [17:10<1:15:24,  2.02it/s]
 15%|█▍        | 1585/10707 [17:11<1:15:20,  2.02it/s]
 15%|█▍        | 1586/10707 [17:11<1:15:21,  2.02it/s]
 15%|█▍        | 1587/10707 [17:12<1:15:14,  2.02it/s]
 15%|█▍        | 1588/10707 [17:12<1:15:22,  2.02it/s]
 15%|█▍        | 1589/10707 [17:13<1:15:18,  2.02it/s]
 15%|█▍        | 1590/10707 [17:13<1:15:18,  2.02it/s]
 15%|█▍        | 1591/10707 [17:14<1:15:21,  2.02it/s]
 15%|█▍        | 1592/10707 [17:14<1:15:28,  2.01it/s]
 15%|█▍        | 1593/10707 [17:15<1:15:27,  2.01it/s]
 15%|█▍        | 1594/10707 [17:15<1:15:22,  2.02it/s]
 15%|█▍        | 1595/10707 [17:16<1:15:19,  2.02it/s]
 15%|█▍        | 1596/10707 [17:16<1:15:15,  2.02it/s]
 15%|█▍        | 1597/10707 [17:17<1:15:12,  2.02it/s]
 15%|█▍        | 1598/10707 [17:17<1:15:14,  2.02it/s]
 15%|█▍        | 1599/10707 [17:18<1:15:10,  2.02it/s]
 15%|█▍        | 1600/10707 [17:18<1:15:14,  2.02it/s]
                                                      
{'loss': 4.1663, 'grad_norm': 0.23800384998321533, 'learning_rate': 0.000992582104134248, 'epoch': 0.15}
-
 15%|█▍        | 1600/10707 [17:18<1:15:14,  2.02it/s]
 15%|█▍        | 1601/10707 [17:19<1:15:15,  2.02it/s]
 15%|█▍        | 1602/10707 [17:19<1:15:21,  2.01it/s]
 15%|█▍        | 1603/10707 [17:20<1:15:21,  2.01it/s]
 15%|█▍        | 1604/10707 [17:20<1:15:16,  2.02it/s]
 15%|█▍        | 1605/10707 [17:21<1:15:11,  2.02it/s]
 15%|█▍        | 1606/10707 [17:21<1:15:11,  2.02it/s]
 15%|█▌        | 1607/10707 [17:22<1:15:11,  2.02it/s]
 15%|█▌        | 1608/10707 [17:22<1:15:11,  2.02it/s]
 15%|█▌        | 1609/10707 [17:23<1:15:13,  2.02it/s]
 15%|█▌        | 1610/10707 [17:23<1:15:09,  2.02it/s]
 15%|█▌        | 1611/10707 [17:24<1:15:13,  2.02it/s]
 15%|█▌        | 1612/10707 [17:24<1:15:13,  2.01it/s]
 15%|█▌        | 1613/10707 [17:25<1:15:16,  2.01it/s]
 15%|█▌        | 1614/10707 [17:25<1:15:16,  2.01it/s]
 15%|█▌        | 1615/10707 [17:26<1:15:14,  2.01it/s]
 15%|█▌        | 1616/10707 [17:26<1:15:18,  2.01it/s]
 15%|█▌        | 1617/10707 [17:27<1:15:13,  2.01it/s]
 15%|█▌        | 1618/10707 [17:27<1:15:10,  2.01it/s]
 15%|█▌        | 1619/10707 [17:28<1:15:07,  2.02it/s]
 15%|█▌        | 1620/10707 [17:28<1:15:06,  2.02it/s]
 15%|█▌        | 1621/10707 [17:29<1:15:07,  2.02it/s]
 15%|█▌        | 1622/10707 [17:29<1:15:02,  2.02it/s]
 15%|█▌        | 1623/10707 [17:30<1:15:05,  2.02it/s]
 15%|█▌        | 1624/10707 [17:30<1:15:00,  2.02it/s]
 15%|█▌        | 1625/10707 [17:31<1:15:02,  2.02it/s]{'loss': 4.1629, 'grad_norm': 0.2743390202522278, 'learning_rate': 0.0009918663644433018, 'epoch': 0.15}
                                                      
-
 15%|█▌        | 1625/10707 [17:31<1:15:02,  2.02it/s]
 15%|█▌        | 1626/10707 [17:31<1:15:10,  2.01it/s]
 15%|█▌        | 1627/10707 [17:32<1:15:14,  2.01it/s]
 15%|█▌        | 1628/10707 [17:32<1:15:11,  2.01it/s]
 15%|█▌        | 1629/10707 [17:33<1:15:07,  2.01it/s]
 15%|█▌        | 1630/10707 [17:33<1:15:06,  2.01it/s]
 15%|█▌        | 1631/10707 [17:34<1:15:08,  2.01it/s]
 15%|█▌        | 1632/10707 [17:34<1:15:05,  2.01it/s]
 15%|█▌        | 1633/10707 [17:35<1:15:03,  2.02it/s]
 15%|█▌        | 1634/10707 [17:35<1:15:02,  2.02it/s]
 15%|█▌        | 1635/10707 [17:36<1:15:01,  2.02it/s]
 15%|█▌        | 1636/10707 [17:36<1:14:58,  2.02it/s]
 15%|█▌        | 1637/10707 [17:37<1:15:01,  2.01it/s]
 15%|█▌        | 1638/10707 [17:37<1:15:00,  2.02it/s]
 15%|█▌        | 1639/10707 [17:38<1:15:00,  2.01it/s]
 15%|█▌        | 1640/10707 [17:38<1:14:55,  2.02it/s]
 15%|█▌        | 1641/10707 [17:39<1:14:56,  2.02it/s]
 15%|█▌        | 1642/10707 [17:39<1:14:53,  2.02it/s]
 15%|█▌        | 1643/10707 [17:40<1:14:57,  2.02it/s]
 15%|█▌        | 1644/10707 [17:40<1:14:52,  2.02it/s]
 15%|█▌        | 1645/10707 [17:41<1:14:51,  2.02it/s]
 15%|█▌        | 1646/10707 [17:41<1:14:52,  2.02it/s]
 15%|█▌        | 1647/10707 [17:42<1:14:56,  2.01it/s]
 15%|█▌        | 1648/10707 [17:42<1:14:57,  2.01it/s]
 15%|█▌        | 1649/10707 [17:42<1:14:56,  2.01it/s]
 15%|█▌        | 1650/10707 [17:43<1:14:55,  2.01it/s]{'loss': 4.1613, 'grad_norm': 0.2371363490819931, 'learning_rate': 0.0009911179486008707, 'epoch': 0.15}
-
                                                      

 15%|█▌        | 1650/10707 [17:43<1:14:55,  2.01it/s]
 15%|█▌        | 1651/10707 [17:43<1:14:56,  2.01it/s]
 15%|█▌        | 1652/10707 [17:44<1:14:56,  2.01it/s]
 15%|█▌        | 1653/10707 [17:44<1:15:00,  2.01it/s]
 15%|█▌        | 1654/10707 [17:45<1:14:55,  2.01it/s]
 15%|█▌        | 1655/10707 [17:45<1:14:56,  2.01it/s]
 15%|█▌        | 1656/10707 [17:46<1:14:50,  2.02it/s]
 15%|█▌        | 1657/10707 [17:46<1:14:50,  2.02it/s]
 15%|█▌        | 1658/10707 [17:47<1:14:51,  2.01it/s]
 15%|█▌        | 1659/10707 [17:47<1:14:49,  2.02it/s]
 16%|█▌        | 1660/10707 [17:48<1:14:49,  2.02it/s]
 16%|█▌        | 1661/10707 [17:48<1:14:49,  2.01it/s]
 16%|█▌        | 1662/10707 [17:49<1:14:50,  2.01it/s]
 16%|█▌        | 1663/10707 [17:49<1:14:49,  2.01it/s]
 16%|█▌        | 1664/10707 [17:50<1:14:50,  2.01it/s]
 16%|█▌        | 1665/10707 [17:50<1:14:50,  2.01it/s]
 16%|█▌        | 1666/10707 [17:51<1:14:50,  2.01it/s]
 16%|█▌        | 1667/10707 [17:51<1:14:53,  2.01it/s]
 16%|█▌        | 1668/10707 [17:52<1:14:48,  2.01it/s]
 16%|█▌        | 1669/10707 [17:52<1:14:45,  2.02it/s]
 16%|█▌        | 1670/10707 [17:53<1:14:41,  2.02it/s]
 16%|█▌        | 1671/10707 [17:53<1:14:40,  2.02it/s]
 16%|█▌        | 1672/10707 [17:54<1:14:40,  2.02it/s]
 16%|█▌        | 1673/10707 [17:54<1:14:37,  2.02it/s]
 16%|█▌        | 1674/10707 [17:55<1:14:43,  2.01it/s]
 16%|█▌        | 1675/10707 [17:55<1:14:41,  2.02it/s]
                                                      
{'loss': 4.1515, 'grad_norm': 0.24666960537433624, 'learning_rate': 0.0009903369063264539, 'epoch': 0.16}
-
 16%|█▌        | 1675/10707 [17:55<1:14:41,  2.02it/s]
 16%|█▌        | 1676/10707 [17:56<1:14:48,  2.01it/s]
 16%|█▌        | 1677/10707 [17:56<1:14:44,  2.01it/s]
 16%|█▌        | 1678/10707 [17:57<1:14:41,  2.01it/s]
 16%|█▌        | 1679/10707 [17:57<1:14:38,  2.02it/s]
 16%|█▌        | 1680/10707 [17:58<1:14:36,  2.02it/s]
 16%|█▌        | 1681/10707 [17:58<1:14:34,  2.02it/s]
 16%|█▌        | 1682/10707 [17:59<1:14:38,  2.02it/s]
 16%|█▌        | 1683/10707 [17:59<1:14:32,  2.02it/s]
 16%|█▌        | 1684/10707 [18:00<1:14:33,  2.02it/s]
 16%|█▌        | 1685/10707 [18:00<1:14:28,  2.02it/s]
 16%|█▌        | 1686/10707 [18:01<1:14:28,  2.02it/s]
 16%|█▌        | 1687/10707 [18:01<1:14:27,  2.02it/s]
 16%|█▌        | 1688/10707 [18:02<1:14:32,  2.02it/s]
 16%|█▌        | 1689/10707 [18:02<1:14:28,  2.02it/s]
 16%|█▌        | 1690/10707 [18:03<1:14:27,  2.02it/s]
 16%|█▌        | 1691/10707 [18:03<1:14:23,  2.02it/s]
 16%|█▌        | 1692/10707 [18:04<1:14:24,  2.02it/s]
 16%|█▌        | 1693/10707 [18:04<1:14:25,  2.02it/s]
 16%|█▌        | 1694/10707 [18:05<1:14:29,  2.02it/s]
 16%|█▌        | 1695/10707 [18:05<1:14:22,  2.02it/s]
 16%|█▌        | 1696/10707 [18:06<1:14:29,  2.02it/s]
 16%|█▌        | 1697/10707 [18:06<1:14:25,  2.02it/s]
 16%|█▌        | 1698/10707 [18:07<1:14:23,  2.02it/s]
 16%|█▌        | 1699/10707 [18:07<1:14:25,  2.02it/s]
 16%|█▌        | 1700/10707 [18:08<1:14:23,  2.02it/s]{'loss': 4.1351, 'grad_norm': 0.2410757839679718, 'learning_rate': 0.000989523289507022, 'epoch': 0.16}
-
                                                      

 16%|█▌        | 1700/10707 [18:08<1:14:23,  2.02it/s]
 16%|█▌        | 1701/10707 [18:08<1:14:33,  2.01it/s]
 16%|█▌        | 1702/10707 [18:09<1:14:31,  2.01it/s]
 16%|█▌        | 1703/10707 [18:09<1:14:30,  2.01it/s]
 16%|█▌        | 1704/10707 [18:10<1:14:29,  2.01it/s]
 16%|█▌        | 1705/10707 [18:10<1:14:24,  2.02it/s]
 16%|█▌        | 1706/10707 [18:11<1:14:23,  2.02it/s]
 16%|█▌        | 1707/10707 [18:11<1:14:19,  2.02it/s]
 16%|█▌        | 1708/10707 [18:12<1:14:17,  2.02it/s]
 16%|█▌        | 1709/10707 [18:12<1:14:18,  2.02it/s]
 16%|█▌        | 1710/10707 [18:13<1:14:21,  2.02it/s]
 16%|█▌        | 1711/10707 [18:13<1:14:16,  2.02it/s]
 16%|█▌        | 1712/10707 [18:14<1:14:16,  2.02it/s]
 16%|█▌        | 1713/10707 [18:14<1:14:16,  2.02it/s]
 16%|█▌        | 1714/10707 [18:15<1:14:23,  2.01it/s]
 16%|█▌        | 1715/10707 [18:15<1:14:21,  2.02it/s]
 16%|█▌        | 1716/10707 [18:16<1:14:22,  2.01it/s]
 16%|█▌        | 1717/10707 [18:16<1:14:19,  2.02it/s]
 16%|█▌        | 1718/10707 [18:17<1:14:21,  2.01it/s]
 16%|█▌        | 1719/10707 [18:17<1:14:17,  2.02it/s]
 16%|█▌        | 1720/10707 [18:18<1:14:18,  2.02it/s]
 16%|█▌        | 1721/10707 [18:18<1:14:20,  2.01it/s]
 16%|█▌        | 1722/10707 [18:19<1:14:19,  2.01it/s]
 16%|█▌        | 1723/10707 [18:19<1:14:15,  2.02it/s]
 16%|█▌        | 1724/10707 [18:20<1:14:18,  2.01it/s]
 16%|█▌        | 1725/10707 [18:20<1:14:10,  2.02it/s]
                                                      {'loss': 4.1279, 'grad_norm': 0.22203852236270905, 'learning_rate': 0.0009886771521935706, 'epoch': 0.16}
-
 16%|█▌        | 1725/10707 [18:20<1:14:10,  2.02it/s]
 16%|█▌        | 1726/10707 [18:21<1:14:16,  2.02it/s]
 16%|█▌        | 1727/10707 [18:21<1:14:15,  2.02it/s]
 16%|█▌        | 1728/10707 [18:22<1:14:11,  2.02it/s]
 16%|█▌        | 1729/10707 [18:22<1:14:08,  2.02it/s]
 16%|█▌        | 1730/10707 [18:23<1:14:07,  2.02it/s]
 16%|█▌        | 1731/10707 [18:23<1:14:08,  2.02it/s]
 16%|█▌        | 1732/10707 [18:24<1:14:03,  2.02it/s]
 16%|█▌        | 1733/10707 [18:24<1:14:13,  2.02it/s]
 16%|█▌        | 1734/10707 [18:25<1:14:04,  2.02it/s]
 16%|█▌        | 1735/10707 [18:25<1:14:06,  2.02it/s]
 16%|█▌        | 1736/10707 [18:26<1:14:04,  2.02it/s]
 16%|█▌        | 1737/10707 [18:26<1:14:03,  2.02it/s]
 16%|█▌        | 1738/10707 [18:27<1:14:01,  2.02it/s]
 16%|█▌        | 1739/10707 [18:27<1:14:02,  2.02it/s]
 16%|█▋        | 1740/10707 [18:28<1:14:01,  2.02it/s]
 16%|█▋        | 1741/10707 [18:28<1:14:13,  2.01it/s]
 16%|█▋        | 1742/10707 [18:29<1:14:14,  2.01it/s]
 16%|█▋        | 1743/10707 [18:29<1:14:07,  2.02it/s]
 16%|█▋        | 1744/10707 [18:30<1:14:08,  2.01it/s]
 16%|█▋        | 1745/10707 [18:30<1:14:02,  2.02it/s]
 16%|█▋        | 1746/10707 [18:31<1:14:05,  2.02it/s]
 16%|█▋        | 1747/10707 [18:31<1:14:07,  2.01it/s]
 16%|█▋        | 1748/10707 [18:32<1:14:03,  2.02it/s]
 16%|█▋        | 1749/10707 [18:32<1:14:07,  2.01it/s]
 16%|█▋        | 1750/10707 [18:33<1:14:01,  2.02it/s]{'loss': 4.1221, 'grad_norm': 0.2390998750925064, 'learning_rate': 0.0009877985505975279, 'epoch': 0.16}
-
                                                      

 16%|█▋        | 1750/10707 [18:33<1:14:01,  2.02it/s]
 16%|█▋        | 1751/10707 [18:33<1:14:08,  2.01it/s]
 16%|█▋        | 1752/10707 [18:34<1:14:08,  2.01it/s]
 16%|█▋        | 1753/10707 [18:34<1:14:08,  2.01it/s]
 16%|█▋        | 1754/10707 [18:35<1:14:06,  2.01it/s]
 16%|█▋        | 1755/10707 [18:35<1:14:02,  2.01it/s]
 16%|█▋        | 1756/10707 [18:36<1:14:06,  2.01it/s]
 16%|█▋        | 1757/10707 [18:36<1:14:04,  2.01it/s]
 16%|█▋        | 1758/10707 [18:37<1:14:00,  2.02it/s]
 16%|█▋        | 1759/10707 [18:37<1:13:56,  2.02it/s]
 16%|█▋        | 1760/10707 [18:38<1:13:58,  2.02it/s]
 16%|█▋        | 1761/10707 [18:38<1:14:04,  2.01it/s]
 16%|█▋        | 1762/10707 [18:39<1:25:40,  1.74it/s]
 16%|█▋        | 1763/10707 [18:39<1:22:08,  1.81it/s]
 16%|█▋        | 1764/10707 [18:40<1:19:40,  1.87it/s]
 16%|█▋        | 1765/10707 [18:40<1:17:58,  1.91it/s]
 16%|█▋        | 1766/10707 [18:41<1:16:44,  1.94it/s]
 17%|█▋        | 1767/10707 [18:41<1:15:52,  1.96it/s]
 17%|█▋        | 1768/10707 [18:42<1:15:17,  1.98it/s]
 17%|█▋        | 1769/10707 [18:43<1:26:58,  1.71it/s]
 17%|█▋        | 1770/10707 [18:43<1:22:59,  1.79it/s]
 17%|█▋        | 1771/10707 [18:44<1:20:15,  1.86it/s]
 17%|█▋        | 1772/10707 [18:44<1:18:19,  1.90it/s]
 17%|█▋        | 1773/10707 [18:45<1:16:57,  1.93it/s]
 17%|█▋        | 1774/10707 [18:45<1:16:00,  1.96it/s]
 17%|█▋        | 1775/10707 [18:46<1:15:20,  1.98it/s]
                                                      
{'loss': 4.126, 'grad_norm': 0.24157662689685822, 'learning_rate': 0.0009868875430870216, 'epoch': 0.17}
-
 17%|█▋        | 1775/10707 [18:46<1:15:20,  1.98it/s]
 17%|█▋        | 1776/10707 [18:46<1:14:57,  1.99it/s]
 17%|█▋        | 1777/10707 [18:47<1:14:38,  1.99it/s]
 17%|█▋        | 1778/10707 [18:47<1:14:23,  2.00it/s]
 17%|█▋        | 1779/10707 [18:48<1:14:10,  2.01it/s]
 17%|█▋        | 1780/10707 [18:48<1:14:04,  2.01it/s]
 17%|█▋        | 1781/10707 [18:49<1:13:57,  2.01it/s]
 17%|█▋        | 1782/10707 [18:49<1:13:53,  2.01it/s]
 17%|█▋        | 1783/10707 [18:50<1:13:50,  2.01it/s]
 17%|█▋        | 1784/10707 [18:50<1:13:46,  2.02it/s]
 17%|█▋        | 1785/10707 [18:50<1:13:42,  2.02it/s]
 17%|█▋        | 1786/10707 [18:51<1:13:41,  2.02it/s]
 17%|█▋        | 1787/10707 [18:51<1:13:41,  2.02it/s]
 17%|█▋        | 1788/10707 [18:52<1:13:42,  2.02it/s]
 17%|█▋        | 1789/10707 [18:52<1:13:44,  2.02it/s]
 17%|█▋        | 1790/10707 [18:53<1:13:46,  2.01it/s]
 17%|█▋        | 1791/10707 [18:53<1:13:47,  2.01it/s]
 17%|█▋        | 1792/10707 [18:54<1:13:39,  2.02it/s]
 17%|█▋        | 1793/10707 [18:54<1:13:41,  2.02it/s]
 17%|█▋        | 1794/10707 [18:55<1:13:40,  2.02it/s]
 17%|█▋        | 1795/10707 [18:55<1:13:41,  2.02it/s]
 17%|█▋        | 1796/10707 [18:56<1:13:33,  2.02it/s]
 17%|█▋        | 1797/10707 [18:56<1:13:33,  2.02it/s]
 17%|█▋        | 1798/10707 [18:57<1:13:33,  2.02it/s]
 17%|█▋        | 1799/10707 [18:57<1:13:33,  2.02it/s]
 17%|█▋        | 1800/10707 [18:58<1:13:39,  2.02it/s]{'loss': 4.111, 'grad_norm': 0.2507055401802063, 'learning_rate': 0.0009859441901830016, 'epoch': 0.17}
                                                      
-
 17%|█▋        | 1800/10707 [18:58<1:13:39,  2.02it/s]
 17%|█▋        | 1801/10707 [18:58<1:13:40,  2.01it/s]
 17%|█▋        | 1802/10707 [18:59<1:13:42,  2.01it/s]
 17%|█▋        | 1803/10707 [18:59<1:13:33,  2.02it/s]
 17%|█▋        | 1804/10707 [19:00<1:13:37,  2.02it/s]
 17%|█▋        | 1805/10707 [19:00<1:13:34,  2.02it/s]
 17%|█▋        | 1806/10707 [19:01<1:13:33,  2.02it/s]
 17%|█▋        | 1807/10707 [19:01<1:13:31,  2.02it/s]
 17%|█▋        | 1808/10707 [19:02<1:13:32,  2.02it/s]
 17%|█▋        | 1809/10707 [19:02<1:13:31,  2.02it/s]
 17%|█▋        | 1810/10707 [19:03<1:13:31,  2.02it/s]
 17%|█▋        | 1811/10707 [19:03<1:13:26,  2.02it/s]
 17%|█▋        | 1812/10707 [19:04<1:13:26,  2.02it/s]
 17%|█▋        | 1813/10707 [19:04<1:13:31,  2.02it/s]
 17%|█▋        | 1814/10707 [19:05<1:13:31,  2.02it/s]
 17%|█▋        | 1815/10707 [19:05<1:13:29,  2.02it/s]
 17%|█▋        | 1816/10707 [19:06<1:13:29,  2.02it/s]
 17%|█▋        | 1817/10707 [19:06<1:13:32,  2.01it/s]
 17%|█▋        | 1818/10707 [19:07<1:13:28,  2.02it/s]
 17%|█▋        | 1819/10707 [19:07<1:13:29,  2.02it/s]
 17%|█▋        | 1820/10707 [19:08<1:13:29,  2.02it/s]
 17%|█▋        | 1821/10707 [19:08<1:13:31,  2.01it/s]
 17%|█▋        | 1822/10707 [19:09<1:13:28,  2.02it/s]
 17%|█▋        | 1823/10707 [19:09<1:13:29,  2.01it/s]
 17%|█▋        | 1824/10707 [19:10<1:13:26,  2.02it/s]
 17%|█▋        | 1825/10707 [19:10<1:13:28,  2.01it/s]{'loss': 4.1065, 'grad_norm': 0.22606442868709564, 'learning_rate': 0.0009849685545552183, 'epoch': 0.17}
-
                                                      

 17%|█▋        | 1825/10707 [19:10<1:13:28,  2.01it/s]
 17%|█▋        | 1826/10707 [19:11<1:13:34,  2.01it/s]
 17%|█▋        | 1827/10707 [19:11<1:13:30,  2.01it/s]
 17%|█▋        | 1828/10707 [19:12<1:13:24,  2.02it/s]
 17%|█▋        | 1829/10707 [19:12<1:13:21,  2.02it/s]
 17%|█▋        | 1830/10707 [19:13<1:13:20,  2.02it/s]
 17%|█▋        | 1831/10707 [19:13<1:13:19,  2.02it/s]
 17%|█▋        | 1832/10707 [19:14<1:13:27,  2.01it/s]
 17%|█▋        | 1833/10707 [19:14<1:13:29,  2.01it/s]
 17%|█▋        | 1834/10707 [19:15<1:13:28,  2.01it/s]
 17%|█▋        | 1835/10707 [19:15<1:13:27,  2.01it/s]
 17%|█▋        | 1836/10707 [19:16<1:13:22,  2.02it/s]
 17%|█▋        | 1837/10707 [19:16<1:13:22,  2.01it/s]
 17%|█▋        | 1838/10707 [19:17<1:13:21,  2.02it/s]
 17%|█▋        | 1839/10707 [19:17<1:13:25,  2.01it/s]
 17%|█▋        | 1840/10707 [19:18<1:13:24,  2.01it/s]
 17%|█▋        | 1841/10707 [19:18<1:13:19,  2.02it/s]
 17%|█▋        | 1842/10707 [19:19<1:13:25,  2.01it/s]
 17%|█▋        | 1843/10707 [19:19<1:13:16,  2.02it/s]
 17%|█▋        | 1844/10707 [19:20<1:13:16,  2.02it/s]
 17%|█▋        | 1845/10707 [19:20<1:13:10,  2.02it/s]
 17%|█▋        | 1846/10707 [19:21<1:13:10,  2.02it/s]
 17%|█▋        | 1847/10707 [19:21<1:13:09,  2.02it/s]
 17%|█▋        | 1848/10707 [19:22<1:13:10,  2.02it/s]
 17%|█▋        | 1849/10707 [19:22<1:13:03,  2.02it/s]
 17%|█▋        | 1850/10707 [19:23<1:13:07,  2.02it/s]{'loss': 4.0978, 'grad_norm': 0.23557049036026, 'learning_rate': 0.0009839607010180603, 'epoch': 0.17}
-
                                                      

 17%|█▋        | 1850/10707 [19:23<1:13:07,  2.02it/s]
 17%|█▋        | 1851/10707 [19:23<1:13:11,  2.02it/s]
 17%|█▋        | 1852/10707 [19:24<1:13:16,  2.01it/s]
 17%|█▋        | 1853/10707 [19:24<1:13:10,  2.02it/s]
 17%|█▋        | 1854/10707 [19:25<1:13:09,  2.02it/s]
 17%|█▋        | 1855/10707 [19:25<1:13:08,  2.02it/s]
 17%|█▋        | 1856/10707 [19:26<1:13:08,  2.02it/s]
 17%|█▋        | 1857/10707 [19:26<1:13:09,  2.02it/s]
 17%|█▋        | 1858/10707 [19:27<1:13:07,  2.02it/s]
 17%|█▋        | 1859/10707 [19:27<1:13:06,  2.02it/s]
 17%|█▋        | 1860/10707 [19:28<1:13:05,  2.02it/s]
 17%|█▋        | 1861/10707 [19:28<1:13:05,  2.02it/s]
 17%|█▋        | 1862/10707 [19:29<1:13:04,  2.02it/s]
 17%|█▋        | 1863/10707 [19:29<1:13:02,  2.02it/s]
 17%|█▋        | 1864/10707 [19:30<1:13:03,  2.02it/s]
 17%|█▋        | 1865/10707 [19:30<1:13:00,  2.02it/s]
 17%|█▋        | 1866/10707 [19:31<1:13:02,  2.02it/s]
 17%|█▋        | 1867/10707 [19:31<1:13:03,  2.02it/s]
 17%|█▋        | 1868/10707 [19:32<1:13:02,  2.02it/s]
 17%|█▋        | 1869/10707 [19:32<1:13:04,  2.02it/s]
 17%|█▋        | 1870/10707 [19:33<1:12:59,  2.02it/s]
 17%|█▋        | 1871/10707 [19:33<1:12:58,  2.02it/s]
 17%|█▋        | 1872/10707 [19:34<1:12:53,  2.02it/s]
 17%|█▋        | 1873/10707 [19:34<1:12:56,  2.02it/s]
 18%|█▊        | 1874/10707 [19:35<1:12:51,  2.02it/s]
 18%|█▊        | 1875/10707 [19:35<1:12:50,  2.02it/s]{'loss': 4.0891, 'grad_norm': 0.21315155923366547, 'learning_rate': 0.0009829206965262477, 'epoch': 0.18}
-
                                                      

 18%|█▊        | 1875/10707 [19:35<1:12:50,  2.02it/s]
 18%|█▊        | 1876/10707 [19:36<1:12:56,  2.02it/s]
 18%|█▊        | 1877/10707 [19:36<1:12:58,  2.02it/s]
 18%|█▊        | 1878/10707 [19:37<1:12:57,  2.02it/s]
 18%|█▊        | 1879/10707 [19:37<1:12:54,  2.02it/s]
 18%|█▊        | 1880/10707 [19:38<1:12:52,  2.02it/s]
 18%|█▊        | 1881/10707 [19:38<1:12:51,  2.02it/s]
 18%|█▊        | 1882/10707 [19:39<1:12:50,  2.02it/s]
 18%|█▊        | 1883/10707 [19:39<1:12:49,  2.02it/s]
 18%|█▊        | 1884/10707 [19:40<1:12:55,  2.02it/s]
 18%|█▊        | 1885/10707 [19:40<1:12:52,  2.02it/s]
 18%|█▊        | 1886/10707 [19:41<1:12:52,  2.02it/s]
 18%|█▊        | 1887/10707 [19:41<1:12:48,  2.02it/s]
 18%|█▊        | 1888/10707 [19:42<1:12:49,  2.02it/s]
 18%|█▊        | 1889/10707 [19:42<1:12:44,  2.02it/s]
 18%|█▊        | 1890/10707 [19:43<1:12:50,  2.02it/s]
 18%|█▊        | 1891/10707 [19:43<1:12:49,  2.02it/s]
 18%|█▊        | 1892/10707 [19:44<1:12:54,  2.02it/s]
 18%|█▊        | 1893/10707 [19:44<1:12:50,  2.02it/s]
 18%|█▊        | 1894/10707 [19:45<1:12:51,  2.02it/s]
 18%|█▊        | 1895/10707 [19:45<1:12:49,  2.02it/s]
 18%|█▊        | 1896/10707 [19:46<1:12:46,  2.02it/s]
 18%|█▊        | 1897/10707 [19:46<1:12:45,  2.02it/s]
 18%|█▊        | 1898/10707 [19:47<1:12:48,  2.02it/s]
 18%|█▊        | 1899/10707 [19:47<1:12:45,  2.02it/s]
 18%|█▊        | 1900/10707 [19:48<1:12:48,  2.02it/s]{'loss': 4.0812, 'grad_norm': 0.2221754491329193, 'learning_rate': 0.0009818486101703845, 'epoch': 0.18}
-
                                                      

 18%|█▊        | 1900/10707 [19:48<1:12:48,  2.02it/s]
 18%|█▊        | 1901/10707 [19:48<1:12:48,  2.02it/s]
 18%|█▊        | 1902/10707 [19:49<1:12:48,  2.02it/s]
 18%|█▊        | 1903/10707 [19:49<1:12:49,  2.02it/s]
 18%|█▊        | 1904/10707 [19:50<1:12:45,  2.02it/s]
 18%|█▊        | 1905/10707 [19:50<1:12:48,  2.01it/s]
 18%|█▊        | 1906/10707 [19:50<1:12:44,  2.02it/s]
 18%|█▊        | 1907/10707 [19:51<1:12:45,  2.02it/s]
 18%|█▊        | 1908/10707 [19:51<1:12:42,  2.02it/s]
 18%|█▊        | 1909/10707 [19:52<1:12:44,  2.02it/s]
 18%|█▊        | 1910/10707 [19:52<1:12:41,  2.02it/s]
 18%|█▊        | 1911/10707 [19:53<1:12:39,  2.02it/s]
 18%|█▊        | 1912/10707 [19:53<1:12:36,  2.02it/s]
 18%|█▊        | 1913/10707 [19:54<1:12:35,  2.02it/s]
 18%|█▊        | 1914/10707 [19:54<1:12:36,  2.02it/s]
 18%|█▊        | 1915/10707 [19:55<1:12:37,  2.02it/s]
 18%|█▊        | 1916/10707 [19:55<1:12:38,  2.02it/s]
 18%|█▊        | 1917/10707 [19:56<1:12:33,  2.02it/s]
 18%|█▊        | 1918/10707 [19:56<1:12:38,  2.02it/s]
 18%|█▊        | 1919/10707 [19:57<1:12:35,  2.02it/s]
 18%|█▊        | 1920/10707 [19:57<1:12:39,  2.02it/s]
 18%|█▊        | 1921/10707 [19:58<1:12:35,  2.02it/s]
 18%|█▊        | 1922/10707 [19:58<1:12:37,  2.02it/s]
 18%|█▊        | 1923/10707 [19:59<1:12:35,  2.02it/s]
 18%|█▊        | 1924/10707 [19:59<1:12:34,  2.02it/s]
 18%|█▊        | 1925/10707 [20:00<1:12:34,  2.02it/s]{'loss': 4.0735, 'grad_norm': 0.21324887871742249, 'learning_rate': 0.0009807445131723696, 'epoch': 0.18}
-
                                                      

 18%|█▊        | 1925/10707 [20:00<1:12:34,  2.02it/s]
 18%|█▊        | 1926/10707 [20:00<1:12:37,  2.02it/s]
 18%|█▊        | 1927/10707 [20:01<1:12:38,  2.01it/s]
 18%|█▊        | 1928/10707 [20:01<1:12:29,  2.02it/s]
 18%|█▊        | 1929/10707 [20:02<1:12:33,  2.02it/s]
 18%|█▊        | 1930/10707 [20:02<1:12:29,  2.02it/s]
 18%|█▊        | 1931/10707 [20:03<1:12:34,  2.02it/s]
 18%|█▊        | 1932/10707 [20:03<1:12:32,  2.02it/s]
 18%|█▊        | 1933/10707 [20:04<1:12:29,  2.02it/s]
 18%|█▊        | 1934/10707 [20:04<1:12:27,  2.02it/s]
 18%|█▊        | 1935/10707 [20:05<1:12:30,  2.02it/s]
 18%|█▊        | 1936/10707 [20:05<1:12:32,  2.02it/s]
 18%|█▊        | 1937/10707 [20:06<1:12:28,  2.02it/s]
 18%|█▊        | 1938/10707 [20:06<1:12:31,  2.01it/s]
 18%|█▊        | 1939/10707 [20:07<1:12:27,  2.02it/s]
 18%|█▊        | 1940/10707 [20:07<1:12:23,  2.02it/s]
 18%|█▊        | 1941/10707 [20:08<1:12:22,  2.02it/s]
 18%|█▊        | 1942/10707 [20:08<1:12:26,  2.02it/s]
 18%|█▊        | 1943/10707 [20:09<1:12:20,  2.02it/s]
 18%|█▊        | 1944/10707 [20:09<1:12:20,  2.02it/s]
 18%|█▊        | 1945/10707 [20:10<1:12:17,  2.02it/s]
 18%|█▊        | 1946/10707 [20:10<1:12:20,  2.02it/s]
 18%|█▊        | 1947/10707 [20:11<1:12:15,  2.02it/s]
 18%|█▊        | 1948/10707 [20:11<1:12:24,  2.02it/s]
 18%|█▊        | 1949/10707 [20:12<1:12:18,  2.02it/s]
 18%|█▊        | 1950/10707 [20:12<1:12:17,  2.02it/s]{'loss': 4.077, 'grad_norm': 0.2358422428369522, 'learning_rate': 0.000979608478880663, 'epoch': 0.18}
-
                                                      

 18%|█▊        | 1950/10707 [20:12<1:12:17,  2.02it/s]
 18%|█▊        | 1951/10707 [20:13<1:12:20,  2.02it/s]
 18%|█▊        | 1952/10707 [20:13<1:12:22,  2.02it/s]
 18%|█▊        | 1953/10707 [20:14<1:12:18,  2.02it/s]
 18%|█▊        | 1954/10707 [20:14<1:12:22,  2.02it/s]
 18%|█▊        | 1955/10707 [20:15<1:12:19,  2.02it/s]
 18%|█▊        | 1956/10707 [20:15<1:12:18,  2.02it/s]
 18%|█▊        | 1957/10707 [20:16<1:12:18,  2.02it/s]
 18%|█▊        | 1958/10707 [20:16<1:12:19,  2.02it/s]
 18%|█▊        | 1959/10707 [20:17<1:12:19,  2.02it/s]
 18%|█▊        | 1960/10707 [20:17<1:12:16,  2.02it/s]
 18%|█▊        | 1961/10707 [20:18<1:12:17,  2.02it/s]
 18%|█▊        | 1962/10707 [20:18<1:12:14,  2.02it/s]
 18%|█▊        | 1963/10707 [20:19<1:12:13,  2.02it/s]
 18%|█▊        | 1964/10707 [20:19<1:12:10,  2.02it/s]
 18%|█▊        | 1965/10707 [20:20<1:12:10,  2.02it/s]
 18%|█▊        | 1966/10707 [20:20<1:12:09,  2.02it/s]
 18%|█▊        | 1967/10707 [20:21<1:12:11,  2.02it/s]
 18%|█▊        | 1968/10707 [20:21<1:12:12,  2.02it/s]
 18%|█▊        | 1969/10707 [20:22<1:12:12,  2.02it/s]
 18%|█▊        | 1970/10707 [20:22<1:12:08,  2.02it/s]
 18%|█▊        | 1971/10707 [20:23<1:12:12,  2.02it/s]
 18%|█▊        | 1972/10707 [20:23<1:12:09,  2.02it/s]
 18%|█▊        | 1973/10707 [20:24<1:12:10,  2.02it/s]
 18%|█▊        | 1974/10707 [20:24<1:12:05,  2.02it/s]
 18%|█▊        | 1975/10707 [20:25<1:12:06,  2.02it/s]{'loss': 4.0642, 'grad_norm': 0.22556467354297638, 'learning_rate': 0.0009784405827654163, 'epoch': 0.18}
-
                                                      

 18%|█▊        | 1975/10707 [20:25<1:12:06,  2.02it/s]
 18%|█▊        | 1976/10707 [20:25<1:12:11,  2.02it/s]
 18%|█▊        | 1977/10707 [20:26<1:12:09,  2.02it/s]
 18%|█▊        | 1978/10707 [20:26<1:12:15,  2.01it/s]
 18%|█▊        | 1979/10707 [20:27<1:12:15,  2.01it/s]
 18%|█▊        | 1980/10707 [20:27<1:12:11,  2.01it/s]
 19%|█▊        | 1981/10707 [20:28<1:12:10,  2.01it/s]
 19%|█▊        | 1982/10707 [20:28<1:12:06,  2.02it/s]
 19%|█▊        | 1983/10707 [20:29<1:12:09,  2.01it/s]
 19%|█▊        | 1984/10707 [20:29<1:12:01,  2.02it/s]
 19%|█▊        | 1985/10707 [20:30<1:12:07,  2.02it/s]
 19%|█▊        | 1986/10707 [20:30<1:12:02,  2.02it/s]
 19%|█▊        | 1987/10707 [20:31<1:12:03,  2.02it/s]
 19%|█▊        | 1988/10707 [20:31<1:12:01,  2.02it/s]
 19%|█▊        | 1989/10707 [20:32<1:12:00,  2.02it/s]
 19%|█▊        | 1990/10707 [20:32<1:12:01,  2.02it/s]
 19%|█▊        | 1991/10707 [20:33<1:12:05,  2.01it/s]
 19%|█▊        | 1992/10707 [20:33<1:12:06,  2.01it/s]
 19%|█▊        | 1993/10707 [20:34<1:12:05,  2.01it/s]
 19%|█▊        | 1994/10707 [20:34<1:12:02,  2.02it/s]
 19%|█▊        | 1995/10707 [20:35<1:12:00,  2.02it/s]
 19%|█▊        | 1996/10707 [20:35<1:11:55,  2.02it/s]
 19%|█▊        | 1997/10707 [20:36<1:12:02,  2.02it/s]
 19%|█▊        | 1998/10707 [20:36<1:11:57,  2.02it/s]
 19%|█▊        | 1999/10707 [20:37<1:11:59,  2.02it/s]
 19%|█▊        | 2000/10707 [20:37<1:11:54,  2.02it/s]{'loss': 4.0593, 'grad_norm': 0.23933124542236328, 'learning_rate': 0.0009772409024134555, 'epoch': 0.19}
                                                      
-
 19%|█▊        | 2000/10707 [20:37<1:11:54,  2.02it/s]
 19%|█▊        | 2001/10707 [20:38<1:12:11,  2.01it/s]
 19%|█▊        | 2002/10707 [20:38<1:12:13,  2.01it/s]
 19%|█▊        | 2003/10707 [20:39<1:12:07,  2.01it/s]
 19%|█▊        | 2004/10707 [20:39<1:12:03,  2.01it/s]
 19%|█▊        | 2005/10707 [20:40<1:12:05,  2.01it/s]
 19%|█▊        | 2006/10707 [20:40<1:11:59,  2.01it/s]
 19%|█▊        | 2007/10707 [20:41<1:11:57,  2.01it/s]
 19%|█▉        | 2008/10707 [20:41<1:11:50,  2.02it/s]
 19%|█▉        | 2009/10707 [20:42<1:11:56,  2.02it/s]
 19%|█▉        | 2010/10707 [20:42<1:11:51,  2.02it/s]
 19%|█▉        | 2011/10707 [20:43<1:11:50,  2.02it/s]
 19%|█▉        | 2012/10707 [20:43<1:11:52,  2.02it/s]
 19%|█▉        | 2013/10707 [20:44<1:11:50,  2.02it/s]
 19%|█▉        | 2014/10707 [20:44<1:11:49,  2.02it/s]
 19%|█▉        | 2015/10707 [20:45<1:11:50,  2.02it/s]
 19%|█▉        | 2016/10707 [20:45<1:11:49,  2.02it/s]
 19%|█▉        | 2017/10707 [20:46<1:11:48,  2.02it/s]
 19%|█▉        | 2018/10707 [20:46<1:11:44,  2.02it/s]
 19%|█▉        | 2019/10707 [20:47<1:11:42,  2.02it/s]
 19%|█▉        | 2020/10707 [20:47<1:11:43,  2.02it/s]
 19%|█▉        | 2021/10707 [20:48<1:11:41,  2.02it/s]
 19%|█▉        | 2022/10707 [20:48<1:11:42,  2.02it/s]
 19%|█▉        | 2023/10707 [20:49<1:11:46,  2.02it/s]
 19%|█▉        | 2024/10707 [20:49<1:11:46,  2.02it/s]
 19%|█▉        | 2025/10707 [20:50<1:11:47,  2.02it/s]{'loss': 4.0504, 'grad_norm': 0.23273514211177826, 'learning_rate': 0.0009760095175231292, 'epoch': 0.19}
-
                                                      

 19%|█▉        | 2025/10707 [20:50<1:11:47,  2.02it/s]
 19%|█▉        | 2026/10707 [20:50<1:11:52,  2.01it/s]
 19%|█▉        | 2027/10707 [20:50<1:11:48,  2.01it/s]
 19%|█▉        | 2028/10707 [20:51<1:11:51,  2.01it/s]
 19%|█▉        | 2029/10707 [20:51<1:11:49,  2.01it/s]
 19%|█▉        | 2030/10707 [20:52<1:11:47,  2.01it/s]
 19%|█▉        | 2031/10707 [20:52<1:11:44,  2.02it/s]
 19%|█▉        | 2032/10707 [20:53<1:11:42,  2.02it/s]
 19%|█▉        | 2033/10707 [20:53<1:11:39,  2.02it/s]
 19%|█▉        | 2034/10707 [20:54<1:11:37,  2.02it/s]
 19%|█▉        | 2035/10707 [20:54<1:11:42,  2.02it/s]
 19%|█▉        | 2036/10707 [20:55<1:11:37,  2.02it/s]
 19%|█▉        | 2037/10707 [20:55<1:11:39,  2.02it/s]
 19%|█▉        | 2038/10707 [20:56<1:11:34,  2.02it/s]
 19%|█▉        | 2039/10707 [20:56<1:11:37,  2.02it/s]
 19%|█▉        | 2040/10707 [20:57<1:11:31,  2.02it/s]
 19%|█▉        | 2041/10707 [20:57<1:11:31,  2.02it/s]
 19%|█▉        | 2042/10707 [20:58<1:11:31,  2.02it/s]
 19%|█▉        | 2043/10707 [20:58<1:11:33,  2.02it/s]
 19%|█▉        | 2044/10707 [20:59<1:11:34,  2.02it/s]
 19%|█▉        | 2045/10707 [20:59<1:11:33,  2.02it/s]
 19%|█▉        | 2046/10707 [21:00<1:11:32,  2.02it/s]
 19%|█▉        | 2047/10707 [21:00<1:11:33,  2.02it/s]
 19%|█▉        | 2048/10707 [21:01<1:11:33,  2.02it/s]
 19%|█▉        | 2049/10707 [21:01<1:11:32,  2.02it/s]
 19%|█▉        | 2050/10707 [21:02<1:11:33,  2.02it/s]{'loss': 4.0525, 'grad_norm': 0.22408126294612885, 'learning_rate': 0.0009747465098990131, 'epoch': 0.19}
-
                                                      

 19%|█▉        | 2050/10707 [21:02<1:11:33,  2.02it/s]
 19%|█▉        | 2051/10707 [21:02<1:11:35,  2.02it/s]
 19%|█▉        | 2052/10707 [21:03<1:11:38,  2.01it/s]
 19%|█▉        | 2053/10707 [21:03<1:11:34,  2.02it/s]
 19%|█▉        | 2054/10707 [21:04<1:11:32,  2.02it/s]
 19%|█▉        | 2055/10707 [21:04<1:11:33,  2.02it/s]
 19%|█▉        | 2056/10707 [21:05<1:11:30,  2.02it/s]
 19%|█▉        | 2057/10707 [21:05<1:11:32,  2.02it/s]
 19%|█▉        | 2058/10707 [21:06<1:11:25,  2.02it/s]
 19%|█▉        | 2059/10707 [21:06<1:11:34,  2.01it/s]
 19%|█▉        | 2060/10707 [21:07<1:11:27,  2.02it/s]
 19%|█▉        | 2061/10707 [21:07<1:11:26,  2.02it/s]
 19%|█▉        | 2062/10707 [21:08<1:11:26,  2.02it/s]
 19%|█▉        | 2063/10707 [21:08<1:11:27,  2.02it/s]
 19%|█▉        | 2064/10707 [21:09<1:11:27,  2.02it/s]
 19%|█▉        | 2065/10707 [21:09<1:11:23,  2.02it/s]
 19%|█▉        | 2066/10707 [21:10<1:11:27,  2.02it/s]
 19%|█▉        | 2067/10707 [21:10<1:11:26,  2.02it/s]
 19%|█▉        | 2068/10707 [21:11<1:11:26,  2.02it/s]
 19%|█▉        | 2069/10707 [21:11<1:11:21,  2.02it/s]
 19%|█▉        | 2070/10707 [21:12<1:11:24,  2.02it/s]
 19%|█▉        | 2071/10707 [21:12<1:11:19,  2.02it/s]
 19%|█▉        | 2072/10707 [21:13<1:11:21,  2.02it/s]
 19%|█▉        | 2073/10707 [21:13<1:11:22,  2.02it/s]
 19%|█▉        | 2074/10707 [21:14<1:11:19,  2.02it/s]
 19%|█▉        | 2075/10707 [21:14<1:11:19,  2.02it/s]{'loss': 4.0475, 'grad_norm': 0.234641894698143, 'learning_rate': 0.0009734519634464753, 'epoch': 0.19}
-
                                                      

 19%|█▉        | 2075/10707 [21:14<1:11:19,  2.02it/s]
 19%|█▉        | 2076/10707 [21:15<1:11:23,  2.02it/s]
 19%|█▉        | 2077/10707 [21:15<1:11:27,  2.01it/s]
 19%|█▉        | 2078/10707 [21:16<1:11:23,  2.01it/s]
 19%|█▉        | 2079/10707 [21:16<1:11:22,  2.01it/s]
 19%|█▉        | 2080/10707 [21:17<1:11:22,  2.01it/s]
 19%|█▉        | 2081/10707 [21:17<1:11:18,  2.02it/s]
 19%|█▉        | 2082/10707 [21:18<1:11:24,  2.01it/s]
 19%|█▉        | 2083/10707 [21:18<1:11:19,  2.02it/s]
 19%|█▉        | 2084/10707 [21:19<1:11:25,  2.01it/s]
 19%|█▉        | 2085/10707 [21:19<1:11:21,  2.01it/s]
 19%|█▉        | 2086/10707 [21:20<1:11:27,  2.01it/s]
 19%|█▉        | 2087/10707 [21:20<1:11:22,  2.01it/s]
 20%|█▉        | 2088/10707 [21:21<1:11:23,  2.01it/s]
 20%|█▉        | 2089/10707 [21:21<1:11:22,  2.01it/s]
 20%|█▉        | 2090/10707 [21:22<1:11:19,  2.01it/s]
 20%|█▉        | 2091/10707 [21:22<1:11:23,  2.01it/s]
 20%|█▉        | 2092/10707 [21:23<1:11:14,  2.02it/s]
 20%|█▉        | 2093/10707 [21:23<1:11:17,  2.01it/s]
 20%|█▉        | 2094/10707 [21:24<1:11:15,  2.01it/s]
 20%|█▉        | 2095/10707 [21:24<1:11:14,  2.01it/s]
 20%|█▉        | 2096/10707 [21:25<1:11:14,  2.01it/s]
 20%|█▉        | 2097/10707 [21:25<1:11:13,  2.01it/s]
 20%|█▉        | 2098/10707 [21:26<1:11:13,  2.01it/s]
 20%|█▉        | 2099/10707 [21:26<1:11:09,  2.02it/s]
 20%|█▉        | 2100/10707 [21:27<1:11:08,  2.02it/s]{'loss': 4.0443, 'grad_norm': 0.2347092479467392, 'learning_rate': 0.0009721259641661031, 'epoch': 0.2}
-
                                                      

 20%|█▉        | 2100/10707 [21:27<1:11:08,  2.02it/s]
 20%|█▉        | 2101/10707 [21:27<1:11:13,  2.01it/s]
 20%|█▉        | 2102/10707 [21:28<1:11:11,  2.01it/s]
 20%|█▉        | 2103/10707 [21:28<1:11:13,  2.01it/s]
 20%|█▉        | 2104/10707 [21:29<1:11:09,  2.01it/s]
 20%|█▉        | 2105/10707 [21:29<1:11:07,  2.02it/s]
 20%|█▉        | 2106/10707 [21:30<1:11:04,  2.02it/s]
 20%|█▉        | 2107/10707 [21:30<1:11:05,  2.02it/s]
 20%|█▉        | 2108/10707 [21:31<1:11:03,  2.02it/s]
 20%|█▉        | 2109/10707 [21:31<1:11:04,  2.02it/s]
 20%|█▉        | 2110/10707 [21:32<1:10:58,  2.02it/s]
 20%|█▉        | 2111/10707 [21:32<1:11:01,  2.02it/s]
 20%|█▉        | 2112/10707 [21:33<1:10:57,  2.02it/s]
 20%|█▉        | 2113/10707 [21:33<1:10:59,  2.02it/s]
 20%|█▉        | 2114/10707 [21:34<1:10:55,  2.02it/s]
 20%|█▉        | 2115/10707 [21:34<1:10:58,  2.02it/s]
 20%|█▉        | 2116/10707 [21:35<1:10:55,  2.02it/s]
 20%|█▉        | 2117/10707 [21:35<1:11:00,  2.02it/s]
 20%|█▉        | 2118/10707 [21:36<1:10:56,  2.02it/s]
 20%|█▉        | 2119/10707 [21:36<1:10:56,  2.02it/s]
 20%|█▉        | 2120/10707 [21:37<1:10:55,  2.02it/s]
 20%|█▉        | 2121/10707 [21:37<1:10:56,  2.02it/s]
 20%|█▉        | 2122/10707 [21:38<1:10:58,  2.02it/s]
 20%|█▉        | 2123/10707 [21:38<1:10:57,  2.02it/s]
 20%|█▉        | 2124/10707 [21:39<1:10:54,  2.02it/s]
 20%|█▉        | 2125/10707 [21:39<1:10:57,  2.02it/s]{'loss': 4.0384, 'grad_norm': 0.210103839635849, 'learning_rate': 0.0009707686001479882, 'epoch': 0.2}
-
                                                      

 20%|█▉        | 2125/10707 [21:39<1:10:57,  2.02it/s]
 20%|█▉        | 2126/10707 [21:40<1:11:02,  2.01it/s]
 20%|█▉        | 2127/10707 [21:40<1:11:00,  2.01it/s]
 20%|█▉        | 2128/10707 [21:41<1:10:54,  2.02it/s]
 20%|█▉        | 2129/10707 [21:41<1:10:53,  2.02it/s]
 20%|█▉        | 2130/10707 [21:42<1:10:48,  2.02it/s]
 20%|█▉        | 2131/10707 [21:42<1:10:47,  2.02it/s]
 20%|█▉        | 2132/10707 [21:43<1:10:46,  2.02it/s]
 20%|█▉        | 2133/10707 [21:43<1:10:52,  2.02it/s]
 20%|█▉        | 2134/10707 [21:44<1:10:53,  2.02it/s]
 20%|█▉        | 2135/10707 [21:44<1:10:50,  2.02it/s]
 20%|█▉        | 2136/10707 [21:45<1:10:55,  2.01it/s]
 20%|█▉        | 2137/10707 [21:45<1:10:47,  2.02it/s]
 20%|█▉        | 2138/10707 [21:46<1:10:50,  2.02it/s]
 20%|█▉        | 2139/10707 [21:46<1:10:46,  2.02it/s]
 20%|█▉        | 2140/10707 [21:47<1:10:49,  2.02it/s]
 20%|█▉        | 2141/10707 [21:47<1:10:52,  2.01it/s]
 20%|██        | 2142/10707 [21:48<1:10:46,  2.02it/s]
 20%|██        | 2143/10707 [21:48<1:10:47,  2.02it/s]
 20%|██        | 2144/10707 [21:49<1:10:43,  2.02it/s]
 20%|██        | 2145/10707 [21:49<1:10:44,  2.02it/s]
 20%|██        | 2146/10707 [21:50<1:10:39,  2.02it/s]
 20%|██        | 2147/10707 [21:50<1:10:43,  2.02it/s]
 20%|██        | 2148/10707 [21:51<1:10:40,  2.02it/s]
 20%|██        | 2149/10707 [21:51<1:10:42,  2.02it/s]
 20%|██        | 2150/10707 [21:52<1:10:42,  2.02it/s]{'loss': 4.0302, 'grad_norm': 0.24162980914115906, 'learning_rate': 0.000969379961565876, 'epoch': 0.2}
-
                                                      

 20%|██        | 2150/10707 [21:52<1:10:42,  2.02it/s]
 20%|██        | 2151/10707 [21:52<1:10:44,  2.02it/s]
 20%|██        | 2152/10707 [21:52<1:10:46,  2.01it/s]
 20%|██        | 2153/10707 [21:53<1:10:41,  2.02it/s]
 20%|██        | 2154/10707 [21:53<1:10:42,  2.02it/s]
 20%|██        | 2155/10707 [21:54<1:10:39,  2.02it/s]
 20%|██        | 2156/10707 [21:54<1:10:44,  2.01it/s]
 20%|██        | 2157/10707 [21:55<1:10:43,  2.02it/s]
 20%|██        | 2158/10707 [21:55<1:10:38,  2.02it/s]
 20%|██        | 2159/10707 [21:56<1:10:38,  2.02it/s]
 20%|██        | 2160/10707 [21:56<1:10:37,  2.02it/s]
 20%|██        | 2161/10707 [21:57<1:10:38,  2.02it/s]
 20%|██        | 2162/10707 [21:57<1:10:32,  2.02it/s]
 20%|██        | 2163/10707 [21:58<1:10:32,  2.02it/s]
 20%|██        | 2164/10707 [21:58<1:10:29,  2.02it/s]
 20%|██        | 2165/10707 [21:59<1:10:33,  2.02it/s]
 20%|██        | 2166/10707 [21:59<1:10:26,  2.02it/s]
 20%|██        | 2167/10707 [22:00<1:10:37,  2.02it/s]
 20%|██        | 2168/10707 [22:00<1:10:35,  2.02it/s]
 20%|██        | 2169/10707 [22:01<1:10:37,  2.02it/s]
 20%|██        | 2170/10707 [22:01<1:10:36,  2.02it/s]
 20%|██        | 2171/10707 [22:02<1:10:33,  2.02it/s]
 20%|██        | 2172/10707 [22:02<1:10:37,  2.01it/s]
 20%|██        | 2173/10707 [22:03<1:10:31,  2.02it/s]
 20%|██        | 2174/10707 [22:03<1:10:29,  2.02it/s]
 20%|██        | 2175/10707 [22:04<1:10:32,  2.02it/s]{'loss': 4.0271, 'grad_norm': 0.21163760125637054, 'learning_rate': 0.0009679601406711746, 'epoch': 0.2}
-
                                                      

 20%|██        | 2175/10707 [22:04<1:10:32,  2.02it/s]
 20%|██        | 2176/10707 [22:04<1:10:39,  2.01it/s]
 20%|██        | 2177/10707 [22:05<1:10:41,  2.01it/s]
 20%|██        | 2178/10707 [22:05<1:10:40,  2.01it/s]
 20%|██        | 2179/10707 [22:06<1:10:41,  2.01it/s]
 20%|██        | 2180/10707 [22:06<1:10:41,  2.01it/s]
 20%|██        | 2181/10707 [22:07<1:10:38,  2.01it/s]
 20%|██        | 2182/10707 [22:07<1:10:34,  2.01it/s]
 20%|██        | 2183/10707 [22:08<1:10:30,  2.01it/s]
 20%|██        | 2184/10707 [22:08<1:10:29,  2.02it/s]
 20%|██        | 2185/10707 [22:09<1:10:32,  2.01it/s]
 20%|██        | 2186/10707 [22:09<1:10:28,  2.02it/s]
 20%|██        | 2187/10707 [22:10<1:10:25,  2.02it/s]
 20%|██        | 2188/10707 [22:10<1:10:25,  2.02it/s]
 20%|██        | 2189/10707 [22:11<1:10:24,  2.02it/s]
 20%|██        | 2190/10707 [22:11<1:10:22,  2.02it/s]
 20%|██        | 2191/10707 [22:12<1:10:17,  2.02it/s]
 20%|██        | 2192/10707 [22:12<1:10:23,  2.02it/s]
 20%|██        | 2193/10707 [22:13<1:10:17,  2.02it/s]
 20%|██        | 2194/10707 [22:13<1:10:23,  2.02it/s]
 21%|██        | 2195/10707 [22:14<1:10:22,  2.02it/s]
 21%|██        | 2196/10707 [22:14<1:10:22,  2.02it/s]
 21%|██        | 2197/10707 [22:15<1:10:16,  2.02it/s]
 21%|██        | 2198/10707 [22:15<1:10:18,  2.02it/s]
 21%|██        | 2199/10707 [22:16<1:10:16,  2.02it/s]
 21%|██        | 2200/10707 [22:16<1:10:19,  2.02it/s]{'loss': 4.0272, 'grad_norm': 0.2401537299156189, 'learning_rate': 0.000966509231786826, 'epoch': 0.21}
-
                                                      

 21%|██        | 2200/10707 [22:16<1:10:19,  2.02it/s]
 21%|██        | 2201/10707 [22:17<1:10:26,  2.01it/s]
 21%|██        | 2202/10707 [22:17<1:10:20,  2.02it/s]
 21%|██        | 2203/10707 [22:18<1:10:20,  2.01it/s]
 21%|██        | 2204/10707 [22:18<1:10:14,  2.02it/s]
 21%|██        | 2205/10707 [22:19<1:10:16,  2.02it/s]
 21%|██        | 2206/10707 [22:19<1:10:09,  2.02it/s]
 21%|██        | 2207/10707 [22:20<1:10:14,  2.02it/s]
 21%|██        | 2208/10707 [22:20<1:10:12,  2.02it/s]
 21%|██        | 2209/10707 [22:21<1:10:11,  2.02it/s]
 21%|██        | 2210/10707 [22:21<1:10:09,  2.02it/s]
 21%|██        | 2211/10707 [22:22<1:10:09,  2.02it/s]
 21%|██        | 2212/10707 [22:22<1:10:07,  2.02it/s]
 21%|██        | 2213/10707 [22:23<1:10:07,  2.02it/s]
 21%|██        | 2214/10707 [22:23<1:10:08,  2.02it/s]
 21%|██        | 2215/10707 [22:24<1:10:05,  2.02it/s]
 21%|██        | 2216/10707 [22:24<1:10:11,  2.02it/s]
 21%|██        | 2217/10707 [22:25<1:10:09,  2.02it/s]
 21%|██        | 2218/10707 [22:25<1:10:08,  2.02it/s]
 21%|██        | 2219/10707 [22:26<1:10:09,  2.02it/s]
 21%|██        | 2220/10707 [22:26<1:10:04,  2.02it/s]
 21%|██        | 2221/10707 [22:27<1:10:03,  2.02it/s]
 21%|██        | 2222/10707 [22:27<1:10:06,  2.02it/s]
 21%|██        | 2223/10707 [22:28<1:10:02,  2.02it/s]
 21%|██        | 2224/10707 [22:28<1:10:05,  2.02it/s]
 21%|██        | 2225/10707 [22:29<1:10:04,  2.02it/s]{'loss': 4.0108, 'grad_norm': 0.23768597841262817, 'learning_rate': 0.0009650273313010404, 'epoch': 0.21}
                                                      
-
 21%|██        | 2225/10707 [22:29<1:10:04,  2.02it/s]
 21%|██        | 2226/10707 [22:29<1:10:14,  2.01it/s]
 21%|██        | 2227/10707 [22:30<1:10:05,  2.02it/s]
 21%|██        | 2228/10707 [22:30<1:10:05,  2.02it/s]
 21%|██        | 2229/10707 [22:31<1:10:05,  2.02it/s]
 21%|██        | 2230/10707 [22:31<1:10:03,  2.02it/s]
 21%|██        | 2231/10707 [22:32<1:10:03,  2.02it/s]
 21%|██        | 2232/10707 [22:32<1:10:03,  2.02it/s]
 21%|██        | 2233/10707 [22:33<1:10:02,  2.02it/s]
 21%|██        | 2234/10707 [22:33<1:10:06,  2.01it/s]
 21%|██        | 2235/10707 [22:34<1:10:05,  2.01it/s]
 21%|██        | 2236/10707 [22:34<1:10:03,  2.02it/s]
 21%|██        | 2237/10707 [22:35<1:10:00,  2.02it/s]
 21%|██        | 2238/10707 [22:35<1:10:01,  2.02it/s]
 21%|██        | 2239/10707 [22:36<1:10:03,  2.01it/s]
 21%|██        | 2240/10707 [22:36<1:10:02,  2.01it/s]
 21%|██        | 2241/10707 [22:37<1:10:03,  2.01it/s]
 21%|██        | 2242/10707 [22:37<1:10:01,  2.01it/s]
 21%|██        | 2243/10707 [22:38<1:09:58,  2.02it/s]
 21%|██        | 2244/10707 [22:38<1:09:54,  2.02it/s]
 21%|██        | 2245/10707 [22:39<1:09:53,  2.02it/s]
 21%|██        | 2246/10707 [22:39<1:09:52,  2.02it/s]
 21%|██        | 2247/10707 [22:40<1:09:53,  2.02it/s]
 21%|██        | 2248/10707 [22:40<1:09:49,  2.02it/s]
 21%|██        | 2249/10707 [22:41<1:09:50,  2.02it/s]
 21%|██        | 2250/10707 [22:41<1:09:51,  2.02it/s]{'loss': 4.0119, 'grad_norm': 0.2131512612104416, 'learning_rate': 0.0009635145376608926, 'epoch': 0.21}
-
                                                      

 21%|██        | 2250/10707 [22:41<1:09:51,  2.02it/s]
 21%|██        | 2251/10707 [22:42<1:09:55,  2.02it/s]
 21%|██        | 2252/10707 [22:42<1:09:54,  2.02it/s]
 21%|██        | 2253/10707 [22:43<1:09:53,  2.02it/s]
 21%|██        | 2254/10707 [22:43<1:09:52,  2.02it/s]
 21%|██        | 2255/10707 [22:44<1:09:52,  2.02it/s]
 21%|██        | 2256/10707 [22:44<1:09:49,  2.02it/s]
 21%|██        | 2257/10707 [22:45<1:09:52,  2.02it/s]
 21%|██        | 2258/10707 [22:45<1:09:51,  2.02it/s]
 21%|██        | 2259/10707 [22:46<1:09:52,  2.01it/s]
 21%|██        | 2260/10707 [22:46<1:09:53,  2.01it/s]
 21%|██        | 2261/10707 [22:47<1:09:52,  2.01it/s]
 21%|██        | 2262/10707 [22:47<1:09:52,  2.01it/s]
 21%|██        | 2263/10707 [22:48<1:09:55,  2.01it/s]
 21%|██        | 2264/10707 [22:48<1:09:55,  2.01it/s]
 21%|██        | 2265/10707 [22:49<1:09:50,  2.01it/s]
 21%|██        | 2266/10707 [22:49<1:09:47,  2.02it/s]
 21%|██        | 2267/10707 [22:50<1:09:47,  2.02it/s]
 21%|██        | 2268/10707 [22:50<1:09:45,  2.02it/s]
 21%|██        | 2269/10707 [22:51<1:09:45,  2.02it/s]
 21%|██        | 2270/10707 [22:51<1:09:38,  2.02it/s]
 21%|██        | 2271/10707 [22:52<1:09:40,  2.02it/s]
 21%|██        | 2272/10707 [22:52<1:09:39,  2.02it/s]
 21%|██        | 2273/10707 [22:53<1:09:41,  2.02it/s]
 21%|██        | 2274/10707 [22:53<1:09:41,  2.02it/s]
 21%|██        | 2275/10707 [22:53<1:09:38,  2.02it/s]{'loss': 4.0086, 'grad_norm': 0.21913419663906097, 'learning_rate': 0.0009619709513657822, 'epoch': 0.21}
-
                                                      

 21%|██        | 2275/10707 [22:54<1:09:38,  2.02it/s]
 21%|██▏       | 2276/10707 [22:54<1:10:00,  2.01it/s]
 21%|██▏       | 2277/10707 [22:55<1:09:55,  2.01it/s]
 21%|██▏       | 2278/10707 [22:55<1:09:48,  2.01it/s]
 21%|██▏       | 2279/10707 [22:55<1:09:44,  2.01it/s]
 21%|██▏       | 2280/10707 [22:56<1:09:35,  2.02it/s]
 21%|██▏       | 2281/10707 [22:56<1:09:37,  2.02it/s]
 21%|██▏       | 2282/10707 [22:57<1:09:33,  2.02it/s]
 21%|██▏       | 2283/10707 [22:57<1:09:33,  2.02it/s]
 21%|██▏       | 2284/10707 [22:58<1:09:29,  2.02it/s]
 21%|██▏       | 2285/10707 [22:58<1:09:31,  2.02it/s]
 21%|██▏       | 2286/10707 [22:59<1:09:33,  2.02it/s]
 21%|██▏       | 2287/10707 [22:59<1:09:35,  2.02it/s]
 21%|██▏       | 2288/10707 [23:00<1:09:38,  2.02it/s]
 21%|██▏       | 2289/10707 [23:00<1:09:35,  2.02it/s]
 21%|██▏       | 2290/10707 [23:01<1:09:37,  2.01it/s]
 21%|██▏       | 2291/10707 [23:01<1:09:33,  2.02it/s]
 21%|██▏       | 2292/10707 [23:02<1:09:36,  2.02it/s]
 21%|██▏       | 2293/10707 [23:02<1:09:30,  2.02it/s]
 21%|██▏       | 2294/10707 [23:03<1:09:31,  2.02it/s]
 21%|██▏       | 2295/10707 [23:03<1:09:34,  2.02it/s]
 21%|██▏       | 2296/10707 [23:04<1:09:33,  2.02it/s]
 21%|██▏       | 2297/10707 [23:04<1:09:32,  2.02it/s]
 21%|██▏       | 2298/10707 [23:05<1:09:28,  2.02it/s]
 21%|██▏       | 2299/10707 [23:05<1:09:28,  2.02it/s]
 21%|██▏       | 2300/10707 [23:06<1:09:29,  2.02it/s]{'loss': 4.0068, 'grad_norm': 0.2029961794614792, 'learning_rate': 0.0009603966749607558, 'epoch': 0.21}
                                                      
-
 21%|██▏       | 2300/10707 [23:06<1:09:29,  2.02it/s]
 21%|██▏       | 2301/10707 [23:06<1:09:33,  2.01it/s]
 21%|██▏       | 2302/10707 [23:07<1:09:31,  2.02it/s]
 22%|██▏       | 2303/10707 [23:07<1:09:38,  2.01it/s]
 22%|██▏       | 2304/10707 [23:08<1:09:32,  2.01it/s]
 22%|██▏       | 2305/10707 [23:08<1:09:30,  2.01it/s]
 22%|██▏       | 2306/10707 [23:09<1:09:27,  2.02it/s]
 22%|██▏       | 2307/10707 [23:09<1:09:28,  2.02it/s]
 22%|██▏       | 2308/10707 [23:10<1:09:29,  2.01it/s]
 22%|██▏       | 2309/10707 [23:10<1:09:26,  2.02it/s]
 22%|██▏       | 2310/10707 [23:11<1:09:26,  2.02it/s]
 22%|██▏       | 2311/10707 [23:11<1:09:24,  2.02it/s]
 22%|██▏       | 2312/10707 [23:12<1:09:25,  2.02it/s]
 22%|██▏       | 2313/10707 [23:12<1:09:24,  2.02it/s]
 22%|██▏       | 2314/10707 [23:13<1:09:26,  2.01it/s]
 22%|██▏       | 2315/10707 [23:13<1:09:20,  2.02it/s]
 22%|██▏       | 2316/10707 [23:14<1:09:23,  2.02it/s]
 22%|██▏       | 2317/10707 [23:14<1:09:20,  2.02it/s]
 22%|██▏       | 2318/10707 [23:15<1:09:21,  2.02it/s]
 22%|██▏       | 2319/10707 [23:15<1:09:18,  2.02it/s]
 22%|██▏       | 2320/10707 [23:16<1:09:18,  2.02it/s]
 22%|██▏       | 2321/10707 [23:16<1:09:21,  2.02it/s]
 22%|██▏       | 2322/10707 [23:17<1:09:16,  2.02it/s]
 22%|██▏       | 2323/10707 [23:17<1:09:14,  2.02it/s]
 22%|██▏       | 2324/10707 [23:18<1:09:12,  2.02it/s]
 22%|██▏       | 2325/10707 [23:18<1:09:14,  2.02it/s]
                                                      {'loss': 4.0018, 'grad_norm': 0.21629446744918823, 'learning_rate': 0.0009587918130296969, 'epoch': 0.22}
-

 22%|██▏       | 2325/10707 [23:18<1:09:14,  2.02it/s]
 22%|██▏       | 2326/10707 [23:19<1:09:14,  2.02it/s]
 22%|██▏       | 2327/10707 [23:19<1:09:18,  2.02it/s]
 22%|██▏       | 2328/10707 [23:20<1:09:18,  2.01it/s]
 22%|██▏       | 2329/10707 [23:20<1:09:15,  2.02it/s]
 22%|██▏       | 2330/10707 [23:21<1:09:15,  2.02it/s]
 22%|██▏       | 2331/10707 [23:21<1:09:12,  2.02it/s]
 22%|██▏       | 2332/10707 [23:22<1:09:15,  2.02it/s]
 22%|██▏       | 2333/10707 [23:22<1:09:09,  2.02it/s]
 22%|██▏       | 2334/10707 [23:23<1:09:17,  2.01it/s]
 22%|██▏       | 2335/10707 [23:23<1:09:13,  2.02it/s]
 22%|██▏       | 2336/10707 [23:24<1:09:10,  2.02it/s]
 22%|██▏       | 2337/10707 [23:24<1:09:09,  2.02it/s]
 22%|██▏       | 2338/10707 [23:25<1:09:08,  2.02it/s]
 22%|██▏       | 2339/10707 [23:25<1:09:06,  2.02it/s]
 22%|██▏       | 2340/10707 [23:26<1:09:06,  2.02it/s]
 22%|██▏       | 2341/10707 [23:26<1:09:08,  2.02it/s]
 22%|██▏       | 2342/10707 [23:27<1:09:07,  2.02it/s]
 22%|██▏       | 2343/10707 [23:27<1:09:09,  2.02it/s]
 22%|██▏       | 2344/10707 [23:28<1:09:04,  2.02it/s]
 22%|██▏       | 2345/10707 [23:28<1:09:08,  2.02it/s]
 22%|██▏       | 2346/10707 [23:29<1:09:09,  2.01it/s]
 22%|██▏       | 2347/10707 [23:29<1:09:09,  2.01it/s]
 22%|██▏       | 2348/10707 [23:30<1:09:08,  2.01it/s]
 22%|██▏       | 2349/10707 [23:30<1:09:05,  2.02it/s]
 22%|██▏       | 2350/10707 [23:31<1:09:05,  2.02it/s]{'loss': 3.9967, 'grad_norm': 0.21959742903709412, 'learning_rate': 0.0009571564721883761, 'epoch': 0.22}
-
                                                      

 22%|██▏       | 2350/10707 [23:31<1:09:05,  2.02it/s]
 22%|██▏       | 2351/10707 [23:31<1:09:10,  2.01it/s]
 22%|██▏       | 2352/10707 [23:32<1:09:11,  2.01it/s]
 22%|██▏       | 2353/10707 [23:32<1:09:04,  2.02it/s]
 22%|██▏       | 2354/10707 [23:33<1:09:08,  2.01it/s]
 22%|██▏       | 2355/10707 [23:33<1:09:01,  2.02it/s]
 22%|██▏       | 2356/10707 [23:34<1:09:03,  2.02it/s]
 22%|██▏       | 2357/10707 [23:34<1:08:58,  2.02it/s]
 22%|██▏       | 2358/10707 [23:35<1:08:59,  2.02it/s]
 22%|██▏       | 2359/10707 [23:35<1:08:54,  2.02it/s]
 22%|██▏       | 2360/10707 [23:36<1:08:56,  2.02it/s]
 22%|██▏       | 2361/10707 [23:36<1:08:55,  2.02it/s]
 22%|██▏       | 2362/10707 [23:37<1:08:56,  2.02it/s]
 22%|██▏       | 2363/10707 [23:37<1:08:56,  2.02it/s]
 22%|██▏       | 2364/10707 [23:38<1:08:57,  2.02it/s]
 22%|██▏       | 2365/10707 [23:38<1:08:55,  2.02it/s]
 22%|██▏       | 2366/10707 [23:39<1:08:57,  2.02it/s]
 22%|██▏       | 2367/10707 [23:39<1:08:56,  2.02it/s]
 22%|██▏       | 2368/10707 [23:40<1:08:56,  2.02it/s]
 22%|██▏       | 2369/10707 [23:40<1:09:00,  2.01it/s]
 22%|██▏       | 2370/10707 [23:41<1:08:58,  2.01it/s]
 22%|██▏       | 2371/10707 [23:41<1:08:58,  2.01it/s]
 22%|██▏       | 2372/10707 [23:42<1:09:00,  2.01it/s]
 22%|██▏       | 2373/10707 [23:42<1:08:58,  2.01it/s]
 22%|██▏       | 2374/10707 [23:43<1:08:54,  2.02it/s]
 22%|██��       | 2375/10707 [23:43<1:08:54,  2.02it/s]{'loss': 3.9994, 'grad_norm': 0.2141597419977188, 'learning_rate': 0.0009554907610773692, 'epoch': 0.22}
-
                                                      

 22%|██▏       | 2375/10707 [23:43<1:08:54,  2.02it/s]
 22%|██▏       | 2376/10707 [23:44<1:09:01,  2.01it/s]
 22%|██▏       | 2377/10707 [23:44<1:08:59,  2.01it/s]
 22%|██▏       | 2378/10707 [23:45<1:08:56,  2.01it/s]
 22%|██▏       | 2379/10707 [23:45<1:08:53,  2.01it/s]
 22%|██▏       | 2380/10707 [23:46<1:08:50,  2.02it/s]
 22%|██▏       | 2381/10707 [23:46<1:08:46,  2.02it/s]
 22%|██▏       | 2382/10707 [23:47<1:08:45,  2.02it/s]
 22%|██▏       | 2383/10707 [23:47<1:08:43,  2.02it/s]
 22%|██▏       | 2384/10707 [23:48<1:08:42,  2.02it/s]
 22%|██▏       | 2385/10707 [23:48<1:08:41,  2.02it/s]
 22%|██▏       | 2386/10707 [23:49<1:08:40,  2.02it/s]
 22%|██▏       | 2387/10707 [23:49<1:08:42,  2.02it/s]
 22%|██▏       | 2388/10707 [23:50<1:08:41,  2.02it/s]
 22%|██▏       | 2389/10707 [23:50<1:08:46,  2.02it/s]
 22%|██▏       | 2390/10707 [23:51<1:08:41,  2.02it/s]
 22%|██▏       | 2391/10707 [23:51<1:08:44,  2.02it/s]
 22%|██▏       | 2392/10707 [23:52<1:08:41,  2.02it/s]
 22%|██▏       | 2393/10707 [23:52<1:08:42,  2.02it/s]
 22%|██▏       | 2394/10707 [23:53<1:08:40,  2.02it/s]
 22%|██▏       | 2395/10707 [23:53<1:08:39,  2.02it/s]
 22%|██▏       | 2396/10707 [23:54<1:08:36,  2.02it/s]
 22%|██▏       | 2397/10707 [23:54<1:08:38,  2.02it/s]
 22%|██▏       | 2398/10707 [23:55<1:08:41,  2.02it/s]
 22%|██▏       | 2399/10707 [23:55<1:08:40,  2.02it/s]
 22%|██▏       | 2400/10707 [23:56<1:08:45,  2.01it/s]{'loss': 3.9946, 'grad_norm': 0.2102956920862198, 'learning_rate': 0.0009537947903548398, 'epoch': 0.22}
-
                                                      

 22%|██▏       | 2400/10707 [23:56<1:08:45,  2.01it/s]
 22%|██▏       | 2401/10707 [23:56<1:08:45,  2.01it/s]
 22%|██▏       | 2402/10707 [23:56<1:08:41,  2.01it/s]
 22%|██▏       | 2403/10707 [23:57<1:08:42,  2.01it/s]
 22%|██▏       | 2404/10707 [23:57<1:08:39,  2.02it/s]
 22%|██▏       | 2405/10707 [23:58<1:08:39,  2.02it/s]
 22%|██▏       | 2406/10707 [23:58<1:08:37,  2.02it/s]
 22%|██▏       | 2407/10707 [23:59<1:08:38,  2.02it/s]
 22%|██▏       | 2408/10707 [23:59<1:08:37,  2.02it/s]
 22%|██▏       | 2409/10707 [24:00<1:08:35,  2.02it/s]
 23%|██▎       | 2410/10707 [24:00<1:08:32,  2.02it/s]
 23%|██▎       | 2411/10707 [24:01<1:08:30,  2.02it/s]
 23%|██▎       | 2412/10707 [24:01<1:08:34,  2.02it/s]
 23%|██▎       | 2413/10707 [24:02<1:08:33,  2.02it/s]
 23%|██▎       | 2414/10707 [24:02<1:08:33,  2.02it/s]
 23%|██▎       | 2415/10707 [24:03<1:08:31,  2.02it/s]
 23%|██▎       | 2416/10707 [24:03<1:08:30,  2.02it/s]
 23%|██▎       | 2417/10707 [24:04<1:08:29,  2.02it/s]
 23%|██▎       | 2418/10707 [24:04<1:08:32,  2.02it/s]
 23%|██▎       | 2419/10707 [24:05<1:08:27,  2.02it/s]
 23%|██▎       | 2420/10707 [24:05<1:08:26,  2.02it/s]
 23%|██▎       | 2421/10707 [24:06<1:08:26,  2.02it/s]
 23%|██▎       | 2422/10707 [24:06<1:08:26,  2.02it/s]
 23%|██▎       | 2423/10707 [24:07<1:08:26,  2.02it/s]
 23%|██▎       | 2424/10707 [24:07<1:08:25,  2.02it/s]
 23%|██▎       | 2425/10707 [24:08<1:08:29,  2.02it/s]
                                                      {'loss': 3.9798, 'grad_norm': 0.21812114119529724, 'learning_rate': 0.0009520686726891878, 'epoch': 0.23}
-
 23%|██▎       | 2425/10707 [24:08<1:08:29,  2.02it/s]
 23%|██▎       | 2426/10707 [24:08<1:08:31,  2.01it/s]
 23%|██▎       | 2427/10707 [24:09<1:08:32,  2.01it/s]
 23%|██▎       | 2428/10707 [24:09<1:08:30,  2.01it/s]
 23%|██▎       | 2429/10707 [24:10<1:08:29,  2.01it/s]
 23%|██▎       | 2430/10707 [24:10<1:08:30,  2.01it/s]
 23%|██▎       | 2431/10707 [24:11<1:08:23,  2.02it/s]
 23%|██▎       | 2432/10707 [24:11<1:08:28,  2.01it/s]
 23%|██▎       | 2433/10707 [24:12<1:08:19,  2.02it/s]
 23%|██▎       | 2434/10707 [24:12<1:08:20,  2.02it/s]
 23%|██▎       | 2435/10707 [24:13<1:08:16,  2.02it/s]
 23%|██▎       | 2436/10707 [24:13<1:08:19,  2.02it/s]
 23%|██▎       | 2437/10707 [24:14<1:08:16,  2.02it/s]
 23%|██▎       | 2438/10707 [24:14<1:08:18,  2.02it/s]
 23%|██▎       | 2439/10707 [24:15<1:08:15,  2.02it/s]
 23%|██▎       | 2440/10707 [24:15<1:08:19,  2.02it/s]
 23%|██▎       | 2441/10707 [24:16<1:08:21,  2.02it/s]
 23%|██▎       | 2442/10707 [24:16<1:08:15,  2.02it/s]
 23%|██▎       | 2443/10707 [24:17<1:08:17,  2.02it/s]
 23%|██▎       | 2444/10707 [24:17<1:08:16,  2.02it/s]
 23%|██▎       | 2445/10707 [24:18<1:08:22,  2.01it/s]
 23%|██▎       | 2446/10707 [24:18<1:08:15,  2.02it/s]
 23%|██▎       | 2447/10707 [24:19<1:08:15,  2.02it/s]
 23%|█��▎       | 2448/10707 [24:19<1:08:15,  2.02it/s]
 23%|██▎       | 2449/10707 [24:20<1:08:17,  2.02it/s]
 23%|██▎       | 2450/10707 [24:20<1:08:19,  2.01it/s]{'loss': 3.9915, 'grad_norm': 0.20725862681865692, 'learning_rate': 0.0009503125227515645, 'epoch': 0.23}
-
                                                      

 23%|██▎       | 2450/10707 [24:20<1:08:19,  2.01it/s]
 23%|██▎       | 2451/10707 [24:21<1:08:22,  2.01it/s]
 23%|██▎       | 2452/10707 [24:21<1:08:27,  2.01it/s]
 23%|██▎       | 2453/10707 [24:22<1:08:24,  2.01it/s]
 23%|██▎       | 2454/10707 [24:22<1:08:17,  2.01it/s]
 23%|██▎       | 2455/10707 [24:23<1:08:18,  2.01it/s]
 23%|██▎       | 2456/10707 [24:23<1:08:14,  2.02it/s]
 23%|██▎       | 2457/10707 [24:24<1:08:11,  2.02it/s]
 23%|██▎       | 2458/10707 [24:24<1:08:08,  2.02it/s]
 23%|██▎       | 2459/10707 [24:25<1:08:08,  2.02it/s]
 23%|██▎       | 2460/10707 [24:25<1:08:09,  2.02it/s]
 23%|██▎       | 2461/10707 [24:26<1:08:11,  2.02it/s]
 23%|██▎       | 2462/10707 [24:26<1:08:09,  2.02it/s]
 23%|██▎       | 2463/10707 [24:27<1:08:09,  2.02it/s]
 23%|██▎       | 2464/10707 [24:27<1:08:07,  2.02it/s]
 23%|██▎       | 2465/10707 [24:28<1:08:03,  2.02it/s]
 23%|██▎       | 2466/10707 [24:28<1:08:04,  2.02it/s]
 23%|██▎       | 2467/10707 [24:29<1:08:07,  2.02it/s]
 23%|██▎       | 2468/10707 [24:29<1:08:06,  2.02it/s]
 23%|██▎       | 2469/10707 [24:30<1:08:06,  2.02it/s]
 23%|██▎       | 2470/10707 [24:30<1:08:04,  2.02it/s]
 23%|██▎       | 2471/10707 [24:31<1:08:04,  2.02it/s]
 23%|██▎       | 2472/10707 [24:31<1:08:07,  2.01it/s]
 23%|██▎       | 2473/10707 [24:32<1:08:04,  2.02it/s]
 23%|██▎       | 2474/10707 [24:32<1:08:01,  2.02it/s]
 23%|██▎       | 2475/10707 [24:33<1:08:03,  2.02it/s]{'loss': 3.9847, 'grad_norm': 0.22244808077812195, 'learning_rate': 0.0009485264572082551, 'epoch': 0.23}
-
                                                      

 23%|██▎       | 2475/10707 [24:33<1:08:03,  2.02it/s]
 23%|██▎       | 2476/10707 [24:33<1:08:02,  2.02it/s]
 23%|██▎       | 2477/10707 [24:34<1:18:47,  1.74it/s]
 23%|██▎       | 2478/10707 [24:34<1:15:27,  1.82it/s]
 23%|██▎       | 2479/10707 [24:35<1:13:13,  1.87it/s]
 23%|██▎       | 2480/10707 [24:35<1:11:33,  1.92it/s]
 23%|██▎       | 2481/10707 [24:36<1:10:31,  1.94it/s]
 23%|██▎       | 2482/10707 [24:36<1:09:39,  1.97it/s]
 23%|██▎       | 2483/10707 [24:37<1:09:09,  1.98it/s]
 23%|██▎       | 2484/10707 [24:37<1:08:41,  2.00it/s]
 23%|██▎       | 2485/10707 [24:38<1:08:30,  2.00it/s]
 23%|██▎       | 2486/10707 [24:38<1:08:18,  2.01it/s]
 23%|██▎       | 2487/10707 [24:39<1:08:11,  2.01it/s]
 23%|██▎       | 2488/10707 [24:39<1:08:01,  2.01it/s]
 23%|██▎       | 2489/10707 [24:40<1:19:24,  1.72it/s]
 23%|██▎       | 2490/10707 [24:41<1:15:56,  1.80it/s]
 23%|██▎       | 2491/10707 [24:41<1:13:31,  1.86it/s]
 23%|██▎       | 2492/10707 [24:42<1:11:49,  1.91it/s]
 23%|██▎       | 2493/10707 [24:42<1:10:35,  1.94it/s]
 23%|██▎       | 2494/10707 [24:43<1:09:46,  1.96it/s]
 23%|██▎       | 2495/10707 [24:43<1:09:07,  1.98it/s]
 23%|██▎       | 2496/10707 [24:44<1:08:46,  1.99it/s]
 23%|██▎       | 2497/10707 [24:44<1:08:27,  2.00it/s]
 23%|██▎       | 2498/10707 [24:45<1:08:17,  2.00it/s]
 23%|██▎       | 2499/10707 [24:45<1:08:11,  2.01it/s]
 23%|██▎       | 2500/10707 [24:46<1:08:05,  2.01it/s]{'loss': 3.9755, 'grad_norm': 0.21792393922805786, 'learning_rate': 0.0009467105947129268, 'epoch': 0.23}
-
                                                      

 23%|██▎       | 2500/10707 [24:46<1:08:05,  2.01it/s]
 23%|██▎       | 2501/10707 [24:46<1:08:03,  2.01it/s]
 23%|██▎       | 2502/10707 [24:47<1:08:01,  2.01it/s]
 23%|██▎       | 2503/10707 [24:47<1:08:00,  2.01it/s]
 23%|██▎       | 2504/10707 [24:48<1:07:54,  2.01it/s]
 23%|██▎       | 2505/10707 [24:48<1:07:50,  2.01it/s]
 23%|██▎       | 2506/10707 [24:49<1:07:47,  2.02it/s]
 23%|██▎       | 2507/10707 [24:49<1:07:47,  2.02it/s]
 23%|██▎       | 2508/10707 [24:50<1:07:42,  2.02it/s]
 23%|██▎       | 2509/10707 [24:50<1:07:39,  2.02it/s]
 23%|██▎       | 2510/10707 [24:51<1:07:36,  2.02it/s]
 23%|██▎       | 2511/10707 [24:51<1:07:39,  2.02it/s]
 23%|██▎       | 2512/10707 [24:52<1:07:36,  2.02it/s]
 23%|██▎       | 2513/10707 [24:52<1:07:33,  2.02it/s]
 23%|██▎       | 2514/10707 [24:53<1:07:34,  2.02it/s]
 23%|██▎       | 2515/10707 [24:53<1:07:32,  2.02it/s]
 23%|██▎       | 2516/10707 [24:54<1:07:31,  2.02it/s]
 24%|██▎       | 2517/10707 [24:54<1:07:35,  2.02it/s]
 24%|██▎       | 2518/10707 [24:55<1:07:32,  2.02it/s]
 24%|██▎       | 2519/10707 [24:55<1:07:35,  2.02it/s]
 24%|██▎       | 2520/10707 [24:56<1:07:33,  2.02it/s]
 24%|██▎       | 2521/10707 [24:56<1:07:38,  2.02it/s]
 24%|██▎       | 2522/10707 [24:57<1:07:33,  2.02it/s]
 24%|██▎       | 2523/10707 [24:57<1:07:38,  2.02it/s]
 24%|██▎       | 2524/10707 [24:58<1:07:37,  2.02it/s]
 24%|██▎       | 2525/10707 [24:58<1:07:33,  2.02it/s]
                                                      {'loss': 3.9714, 'grad_norm': 0.21393939852714539, 'learning_rate': 0.0009448650558987487, 'epoch': 0.24}
-
 24%|██▎       | 2525/10707 [24:58<1:07:33,  2.02it/s]
 24%|██▎       | 2526/10707 [24:59<1:07:39,  2.02it/s]
 24%|██▎       | 2527/10707 [24:59<1:07:32,  2.02it/s]
 24%|██▎       | 2528/10707 [25:00<1:07:40,  2.01it/s]
 24%|██▎       | 2529/10707 [25:00<1:07:35,  2.02it/s]
 24%|██▎       | 2530/10707 [25:00<1:07:34,  2.02it/s]
 24%|██▎       | 2531/10707 [25:01<1:07:31,  2.02it/s]
 24%|██▎       | 2532/10707 [25:01<1:07:31,  2.02it/s]
 24%|██▎       | 2533/10707 [25:02<1:07:24,  2.02it/s]
 24%|██▎       | 2534/10707 [25:02<1:07:27,  2.02it/s]
 24%|██▎       | 2535/10707 [25:03<1:07:22,  2.02it/s]
 24%|██▎       | 2536/10707 [25:03<1:07:27,  2.02it/s]
 24%|██▎       | 2537/10707 [25:04<1:07:21,  2.02it/s]
 24%|██▎       | 2538/10707 [25:04<1:07:24,  2.02it/s]
 24%|██▎       | 2539/10707 [25:05<1:07:18,  2.02it/s]
 24%|██▎       | 2540/10707 [25:05<1:07:23,  2.02it/s]
 24%|██▎       | 2541/10707 [25:06<1:07:22,  2.02it/s]
 24%|██▎       | 2542/10707 [25:06<1:07:21,  2.02it/s]
 24%|██▍       | 2543/10707 [25:07<1:07:17,  2.02it/s]
 24%|██▍       | 2544/10707 [25:07<1:07:18,  2.02it/s]
 24%|██▍       | 2545/10707 [25:08<1:07:15,  2.02it/s]
 24%|██▍       | 2546/10707 [25:08<1:07:17,  2.02it/s]
 24%|██▍       | 2547/10707 [25:09<1:07:18,  2.02it/s]
 24%|██▍       | 2548/10707 [25:09<1:07:21,  2.02it/s]
 24%|██▍       | 2549/10707 [25:10<1:07:17,  2.02it/s]
 24%|██▍       | 2550/10707 [25:10<1:07:19,  2.02it/s]
                                                      
{'loss': 3.9649, 'grad_norm': 0.21261681616306305, 'learning_rate': 0.0009429899633703751, 'epoch': 0.24}
-
 24%|██▍       | 2550/10707 [25:10<1:07:19,  2.02it/s]
 24%|██▍       | 2551/10707 [25:11<1:07:22,  2.02it/s]
 24%|██▍       | 2552/10707 [25:11<1:07:19,  2.02it/s]
 24%|██▍       | 2553/10707 [25:12<1:07:25,  2.02it/s]
 24%|██▍       | 2554/10707 [25:12<1:07:16,  2.02it/s]
 24%|██▍       | 2555/10707 [25:13<1:07:17,  2.02it/s]
 24%|██▍       | 2556/10707 [25:13<1:07:15,  2.02it/s]
 24%|██▍       | 2557/10707 [25:14<1:07:17,  2.02it/s]
 24%|██▍       | 2558/10707 [25:14<1:07:12,  2.02it/s]
 24%|██▍       | 2559/10707 [25:15<1:07:17,  2.02it/s]
 24%|██▍       | 2560/10707 [25:15<1:07:10,  2.02it/s]
 24%|██▍       | 2561/10707 [25:16<1:07:12,  2.02it/s]
 24%|██▍       | 2562/10707 [25:16<1:07:10,  2.02it/s]
 24%|██▍       | 2563/10707 [25:17<1:07:10,  2.02it/s]
 24%|██▍       | 2564/10707 [25:17<1:07:09,  2.02it/s]
 24%|██▍       | 2565/10707 [25:18<1:07:12,  2.02it/s]
 24%|██▍       | 2566/10707 [25:18<1:07:08,  2.02it/s]
 24%|██▍       | 2567/10707 [25:19<1:07:11,  2.02it/s]
 24%|██▍       | 2568/10707 [25:19<1:07:07,  2.02it/s]
 24%|██▍       | 2569/10707 [25:20<1:07:08,  2.02it/s]
 24%|██▍       | 2570/10707 [25:20<1:07:07,  2.02it/s]
 24%|██▍       | 2571/10707 [25:21<1:07:10,  2.02it/s]
 24%|██▍       | 2572/10707 [25:21<1:07:05,  2.02it/s]
 24%|██▍       | 2573/10707 [25:22<1:07:08,  2.02it/s]
 24%|██▍       | 2574/10707 [25:22<1:07:07,  2.02it/s]
 24%|██▍       | 2575/10707 [25:23<1:07:06,  2.02it/s]
                                                      
{'loss': 3.9626, 'grad_norm': 0.22643086314201355, 'learning_rate': 0.0009410854416958028, 'epoch': 0.24}
-
 24%|██▍       | 2575/10707 [25:23<1:07:06,  2.02it/s]
 24%|██▍       | 2576/10707 [25:23<1:07:14,  2.02it/s]
 24%|██▍       | 2577/10707 [25:24<1:07:11,  2.02it/s]
 24%|██▍       | 2578/10707 [25:24<1:07:09,  2.02it/s]
 24%|██▍       | 2579/10707 [25:25<1:07:06,  2.02it/s]
 24%|██▍       | 2580/10707 [25:25<1:07:10,  2.02it/s]
 24%|██▍       | 2581/10707 [25:26<1:07:06,  2.02it/s]
 24%|██▍       | 2582/10707 [25:26<1:07:05,  2.02it/s]
 24%|██▍       | 2583/10707 [25:27<1:07:00,  2.02it/s]
 24%|██▍       | 2584/10707 [25:27<1:07:03,  2.02it/s]
 24%|██▍       | 2585/10707 [25:28<1:06:59,  2.02it/s]
 24%|██▍       | 2586/10707 [25:28<1:07:03,  2.02it/s]
 24%|██▍       | 2587/10707 [25:29<1:07:00,  2.02it/s]
 24%|██▍       | 2588/10707 [25:29<1:07:02,  2.02it/s]
 24%|██▍       | 2589/10707 [25:30<1:06:57,  2.02it/s]
 24%|██▍       | 2590/10707 [25:30<1:06:57,  2.02it/s]
 24%|██▍       | 2591/10707 [25:31<1:07:01,  2.02it/s]
 24%|██▍       | 2592/10707 [25:31<1:06:59,  2.02it/s]
 24%|██▍       | 2593/10707 [25:32<1:06:58,  2.02it/s]
 24%|██▍       | 2594/10707 [25:32<1:07:00,  2.02it/s]
 24%|██▍       | 2595/10707 [25:33<1:06:58,  2.02it/s]
 24%|██▍       | 2596/10707 [25:33<1:07:02,  2.02it/s]
 24%|██▍       | 2597/10707 [25:34<1:06:59,  2.02it/s]
 24%|██▍       | 2598/10707 [25:34<1:07:04,  2.01it/s]
 24%|██▍       | 2599/10707 [25:35<1:06:59,  2.02it/s]
 24%|██▍       | 2600/10707 [25:35<1:06:55,  2.02it/s]{'loss': 3.9645, 'grad_norm': 0.2145058959722519, 'learning_rate': 0.0009391516173980942, 'epoch': 0.24}
-
                                                      

 24%|██▍       | 2600/10707 [25:35<1:06:55,  2.02it/s]
 24%|██▍       | 2601/10707 [25:36<1:07:03,  2.01it/s]
 24%|██▍       | 2602/10707 [25:36<1:07:02,  2.01it/s]
 24%|██▍       | 2603/10707 [25:37<1:07:02,  2.01it/s]
 24%|██▍       | 2604/10707 [25:37<1:07:01,  2.02it/s]
 24%|██▍       | 2605/10707 [25:38<1:06:57,  2.02it/s]
 24%|██▍       | 2606/10707 [25:38<1:06:58,  2.02it/s]
 24%|██▍       | 2607/10707 [25:39<1:06:51,  2.02it/s]
 24%|██▍       | 2608/10707 [25:39<1:06:50,  2.02it/s]
 24%|██▍       | 2609/10707 [25:40<1:06:47,  2.02it/s]
 24%|██▍       | 2610/10707 [25:40<1:06:51,  2.02it/s]
 24%|██▍       | 2611/10707 [25:41<1:06:47,  2.02it/s]
 24%|██▍       | 2612/10707 [25:41<1:06:49,  2.02it/s]
 24%|██▍       | 2613/10707 [25:42<1:06:47,  2.02it/s]
 24%|██▍       | 2614/10707 [25:42<1:06:51,  2.02it/s]
 24%|██▍       | 2615/10707 [25:43<1:06:45,  2.02it/s]
 24%|██▍       | 2616/10707 [25:43<1:06:46,  2.02it/s]
 24%|██▍       | 2617/10707 [25:44<1:06:45,  2.02it/s]
 24%|██▍       | 2618/10707 [25:44<1:06:47,  2.02it/s]
 24%|██▍       | 2619/10707 [25:45<1:06:43,  2.02it/s]
 24%|██▍       | 2620/10707 [25:45<1:06:46,  2.02it/s]
 24%|██▍       | 2621/10707 [25:46<1:06:41,  2.02it/s]
 24%|██▍       | 2622/10707 [25:46<1:06:42,  2.02it/s]
 24%|██▍       | 2623/10707 [25:47<1:06:38,  2.02it/s]
 25%|██▍       | 2624/10707 [25:47<1:06:38,  2.02it/s]
 25%|██▍       | 2625/10707 [25:48<1:06:36,  2.02it/s]{'loss': 3.9579, 'grad_norm': 0.24217671155929565, 'learning_rate': 0.0009371886189469724, 'epoch': 0.25}
-
                                                      

 25%|██▍       | 2625/10707 [25:48<1:06:36,  2.02it/s]
 25%|██▍       | 2626/10707 [25:48<1:06:46,  2.02it/s]
 25%|██▍       | 2627/10707 [25:49<1:06:40,  2.02it/s]
 25%|██▍       | 2628/10707 [25:49<1:06:42,  2.02it/s]
 25%|██▍       | 2629/10707 [25:50<1:06:39,  2.02it/s]
 25%|██▍       | 2630/10707 [25:50<1:06:38,  2.02it/s]
 25%|██▍       | 2631/10707 [25:51<1:06:36,  2.02it/s]
 25%|██▍       | 2632/10707 [25:51<1:06:35,  2.02it/s]
 25%|██▍       | 2633/10707 [25:52<1:06:34,  2.02it/s]
 25%|██▍       | 2634/10707 [25:52<1:06:35,  2.02it/s]
 25%|██▍       | 2635/10707 [25:52<1:06:39,  2.02it/s]
 25%|██▍       | 2636/10707 [25:53<1:06:36,  2.02it/s]
 25%|██▍       | 2637/10707 [25:53<1:06:34,  2.02it/s]
 25%|██▍       | 2638/10707 [25:54<1:06:31,  2.02it/s]
 25%|██▍       | 2639/10707 [25:54<1:06:29,  2.02it/s]
 25%|██▍       | 2640/10707 [25:55<1:06:33,  2.02it/s]
 25%|██▍       | 2641/10707 [25:55<1:06:34,  2.02it/s]
 25%|██▍       | 2642/10707 [25:56<1:06:31,  2.02it/s]
 25%|██▍       | 2643/10707 [25:56<1:06:33,  2.02it/s]
 25%|██▍       | 2644/10707 [25:57<1:06:27,  2.02it/s]
 25%|██▍       | 2645/10707 [25:57<1:06:31,  2.02it/s]
 25%|██▍       | 2646/10707 [25:58<1:06:27,  2.02it/s]
 25%|██▍       | 2647/10707 [25:58<1:06:29,  2.02it/s]
 25%|██▍       | 2648/10707 [25:59<1:06:27,  2.02it/s]
 25%|██▍       | 2649/10707 [25:59<1:06:31,  2.02it/s]
 25%|██▍       | 2650/10707 [26:00<1:06:28,  2.02it/s]
                                                      {'loss': 3.9569, 'grad_norm': 0.22320479154586792, 'learning_rate': 0.0009351965767502868, 'epoch': 0.25}
-
 25%|██▍       | 2650/10707 [26:00<1:06:28,  2.02it/s]
 25%|██▍       | 2651/10707 [26:00<1:06:39,  2.01it/s]
 25%|██▍       | 2652/10707 [26:01<1:06:30,  2.02it/s]
 25%|██▍       | 2653/10707 [26:01<1:06:33,  2.02it/s]
 25%|██▍       | 2654/10707 [26:02<1:06:33,  2.02it/s]
 25%|██▍       | 2655/10707 [26:02<1:06:34,  2.02it/s]
 25%|██▍       | 2656/10707 [26:03<1:06:34,  2.02it/s]
 25%|██▍       | 2657/10707 [26:03<1:06:30,  2.02it/s]
 25%|██▍       | 2658/10707 [26:04<1:06:29,  2.02it/s]
 25%|██▍       | 2659/10707 [26:04<1:06:26,  2.02it/s]
 25%|██▍       | 2660/10707 [26:05<1:06:31,  2.02it/s]
 25%|██▍       | 2661/10707 [26:05<1:06:29,  2.02it/s]
 25%|██▍       | 2662/10707 [26:06<1:06:29,  2.02it/s]
 25%|██▍       | 2663/10707 [26:06<1:06:30,  2.02it/s]
 25%|██▍       | 2664/10707 [26:07<1:06:26,  2.02it/s]
 25%|██▍       | 2665/10707 [26:07<1:06:26,  2.02it/s]
 25%|██▍       | 2666/10707 [26:08<1:06:20,  2.02it/s]
 25%|██▍       | 2667/10707 [26:08<1:06:18,  2.02it/s]
 25%|██▍       | 2668/10707 [26:09<1:06:17,  2.02it/s]
 25%|██▍       | 2669/10707 [26:09<1:06:20,  2.02it/s]
 25%|██▍       | 2670/10707 [26:10<1:06:16,  2.02it/s]
 25%|██▍       | 2671/10707 [26:10<1:06:16,  2.02it/s]
 25%|██▍       | 2672/10707 [26:11<1:06:15,  2.02it/s]
 25%|██▍       | 2673/10707 [26:11<1:06:17,  2.02it/s]
 25%|██▍       | 2674/10707 [26:12<1:06:12,  2.02it/s]
 25%|██▍       | 2675/10707 [26:12<1:06:16,  2.02it/s]
                                                      
{'loss': 3.9517, 'grad_norm': 0.22064341604709625, 'learning_rate': 0.0009331756231453494, 'epoch': 0.25}
-
 25%|██▍       | 2675/10707 [26:12<1:06:16,  2.02it/s]
 25%|██▍       | 2676/10707 [26:13<1:06:18,  2.02it/s]
 25%|██▌       | 2677/10707 [26:13<1:06:19,  2.02it/s]
 25%|██▌       | 2678/10707 [26:14<1:06:15,  2.02it/s]
 25%|██▌       | 2679/10707 [26:14<1:06:15,  2.02it/s]
 25%|██▌       | 2680/10707 [26:15<1:06:12,  2.02it/s]
 25%|██▌       | 2681/10707 [26:15<1:06:13,  2.02it/s]
 25%|██▌       | 2682/10707 [26:16<1:06:09,  2.02it/s]
 25%|██▌       | 2683/10707 [26:16<1:06:14,  2.02it/s]
 25%|██▌       | 2684/10707 [26:17<1:06:08,  2.02it/s]
 25%|██▌       | 2685/10707 [26:17<1:06:12,  2.02it/s]
 25%|██▌       | 2686/10707 [26:18<1:06:12,  2.02it/s]
 25%|██▌       | 2687/10707 [26:18<1:06:11,  2.02it/s]
 25%|██▌       | 2688/10707 [26:19<1:06:12,  2.02it/s]
 25%|██▌       | 2689/10707 [26:19<1:06:09,  2.02it/s]
 25%|██▌       | 2690/10707 [26:20<1:06:10,  2.02it/s]
 25%|██▌       | 2691/10707 [26:20<1:06:08,  2.02it/s]
 25%|██▌       | 2692/10707 [26:21<1:06:12,  2.02it/s]
 25%|██▌       | 2693/10707 [26:21<1:06:08,  2.02it/s]
 25%|██▌       | 2694/10707 [26:22<1:06:12,  2.02it/s]
 25%|██▌       | 2695/10707 [26:22<1:06:09,  2.02it/s]
 25%|██▌       | 2696/10707 [26:23<1:06:10,  2.02it/s]
 25%|██▌       | 2697/10707 [26:23<1:06:08,  2.02it/s]
 25%|██▌       | 2698/10707 [26:24<1:06:09,  2.02it/s]
 25%|██▌       | 2699/10707 [26:24<1:06:17,  2.01it/s]
 25%|██▌       | 2700/10707 [26:25<1:06:17,  2.01it/s]{'loss': 3.9449, 'grad_norm': 0.22764825820922852, 'learning_rate': 0.0009311258923901436, 'epoch': 0.25}
-
                                                      

 25%|██▌       | 2700/10707 [26:25<1:06:17,  2.01it/s]
 25%|██▌       | 2701/10707 [26:25<1:06:20,  2.01it/s]
 25%|██▌       | 2702/10707 [26:26<1:06:16,  2.01it/s]
 25%|██▌       | 2703/10707 [26:26<1:06:12,  2.01it/s]
 25%|██▌       | 2704/10707 [26:27<1:06:09,  2.02it/s]
 25%|██▌       | 2705/10707 [26:27<1:06:09,  2.02it/s]
 25%|██▌       | 2706/10707 [26:28<1:06:07,  2.02it/s]
 25%|██▌       | 2707/10707 [26:28<1:06:04,  2.02it/s]
 25%|██▌       | 2708/10707 [26:29<1:06:05,  2.02it/s]
 25%|██▌       | 2709/10707 [26:29<1:06:06,  2.02it/s]
 25%|██▌       | 2710/10707 [26:30<1:06:08,  2.02it/s]
 25%|██▌       | 2711/10707 [26:30<1:06:04,  2.02it/s]
 25%|██▌       | 2712/10707 [26:31<1:06:01,  2.02it/s]
 25%|██▌       | 2713/10707 [26:31<1:05:59,  2.02it/s]
 25%|██▌       | 2714/10707 [26:32<1:05:58,  2.02it/s]
 25%|██▌       | 2715/10707 [26:32<1:06:00,  2.02it/s]
 25%|██▌       | 2716/10707 [26:33<1:06:00,  2.02it/s]
 25%|██▌       | 2717/10707 [26:33<1:05:59,  2.02it/s]
 25%|██▌       | 2718/10707 [26:34<1:05:59,  2.02it/s]
 25%|██▌       | 2719/10707 [26:34<1:05:58,  2.02it/s]
 25%|██▌       | 2720/10707 [26:35<1:05:59,  2.02it/s]
 25%|██▌       | 2721/10707 [26:35<1:05:58,  2.02it/s]
 25%|██▌       | 2722/10707 [26:36<1:05:56,  2.02it/s]
 25%|██▌       | 2723/10707 [26:36<1:05:54,  2.02it/s]
 25%|██▌       | 2724/10707 [26:37<1:05:52,  2.02it/s]
 25%|██▌       | 2725/10707 [26:37<1:05:56,  2.02it/s]{'loss': 3.9421, 'grad_norm': 0.20620965957641602, 'learning_rate': 0.0009290475206544048, 'epoch': 0.25}
                                                      
-
 25%|██▌       | 2725/10707 [26:37<1:05:56,  2.02it/s]
 25%|██▌       | 2726/10707 [26:38<1:05:55,  2.02it/s]
 25%|██▌       | 2727/10707 [26:38<1:05:56,  2.02it/s]
 25%|██▌       | 2728/10707 [26:39<1:05:51,  2.02it/s]
 25%|██▌       | 2729/10707 [26:39<1:05:52,  2.02it/s]
 25%|██▌       | 2730/10707 [26:40<1:05:49,  2.02it/s]
 26%|██▌       | 2731/10707 [26:40<1:05:52,  2.02it/s]
 26%|██▌       | 2732/10707 [26:41<1:05:46,  2.02it/s]
 26%|██▌       | 2733/10707 [26:41<1:05:47,  2.02it/s]
 26%|██▌       | 2734/10707 [26:42<1:05:43,  2.02it/s]
 26%|██▌       | 2735/10707 [26:42<1:05:48,  2.02it/s]
 26%|██▌       | 2736/10707 [26:43<1:05:45,  2.02it/s]
 26%|██▌       | 2737/10707 [26:43<1:05:45,  2.02it/s]
 26%|██▌       | 2738/10707 [26:44<1:05:43,  2.02it/s]
 26%|██▌       | 2739/10707 [26:44<1:05:45,  2.02it/s]
 26%|██▌       | 2740/10707 [26:45<1:05:40,  2.02it/s]
 26%|██▌       | 2741/10707 [26:45<1:05:42,  2.02it/s]
 26%|██▌       | 2742/10707 [26:46<1:05:39,  2.02it/s]
 26%|██▌       | 2743/10707 [26:46<1:05:41,  2.02it/s]
 26%|██▌       | 2744/10707 [26:46<1:05:40,  2.02it/s]
 26%|██▌       | 2745/10707 [26:47<1:05:43,  2.02it/s]
 26%|██▌       | 2746/10707 [26:47<1:05:40,  2.02it/s]
 26%|██▌       | 2747/10707 [26:48<1:05:43,  2.02it/s]
 26%|██▌       | 2748/10707 [26:48<1:05:40,  2.02it/s]
 26%|██▌       | 2749/10707 [26:49<1:05:41,  2.02it/s]
 26%|██▌       | 2750/10707 [26:49<1:05:37,  2.02it/s]{'loss': 3.9349, 'grad_norm': 0.21151551604270935, 'learning_rate': 0.000926940646010574, 'epoch': 0.26}
                                                      
-
 26%|██▌       | 2750/10707 [26:49<1:05:37,  2.02it/s]
 26%|██▌       | 2751/10707 [26:50<1:05:43,  2.02it/s]
 26%|██▌       | 2752/10707 [26:50<1:05:45,  2.02it/s]
 26%|██▌       | 2753/10707 [26:51<1:05:38,  2.02it/s]
 26%|██▌       | 2754/10707 [26:51<1:05:38,  2.02it/s]
 26%|██▌       | 2755/10707 [26:52<1:05:34,  2.02it/s]
 26%|██▌       | 2756/10707 [26:52<1:05:41,  2.02it/s]
 26%|██▌       | 2757/10707 [26:53<1:05:41,  2.02it/s]
 26%|██▌       | 2758/10707 [26:53<1:05:41,  2.02it/s]
 26%|██▌       | 2759/10707 [26:54<1:05:36,  2.02it/s]
 26%|██▌       | 2760/10707 [26:54<1:05:39,  2.02it/s]
 26%|██▌       | 2761/10707 [26:55<1:05:36,  2.02it/s]
 26%|██▌       | 2762/10707 [26:55<1:05:39,  2.02it/s]
 26%|██▌       | 2763/10707 [26:56<1:05:37,  2.02it/s]
 26%|██▌       | 2764/10707 [26:56<1:05:35,  2.02it/s]
 26%|██▌       | 2765/10707 [26:57<1:05:35,  2.02it/s]
 26%|██▌       | 2766/10707 [26:57<1:05:33,  2.02it/s]
 26%|██▌       | 2767/10707 [26:58<1:05:34,  2.02it/s]
 26%|██▌       | 2768/10707 [26:58<1:05:28,  2.02it/s]
 26%|██▌       | 2769/10707 [26:59<1:05:35,  2.02it/s]
 26%|██▌       | 2770/10707 [26:59<1:05:29,  2.02it/s]
 26%|██▌       | 2771/10707 [27:00<1:05:31,  2.02it/s]
 26%|██▌       | 2772/10707 [27:00<1:05:27,  2.02it/s]
 26%|██▌       | 2773/10707 [27:01<1:05:32,  2.02it/s]
 26%|██▌       | 2774/10707 [27:01<1:05:29,  2.02it/s]
 26%|██▌       | 2775/10707 [27:02<1:05:31,  2.02it/s]{'loss': 3.945, 'grad_norm': 0.21159379184246063, 'learning_rate': 0.0009248054084246259, 'epoch': 0.26}
-
                                                      

 26%|██▌       | 2775/10707 [27:02<1:05:31,  2.02it/s]
 26%|██▌       | 2776/10707 [27:02<1:05:35,  2.02it/s]
 26%|██▌       | 2777/10707 [27:03<1:05:30,  2.02it/s]
 26%|██▌       | 2778/10707 [27:03<1:05:31,  2.02it/s]
 26%|██▌       | 2779/10707 [27:04<1:05:30,  2.02it/s]
 26%|██▌       | 2780/10707 [27:04<1:05:28,  2.02it/s]
 26%|██▌       | 2781/10707 [27:05<1:05:24,  2.02it/s]
 26%|██▌       | 2782/10707 [27:05<1:05:24,  2.02it/s]
 26%|██▌       | 2783/10707 [27:06<1:05:21,  2.02it/s]
 26%|██▌       | 2784/10707 [27:06<1:05:25,  2.02it/s]
 26%|██▌       | 2785/10707 [27:07<1:05:22,  2.02it/s]
 26%|██▌       | 2786/10707 [27:07<1:05:25,  2.02it/s]
 26%|██▌       | 2787/10707 [27:08<1:05:24,  2.02it/s]
 26%|██▌       | 2788/10707 [27:08<1:05:24,  2.02it/s]
 26%|██▌       | 2789/10707 [27:09<1:05:21,  2.02it/s]
 26%|██▌       | 2790/10707 [27:09<1:05:20,  2.02it/s]
 26%|██▌       | 2791/10707 [27:10<1:05:25,  2.02it/s]
 26%|██▌       | 2792/10707 [27:10<1:05:25,  2.02it/s]
 26%|██▌       | 2793/10707 [27:11<1:05:27,  2.02it/s]
 26%|██▌       | 2794/10707 [27:11<1:05:23,  2.02it/s]
 26%|██▌       | 2795/10707 [27:12<1:05:24,  2.02it/s]
 26%|██▌       | 2796/10707 [27:12<1:05:21,  2.02it/s]
 26%|██▌       | 2797/10707 [27:13<1:05:22,  2.02it/s]
 26%|██▌       | 2798/10707 [27:13<1:05:20,  2.02it/s]
 26%|██▌       | 2799/10707 [27:14<1:05:23,  2.02it/s]
 26%|██▌       | 2800/10707 [27:14<1:05:17,  2.02it/s]{'loss': 3.9338, 'grad_norm': 0.21858742833137512, 'learning_rate': 0.0009226419497467697, 'epoch': 0.26}
                                                      
-
 26%|██▌       | 2800/10707 [27:14<1:05:17,  2.02it/s]
 26%|██▌       | 2801/10707 [27:15<1:05:26,  2.01it/s]
 26%|██▌       | 2802/10707 [27:15<1:05:22,  2.02it/s]
 26%|██▌       | 2803/10707 [27:16<1:05:25,  2.01it/s]
 26%|██▌       | 2804/10707 [27:16<1:05:24,  2.01it/s]
 26%|██▌       | 2805/10707 [27:17<1:05:21,  2.01it/s]
 26%|██▌       | 2806/10707 [27:17<1:05:25,  2.01it/s]
 26%|██▌       | 2807/10707 [27:18<1:05:22,  2.01it/s]
 26%|██▌       | 2808/10707 [27:18<1:05:18,  2.02it/s]
 26%|██▌       | 2809/10707 [27:19<1:05:18,  2.02it/s]
 26%|██▌       | 2810/10707 [27:19<1:05:20,  2.01it/s]
 26%|██▋       | 2811/10707 [27:20<1:05:16,  2.02it/s]
 26%|██▋       | 2812/10707 [27:20<1:05:19,  2.01it/s]
 26%|██▋       | 2813/10707 [27:21<1:05:16,  2.02it/s]
 26%|██▋       | 2814/10707 [27:21<1:05:18,  2.01it/s]
 26%|██▋       | 2815/10707 [27:22<1:05:17,  2.01it/s]
 26%|██▋       | 2816/10707 [27:22<1:05:16,  2.01it/s]
 26%|██▋       | 2817/10707 [27:23<1:05:16,  2.01it/s]
 26%|██▋       | 2818/10707 [27:23<1:05:15,  2.01it/s]
 26%|██▋       | 2819/10707 [27:24<1:05:16,  2.01it/s]
 26%|██▋       | 2820/10707 [27:24<1:05:10,  2.02it/s]
 26%|██▋       | 2821/10707 [27:25<1:05:09,  2.02it/s]
 26%|██▋       | 2822/10707 [27:25<1:05:08,  2.02it/s]
 26%|██▋       | 2823/10707 [27:26<1:05:05,  2.02it/s]
 26%|██▋       | 2824/10707 [27:26<1:05:07,  2.02it/s]
 26%|██▋       | 2825/10707 [27:27<1:05:04,  2.02it/s]{'loss': 3.9283, 'grad_norm': 0.21338629722595215, 'learning_rate': 0.0009204504137020267, 'epoch': 0.26}
-
                                                      

 26%|██▋       | 2825/10707 [27:27<1:05:04,  2.02it/s]
 26%|██▋       | 2826/10707 [27:27<1:05:13,  2.01it/s]
 26%|██▋       | 2827/10707 [27:28<1:05:09,  2.02it/s]
 26%|██▋       | 2828/10707 [27:28<1:05:09,  2.02it/s]
 26%|██▋       | 2829/10707 [27:29<1:05:04,  2.02it/s]
 26%|██▋       | 2830/10707 [27:29<1:05:04,  2.02it/s]
 26%|██▋       | 2831/10707 [27:30<1:05:01,  2.02it/s]
 26%|██▋       | 2832/10707 [27:30<1:05:02,  2.02it/s]
 26%|██▋       | 2833/10707 [27:31<1:05:03,  2.02it/s]
 26%|██▋       | 2834/10707 [27:31<1:04:59,  2.02it/s]
 26%|██▋       | 2835/10707 [27:32<1:04:57,  2.02it/s]
 26%|██▋       | 2836/10707 [27:32<1:04:57,  2.02it/s]
 26%|██▋       | 2837/10707 [27:33<1:04:57,  2.02it/s]
 27%|██▋       | 2838/10707 [27:33<1:04:54,  2.02it/s]
 27%|██▋       | 2839/10707 [27:34<1:04:56,  2.02it/s]
 27%|██▋       | 2840/10707 [27:34<1:04:52,  2.02it/s]
 27%|██▋       | 2841/10707 [27:35<1:04:51,  2.02it/s]
 27%|██▋       | 2842/10707 [27:35<1:04:50,  2.02it/s]
 27%|██▋       | 2843/10707 [27:36<1:04:49,  2.02it/s]
 27%|██▋       | 2844/10707 [27:36<1:04:50,  2.02it/s]
 27%|██▋       | 2845/10707 [27:37<1:04:48,  2.02it/s]
 27%|██▋       | 2846/10707 [27:37<1:04:51,  2.02it/s]
 27%|██▋       | 2847/10707 [27:38<1:04:47,  2.02it/s]
 27%|██▋       | 2848/10707 [27:38<1:04:50,  2.02it/s]
 27%|██▋       | 2849/10707 [27:39<1:04:47,  2.02it/s]
 27%|██▋       | 2850/10707 [27:39<1:04:46,  2.02it/s]{'loss': 3.9261, 'grad_norm': 0.21374352276325226, 'learning_rate': 0.0009182309458806804, 'epoch': 0.27}
                                                      
-
 27%|██▋       | 2850/10707 [27:39<1:04:46,  2.02it/s]
 27%|██▋       | 2851/10707 [27:40<1:04:48,  2.02it/s]
 27%|██▋       | 2852/10707 [27:40<1:04:52,  2.02it/s]
 27%|██▋       | 2853/10707 [27:41<1:04:46,  2.02it/s]
 27%|██▋       | 2854/10707 [27:41<1:04:50,  2.02it/s]
 27%|██▋       | 2855/10707 [27:42<1:04:45,  2.02it/s]
 27%|██▋       | 2856/10707 [27:42<1:04:49,  2.02it/s]
 27%|██▋       | 2857/10707 [27:42<1:04:42,  2.02it/s]
 27%|██▋       | 2858/10707 [27:43<1:04:47,  2.02it/s]
 27%|██▋       | 2859/10707 [27:43<1:04:45,  2.02it/s]
 27%|██▋       | 2860/10707 [27:44<1:04:45,  2.02it/s]
 27%|██▋       | 2861/10707 [27:44<1:04:45,  2.02it/s]
 27%|██▋       | 2862/10707 [27:45<1:04:44,  2.02it/s]
 27%|██▋       | 2863/10707 [27:45<1:04:40,  2.02it/s]
 27%|██▋       | 2864/10707 [27:46<1:04:40,  2.02it/s]
 27%|██▋       | 2865/10707 [27:46<1:04:44,  2.02it/s]
 27%|██▋       | 2866/10707 [27:47<1:04:45,  2.02it/s]
 27%|██▋       | 2867/10707 [27:47<1:04:43,  2.02it/s]
 27%|██▋       | 2868/10707 [27:48<1:04:40,  2.02it/s]
 27%|██▋       | 2869/10707 [27:48<1:04:43,  2.02it/s]
 27%|██▋       | 2870/10707 [27:49<1:04:39,  2.02it/s]
 27%|██▋       | 2871/10707 [27:49<1:04:41,  2.02it/s]
 27%|██▋       | 2872/10707 [27:50<1:04:36,  2.02it/s]
 27%|██▋       | 2873/10707 [27:50<1:04:40,  2.02it/s]
 27%|██▋       | 2874/10707 [27:51<1:04:39,  2.02it/s]
 27%|██▋       | 2875/10707 [27:51<1:04:42,  2.02it/s]{'loss': 3.9199, 'grad_norm': 0.21764802932739258, 'learning_rate': 0.0009159836937286069, 'epoch': 0.27}
-
                                                      

 27%|██▋       | 2875/10707 [27:51<1:04:42,  2.02it/s]
 27%|██▋       | 2876/10707 [27:52<1:04:48,  2.01it/s]
 27%|██▋       | 2877/10707 [27:52<1:04:42,  2.02it/s]
 27%|██▋       | 2878/10707 [27:53<1:04:42,  2.02it/s]
 27%|██▋       | 2879/10707 [27:53<1:04:40,  2.02it/s]
 27%|██▋       | 2880/10707 [27:54<1:04:39,  2.02it/s]
 27%|██▋       | 2881/10707 [27:54<1:04:37,  2.02it/s]
 27%|██▋       | 2882/10707 [27:55<1:04:35,  2.02it/s]
 27%|██▋       | 2883/10707 [27:55<1:04:32,  2.02it/s]
 27%|██▋       | 2884/10707 [27:56<1:04:34,  2.02it/s]
 27%|██▋       | 2885/10707 [27:56<1:04:31,  2.02it/s]
 27%|██▋       | 2886/10707 [27:57<1:04:31,  2.02it/s]
 27%|██▋       | 2887/10707 [27:57<1:04:30,  2.02it/s]
 27%|██▋       | 2888/10707 [27:58<1:04:31,  2.02it/s]
 27%|██▋       | 2889/10707 [27:58<1:04:30,  2.02it/s]
 27%|██▋       | 2890/10707 [27:59<1:04:33,  2.02it/s]
 27%|██▋       | 2891/10707 [27:59<1:04:31,  2.02it/s]
 27%|██▋       | 2892/10707 [28:00<1:04:33,  2.02it/s]
 27%|██▋       | 2893/10707 [28:00<1:04:31,  2.02it/s]
 27%|██▋       | 2894/10707 [28:01<1:04:31,  2.02it/s]
 27%|██▋       | 2895/10707 [28:01<1:04:32,  2.02it/s]
 27%|██▋       | 2896/10707 [28:02<1:04:32,  2.02it/s]
 27%|██▋       | 2897/10707 [28:02<1:04:30,  2.02it/s]
 27%|██▋       | 2898/10707 [28:03<1:04:36,  2.01it/s]
 27%|██▋       | 2899/10707 [28:03<1:04:32,  2.02it/s]
 27%|██▋       | 2900/10707 [28:04<1:04:28,  2.02it/s]{'loss': 3.9289, 'grad_norm': 0.21996337175369263, 'learning_rate': 0.0009137088065374776, 'epoch': 0.27}
-
                                                      

 27%|██▋       | 2900/10707 [28:04<1:04:28,  2.02it/s]
 27%|██▋       | 2901/10707 [28:04<1:04:31,  2.02it/s]
 27%|██▋       | 2902/10707 [28:05<1:04:31,  2.02it/s]
 27%|██▋       | 2903/10707 [28:05<1:04:24,  2.02it/s]
 27%|██▋       | 2904/10707 [28:06<1:04:24,  2.02it/s]
 27%|██▋       | 2905/10707 [28:06<1:04:22,  2.02it/s]
 27%|██▋       | 2906/10707 [28:07<1:04:22,  2.02it/s]
 27%|██▋       | 2907/10707 [28:07<1:04:19,  2.02it/s]
 27%|██▋       | 2908/10707 [28:08<1:04:20,  2.02it/s]
 27%|██▋       | 2909/10707 [28:08<1:04:21,  2.02it/s]
 27%|██▋       | 2910/10707 [28:09<1:04:21,  2.02it/s]
 27%|██▋       | 2911/10707 [28:09<1:04:19,  2.02it/s]
 27%|██▋       | 2912/10707 [28:10<1:04:20,  2.02it/s]
 27%|██▋       | 2913/10707 [28:10<1:04:20,  2.02it/s]
 27%|██▋       | 2914/10707 [28:11<1:04:16,  2.02it/s]
 27%|██▋       | 2915/10707 [28:11<1:04:16,  2.02it/s]
 27%|██▋       | 2916/10707 [28:12<1:04:14,  2.02it/s]
 27%|██▋       | 2917/10707 [28:12<1:04:23,  2.02it/s]
 27%|██▋       | 2918/10707 [28:13<1:04:19,  2.02it/s]
 27%|██▋       | 2919/10707 [28:13<1:04:18,  2.02it/s]
 27%|██▋       | 2920/10707 [28:14<1:04:18,  2.02it/s]
 27%|██▋       | 2921/10707 [28:14<1:04:17,  2.02it/s]
 27%|██▋       | 2922/10707 [28:15<1:04:16,  2.02it/s]
 27%|██▋       | 2923/10707 [28:15<1:04:20,  2.02it/s]
 27%|██▋       | 2924/10707 [28:16<1:04:20,  2.02it/s]
 27%|██▋       | 2925/10707 [28:16<1:04:17,  2.02it/s]{'loss': 3.9216, 'grad_norm': 0.21794037520885468, 'learning_rate': 0.0009114064354348422, 'epoch': 0.27}
-
                                                      

 27%|██▋       | 2925/10707 [28:16<1:04:17,  2.02it/s]
 27%|██▋       | 2926/10707 [28:17<1:04:22,  2.01it/s]
 27%|██▋       | 2927/10707 [28:17<1:04:21,  2.01it/s]
 27%|██▋       | 2928/10707 [28:18<1:04:20,  2.02it/s]
 27%|██▋       | 2929/10707 [28:18<1:04:16,  2.02it/s]
 27%|██▋       | 2930/10707 [28:19<1:04:13,  2.02it/s]
 27%|██▋       | 2931/10707 [28:19<1:04:11,  2.02it/s]
 27%|██▋       | 2932/10707 [28:20<1:04:09,  2.02it/s]
 27%|██▋       | 2933/10707 [28:20<1:04:08,  2.02it/s]
 27%|██▋       | 2934/10707 [28:21<1:04:09,  2.02it/s]
 27%|██▋       | 2935/10707 [28:21<1:04:11,  2.02it/s]
 27%|██▋       | 2936/10707 [28:22<1:04:15,  2.02it/s]
 27%|██▋       | 2937/10707 [28:22<1:04:11,  2.02it/s]
 27%|██▋       | 2938/10707 [28:23<1:04:14,  2.02it/s]
 27%|██▋       | 2939/10707 [28:23<1:04:11,  2.02it/s]
 27%|██▋       | 2940/10707 [28:24<1:04:09,  2.02it/s]
 27%|██▋       | 2941/10707 [28:24<1:04:07,  2.02it/s]
 27%|██▋       | 2942/10707 [28:25<1:04:04,  2.02it/s]
 27%|██▋       | 2943/10707 [28:25<1:04:05,  2.02it/s]
 27%|██▋       | 2944/10707 [28:26<1:04:06,  2.02it/s]
 28%|██▊       | 2945/10707 [28:26<1:04:08,  2.02it/s]
 28%|██▊       | 2946/10707 [28:27<1:04:06,  2.02it/s]
 28%|██▊       | 2947/10707 [28:27<1:04:04,  2.02it/s]
 28%|██▊       | 2948/10707 [28:28<1:04:02,  2.02it/s]
 28%|██▊       | 2949/10707 [28:28<1:04:05,  2.02it/s]
 28%|██▊       | 2950/10707 [28:29<1:04:04,  2.02it/s]{'loss': 3.9185, 'grad_norm': 0.19962403178215027, 'learning_rate': 0.0009090767333740889, 'epoch': 0.28}
-
                                                      

 28%|██▊       | 2950/10707 [28:29<1:04:04,  2.02it/s]
 28%|██▊       | 2951/10707 [28:29<1:04:09,  2.01it/s]
 28%|██▊       | 2952/10707 [28:30<1:04:07,  2.02it/s]
 28%|██▊       | 2953/10707 [28:30<1:04:05,  2.02it/s]
 28%|██▊       | 2954/10707 [28:31<1:04:02,  2.02it/s]
 28%|██▊       | 2955/10707 [28:31<1:03:59,  2.02it/s]
 28%|██▊       | 2956/10707 [28:32<1:03:58,  2.02it/s]
 28%|██▊       | 2957/10707 [28:32<1:03:56,  2.02it/s]
 28%|██▊       | 2958/10707 [28:33<1:03:57,  2.02it/s]
 28%|██▊       | 2959/10707 [28:33<1:03:56,  2.02it/s]
 28%|██▊       | 2960/10707 [28:34<1:03:58,  2.02it/s]
 28%|██▊       | 2961/10707 [28:34<1:03:57,  2.02it/s]
 28%|██▊       | 2962/10707 [28:35<1:03:57,  2.02it/s]
 28%|██▊       | 2963/10707 [28:35<1:03:55,  2.02it/s]
 28%|██▊       | 2964/10707 [28:36<1:03:56,  2.02it/s]
 28%|██▊       | 2965/10707 [28:36<1:03:53,  2.02it/s]
 28%|██▊       | 2966/10707 [28:36<1:03:52,  2.02it/s]
 28%|██▊       | 2967/10707 [28:37<1:03:51,  2.02it/s]
 28%|██▊       | 2968/10707 [28:37<1:03:54,  2.02it/s]
 28%|██▊       | 2969/10707 [28:38<1:03:48,  2.02it/s]
 28%|██▊       | 2970/10707 [28:38<1:03:51,  2.02it/s]
 28%|██▊       | 2971/10707 [28:39<1:03:47,  2.02it/s]
 28%|██▊       | 2972/10707 [28:39<1:03:47,  2.02it/s]
 28%|██▊       | 2973/10707 [28:40<1:03:52,  2.02it/s]
 28%|██▊       | 2974/10707 [28:40<1:03:53,  2.02it/s]
 28%|██▊       | 2975/10707 [28:41<1:03:53,  2.02it/s]{'loss': 3.9109, 'grad_norm': 0.21586069464683533, 'learning_rate': 0.0009067198551242827, 'epoch': 0.28}
-
                                                      

 28%|██▊       | 2975/10707 [28:41<1:03:53,  2.02it/s]
 28%|██▊       | 2976/10707 [28:41<1:03:55,  2.02it/s]
 28%|██▊       | 2977/10707 [28:42<1:03:57,  2.01it/s]
 28%|██▊       | 2978/10707 [28:42<1:03:51,  2.02it/s]
 28%|██▊       | 2979/10707 [28:43<1:03:52,  2.02it/s]
 28%|██▊       | 2980/10707 [28:43<1:03:51,  2.02it/s]
 28%|██▊       | 2981/10707 [28:44<1:03:50,  2.02it/s]
 28%|██▊       | 2982/10707 [28:44<1:03:46,  2.02it/s]
 28%|██▊       | 2983/10707 [28:45<1:03:49,  2.02it/s]
 28%|██▊       | 2984/10707 [28:45<1:03:45,  2.02it/s]
 28%|██▊       | 2985/10707 [28:46<1:03:45,  2.02it/s]
 28%|██▊       | 2986/10707 [28:46<1:03:41,  2.02it/s]
 28%|██▊       | 2987/10707 [28:47<1:03:41,  2.02it/s]
 28%|██▊       | 2988/10707 [28:47<1:03:43,  2.02it/s]
 28%|██▊       | 2989/10707 [28:48<1:03:44,  2.02it/s]
 28%|██▊       | 2990/10707 [28:48<1:03:46,  2.02it/s]
 28%|██▊       | 2991/10707 [28:49<1:03:42,  2.02it/s]
 28%|██▊       | 2992/10707 [28:49<1:03:43,  2.02it/s]
 28%|██▊       | 2993/10707 [28:50<1:03:40,  2.02it/s]
 28%|██▊       | 2994/10707 [28:50<1:03:42,  2.02it/s]
 28%|██▊       | 2995/10707 [28:51<1:03:40,  2.02it/s]
 28%|██▊       | 2996/10707 [28:51<1:03:39,  2.02it/s]
 28%|██▊       | 2997/10707 [28:52<1:03:40,  2.02it/s]
 28%|██▊       | 2998/10707 [28:52<1:03:41,  2.02it/s]
 28%|██▊       | 2999/10707 [28:53<1:03:39,  2.02it/s]
 28%|██▊       | 3000/10707 [28:53<1:03:38,  2.02it/s]
                                                      {'loss': 3.9033, 'grad_norm': 0.2383001446723938, 'learning_rate': 0.0009043359572598848, 'epoch': 0.28}
-
 28%|██▊       | 3000/10707 [28:53<1:03:38,  2.02it/s]
 28%|██▊       | 3001/10707 [28:54<1:03:40,  2.02it/s]
 28%|██▊       | 3002/10707 [28:54<1:03:45,  2.01it/s]
 28%|██▊       | 3003/10707 [28:55<1:03:39,  2.02it/s]
 28%|██▊       | 3004/10707 [28:55<1:03:42,  2.02it/s]
 28%|██▊       | 3005/10707 [28:56<1:03:36,  2.02it/s]
 28%|██▊       | 3006/10707 [28:56<1:03:34,  2.02it/s]
 28%|██▊       | 3007/10707 [28:57<1:03:29,  2.02it/s]
 28%|██▊       | 3008/10707 [28:57<1:03:30,  2.02it/s]
 28%|██▊       | 3009/10707 [28:58<1:03:27,  2.02it/s]
 28%|██▊       | 3010/10707 [28:58<1:03:30,  2.02it/s]
 28%|██▊       | 3011/10707 [28:59<1:03:25,  2.02it/s]
 28%|██▊       | 3012/10707 [28:59<1:03:27,  2.02it/s]
 28%|██▊       | 3013/10707 [29:00<1:03:25,  2.02it/s]
 28%|██▊       | 3014/10707 [29:00<1:03:29,  2.02it/s]
 28%|██▊       | 3015/10707 [29:01<1:03:24,  2.02it/s]
 28%|██▊       | 3016/10707 [29:01<1:03:24,  2.02it/s]
 28%|██▊       | 3017/10707 [29:02<1:03:22,  2.02it/s]
 28%|██▊       | 3018/10707 [29:02<1:03:24,  2.02it/s]
 28%|██▊       | 3019/10707 [29:03<1:03:31,  2.02it/s]
 28%|██▊       | 3020/10707 [29:03<1:03:28,  2.02it/s]
 28%|██▊       | 3021/10707 [29:04<1:03:30,  2.02it/s]
 28%|██▊       | 3022/10707 [29:04<1:03:28,  2.02it/s]
 28%|██▊       | 3023/10707 [29:05<1:03:27,  2.02it/s]
 28%|██▊       | 3024/10707 [29:05<1:03:29,  2.02it/s]
 28%|██▊       | 3025/10707 [29:06<1:03:27,  2.02it/s]
                                                      
{'loss': 3.9182, 'grad_norm': 0.20144227147102356, 'learning_rate': 0.000901925198150349, 'epoch': 0.28}
-
 28%|██▊       | 3025/10707 [29:06<1:03:27,  2.02it/s]
 28%|██▊       | 3026/10707 [29:06<1:03:30,  2.02it/s]
 28%|██▊       | 3027/10707 [29:07<1:03:27,  2.02it/s]
 28%|██▊       | 3028/10707 [29:07<1:03:21,  2.02it/s]
 28%|██▊       | 3029/10707 [29:08<1:03:23,  2.02it/s]
 28%|██▊       | 3030/10707 [29:08<1:03:19,  2.02it/s]
 28%|██▊       | 3031/10707 [29:09<1:03:20,  2.02it/s]
 28%|██▊       | 3032/10707 [29:09<1:03:20,  2.02it/s]
 28%|██▊       | 3033/10707 [29:10<1:03:24,  2.02it/s]
 28%|██▊       | 3034/10707 [29:10<1:03:20,  2.02it/s]
 28%|██▊       | 3035/10707 [29:11<1:03:31,  2.01it/s]
 28%|██▊       | 3036/10707 [29:11<1:03:27,  2.01it/s]
 28%|██▊       | 3037/10707 [29:12<1:03:25,  2.02it/s]
 28%|██▊       | 3038/10707 [29:12<1:03:19,  2.02it/s]
 28%|██▊       | 3039/10707 [29:13<1:03:19,  2.02it/s]
 28%|██▊       | 3040/10707 [29:13<1:03:18,  2.02it/s]
 28%|██▊       | 3041/10707 [29:14<1:03:18,  2.02it/s]
 28%|██▊       | 3042/10707 [29:14<1:03:15,  2.02it/s]
 28%|██▊       | 3043/10707 [29:15<1:03:18,  2.02it/s]
 28%|██▊       | 3044/10707 [29:15<1:03:22,  2.02it/s]
 28%|██▊       | 3045/10707 [29:16<1:03:19,  2.02it/s]
 28%|██▊       | 3046/10707 [29:16<1:03:16,  2.02it/s]
 28%|██▊       | 3047/10707 [29:17<1:03:17,  2.02it/s]
 28%|██▊       | 3048/10707 [29:17<1:03:13,  2.02it/s]
 28%|██▊       | 3049/10707 [29:18<1:03:14,  2.02it/s]
 28%|██▊       | 3050/10707 [29:18<1:03:11,  2.02it/s]{'loss': 3.8999, 'grad_norm': 0.2004518061876297, 'learning_rate': 0.0008994877379496025, 'epoch': 0.28}
-
                                                      

 28%|██▊       | 3050/10707 [29:18<1:03:11,  2.02it/s]
 28%|██▊       | 3051/10707 [29:19<1:03:17,  2.02it/s]
 29%|██▊       | 3052/10707 [29:19<1:03:09,  2.02it/s]
 29%|██▊       | 3053/10707 [29:20<1:03:13,  2.02it/s]
 29%|██▊       | 3054/10707 [29:20<1:03:10,  2.02it/s]
 29%|██▊       | 3055/10707 [29:21<1:03:13,  2.02it/s]
 29%|██▊       | 3056/10707 [29:21<1:03:13,  2.02it/s]
 29%|██▊       | 3057/10707 [29:22<1:03:12,  2.02it/s]
 29%|██▊       | 3058/10707 [29:22<1:03:11,  2.02it/s]
 29%|██▊       | 3059/10707 [29:23<1:03:11,  2.02it/s]
 29%|██▊       | 3060/10707 [29:23<1:03:08,  2.02it/s]
 29%|██▊       | 3061/10707 [29:24<1:03:07,  2.02it/s]
 29%|██▊       | 3062/10707 [29:24<1:03:06,  2.02it/s]
 29%|██▊       | 3063/10707 [29:25<1:03:03,  2.02it/s]
 29%|██▊       | 3064/10707 [29:25<1:03:04,  2.02it/s]
 29%|██▊       | 3065/10707 [29:26<1:03:00,  2.02it/s]
 29%|██▊       | 3066/10707 [29:26<1:03:00,  2.02it/s]
 29%|██▊       | 3067/10707 [29:27<1:03:02,  2.02it/s]
 29%|██▊       | 3068/10707 [29:27<1:03:00,  2.02it/s]
 29%|██▊       | 3069/10707 [29:28<1:02:59,  2.02it/s]
 29%|██▊       | 3070/10707 [29:28<1:02:59,  2.02it/s]
 29%|██▊       | 3071/10707 [29:29<1:03:03,  2.02it/s]
 29%|██▊       | 3072/10707 [29:29<1:03:01,  2.02it/s]
 29%|██▊       | 3073/10707 [29:30<1:03:00,  2.02it/s]
 29%|██▊       | 3074/10707 [29:30<1:02:59,  2.02it/s]
 29%|██▊       | 3075/10707 [29:30<1:03:00,  2.02it/s]{'loss': 3.9046, 'grad_norm': 0.2321648746728897, 'learning_rate': 0.0008970237385854058, 'epoch': 0.29}
-
                                                      

 29%|██▊       | 3075/10707 [29:31<1:03:00,  2.02it/s]
 29%|██▊       | 3076/10707 [29:31<1:03:01,  2.02it/s]
 29%|██▊       | 3077/10707 [29:31<1:03:03,  2.02it/s]
 29%|██▊       | 3078/10707 [29:32<1:03:01,  2.02it/s]
 29%|██▉       | 3079/10707 [29:32<1:03:02,  2.02it/s]
 29%|██▉       | 3080/10707 [29:33<1:02:55,  2.02it/s]
 29%|██▉       | 3081/10707 [29:33<1:02:56,  2.02it/s]
 29%|██▉       | 3082/10707 [29:34<1:02:54,  2.02it/s]
 29%|██▉       | 3083/10707 [29:34<1:02:53,  2.02it/s]
 29%|██▉       | 3084/10707 [29:35<1:02:53,  2.02it/s]
 29%|██▉       | 3085/10707 [29:35<1:02:53,  2.02it/s]
 29%|██▉       | 3086/10707 [29:36<1:02:55,  2.02it/s]
 29%|██▉       | 3087/10707 [29:36<1:02:56,  2.02it/s]
 29%|██▉       | 3088/10707 [29:37<1:03:00,  2.02it/s]
 29%|██▉       | 3089/10707 [29:37<1:02:54,  2.02it/s]
 29%|██▉       | 3090/10707 [29:38<1:02:55,  2.02it/s]
 29%|██▉       | 3091/10707 [29:38<1:02:50,  2.02it/s]
 29%|██▉       | 3092/10707 [29:39<1:02:48,  2.02it/s]
 29%|██▉       | 3093/10707 [29:39<1:02:45,  2.02it/s]
 29%|██▉       | 3094/10707 [29:40<1:02:54,  2.02it/s]
 29%|██▉       | 3095/10707 [29:40<1:02:53,  2.02it/s]
 29%|██▉       | 3096/10707 [29:41<1:02:52,  2.02it/s]
 29%|██▉       | 3097/10707 [29:41<1:02:52,  2.02it/s]
 29%|██▉       | 3098/10707 [29:42<1:02:51,  2.02it/s]
 29%|██▉       | 3099/10707 [29:42<1:02:51,  2.02it/s]
 29%|██▉       | 3100/10707 [29:43<1:02:47,  2.02it/s]{'loss': 3.9059, 'grad_norm': 0.21371042728424072, 'learning_rate': 0.0008945333637485947, 'epoch': 0.29}
                                                      
-
 29%|██▉       | 3100/10707 [29:43<1:02:47,  2.02it/s]
 29%|██▉       | 3101/10707 [29:43<1:02:54,  2.02it/s]
 29%|██▉       | 3102/10707 [29:44<1:02:50,  2.02it/s]
 29%|██▉       | 3103/10707 [29:44<1:02:50,  2.02it/s]
 29%|██▉       | 3104/10707 [29:45<1:02:53,  2.01it/s]
 29%|██▉       | 3105/10707 [29:45<1:02:53,  2.01it/s]
 29%|██▉       | 3106/10707 [29:46<1:02:50,  2.02it/s]
 29%|██▉       | 3107/10707 [29:46<1:02:45,  2.02it/s]
 29%|██▉       | 3108/10707 [29:47<1:02:45,  2.02it/s]
 29%|██▉       | 3109/10707 [29:47<1:02:41,  2.02it/s]
 29%|██▉       | 3110/10707 [29:48<1:02:42,  2.02it/s]
 29%|██▉       | 3111/10707 [29:48<1:02:39,  2.02it/s]
 29%|██▉       | 3112/10707 [29:49<1:02:41,  2.02it/s]
 29%|██▉       | 3113/10707 [29:49<1:02:40,  2.02it/s]
 29%|██▉       | 3114/10707 [29:50<1:02:41,  2.02it/s]
 29%|██▉       | 3115/10707 [29:50<1:02:37,  2.02it/s]
 29%|██▉       | 3116/10707 [29:51<1:02:38,  2.02it/s]
 29%|██▉       | 3117/10707 [29:51<1:02:34,  2.02it/s]
 29%|██▉       | 3118/10707 [29:52<1:02:41,  2.02it/s]
 29%|██▉       | 3119/10707 [29:52<1:02:38,  2.02it/s]
 29%|██▉       | 3120/10707 [29:53<1:02:38,  2.02it/s]
 29%|██▉       | 3121/10707 [29:53<1:02:34,  2.02it/s]
 29%|██▉       | 3122/10707 [29:54<1:02:35,  2.02it/s]
 29%|██▉       | 3123/10707 [29:54<1:02:35,  2.02it/s]
 29%|██▉       | 3124/10707 [29:55<1:02:34,  2.02it/s]
 29%|██▉       | 3125/10707 [29:55<1:02:30,  2.02it/s]{'loss': 3.9033, 'grad_norm': 0.19821694493293762, 'learning_rate': 0.0008920167788822071, 'epoch': 0.29}
-
                                                      

 29%|██▉       | 3125/10707 [29:55<1:02:30,  2.02it/s]
 29%|██▉       | 3126/10707 [29:56<1:02:36,  2.02it/s]
 29%|██▉       | 3127/10707 [29:56<1:02:35,  2.02it/s]
 29%|██▉       | 3128/10707 [29:57<1:02:34,  2.02it/s]
 29%|██▉       | 3129/10707 [29:57<1:02:33,  2.02it/s]
 29%|██▉       | 3130/10707 [29:58<1:02:33,  2.02it/s]
 29%|██▉       | 3131/10707 [29:58<1:02:36,  2.02it/s]
 29%|██▉       | 3132/10707 [29:59<1:02:31,  2.02it/s]
 29%|██▉       | 3133/10707 [29:59<1:02:29,  2.02it/s]
 29%|██▉       | 3134/10707 [30:00<1:02:25,  2.02it/s]
 29%|██▉       | 3135/10707 [30:00<1:02:31,  2.02it/s]
 29%|██▉       | 3136/10707 [30:01<1:02:31,  2.02it/s]
 29%|██▉       | 3137/10707 [30:01<1:02:32,  2.02it/s]
 29%|██▉       | 3138/10707 [30:02<1:02:32,  2.02it/s]
 29%|██▉       | 3139/10707 [30:02<1:02:34,  2.02it/s]
 29%|██▉       | 3140/10707 [30:03<1:02:31,  2.02it/s]
 29%|██▉       | 3141/10707 [30:03<1:02:29,  2.02it/s]
 29%|██▉       | 3142/10707 [30:04<1:02:31,  2.02it/s]
 29%|██▉       | 3143/10707 [30:04<1:02:35,  2.01it/s]
 29%|██▉       | 3144/10707 [30:05<1:02:35,  2.01it/s]
 29%|██▉       | 3145/10707 [30:05<1:02:33,  2.01it/s]
 29%|██▉       | 3146/10707 [30:06<1:02:31,  2.02it/s]
 29%|██▉       | 3147/10707 [30:06<1:02:34,  2.01it/s]
 29%|██▉       | 3148/10707 [30:07<1:02:34,  2.01it/s]
 29%|██▉       | 3149/10707 [30:07<1:02:31,  2.01it/s]
 29%|██▉       | 3150/10707 [30:08<1:02:33,  2.01it/s]{'loss': 3.8906, 'grad_norm': 0.20303502678871155, 'learning_rate': 0.0008894741511704912, 'epoch': 0.29}
-
                                                      

 29%|██▉       | 3150/10707 [30:08<1:02:33,  2.01it/s]
 29%|██▉       | 3151/10707 [30:08<1:02:33,  2.01it/s]
 29%|██▉       | 3152/10707 [30:09<1:02:31,  2.01it/s]
 29%|██▉       | 3153/10707 [30:09<1:02:30,  2.01it/s]
 29%|██▉       | 3154/10707 [30:10<1:02:31,  2.01it/s]
 29%|██▉       | 3155/10707 [30:10<1:02:29,  2.01it/s]
 29%|██▉       | 3156/10707 [30:11<1:02:27,  2.01it/s]
 29%|██▉       | 3157/10707 [30:11<1:02:25,  2.02it/s]
 29%|██▉       | 3158/10707 [30:12<1:02:23,  2.02it/s]
 30%|██▉       | 3159/10707 [30:12<1:02:21,  2.02it/s]
 30%|██▉       | 3160/10707 [30:13<1:02:18,  2.02it/s]
 30%|██▉       | 3161/10707 [30:13<1:02:16,  2.02it/s]
 30%|██▉       | 3162/10707 [30:14<1:02:18,  2.02it/s]
 30%|██▉       | 3163/10707 [30:14<1:02:18,  2.02it/s]
 30%|██▉       | 3164/10707 [30:15<1:02:18,  2.02it/s]
 30%|██▉       | 3165/10707 [30:15<1:02:18,  2.02it/s]
 30%|██▉       | 3166/10707 [30:16<1:02:17,  2.02it/s]
 30%|██▉       | 3167/10707 [30:16<1:02:15,  2.02it/s]
 30%|██▉       | 3168/10707 [30:17<1:02:14,  2.02it/s]
 30%|██▉       | 3169/10707 [30:17<1:02:17,  2.02it/s]
 30%|██▉       | 3170/10707 [30:18<1:02:19,  2.02it/s]
 30%|██▉       | 3171/10707 [30:18<1:02:21,  2.01it/s]
 30%|██▉       | 3172/10707 [30:19<1:02:16,  2.02it/s]
 30%|██▉       | 3173/10707 [30:19<1:02:18,  2.02it/s]
 30%|██▉       | 3174/10707 [30:20<1:02:12,  2.02it/s]
 30%|██▉       | 3175/10707 [30:20<1:02:17,  2.02it/s]{'loss': 3.8958, 'grad_norm': 0.21060453355312347, 'learning_rate': 0.0008869056495277987, 'epoch': 0.3}
-
                                                      

 30%|██▉       | 3175/10707 [30:20<1:02:17,  2.02it/s]
 30%|██▉       | 3176/10707 [30:21<1:02:16,  2.02it/s]
 30%|██▉       | 3177/10707 [30:21<1:02:16,  2.02it/s]
 30%|██▉       | 3178/10707 [30:22<1:02:16,  2.02it/s]
 30%|██▉       | 3179/10707 [30:22<1:02:15,  2.02it/s]
 30%|██▉       | 3180/10707 [30:23<1:02:16,  2.01it/s]
 30%|██▉       | 3181/10707 [30:23<1:02:15,  2.01it/s]
 30%|██▉       | 3182/10707 [30:24<1:02:08,  2.02it/s]
 30%|██▉       | 3183/10707 [30:24<1:02:08,  2.02it/s]
 30%|██▉       | 3184/10707 [30:25<1:02:06,  2.02it/s]
 30%|██▉       | 3185/10707 [30:25<1:02:06,  2.02it/s]
 30%|██▉       | 3186/10707 [30:26<1:02:04,  2.02it/s]
 30%|██▉       | 3187/10707 [30:26<1:02:07,  2.02it/s]
 30%|██▉       | 3188/10707 [30:27<1:02:00,  2.02it/s]
 30%|██▉       | 3189/10707 [30:27<1:02:00,  2.02it/s]
 30%|██▉       | 3190/10707 [30:27<1:01:58,  2.02it/s]
 30%|██▉       | 3191/10707 [30:28<1:02:00,  2.02it/s]
 30%|██▉       | 3192/10707 [30:28<1:01:54,  2.02it/s]
 30%|██▉       | 3193/10707 [30:29<1:01:58,  2.02it/s]
 30%|██▉       | 3194/10707 [30:29<1:01:55,  2.02it/s]
 30%|██▉       | 3195/10707 [30:30<1:01:58,  2.02it/s]
 30%|██▉       | 3196/10707 [30:30<1:01:56,  2.02it/s]
 30%|██▉       | 3197/10707 [30:31<1:11:55,  1.74it/s]
 30%|██▉       | 3198/10707 [30:32<1:08:56,  1.82it/s]
 30%|██▉       | 3199/10707 [30:32<1:06:52,  1.87it/s]
 30%|██▉       | 3200/10707 [30:33<1:05:24,  1.91it/s]{'loss': 3.8931, 'grad_norm': 0.2036726474761963, 'learning_rate': 0.0008843114445873646, 'epoch': 0.3}
-
                                                      

 30%|██▉       | 3200/10707 [30:33<1:05:24,  1.91it/s]
 30%|██▉       | 3201/10707 [30:33<1:04:25,  1.94it/s]
 30%|██▉       | 3202/10707 [30:34<1:03:41,  1.96it/s]
 30%|██▉       | 3203/10707 [30:34<1:03:10,  1.98it/s]
 30%|██▉       | 3204/10707 [30:35<1:02:47,  1.99it/s]
 30%|██▉       | 3205/10707 [30:35<1:02:33,  2.00it/s]
 30%|██▉       | 3206/10707 [30:36<1:02:19,  2.01it/s]
 30%|██▉       | 3207/10707 [30:36<1:02:15,  2.01it/s]
 30%|██▉       | 3208/10707 [30:37<1:02:05,  2.01it/s]
 30%|██▉       | 3209/10707 [30:37<1:02:05,  2.01it/s]
 30%|██▉       | 3210/10707 [30:38<1:12:18,  1.73it/s]
 30%|██▉       | 3211/10707 [30:38<1:09:06,  1.81it/s]
 30%|██▉       | 3212/10707 [30:39<1:06:56,  1.87it/s]
 30%|███       | 3213/10707 [30:39<1:05:24,  1.91it/s]
 30%|███       | 3214/10707 [30:40<1:04:19,  1.94it/s]
 30%|███       | 3215/10707 [30:40<1:03:36,  1.96it/s]
 30%|███       | 3216/10707 [30:41<1:03:05,  1.98it/s]
 30%|███       | 3217/10707 [30:41<1:02:42,  1.99it/s]
 30%|███       | 3218/10707 [30:42<1:02:25,  2.00it/s]
 30%|███       | 3219/10707 [30:42<1:02:16,  2.00it/s]
 30%|███       | 3220/10707 [30:43<1:02:09,  2.01it/s]
 30%|███       | 3221/10707 [30:43<1:02:05,  2.01it/s]
 30%|███       | 3222/10707 [30:44<1:01:56,  2.01it/s]
 30%|███       | 3223/10707 [30:44<1:01:52,  2.02it/s]
 30%|███       | 3224/10707 [30:45<1:01:46,  2.02it/s]
 30%|███       | 3225/10707 [30:45<1:01:47,  2.02it/s]
                                                      {'loss': 3.8887, 'grad_norm': 0.2017975151538849, 'learning_rate': 0.0008816917086899701, 'epoch': 0.3}
-

 30%|███       | 3225/10707 [30:45<1:01:47,  2.02it/s]
 30%|███       | 3226/10707 [30:46<1:01:49,  2.02it/s]
 30%|███       | 3227/10707 [30:46<1:01:48,  2.02it/s]
 30%|███       | 3228/10707 [30:47<1:01:43,  2.02it/s]
 30%|███       | 3229/10707 [30:47<1:01:42,  2.02it/s]
 30%|███       | 3230/10707 [30:48<1:01:38,  2.02it/s]
 30%|███       | 3231/10707 [30:48<1:01:41,  2.02it/s]
 30%|███       | 3232/10707 [30:49<1:01:44,  2.02it/s]
 30%|███       | 3233/10707 [30:49<1:01:44,  2.02it/s]
 30%|███       | 3234/10707 [30:50<1:01:38,  2.02it/s]
 30%|███       | 3235/10707 [30:50<1:01:40,  2.02it/s]
 30%|███       | 3236/10707 [30:51<1:01:42,  2.02it/s]
 30%|███       | 3237/10707 [30:51<1:01:39,  2.02it/s]
 30%|███       | 3238/10707 [30:52<1:01:39,  2.02it/s]
 30%|███       | 3239/10707 [30:52<1:01:39,  2.02it/s]
 30%|███       | 3240/10707 [30:53<1:01:40,  2.02it/s]
 30%|███       | 3241/10707 [30:53<1:01:35,  2.02it/s]
 30%|███       | 3242/10707 [30:54<1:01:36,  2.02it/s]
 30%|███       | 3243/10707 [30:54<1:01:32,  2.02it/s]
 30%|███       | 3244/10707 [30:55<1:01:43,  2.02it/s]
 30%|███       | 3245/10707 [30:55<1:01:39,  2.02it/s]
 30%|███       | 3246/10707 [30:56<1:01:42,  2.01it/s]
 30%|███       | 3247/10707 [30:56<1:01:38,  2.02it/s]
 30%|███       | 3248/10707 [30:57<1:01:34,  2.02it/s]
 30%|███       | 3249/10707 [30:57<1:01:35,  2.02it/s]
 30%|███       | 3250/10707 [30:58<1:01:32,  2.02it/s]
                                                      
{'loss': 3.8851, 'grad_norm': 0.20317339897155762, 'learning_rate': 0.0008790466158724946, 'epoch': 0.3}
-
 30%|███       | 3250/10707 [30:58<1:01:32,  2.02it/s]
 30%|███       | 3251/10707 [30:58<1:01:41,  2.01it/s]
 30%|███       | 3252/10707 [30:59<1:01:36,  2.02it/s]
 30%|███       | 3253/10707 [30:59<1:01:36,  2.02it/s]
 30%|███       | 3254/10707 [31:00<1:01:31,  2.02it/s]
 30%|███       | 3255/10707 [31:00<1:01:32,  2.02it/s]
 30%|███       | 3256/10707 [31:01<1:01:30,  2.02it/s]
 30%|███       | 3257/10707 [31:01<1:01:30,  2.02it/s]
 30%|███       | 3258/10707 [31:02<1:01:28,  2.02it/s]
 30%|███       | 3259/10707 [31:02<1:01:32,  2.02it/s]
 30%|███       | 3260/10707 [31:03<1:01:30,  2.02it/s]
 30%|███       | 3261/10707 [31:03<1:01:30,  2.02it/s]
 30%|███       | 3262/10707 [31:04<1:01:27,  2.02it/s]
 30%|███       | 3263/10707 [31:04<1:01:29,  2.02it/s]
 30%|███       | 3264/10707 [31:05<1:01:24,  2.02it/s]
 30%|███       | 3265/10707 [31:05<1:01:25,  2.02it/s]
 31%|███       | 3266/10707 [31:06<1:01:23,  2.02it/s]
 31%|███       | 3267/10707 [31:06<1:01:24,  2.02it/s]
 31%|███       | 3268/10707 [31:07<1:01:21,  2.02it/s]
 31%|███       | 3269/10707 [31:07<1:01:26,  2.02it/s]
 31%|███       | 3270/10707 [31:08<1:01:21,  2.02it/s]
 31%|███       | 3271/10707 [31:08<1:01:24,  2.02it/s]
 31%|███       | 3272/10707 [31:09<1:01:19,  2.02it/s]
 31%|███       | 3273/10707 [31:09<1:01:22,  2.02it/s]
 31%|███       | 3274/10707 [31:10<1:01:18,  2.02it/s]
 31%|███       | 3275/10707 [31:10<1:01:19,  2.02it/s]{'loss': 3.881, 'grad_norm': 0.20359613001346588, 'learning_rate': 0.000876376341856353, 'epoch': 0.31}
                                                      
-
 31%|███       | 3275/10707 [31:10<1:01:19,  2.02it/s]
 31%|███       | 3276/10707 [31:11<1:01:22,  2.02it/s]
 31%|███       | 3277/10707 [31:11<1:01:20,  2.02it/s]
 31%|███       | 3278/10707 [31:12<1:01:20,  2.02it/s]
 31%|███       | 3279/10707 [31:12<1:01:20,  2.02it/s]
 31%|███       | 3280/10707 [31:13<1:01:21,  2.02it/s]
 31%|███       | 3281/10707 [31:13<1:01:16,  2.02it/s]
 31%|███       | 3282/10707 [31:14<1:01:16,  2.02it/s]
 31%|███       | 3283/10707 [31:14<1:01:12,  2.02it/s]
 31%|███       | 3284/10707 [31:15<1:01:16,  2.02it/s]
 31%|███       | 3285/10707 [31:15<1:01:13,  2.02it/s]
 31%|███       | 3286/10707 [31:16<1:01:18,  2.02it/s]
 31%|███       | 3287/10707 [31:16<1:01:14,  2.02it/s]
 31%|███       | 3288/10707 [31:17<1:01:17,  2.02it/s]
 31%|███       | 3289/10707 [31:17<1:01:13,  2.02it/s]
 31%|███       | 3290/10707 [31:18<1:01:14,  2.02it/s]
 31%|███       | 3291/10707 [31:18<1:01:11,  2.02it/s]
 31%|███       | 3292/10707 [31:19<1:01:13,  2.02it/s]
 31%|███       | 3293/10707 [31:19<1:01:11,  2.02it/s]
 31%|███       | 3294/10707 [31:20<1:01:11,  2.02it/s]
 31%|███       | 3295/10707 [31:20<1:01:04,  2.02it/s]
 31%|███       | 3296/10707 [31:21<1:01:07,  2.02it/s]
 31%|███       | 3297/10707 [31:21<1:01:02,  2.02it/s]
 31%|███       | 3298/10707 [31:22<1:01:08,  2.02it/s]
 31%|███       | 3299/10707 [31:22<1:01:03,  2.02it/s]
 31%|███       | 3300/10707 [31:23<1:01:10,  2.02it/s]{'loss': 3.882, 'grad_norm': 0.2062896192073822, 'learning_rate': 0.0008736810640358228, 'epoch': 0.31}
-
                                                      

 31%|███       | 3300/10707 [31:23<1:01:10,  2.02it/s]
 31%|███       | 3301/10707 [31:23<1:01:16,  2.01it/s]
 31%|███       | 3302/10707 [31:24<1:01:12,  2.02it/s]
 31%|███       | 3303/10707 [31:24<1:01:09,  2.02it/s]
 31%|███       | 3304/10707 [31:25<1:01:10,  2.02it/s]
 31%|███       | 3305/10707 [31:25<1:01:08,  2.02it/s]
 31%|███       | 3306/10707 [31:25<1:01:08,  2.02it/s]
 31%|███       | 3307/10707 [31:26<1:01:11,  2.02it/s]
 31%|███       | 3308/10707 [31:26<1:01:08,  2.02it/s]
 31%|███       | 3309/10707 [31:27<1:01:10,  2.02it/s]
 31%|███       | 3310/10707 [31:27<1:01:07,  2.02it/s]
 31%|███       | 3311/10707 [31:28<1:01:04,  2.02it/s]
 31%|███       | 3312/10707 [31:28<1:01:04,  2.02it/s]
 31%|███       | 3313/10707 [31:29<1:01:00,  2.02it/s]
 31%|███       | 3314/10707 [31:29<1:01:01,  2.02it/s]
 31%|███       | 3315/10707 [31:30<1:00:56,  2.02it/s]
 31%|███       | 3316/10707 [31:30<1:00:54,  2.02it/s]
 31%|███       | 3317/10707 [31:31<1:00:53,  2.02it/s]
 31%|███       | 3318/10707 [31:31<1:00:53,  2.02it/s]
 31%|███       | 3319/10707 [31:32<1:00:57,  2.02it/s]
 31%|███       | 3320/10707 [31:32<1:00:54,  2.02it/s]
 31%|███       | 3321/10707 [31:33<1:00:58,  2.02it/s]
 31%|███       | 3322/10707 [31:33<1:01:00,  2.02it/s]
 31%|███       | 3323/10707 [31:34<1:00:57,  2.02it/s]
 31%|███       | 3324/10707 [31:34<1:00:57,  2.02it/s]
 31%|███       | 3325/10707 [31:35<1:00:55,  2.02it/s]{'loss': 3.8786, 'grad_norm': 0.20444469153881073, 'learning_rate': 0.0008709609614662585, 'epoch': 0.31}
                                                      
-
 31%|███       | 3325/10707 [31:35<1:00:55,  2.02it/s]
 31%|███       | 3326/10707 [31:35<1:01:03,  2.01it/s]
 31%|███       | 3327/10707 [31:36<1:01:01,  2.02it/s]
 31%|███       | 3328/10707 [31:36<1:01:01,  2.02it/s]
 31%|███       | 3329/10707 [31:37<1:01:02,  2.01it/s]
 31%|███       | 3330/10707 [31:37<1:00:59,  2.02it/s]
 31%|███       | 3331/10707 [31:38<1:00:56,  2.02it/s]
 31%|███       | 3332/10707 [31:38<1:00:55,  2.02it/s]
 31%|███       | 3333/10707 [31:39<1:00:58,  2.02it/s]
 31%|███       | 3334/10707 [31:39<1:00:53,  2.02it/s]
 31%|███       | 3335/10707 [31:40<1:00:53,  2.02it/s]
 31%|███       | 3336/10707 [31:40<1:00:48,  2.02it/s]
 31%|███       | 3337/10707 [31:41<1:00:49,  2.02it/s]
 31%|███       | 3338/10707 [31:41<1:00:46,  2.02it/s]
 31%|███       | 3339/10707 [31:42<1:00:46,  2.02it/s]
 31%|███       | 3340/10707 [31:42<1:00:42,  2.02it/s]
 31%|███       | 3341/10707 [31:43<1:00:43,  2.02it/s]
 31%|███       | 3342/10707 [31:43<1:00:43,  2.02it/s]
 31%|███       | 3343/10707 [31:44<1:00:47,  2.02it/s]
 31%|███       | 3344/10707 [31:44<1:00:46,  2.02it/s]
 31%|███       | 3345/10707 [31:45<1:00:45,  2.02it/s]
 31%|███▏      | 3346/10707 [31:45<1:00:45,  2.02it/s]
 31%|███▏      | 3347/10707 [31:46<1:00:42,  2.02it/s]
 31%|███▏      | 3348/10707 [31:46<1:00:50,  2.02it/s]
 31%|███▏      | 3349/10707 [31:47<1:00:45,  2.02it/s]
 31%|███▏      | 3350/10707 [31:47<1:00:50,  2.02it/s]{'loss': 3.8726, 'grad_norm': 0.2050025910139084, 'learning_rate': 0.0008682162148521972, 'epoch': 0.31}
-
                                                      

 31%|███▏      | 3350/10707 [31:47<1:00:50,  2.02it/s]
 31%|███▏      | 3351/10707 [31:48<1:00:51,  2.01it/s]
 31%|███▏      | 3352/10707 [31:48<1:00:53,  2.01it/s]
 31%|███▏      | 3353/10707 [31:49<1:00:50,  2.01it/s]
 31%|███▏      | 3354/10707 [31:49<1:00:46,  2.02it/s]
 31%|███▏      | 3355/10707 [31:50<1:00:52,  2.01it/s]
 31%|███▏      | 3356/10707 [31:50<1:00:45,  2.02it/s]
 31%|███▏      | 3357/10707 [31:51<1:00:46,  2.02it/s]
 31%|███▏      | 3358/10707 [31:51<1:00:43,  2.02it/s]
 31%|███▏      | 3359/10707 [31:52<1:00:41,  2.02it/s]
 31%|███▏      | 3360/10707 [31:52<1:00:40,  2.02it/s]
 31%|███▏      | 3361/10707 [31:53<1:00:36,  2.02it/s]
 31%|███▏      | 3362/10707 [31:53<1:00:38,  2.02it/s]
 31%|███▏      | 3363/10707 [31:54<1:00:34,  2.02it/s]
 31%|███▏      | 3364/10707 [31:54<1:00:34,  2.02it/s]
 31%|███▏      | 3365/10707 [31:55<1:00:35,  2.02it/s]
 31%|███▏      | 3366/10707 [31:55<1:00:33,  2.02it/s]
 31%|███▏      | 3367/10707 [31:56<1:00:30,  2.02it/s]
 31%|███▏      | 3368/10707 [31:56<1:00:30,  2.02it/s]
 31%|███▏      | 3369/10707 [31:57<1:00:29,  2.02it/s]
 31%|███▏      | 3370/10707 [31:57<1:00:32,  2.02it/s]
 31%|███▏      | 3371/10707 [31:58<1:00:28,  2.02it/s]
 31%|███▏      | 3372/10707 [31:58<1:00:29,  2.02it/s]
 32%|███▏      | 3373/10707 [31:59<1:00:30,  2.02it/s]
 32%|███▏      | 3374/10707 [31:59<1:00:28,  2.02it/s]
 32%|███▏      | 3375/10707 [32:00<1:00:30,  2.02it/s]{'loss': 3.8763, 'grad_norm': 0.20487333834171295, 'learning_rate': 0.0008654470065353535, 'epoch': 0.32}
-
                                                      

 32%|███▏      | 3375/10707 [32:00<1:00:30,  2.02it/s]
 32%|███▏      | 3376/10707 [32:00<1:00:38,  2.01it/s]
 32%|███▏      | 3377/10707 [32:01<1:00:34,  2.02it/s]
 32%|███▏      | 3378/10707 [32:01<1:00:32,  2.02it/s]
 32%|███▏      | 3379/10707 [32:02<1:00:32,  2.02it/s]
 32%|███▏      | 3380/10707 [32:02<1:00:31,  2.02it/s]
 32%|███▏      | 3381/10707 [32:03<1:00:26,  2.02it/s]
 32%|███▏      | 3382/10707 [32:03<1:00:29,  2.02it/s]
 32%|███▏      | 3383/10707 [32:04<1:00:26,  2.02it/s]
 32%|███▏      | 3384/10707 [32:04<1:00:33,  2.02it/s]
 32%|███▏      | 3385/10707 [32:05<1:00:25,  2.02it/s]
 32%|███▏      | 3386/10707 [32:05<1:00:24,  2.02it/s]
 32%|███▏      | 3387/10707 [32:06<1:00:20,  2.02it/s]
 32%|███▏      | 3388/10707 [32:06<1:00:24,  2.02it/s]
 32%|███▏      | 3389/10707 [32:07<1:00:19,  2.02it/s]
 32%|███▏      | 3390/10707 [32:07<1:00:21,  2.02it/s]
 32%|███▏      | 3391/10707 [32:08<1:00:18,  2.02it/s]
 32%|███▏      | 3392/10707 [32:08<1:00:22,  2.02it/s]
 32%|███▏      | 3393/10707 [32:09<1:00:20,  2.02it/s]
 32%|███▏      | 3394/10707 [32:09<1:00:22,  2.02it/s]
 32%|███▏      | 3395/10707 [32:10<1:00:22,  2.02it/s]
 32%|███▏      | 3396/10707 [32:10<1:00:21,  2.02it/s]
 32%|███▏      | 3397/10707 [32:11<1:00:16,  2.02it/s]
 32%|███▏      | 3398/10707 [32:11<1:00:19,  2.02it/s]
 32%|███▏      | 3399/10707 [32:12<1:00:14,  2.02it/s]
 32%|███▏      | 3400/10707 [32:12<1:00:15,  2.02it/s]
                                                      
{'loss': 3.88, 'grad_norm': 0.21246883273124695, 'learning_rate': 0.0008626535204825052, 'epoch': 0.32}
-
 32%|███▏      | 3400/10707 [32:12<1:00:15,  2.02it/s]
 32%|███▏      | 3401/10707 [32:13<1:00:19,  2.02it/s]
 32%|███▏      | 3402/10707 [32:13<1:00:17,  2.02it/s]
 32%|███▏      | 3403/10707 [32:14<1:00:17,  2.02it/s]
 32%|███▏      | 3404/10707 [32:14<1:00:20,  2.02it/s]
 32%|███▏      | 3405/10707 [32:15<1:00:20,  2.02it/s]
 32%|███▏      | 3406/10707 [32:15<1:00:16,  2.02it/s]
 32%|███▏      | 3407/10707 [32:16<1:00:14,  2.02it/s]
 32%|███▏      | 3408/10707 [32:16<1:00:13,  2.02it/s]
 32%|███▏      | 3409/10707 [32:17<1:00:17,  2.02it/s]
 32%|███▏      | 3410/10707 [32:17<1:00:13,  2.02it/s]
 32%|███▏      | 3411/10707 [32:18<1:00:12,  2.02it/s]
 32%|███▏      | 3412/10707 [32:18<1:00:09,  2.02it/s]
 32%|███▏      | 3413/10707 [32:18<1:00:11,  2.02it/s]
 32%|███▏      | 3414/10707 [32:19<1:00:08,  2.02it/s]
 32%|███▏      | 3415/10707 [32:19<1:00:08,  2.02it/s]
 32%|███▏      | 3416/10707 [32:20<1:00:04,  2.02it/s]
 32%|███▏      | 3417/10707 [32:20<1:00:07,  2.02it/s]
 32%|███▏      | 3418/10707 [32:21<1:00:03,  2.02it/s]
 32%|███▏      | 3419/10707 [32:21<1:00:08,  2.02it/s]
 32%|███▏      | 3420/10707 [32:22<1:00:05,  2.02it/s]
 32%|███▏      | 3421/10707 [32:22<1:00:05,  2.02it/s]
 32%|███▏      | 3422/10707 [32:23<1:00:02,  2.02it/s]
 32%|███▏      | 3423/10707 [32:23<1:00:02,  2.02it/s]
 32%|███▏      | 3424/10707 [32:24<1:00:01,  2.02it/s]
 32%|███▏      | 3425/10707 [32:24<1:00:00,  2.02it/s]{'loss': 3.8748, 'grad_norm': 0.20345519483089447, 'learning_rate': 0.0008598359422732736, 'epoch': 0.32}
-
                                                      

 32%|███▏      | 3425/10707 [32:24<1:00:00,  2.02it/s]
 32%|███▏      | 3426/10707 [32:25<1:00:05,  2.02it/s]
 32%|███▏      | 3427/10707 [32:25<1:00:10,  2.02it/s]
 32%|███▏      | 3428/10707 [32:26<1:00:05,  2.02it/s]
 32%|███▏      | 3429/10707 [32:26<1:00:03,  2.02it/s]
 32%|███▏      | 3430/10707 [32:27<1:00:02,  2.02it/s]
 32%|███▏      | 3431/10707 [32:27<1:00:02,  2.02it/s]
 32%|███▏      | 3432/10707 [32:28<59:59,  2.02it/s]  
 32%|███▏      | 3433/10707 [32:28<1:00:02,  2.02it/s]
 32%|███▏      | 3434/10707 [32:29<1:00:02,  2.02it/s]
 32%|███▏      | 3435/10707 [32:29<1:00:02,  2.02it/s]
 32%|███▏      | 3436/10707 [32:30<1:00:02,  2.02it/s]
 32%|███▏      | 3437/10707 [32:30<1:00:03,  2.02it/s]
 32%|███▏      | 3438/10707 [32:31<1:00:01,  2.02it/s]
 32%|███▏      | 3439/10707 [32:31<1:00:03,  2.02it/s]
 32%|███▏      | 3440/10707 [32:32<1:00:01,  2.02it/s]
 32%|███▏      | 3441/10707 [32:32<1:00:01,  2.02it/s]
 32%|███▏      | 3442/10707 [32:33<59:57,  2.02it/s]  
 32%|███▏      | 3443/10707 [32:33<59:58,  2.02it/s]
 32%|███▏      | 3444/10707 [32:34<59:55,  2.02it/s]
 32%|███▏      | 3445/10707 [32:34<59:53,  2.02it/s]
 32%|███▏      | 3446/10707 [32:35<59:54,  2.02it/s]
 32%|███▏      | 3447/10707 [32:35<59:53,  2.02it/s]
 32%|███▏      | 3448/10707 [32:36<59:52,  2.02it/s]
 32%|███▏      | 3449/10707 [32:36<59:51,  2.02it/s]
 32%|███▏      | 3450/10707 [32:37<59:53,  2.02it/s]{'loss': 3.8635, 'grad_norm': 0.21436098217964172, 'learning_rate': 0.0008569944590877934, 'epoch': 0.32}
-
                                                    

 32%|███▏      | 3450/10707 [32:37<59:53,  2.02it/s]
 32%|███▏      | 3451/10707 [32:37<1:00:05,  2.01it/s]
 32%|███▏      | 3452/10707 [32:38<1:00:00,  2.01it/s]
 32%|███▏      | 3453/10707 [32:38<1:00:02,  2.01it/s]
 32%|███▏      | 3454/10707 [32:39<59:59,  2.01it/s]  
 32%|███▏      | 3455/10707 [32:39<59:58,  2.02it/s]
 32%|███▏      | 3456/10707 [32:40<59:59,  2.01it/s]
 32%|███▏      | 3457/10707 [32:40<59:59,  2.01it/s]
 32%|███▏      | 3458/10707 [32:41<59:55,  2.02it/s]
 32%|███▏      | 3459/10707 [32:41<59:55,  2.02it/s]
 32%|███▏      | 3460/10707 [32:42<59:50,  2.02it/s]
 32%|███▏      | 3461/10707 [32:42<59:51,  2.02it/s]
 32%|███▏      | 3462/10707 [32:43<59:53,  2.02it/s]
 32%|███▏      | 3463/10707 [32:43<59:51,  2.02it/s]
 32%|███▏      | 3464/10707 [32:44<59:54,  2.01it/s]
 32%|███▏      | 3465/10707 [32:44<59:51,  2.02it/s]
 32%|███▏      | 3466/10707 [32:45<59:49,  2.02it/s]
 32%|███▏      | 3467/10707 [32:45<59:47,  2.02it/s]
 32%|███▏      | 3468/10707 [32:46<59:46,  2.02it/s]
 32%|███▏      | 3469/10707 [32:46<59:45,  2.02it/s]
 32%|███▏      | 3470/10707 [32:47<59:45,  2.02it/s]
 32%|███▏      | 3471/10707 [32:47<59:43,  2.02it/s]
 32%|███▏      | 3472/10707 [32:48<59:41,  2.02it/s]
 32%|███▏      | 3473/10707 [32:48<59:41,  2.02it/s]
 32%|███▏      | 3474/10707 [32:49<59:40,  2.02it/s]
 32%|███▏      | 3475/10707 [32:49<59:40,  2.02it/s]{'loss': 3.8627, 'grad_norm': 0.21791967749595642, 'learning_rate': 0.0008541292596942783, 'epoch': 0.32}
                                                    
-
 32%|███▏      | 3475/10707 [32:49<59:40,  2.02it/s]
 32%|███▏      | 3476/10707 [32:50<59:43,  2.02it/s]
 32%|███▏      | 3477/10707 [32:50<59:39,  2.02it/s]
 32%|███▏      | 3478/10707 [32:51<59:39,  2.02it/s]
 32%|███▏      | 3479/10707 [32:51<59:40,  2.02it/s]
 33%|███▎      | 3480/10707 [32:52<59:34,  2.02it/s]
 33%|███▎      | 3481/10707 [32:52<59:39,  2.02it/s]
 33%|███▎      | 3482/10707 [32:53<59:35,  2.02it/s]
 33%|███▎      | 3483/10707 [32:53<59:41,  2.02it/s]
 33%|███▎      | 3484/10707 [32:54<59:36,  2.02it/s]
 33%|███▎      | 3485/10707 [32:54<59:38,  2.02it/s]
 33%|███▎      | 3486/10707 [32:55<59:36,  2.02it/s]
 33%|███▎      | 3487/10707 [32:55<59:37,  2.02it/s]
 33%|███▎      | 3488/10707 [32:56<59:32,  2.02it/s]
 33%|███▎      | 3489/10707 [32:56<59:33,  2.02it/s]
 33%|███▎      | 3490/10707 [32:57<59:30,  2.02it/s]
 33%|███▎      | 3491/10707 [32:57<59:32,  2.02it/s]
 33%|███▎      | 3492/10707 [32:58<59:27,  2.02it/s]
 33%|███▎      | 3493/10707 [32:58<59:28,  2.02it/s]
 33%|███▎      | 3494/10707 [32:59<59:27,  2.02it/s]
 33%|███▎      | 3495/10707 [32:59<59:29,  2.02it/s]
 33%|███▎      | 3496/10707 [33:00<59:29,  2.02it/s]
 33%|███▎      | 3497/10707 [33:00<59:29,  2.02it/s]
 33%|███▎      | 3498/10707 [33:01<59:27,  2.02it/s]
 33%|███▎      | 3499/10707 [33:01<59:29,  2.02it/s]
 33%|███▎      | 3500/10707 [33:02<59:30,  2.02it/s]{'loss': 3.8625, 'grad_norm': 0.20983728766441345, 'learning_rate': 0.0008512405344364804, 'epoch': 0.33}
-
                                                    

 33%|███▎      | 3500/10707 [33:02<59:30,  2.02it/s]
 33%|███▎      | 3501/10707 [33:02<59:31,  2.02it/s]
 33%|███▎      | 3502/10707 [33:03<59:31,  2.02it/s]
 33%|███▎      | 3503/10707 [33:03<59:24,  2.02it/s]
 33%|███▎      | 3504/10707 [33:04<59:28,  2.02it/s]
 33%|███▎      | 3505/10707 [33:04<59:27,  2.02it/s]
 33%|███▎      | 3506/10707 [33:05<59:28,  2.02it/s]
 33%|███▎      | 3507/10707 [33:05<59:23,  2.02it/s]
 33%|███▎      | 3508/10707 [33:06<59:29,  2.02it/s]
 33%|███▎      | 3509/10707 [33:06<59:30,  2.02it/s]
 33%|███▎      | 3510/10707 [33:07<59:28,  2.02it/s]
 33%|███▎      | 3511/10707 [33:07<59:29,  2.02it/s]
 33%|███▎      | 3512/10707 [33:08<59:26,  2.02it/s]
 33%|███▎      | 3513/10707 [33:08<59:26,  2.02it/s]
 33%|███▎      | 3514/10707 [33:09<59:22,  2.02it/s]
 33%|███▎      | 3515/10707 [33:09<59:23,  2.02it/s]
 33%|███▎      | 3516/10707 [33:10<59:21,  2.02it/s]
 33%|███▎      | 3517/10707 [33:10<59:23,  2.02it/s]
 33%|███▎      | 3518/10707 [33:11<59:20,  2.02it/s]
 33%|███▎      | 3519/10707 [33:11<59:22,  2.02it/s]
 33%|███▎      | 3520/10707 [33:11<59:21,  2.02it/s]
 33%|███▎      | 3521/10707 [33:12<59:21,  2.02it/s]
 33%|███▎      | 3522/10707 [33:12<59:22,  2.02it/s]
 33%|███▎      | 3523/10707 [33:13<59:22,  2.02it/s]
 33%|███▎      | 3524/10707 [33:13<59:19,  2.02it/s]
 33%|███▎      | 3525/10707 [33:14<59:19,  2.02it/s]{'loss': 3.8608, 'grad_norm': 0.2146988958120346, 'learning_rate': 0.0008483284752210456, 'epoch': 0.33}
-
                                                    

 33%|███▎      | 3525/10707 [33:14<59:19,  2.02it/s]
 33%|███▎      | 3526/10707 [33:14<59:18,  2.02it/s]
 33%|███▎      | 3527/10707 [33:15<59:17,  2.02it/s]
 33%|███▎      | 3528/10707 [33:15<59:11,  2.02it/s]
 33%|███▎      | 3529/10707 [33:16<59:14,  2.02it/s]
 33%|███▎      | 3530/10707 [33:16<59:14,  2.02it/s]
 33%|███▎      | 3531/10707 [33:17<59:16,  2.02it/s]
 33%|███▎      | 3532/10707 [33:17<59:14,  2.02it/s]
 33%|███▎      | 3533/10707 [33:18<59:14,  2.02it/s]
 33%|███▎      | 3534/10707 [33:18<59:08,  2.02it/s]
 33%|███▎      | 3535/10707 [33:19<59:11,  2.02it/s]
 33%|███▎      | 3536/10707 [33:19<59:06,  2.02it/s]
 33%|███▎      | 3537/10707 [33:20<59:08,  2.02it/s]
 33%|███▎      | 3538/10707 [33:20<59:06,  2.02it/s]
 33%|███▎      | 3539/10707 [33:21<59:08,  2.02it/s]
 33%|███▎      | 3540/10707 [33:21<59:04,  2.02it/s]
 33%|███▎      | 3541/10707 [33:22<59:06,  2.02it/s]
 33%|███▎      | 3542/10707 [33:22<59:02,  2.02it/s]
 33%|███▎      | 3543/10707 [33:23<59:07,  2.02it/s]
 33%|███▎      | 3544/10707 [33:23<59:04,  2.02it/s]
 33%|███▎      | 3545/10707 [33:24<59:06,  2.02it/s]
 33%|███▎      | 3546/10707 [33:24<59:02,  2.02it/s]
 33%|███▎      | 3547/10707 [33:25<59:04,  2.02it/s]
 33%|███▎      | 3548/10707 [33:25<59:05,  2.02it/s]
 33%|███▎      | 3549/10707 [33:26<59:04,  2.02it/s]
 33%|███▎      | 3550/10707 [33:26<59:04,  2.02it/s]{'loss': 3.8556, 'grad_norm': 0.20713388919830322, 'learning_rate': 0.0008453932755047642, 'epoch': 0.33}
-
                                                    

 33%|███▎      | 3550/10707 [33:26<59:04,  2.02it/s]
 33%|███▎      | 3551/10707 [33:27<59:09,  2.02it/s]
 33%|███▎      | 3552/10707 [33:27<59:07,  2.02it/s]
 33%|███▎      | 3553/10707 [33:28<59:03,  2.02it/s]
 33%|███▎      | 3554/10707 [33:28<59:11,  2.01it/s]
 33%|███▎      | 3555/10707 [33:29<59:07,  2.02it/s]
 33%|███▎      | 3556/10707 [33:29<59:04,  2.02it/s]
 33%|███▎      | 3557/10707 [33:30<59:07,  2.02it/s]
 33%|███▎      | 3558/10707 [33:30<59:05,  2.02it/s]
 33%|███▎      | 3559/10707 [33:31<59:03,  2.02it/s]
 33%|███▎      | 3560/10707 [33:31<59:00,  2.02it/s]
 33%|███▎      | 3561/10707 [33:32<59:04,  2.02it/s]
 33%|███▎      | 3562/10707 [33:32<59:01,  2.02it/s]
 33%|███▎      | 3563/10707 [33:33<59:04,  2.02it/s]
 33%|███▎      | 3564/10707 [33:33<59:00,  2.02it/s]
 33%|███▎      | 3565/10707 [33:34<59:01,  2.02it/s]
 33%|███▎      | 3566/10707 [33:34<58:56,  2.02it/s]
 33%|███▎      | 3567/10707 [33:35<58:57,  2.02it/s]
 33%|███▎      | 3568/10707 [33:35<58:56,  2.02it/s]
 33%|███▎      | 3569/10707 [33:36<58:56,  2.02it/s]
 33%|███▎      | 3570/10707 [33:36<58:52,  2.02it/s]
 33%|███▎      | 3571/10707 [33:37<58:53,  2.02it/s]
 33%|███▎      | 3572/10707 [33:37<58:50,  2.02it/s]
 33%|███▎      | 3573/10707 [33:38<58:52,  2.02it/s]
 33%|███▎      | 3574/10707 [33:38<58:49,  2.02it/s]
 33%|███▎      | 3575/10707 [33:39<58:54,  2.02it/s]{'loss': 3.8613, 'grad_norm': 0.21902155876159668, 'learning_rate': 0.0008424351302817185, 'epoch': 0.33}
-
                                                    

 33%|███▎      | 3575/10707 [33:39<58:54,  2.02it/s]
 33%|███▎      | 3576/10707 [33:39<58:52,  2.02it/s]
 33%|███▎      | 3577/10707 [33:40<58:52,  2.02it/s]
 33%|███▎      | 3578/10707 [33:40<58:48,  2.02it/s]
 33%|███▎      | 3579/10707 [33:41<58:50,  2.02it/s]
 33%|███▎      | 3580/10707 [33:41<58:51,  2.02it/s]
 33%|███▎      | 3581/10707 [33:42<58:48,  2.02it/s]
 33%|███▎      | 3582/10707 [33:42<58:48,  2.02it/s]
 33%|███▎      | 3583/10707 [33:43<58:45,  2.02it/s]
 33%|███▎      | 3584/10707 [33:43<58:50,  2.02it/s]
 33%|███▎      | 3585/10707 [33:44<58:46,  2.02it/s]
 33%|███▎      | 3586/10707 [33:44<58:47,  2.02it/s]
 34%|███▎      | 3587/10707 [33:45<58:42,  2.02it/s]
 34%|███▎      | 3588/10707 [33:45<58:45,  2.02it/s]
 34%|███▎      | 3589/10707 [33:46<58:39,  2.02it/s]
 34%|███▎      | 3590/10707 [33:46<58:41,  2.02it/s]
 34%|███▎      | 3591/10707 [33:47<58:40,  2.02it/s]
 34%|███▎      | 3592/10707 [33:47<58:43,  2.02it/s]
 34%|███▎      | 3593/10707 [33:48<58:41,  2.02it/s]
 34%|███▎      | 3594/10707 [33:48<58:45,  2.02it/s]
 34%|███▎      | 3595/10707 [33:49<58:42,  2.02it/s]
 34%|███▎      | 3596/10707 [33:49<58:40,  2.02it/s]
 34%|███▎      | 3597/10707 [33:50<58:38,  2.02it/s]
 34%|███▎      | 3598/10707 [33:50<58:40,  2.02it/s]
 34%|███▎      | 3599/10707 [33:51<58:38,  2.02it/s]
 34%|███▎      | 3600/10707 [33:51<58:41,  2.02it/s]{'loss': 3.8526, 'grad_norm': 0.21672873198986053, 'learning_rate': 0.0008394542360703301, 'epoch': 0.34}
-
                                                    

 34%|███▎      | 3600/10707 [33:51<58:41,  2.02it/s]
 34%|███▎      | 3601/10707 [33:52<58:45,  2.02it/s]
 34%|███▎      | 3602/10707 [33:52<58:43,  2.02it/s]
 34%|███▎      | 3603/10707 [33:53<58:42,  2.02it/s]
 34%|███▎      | 3604/10707 [33:53<58:37,  2.02it/s]
 34%|███▎      | 3605/10707 [33:54<58:37,  2.02it/s]
 34%|███▎      | 3606/10707 [33:54<58:35,  2.02it/s]
 34%|███▎      | 3607/10707 [33:55<58:36,  2.02it/s]
 34%|███▎      | 3608/10707 [33:55<58:33,  2.02it/s]
 34%|███▎      | 3609/10707 [33:56<58:35,  2.02it/s]
 34%|███▎      | 3610/10707 [33:56<58:46,  2.01it/s]
 34%|███▎      | 3611/10707 [33:57<58:43,  2.01it/s]
 34%|███▎      | 3612/10707 [33:57<58:39,  2.02it/s]
 34%|███▎      | 3613/10707 [33:58<58:36,  2.02it/s]
 34%|███▍      | 3614/10707 [33:58<58:39,  2.02it/s]
 34%|███▍      | 3615/10707 [33:59<58:33,  2.02it/s]
 34%|███▍      | 3616/10707 [33:59<58:34,  2.02it/s]
 34%|███▍      | 3617/10707 [34:00<58:33,  2.02it/s]
 34%|███▍      | 3618/10707 [34:00<58:34,  2.02it/s]
 34%|███▍      | 3619/10707 [34:01<58:29,  2.02it/s]
 34%|███▍      | 3620/10707 [34:01<58:31,  2.02it/s]
 34%|███▍      | 3621/10707 [34:02<58:28,  2.02it/s]
 34%|███▍      | 3622/10707 [34:02<58:30,  2.02it/s]
 34%|███▍      | 3623/10707 [34:03<58:26,  2.02it/s]
 34%|███▍      | 3624/10707 [34:03<58:31,  2.02it/s]
 34%|███▍      | 3625/10707 [34:04<58:28,  2.02it/s]{'loss': 3.85, 'grad_norm': 0.20538106560707092, 'learning_rate': 0.0008364507909003036, 'epoch': 0.34}
-
                                                    

 34%|███▍      | 3625/10707 [34:04<58:28,  2.02it/s]
 34%|███▍      | 3626/10707 [34:04<58:32,  2.02it/s]
 34%|███▍      | 3627/10707 [34:05<58:31,  2.02it/s]
 34%|███▍      | 3628/10707 [34:05<58:30,  2.02it/s]
 34%|███▍      | 3629/10707 [34:05<58:29,  2.02it/s]
 34%|███▍      | 3630/10707 [34:06<58:29,  2.02it/s]
 34%|███▍      | 3631/10707 [34:06<58:26,  2.02it/s]
 34%|███▍      | 3632/10707 [34:07<58:23,  2.02it/s]
 34%|███▍      | 3633/10707 [34:07<58:22,  2.02it/s]
 34%|███▍      | 3634/10707 [34:08<58:22,  2.02it/s]
 34%|███▍      | 3635/10707 [34:08<58:25,  2.02it/s]
 34%|███▍      | 3636/10707 [34:09<58:24,  2.02it/s]
 34%|███▍      | 3637/10707 [34:09<58:25,  2.02it/s]
 34%|███▍      | 3638/10707 [34:10<58:25,  2.02it/s]
 34%|███▍      | 3639/10707 [34:10<58:24,  2.02it/s]
 34%|███▍      | 3640/10707 [34:11<58:24,  2.02it/s]
 34%|███▍      | 3641/10707 [34:11<58:20,  2.02it/s]
 34%|███▍      | 3642/10707 [34:12<58:22,  2.02it/s]
 34%|███▍      | 3643/10707 [34:12<58:17,  2.02it/s]
 34%|███▍      | 3644/10707 [34:13<58:18,  2.02it/s]
 34%|███▍      | 3645/10707 [34:13<58:16,  2.02it/s]
 34%|███▍      | 3646/10707 [34:14<58:17,  2.02it/s]
 34%|███▍      | 3647/10707 [34:14<58:13,  2.02it/s]
 34%|███▍      | 3648/10707 [34:15<58:15,  2.02it/s]
 34%|███▍      | 3649/10707 [34:15<58:12,  2.02it/s]
 34%|███▍      | 3650/10707 [34:16<58:14,  2.02it/s]{'loss': 3.8554, 'grad_norm': 0.21106836199760437, 'learning_rate': 0.0008334249942994707, 'epoch': 0.34}
-
                                                    

 34%|███▍      | 3650/10707 [34:16<58:14,  2.02it/s]
 34%|███▍      | 3651/10707 [34:16<58:17,  2.02it/s]
 34%|███▍      | 3652/10707 [34:17<58:15,  2.02it/s]
 34%|███▍      | 3653/10707 [34:17<58:12,  2.02it/s]
 34%|███▍      | 3654/10707 [34:18<58:12,  2.02it/s]
 34%|███▍      | 3655/10707 [34:18<58:11,  2.02it/s]
 34%|███▍      | 3656/10707 [34:19<58:11,  2.02it/s]
 34%|███▍      | 3657/10707 [34:19<58:08,  2.02it/s]
 34%|███▍      | 3658/10707 [34:20<58:07,  2.02it/s]
 34%|███▍      | 3659/10707 [34:20<58:11,  2.02it/s]
 34%|███▍      | 3660/10707 [34:21<58:06,  2.02it/s]
 34%|███▍      | 3661/10707 [34:21<58:10,  2.02it/s]
 34%|███▍      | 3662/10707 [34:22<58:09,  2.02it/s]
 34%|███▍      | 3663/10707 [34:22<58:11,  2.02it/s]
 34%|███▍      | 3664/10707 [34:23<58:09,  2.02it/s]
 34%|███▍      | 3665/10707 [34:23<58:08,  2.02it/s]
 34%|███▍      | 3666/10707 [34:24<58:04,  2.02it/s]
 34%|███▍      | 3667/10707 [34:24<58:04,  2.02it/s]
 34%|███▍      | 3668/10707 [34:25<58:19,  2.01it/s]
 34%|███▍      | 3669/10707 [34:25<58:19,  2.01it/s]
 34%|███▍      | 3670/10707 [34:26<58:13,  2.01it/s]
 34%|███▍      | 3671/10707 [34:26<58:11,  2.01it/s]
 34%|███▍      | 3672/10707 [34:27<58:09,  2.02it/s]
 34%|███▍      | 3673/10707 [34:27<58:10,  2.02it/s]
 34%|███▍      | 3674/10707 [34:28<58:12,  2.01it/s]
 34%|███▍      | 3675/10707 [34:28<58:10,  2.01it/s]{'loss': 3.8488, 'grad_norm': 0.21110989153385162, 'learning_rate': 0.0008303770472805361, 'epoch': 0.34}
-
                                                    

 34%|███▍      | 3675/10707 [34:28<58:10,  2.01it/s]
 34%|███▍      | 3676/10707 [34:29<58:14,  2.01it/s]
 34%|███▍      | 3677/10707 [34:29<58:07,  2.02it/s]
 34%|███▍      | 3678/10707 [34:30<58:06,  2.02it/s]
 34%|███▍      | 3679/10707 [34:30<58:03,  2.02it/s]
 34%|███▍      | 3680/10707 [34:31<58:03,  2.02it/s]
 34%|███▍      | 3681/10707 [34:31<58:00,  2.02it/s]
 34%|███▍      | 3682/10707 [34:32<58:01,  2.02it/s]
 34%|███▍      | 3683/10707 [34:32<58:01,  2.02it/s]
 34%|███▍      | 3684/10707 [34:33<58:01,  2.02it/s]
 34%|███▍      | 3685/10707 [34:33<58:01,  2.02it/s]
 34%|███▍      | 3686/10707 [34:34<57:59,  2.02it/s]
 34%|███▍      | 3687/10707 [34:34<57:57,  2.02it/s]
 34%|███▍      | 3688/10707 [34:35<57:55,  2.02it/s]
 34%|███▍      | 3689/10707 [34:35<57:57,  2.02it/s]
 34%|███▍      | 3690/10707 [34:36<57:54,  2.02it/s]
 34%|███▍      | 3691/10707 [34:36<57:56,  2.02it/s]
 34%|███▍      | 3692/10707 [34:37<57:51,  2.02it/s]
 34%|███▍      | 3693/10707 [34:37<57:57,  2.02it/s]
 35%|███▍      | 3694/10707 [34:38<57:52,  2.02it/s]
 35%|███▍      | 3695/10707 [34:38<57:51,  2.02it/s]
 35%|███▍      | 3696/10707 [34:39<57:49,  2.02it/s]
 35%|███▍      | 3697/10707 [34:39<57:51,  2.02it/s]
 35%|███▍      | 3698/10707 [34:40<57:49,  2.02it/s]
 35%|███▍      | 3699/10707 [34:40<57:51,  2.02it/s]
 35%|███▍      | 3700/10707 [34:41<57:48,  2.02it/s]
                                                    {'loss': 3.8482, 'grad_norm': 0.20767275989055634, 'learning_rate': 0.0008273071523277223, 'epoch': 0.35}
-

 35%|███▍      | 3700/10707 [34:41<57:48,  2.02it/s]
 35%|███▍      | 3701/10707 [34:41<57:54,  2.02it/s]
 35%|███▍      | 3702/10707 [34:42<57:50,  2.02it/s]
 35%|███▍      | 3703/10707 [34:42<57:48,  2.02it/s]
 35%|███▍      | 3704/10707 [34:43<57:48,  2.02it/s]
 35%|███▍      | 3705/10707 [34:43<57:45,  2.02it/s]
 35%|███▍      | 3706/10707 [34:44<57:48,  2.02it/s]
 35%|███▍      | 3707/10707 [34:44<57:45,  2.02it/s]
 35%|███▍      | 3708/10707 [34:45<57:48,  2.02it/s]
 35%|███▍      | 3709/10707 [34:45<57:43,  2.02it/s]
 35%|███▍      | 3710/10707 [34:46<57:44,  2.02it/s]
 35%|███▍      | 3711/10707 [34:46<57:44,  2.02it/s]
 35%|███▍      | 3712/10707 [34:47<57:45,  2.02it/s]
 35%|███▍      | 3713/10707 [34:47<57:47,  2.02it/s]
 35%|███▍      | 3714/10707 [34:48<57:45,  2.02it/s]
 35%|███▍      | 3715/10707 [34:48<57:45,  2.02it/s]
 35%|███▍      | 3716/10707 [34:49<57:46,  2.02it/s]
 35%|███▍      | 3717/10707 [34:49<57:42,  2.02it/s]
 35%|███▍      | 3718/10707 [34:50<57:43,  2.02it/s]
 35%|███▍      | 3719/10707 [34:50<57:43,  2.02it/s]
 35%|███▍      | 3720/10707 [34:51<57:39,  2.02it/s]
 35%|███▍      | 3721/10707 [34:51<57:41,  2.02it/s]
 35%|███▍      | 3722/10707 [34:52<57:37,  2.02it/s]
 35%|███▍      | 3723/10707 [34:52<57:42,  2.02it/s]
 35%|███▍      | 3724/10707 [34:53<57:38,  2.02it/s]
 35%|███▍      | 3725/10707 [34:53<57:43,  2.02it/s]{'loss': 3.8414, 'grad_norm': 0.19237089157104492, 'learning_rate': 0.000824215513383319, 'epoch': 0.35}
                                                    
-
 35%|███▍      | 3725/10707 [34:53<57:43,  2.02it/s]
 35%|███▍      | 3726/10707 [34:54<57:45,  2.01it/s]
 35%|███▍      | 3727/10707 [34:54<57:41,  2.02it/s]
 35%|███▍      | 3728/10707 [34:55<57:34,  2.02it/s]
 35%|███▍      | 3729/10707 [34:55<57:36,  2.02it/s]
 35%|███▍      | 3730/10707 [34:56<57:34,  2.02it/s]
 35%|███▍      | 3731/10707 [34:56<57:33,  2.02it/s]
 35%|███▍      | 3732/10707 [34:57<57:31,  2.02it/s]
 35%|███▍      | 3733/10707 [34:57<57:33,  2.02it/s]
 35%|███▍      | 3734/10707 [34:58<57:29,  2.02it/s]
 35%|███▍      | 3735/10707 [34:58<57:29,  2.02it/s]
 35%|███▍      | 3736/10707 [34:59<57:27,  2.02it/s]
 35%|███▍      | 3737/10707 [34:59<57:29,  2.02it/s]
 35%|███▍      | 3738/10707 [34:59<57:27,  2.02it/s]
 35%|███▍      | 3739/10707 [35:00<57:29,  2.02it/s]
 35%|███▍      | 3740/10707 [35:00<57:30,  2.02it/s]
 35%|███▍      | 3741/10707 [35:01<57:27,  2.02it/s]
 35%|███▍      | 3742/10707 [35:01<57:23,  2.02it/s]
 35%|███▍      | 3743/10707 [35:02<57:26,  2.02it/s]
 35%|███▍      | 3744/10707 [35:02<57:20,  2.02it/s]
 35%|███▍      | 3745/10707 [35:03<57:24,  2.02it/s]
 35%|███▍      | 3746/10707 [35:03<57:20,  2.02it/s]
 35%|███▍      | 3747/10707 [35:04<57:23,  2.02it/s]
 35%|███▌      | 3748/10707 [35:04<57:23,  2.02it/s]
 35%|███▌      | 3749/10707 [35:05<57:23,  2.02it/s]
 35%|███▌      | 3750/10707 [35:05<57:20,  2.02it/s]{'loss': 3.8405, 'grad_norm': 0.20959612727165222, 'learning_rate': 0.000821102335834134, 'epoch': 0.35}
-
                                                    

 35%|███▌      | 3750/10707 [35:05<57:20,  2.02it/s]
 35%|███▌      | 3751/10707 [35:06<57:27,  2.02it/s]
 35%|███▌      | 3752/10707 [35:06<57:23,  2.02it/s]
 35%|███▌      | 3753/10707 [35:07<57:24,  2.02it/s]
 35%|███▌      | 3754/10707 [35:07<57:24,  2.02it/s]
 35%|███▌      | 3755/10707 [35:08<57:22,  2.02it/s]
 35%|███▌      | 3756/10707 [35:08<57:22,  2.02it/s]
 35%|███▌      | 3757/10707 [35:09<57:17,  2.02it/s]
 35%|███▌      | 3758/10707 [35:09<57:20,  2.02it/s]
 35%|███▌      | 3759/10707 [35:10<57:29,  2.01it/s]
 35%|███▌      | 3760/10707 [35:10<57:25,  2.02it/s]
 35%|███▌      | 3761/10707 [35:11<57:25,  2.02it/s]
 35%|███▌      | 3762/10707 [35:11<57:23,  2.02it/s]
 35%|███▌      | 3763/10707 [35:12<57:22,  2.02it/s]
 35%|███▌      | 3764/10707 [35:12<57:19,  2.02it/s]
 35%|███▌      | 3765/10707 [35:13<57:20,  2.02it/s]
 35%|███▌      | 3766/10707 [35:13<57:17,  2.02it/s]
 35%|███▌      | 3767/10707 [35:14<57:26,  2.01it/s]
 35%|███▌      | 3768/10707 [35:14<57:24,  2.01it/s]
 35%|███▌      | 3769/10707 [35:15<57:22,  2.02it/s]
 35%|███▌      | 3770/10707 [35:15<57:20,  2.02it/s]
 35%|███▌      | 3771/10707 [35:16<57:18,  2.02it/s]
 35%|███▌      | 3772/10707 [35:16<57:17,  2.02it/s]
 35%|███▌      | 3773/10707 [35:17<57:13,  2.02it/s]
 35%|███▌      | 3774/10707 [35:17<57:11,  2.02it/s]
 35%|███▌      | 3775/10707 [35:18<57:11,  2.02it/s]{'loss': 3.8429, 'grad_norm': 0.20905472338199615, 'learning_rate': 0.0008179678264978491, 'epoch': 0.35}
-
                                                    

 35%|███▌      | 3775/10707 [35:18<57:11,  2.02it/s]
 35%|███▌      | 3776/10707 [35:18<57:12,  2.02it/s]
 35%|███▌      | 3777/10707 [35:19<57:11,  2.02it/s]
 35%|███▌      | 3778/10707 [35:19<57:09,  2.02it/s]
 35%|███▌      | 3779/10707 [35:20<57:09,  2.02it/s]
 35%|███▌      | 3780/10707 [35:20<57:06,  2.02it/s]
 35%|███▌      | 3781/10707 [35:21<57:07,  2.02it/s]
 35%|███▌      | 3782/10707 [35:21<57:08,  2.02it/s]
 35%|███▌      | 3783/10707 [35:22<57:07,  2.02it/s]
 35%|███▌      | 3784/10707 [35:22<57:04,  2.02it/s]
 35%|███▌      | 3785/10707 [35:23<57:05,  2.02it/s]
 35%|███▌      | 3786/10707 [35:23<57:03,  2.02it/s]
 35%|███▌      | 3787/10707 [35:24<57:05,  2.02it/s]
 35%|███▌      | 3788/10707 [35:24<57:03,  2.02it/s]
 35%|███▌      | 3789/10707 [35:25<57:04,  2.02it/s]
 35%|███▌      | 3790/10707 [35:25<57:04,  2.02it/s]
 35%|███▌      | 3791/10707 [35:26<57:04,  2.02it/s]
 35%|███▌      | 3792/10707 [35:26<57:00,  2.02it/s]
 35%|███▌      | 3793/10707 [35:27<57:01,  2.02it/s]
 35%|███▌      | 3794/10707 [35:27<56:59,  2.02it/s]
 35%|███▌      | 3795/10707 [35:28<56:59,  2.02it/s]
 35%|███▌      | 3796/10707 [35:28<57:01,  2.02it/s]
 35%|███▌      | 3797/10707 [35:29<57:02,  2.02it/s]
 35%|███▌      | 3798/10707 [35:29<57:01,  2.02it/s]
 35%|███▌      | 3799/10707 [35:30<57:04,  2.02it/s]
 35%|███▌      | 3800/10707 [35:30<57:03,  2.02it/s]{'loss': 3.8344, 'grad_norm': 0.19926698505878448, 'learning_rate': 0.0008148121936092801, 'epoch': 0.35}
-
                                                    

 35%|███▌      | 3800/10707 [35:30<57:03,  2.02it/s]
 36%|███▌      | 3801/10707 [35:31<57:06,  2.02it/s]
 36%|███▌      | 3802/10707 [35:31<57:06,  2.01it/s]
 36%|███▌      | 3803/10707 [35:32<57:06,  2.01it/s]
 36%|███▌      | 3804/10707 [35:32<57:06,  2.01it/s]
 36%|███▌      | 3805/10707 [35:33<57:04,  2.02it/s]
 36%|███▌      | 3806/10707 [35:33<57:01,  2.02it/s]
 36%|███▌      | 3807/10707 [35:34<56:59,  2.02it/s]
 36%|███▌      | 3808/10707 [35:34<56:57,  2.02it/s]
 36%|███▌      | 3809/10707 [35:35<57:01,  2.02it/s]
 36%|███▌      | 3810/10707 [35:35<56:59,  2.02it/s]
 36%|███▌      | 3811/10707 [35:36<56:54,  2.02it/s]
 36%|███▌      | 3812/10707 [35:36<56:55,  2.02it/s]
 36%|███▌      | 3813/10707 [35:37<56:52,  2.02it/s]
 36%|███▌      | 3814/10707 [35:37<56:56,  2.02it/s]
 36%|███▌      | 3815/10707 [35:38<56:52,  2.02it/s]
 36%|███▌      | 3816/10707 [35:38<56:52,  2.02it/s]
 36%|███▌      | 3817/10707 [35:39<56:59,  2.02it/s]
 36%|███▌      | 3818/10707 [35:39<56:55,  2.02it/s]
 36%|███▌      | 3819/10707 [35:40<56:57,  2.02it/s]
 36%|███▌      | 3820/10707 [35:40<56:53,  2.02it/s]
 36%|███▌      | 3821/10707 [35:41<56:54,  2.02it/s]
 36%|███▌      | 3822/10707 [35:41<56:51,  2.02it/s]
 36%|███▌      | 3823/10707 [35:42<56:55,  2.02it/s]
 36%|███▌      | 3824/10707 [35:42<56:51,  2.02it/s]
 36%|███▌      | 3825/10707 [35:43<56:52,  2.02it/s]{'loss': 3.8361, 'grad_norm': 0.1980392038822174, 'learning_rate': 0.000811635646806544, 'epoch': 0.36}
-
                                                    

 36%|███▌      | 3825/10707 [35:43<56:52,  2.02it/s]
 36%|███▌      | 3826/10707 [35:43<56:58,  2.01it/s]
 36%|███▌      | 3827/10707 [35:44<56:53,  2.02it/s]
 36%|███▌      | 3828/10707 [35:44<56:54,  2.01it/s]
 36%|███▌      | 3829/10707 [35:45<56:50,  2.02it/s]
 36%|███▌      | 3830/10707 [35:45<56:51,  2.02it/s]
 36%|███▌      | 3831/10707 [35:46<56:48,  2.02it/s]
 36%|███▌      | 3832/10707 [35:46<56:49,  2.02it/s]
 36%|███▌      | 3833/10707 [35:47<56:42,  2.02it/s]
 36%|███▌      | 3834/10707 [35:47<56:45,  2.02it/s]
 36%|███▌      | 3835/10707 [35:48<56:42,  2.02it/s]
 36%|███▌      | 3836/10707 [35:48<56:44,  2.02it/s]
 36%|███▌      | 3837/10707 [35:49<56:45,  2.02it/s]
 36%|███▌      | 3838/10707 [35:49<56:42,  2.02it/s]
 36%|███▌      | 3839/10707 [35:50<56:40,  2.02it/s]
 36%|███▌      | 3840/10707 [35:50<56:40,  2.02it/s]
 36%|███▌      | 3841/10707 [35:51<56:40,  2.02it/s]
 36%|███▌      | 3842/10707 [35:51<56:41,  2.02it/s]
 36%|███▌      | 3843/10707 [35:52<56:38,  2.02it/s]
 36%|███▌      | 3844/10707 [35:52<56:39,  2.02it/s]
 36%|███▌      | 3845/10707 [35:53<56:35,  2.02it/s]
 36%|███▌      | 3846/10707 [35:53<56:37,  2.02it/s]
 36%|███▌      | 3847/10707 [35:53<56:34,  2.02it/s]
 36%|███▌      | 3848/10707 [35:54<56:36,  2.02it/s]
 36%|███▌      | 3849/10707 [35:54<56:36,  2.02it/s]
 36%|███▌      | 3850/10707 [35:55<56:35,  2.02it/s]
                                                    
{'loss': 3.8333, 'grad_norm': 0.1961681842803955, 'learning_rate': 0.0008084383971171308, 'epoch': 0.36}
-
 36%|███▌      | 3850/10707 [35:55<56:35,  2.02it/s]
 36%|███▌      | 3851/10707 [35:55<56:39,  2.02it/s]
 36%|███▌      | 3852/10707 [35:56<56:38,  2.02it/s]
 36%|███▌      | 3853/10707 [35:56<56:34,  2.02it/s]
 36%|███▌      | 3854/10707 [35:57<56:35,  2.02it/s]
 36%|███▌      | 3855/10707 [35:57<56:34,  2.02it/s]
 36%|███▌      | 3856/10707 [35:58<56:32,  2.02it/s]
 36%|███▌      | 3857/10707 [35:58<56:33,  2.02it/s]
 36%|███▌      | 3858/10707 [35:59<56:26,  2.02it/s]
 36%|███▌      | 3859/10707 [35:59<56:30,  2.02it/s]
 36%|███▌      | 3860/10707 [36:00<56:28,  2.02it/s]
 36%|███▌      | 3861/10707 [36:00<56:28,  2.02it/s]
 36%|███▌      | 3862/10707 [36:01<56:27,  2.02it/s]
 36%|███▌      | 3863/10707 [36:01<56:29,  2.02it/s]
 36%|███▌      | 3864/10707 [36:02<56:26,  2.02it/s]
 36%|███▌      | 3865/10707 [36:02<56:29,  2.02it/s]
 36%|███▌      | 3866/10707 [36:03<56:26,  2.02it/s]
 36%|███▌      | 3867/10707 [36:03<56:28,  2.02it/s]
 36%|███▌      | 3868/10707 [36:04<56:27,  2.02it/s]
 36%|███▌      | 3869/10707 [36:04<56:28,  2.02it/s]
 36%|███▌      | 3870/10707 [36:05<56:30,  2.02it/s]
 36%|███▌      | 3871/10707 [36:05<56:25,  2.02it/s]
 36%|███▌      | 3872/10707 [36:06<56:23,  2.02it/s]
 36%|███▌      | 3873/10707 [36:06<56:21,  2.02it/s]
 36%|███▌      | 3874/10707 [36:07<56:23,  2.02it/s]
 36%|███▌      | 3875/10707 [36:07<56:23,  2.02it/s]{'loss': 3.8301, 'grad_norm': 0.2157057374715805, 'learning_rate': 0.0008052206569438861, 'epoch': 0.36}
-
                                                    

 36%|███▌      | 3875/10707 [36:07<56:23,  2.02it/s]
 36%|███▌      | 3876/10707 [36:08<56:34,  2.01it/s]
 36%|███▌      | 3877/10707 [36:08<56:31,  2.01it/s]
 36%|███▌      | 3878/10707 [36:09<56:25,  2.02it/s]
 36%|███▌      | 3879/10707 [36:09<56:24,  2.02it/s]
 36%|███▌      | 3880/10707 [36:10<56:23,  2.02it/s]
 36%|███▌      | 3881/10707 [36:10<56:21,  2.02it/s]
 36%|███▋      | 3882/10707 [36:11<56:20,  2.02it/s]
 36%|███▋      | 3883/10707 [36:11<56:18,  2.02it/s]
 36%|███▋      | 3884/10707 [36:12<56:20,  2.02it/s]
 36%|███▋      | 3885/10707 [36:12<56:19,  2.02it/s]
 36%|███▋      | 3886/10707 [36:13<56:19,  2.02it/s]
 36%|███▋      | 3887/10707 [36:13<56:20,  2.02it/s]
 36%|███▋      | 3888/10707 [36:14<56:17,  2.02it/s]
 36%|███▋      | 3889/10707 [36:14<56:17,  2.02it/s]
 36%|███▋      | 3890/10707 [36:15<56:21,  2.02it/s]
 36%|███▋      | 3891/10707 [36:15<56:20,  2.02it/s]
 36%|███▋      | 3892/10707 [36:16<56:17,  2.02it/s]
 36%|███▋      | 3893/10707 [36:16<56:12,  2.02it/s]
 36%|███▋      | 3894/10707 [36:17<56:10,  2.02it/s]
 36%|███▋      | 3895/10707 [36:17<56:08,  2.02it/s]
 36%|███▋      | 3896/10707 [36:18<56:11,  2.02it/s]
 36%|███▋      | 3897/10707 [36:18<56:09,  2.02it/s]
 36%|███▋      | 3898/10707 [36:19<56:13,  2.02it/s]
 36%|███▋      | 3899/10707 [36:19<56:08,  2.02it/s]
 36%|███▋      | 3900/10707 [36:20<56:10,  2.02it/s]
                                                    {'loss': 3.8247, 'grad_norm': 0.2183692306280136, 'learning_rate': 0.0008019826400508984, 'epoch': 0.36}
-
 36%|███▋      | 3900/10707 [36:20<56:10,  2.02it/s]
 36%|███▋      | 3901/10707 [36:20<56:12,  2.02it/s]
 36%|███▋      | 3902/10707 [36:21<56:13,  2.02it/s]
 36%|███▋      | 3903/10707 [36:21<56:10,  2.02it/s]
 36%|███▋      | 3904/10707 [36:22<56:11,  2.02it/s]
 36%|███▋      | 3905/10707 [36:22<56:12,  2.02it/s]
 36%|███▋      | 3906/10707 [36:23<56:10,  2.02it/s]
 36%|███▋      | 3907/10707 [36:23<56:11,  2.02it/s]
 36%|███▋      | 3908/10707 [36:24<56:11,  2.02it/s]
 37%|███▋      | 3909/10707 [36:24<56:10,  2.02it/s]
 37%|███▋      | 3910/10707 [36:25<56:05,  2.02it/s]
 37%|███▋      | 3911/10707 [36:25<56:05,  2.02it/s]
 37%|███▋      | 3912/10707 [36:26<1:05:11,  1.74it/s]
 37%|███▋      | 3913/10707 [36:26<1:02:24,  1.81it/s]
 37%|███▋      | 3914/10707 [36:27<1:00:31,  1.87it/s]
 37%|███▋      | 3915/10707 [36:27<59:09,  1.91it/s]  
 37%|███▋      | 3916/10707 [36:28<58:13,  1.94it/s]
 37%|███▋      | 3917/10707 [36:28<57:30,  1.97it/s]
 37%|███▋      | 3918/10707 [36:29<57:02,  1.98it/s]
 37%|███▋      | 3919/10707 [36:29<56:42,  1.99it/s]
 37%|███▋      | 3920/10707 [36:30<56:31,  2.00it/s]
 37%|███▋      | 3921/10707 [36:30<56:20,  2.01it/s]
 37%|███▋      | 3922/10707 [36:31<56:11,  2.01it/s]
 37%|███▋      | 3923/10707 [36:31<56:08,  2.01it/s]
 37%|███▋      | 3924/10707 [36:32<56:03,  2.02it/s]
 37%|███▋      | 3925/10707 [36:32<55:59,  2.02it/s]{'loss': 3.8308, 'grad_norm': 0.21561232209205627, 'learning_rate': 0.0007987245615493002, 'epoch': 0.37}
-
                                                    

 37%|███▋      | 3925/10707 [36:32<55:59,  2.02it/s]
 37%|███▋      | 3926/10707 [36:33<56:07,  2.01it/s]
 37%|███▋      | 3927/10707 [36:33<56:06,  2.01it/s]
 37%|███▋      | 3928/10707 [36:34<56:04,  2.02it/s]
 37%|███▋      | 3929/10707 [36:34<56:00,  2.02it/s]
 37%|███▋      | 3930/10707 [36:35<1:04:54,  1.74it/s]
 37%|███▋      | 3931/10707 [36:36<1:02:15,  1.81it/s]
 37%|███▋      | 3932/10707 [36:36<1:00:31,  1.87it/s]
 37%|███▋      | 3933/10707 [36:37<59:07,  1.91it/s]  
 37%|███▋      | 3934/10707 [36:37<58:09,  1.94it/s]
 37%|███▋      | 3935/10707 [36:38<57:32,  1.96it/s]
 37%|███▋      | 3936/10707 [36:38<56:59,  1.98it/s]
 37%|███▋      | 3937/10707 [36:39<56:39,  1.99it/s]
 37%|███▋      | 3938/10707 [36:39<56:22,  2.00it/s]
 37%|███▋      | 3939/10707 [36:40<56:16,  2.00it/s]
 37%|███▋      | 3940/10707 [36:40<56:05,  2.01it/s]
 37%|███▋      | 3941/10707 [36:41<56:03,  2.01it/s]
 37%|███▋      | 3942/10707 [36:41<55:58,  2.01it/s]
 37%|███▋      | 3943/10707 [36:42<55:54,  2.02it/s]
 37%|███▋      | 3944/10707 [36:42<55:50,  2.02it/s]
 37%|███▋      | 3945/10707 [36:43<55:53,  2.02it/s]
 37%|███▋      | 3946/10707 [36:43<55:49,  2.02it/s]
 37%|███▋      | 3947/10707 [36:44<55:49,  2.02it/s]
 37%|███▋      | 3948/10707 [36:44<55:47,  2.02it/s]
 37%|███▋      | 3949/10707 [36:45<55:47,  2.02it/s]
 37%|███▋      | 3950/10707 [36:45<55:42,  2.02it/s]{'loss': 3.8204, 'grad_norm': 0.20065820217132568, 'learning_rate': 0.000795446637882976, 'epoch': 0.37}
-
                                                    

 37%|███▋      | 3950/10707 [36:45<55:42,  2.02it/s]
 37%|███▋      | 3951/10707 [36:46<55:48,  2.02it/s]
 37%|███▋      | 3952/10707 [36:46<55:46,  2.02it/s]
 37%|███▋      | 3953/10707 [36:47<55:46,  2.02it/s]
 37%|███▋      | 3954/10707 [36:47<55:43,  2.02it/s]
 37%|███▋      | 3955/10707 [36:48<55:42,  2.02it/s]
 37%|███▋      | 3956/10707 [36:48<55:39,  2.02it/s]
 37%|███▋      | 3957/10707 [36:49<55:41,  2.02it/s]
 37%|███▋      | 3958/10707 [36:49<55:44,  2.02it/s]
 37%|███▋      | 3959/10707 [36:50<55:43,  2.02it/s]
 37%|███▋      | 3960/10707 [36:50<55:43,  2.02it/s]
 37%|███▋      | 3961/10707 [36:50<55:38,  2.02it/s]
 37%|███▋      | 3962/10707 [36:51<55:39,  2.02it/s]
 37%|███▋      | 3963/10707 [36:51<55:37,  2.02it/s]
 37%|███▋      | 3964/10707 [36:52<55:41,  2.02it/s]
 37%|███▋      | 3965/10707 [36:52<55:35,  2.02it/s]
 37%|███▋      | 3966/10707 [36:53<55:38,  2.02it/s]
 37%|███▋      | 3967/10707 [36:53<55:34,  2.02it/s]
 37%|███▋      | 3968/10707 [36:54<55:37,  2.02it/s]
 37%|███▋      | 3969/10707 [36:54<55:32,  2.02it/s]
 37%|███▋      | 3970/10707 [36:55<55:35,  2.02it/s]
 37%|███▋      | 3971/10707 [36:55<55:31,  2.02it/s]
 37%|███▋      | 3972/10707 [36:56<55:34,  2.02it/s]
 37%|███▋      | 3973/10707 [36:56<55:30,  2.02it/s]
 37%|███▋      | 3974/10707 [36:57<55:33,  2.02it/s]
 37%|███▋      | 3975/10707 [36:57<55:29,  2.02it/s]{'loss': 3.8225, 'grad_norm': 0.20580913126468658, 'learning_rate': 0.0007921490868141843, 'epoch': 0.37}
-
                                                    

 37%|███▋      | 3975/10707 [36:57<55:29,  2.02it/s]
 37%|███▋      | 3976/10707 [36:58<55:35,  2.02it/s]
 37%|███▋      | 3977/10707 [36:58<55:34,  2.02it/s]
 37%|███▋      | 3978/10707 [36:59<55:34,  2.02it/s]
 37%|███▋      | 3979/10707 [36:59<55:31,  2.02it/s]
 37%|███▋      | 3980/10707 [37:00<55:28,  2.02it/s]
 37%|███▋      | 3981/10707 [37:00<55:31,  2.02it/s]
 37%|███▋      | 3982/10707 [37:01<55:30,  2.02it/s]
 37%|███▋      | 3983/10707 [37:01<55:32,  2.02it/s]
 37%|███▋      | 3984/10707 [37:02<55:30,  2.02it/s]
 37%|███▋      | 3985/10707 [37:02<55:27,  2.02it/s]
 37%|███▋      | 3986/10707 [37:03<55:27,  2.02it/s]
 37%|███▋      | 3987/10707 [37:03<55:24,  2.02it/s]
 37%|███▋      | 3988/10707 [37:04<55:30,  2.02it/s]
 37%|███▋      | 3989/10707 [37:04<55:29,  2.02it/s]
 37%|███▋      | 3990/10707 [37:05<55:29,  2.02it/s]
 37%|███▋      | 3991/10707 [37:05<55:26,  2.02it/s]
 37%|███▋      | 3992/10707 [37:06<55:25,  2.02it/s]
 37%|███▋      | 3993/10707 [37:06<55:23,  2.02it/s]
 37%|███▋      | 3994/10707 [37:07<55:24,  2.02it/s]
 37%|███▋      | 3995/10707 [37:07<55:25,  2.02it/s]
 37%|███▋      | 3996/10707 [37:08<55:26,  2.02it/s]
 37%|███▋      | 3997/10707 [37:08<55:22,  2.02it/s]
 37%|███▋      | 3998/10707 [37:09<55:25,  2.02it/s]
 37%|███▋      | 3999/10707 [37:09<55:19,  2.02it/s]
 37%|███▋      | 4000/10707 [37:10<55:20,  2.02it/s]
                                                    {'loss': 3.8278, 'grad_norm': 0.19656354188919067, 'learning_rate': 0.00078883212740909, 'epoch': 0.37}
-
 37%|███▋      | 4000/10707 [37:10<55:20,  2.02it/s]
 37%|███▋      | 4001/10707 [37:10<55:23,  2.02it/s]
 37%|███▋      | 4002/10707 [37:11<55:29,  2.01it/s]
 37%|███▋      | 4003/10707 [37:11<55:24,  2.02it/s]
 37%|███▋      | 4004/10707 [37:12<55:20,  2.02it/s]
 37%|███▋      | 4005/10707 [37:12<55:19,  2.02it/s]
 37%|███▋      | 4006/10707 [37:13<55:19,  2.02it/s]
 37%|███▋      | 4007/10707 [37:13<55:17,  2.02it/s]
 37%|███▋      | 4008/10707 [37:14<55:16,  2.02it/s]
 37%|███▋      | 4009/10707 [37:14<55:16,  2.02it/s]
 37%|███▋      | 4010/10707 [37:15<55:16,  2.02it/s]
 37%|███▋      | 4011/10707 [37:15<55:20,  2.02it/s]
 37%|███▋      | 4012/10707 [37:16<55:15,  2.02it/s]
 37%|███▋      | 4013/10707 [37:16<55:18,  2.02it/s]
 37%|███▋      | 4014/10707 [37:17<55:15,  2.02it/s]
 37%|███▋      | 4015/10707 [37:17<55:15,  2.02it/s]
 38%|███▊      | 4016/10707 [37:18<55:11,  2.02it/s]
 38%|███▊      | 4017/10707 [37:18<55:13,  2.02it/s]
 38%|███▊      | 4018/10707 [37:19<55:09,  2.02it/s]
 38%|███▊      | 4019/10707 [37:19<55:12,  2.02it/s]
 38%|███▊      | 4020/10707 [37:20<55:08,  2.02it/s]
 38%|███▊      | 4021/10707 [37:20<55:09,  2.02it/s]
 38%|███▊      | 4022/10707 [37:21<55:07,  2.02it/s]
 38%|███▊      | 4023/10707 [37:21<55:07,  2.02it/s]
 38%|███▊      | 4024/10707 [37:22<55:03,  2.02it/s]
 38%|███▊      | 4025/10707 [37:22<55:04,  2.02it/s]{'loss': 3.8178, 'grad_norm': 0.1954190582036972, 'learning_rate': 0.0007854959800232122, 'epoch': 0.38}
-
                                                    

 38%|███▊      | 4025/10707 [37:22<55:04,  2.02it/s]
 38%|███▊      | 4026/10707 [37:23<55:10,  2.02it/s]
 38%|███▊      | 4027/10707 [37:23<55:05,  2.02it/s]
 38%|███▊      | 4028/10707 [37:24<55:09,  2.02it/s]
 38%|███▊      | 4029/10707 [37:24<55:06,  2.02it/s]
 38%|███▊      | 4030/10707 [37:25<55:07,  2.02it/s]
 38%|███▊      | 4031/10707 [37:25<55:06,  2.02it/s]
 38%|███▊      | 4032/10707 [37:26<55:09,  2.02it/s]
 38%|███▊      | 4033/10707 [37:26<55:04,  2.02it/s]
 38%|███▊      | 4034/10707 [37:27<55:06,  2.02it/s]
 38%|███▊      | 4035/10707 [37:27<55:01,  2.02it/s]
 38%|███▊      | 4036/10707 [37:28<55:03,  2.02it/s]
 38%|███▊      | 4037/10707 [37:28<54:57,  2.02it/s]
 38%|███▊      | 4038/10707 [37:29<55:02,  2.02it/s]
 38%|███▊      | 4039/10707 [37:29<55:01,  2.02it/s]
 38%|███▊      | 4040/10707 [37:30<55:02,  2.02it/s]
 38%|███▊      | 4041/10707 [37:30<55:00,  2.02it/s]
 38%|███▊      | 4042/10707 [37:31<55:01,  2.02it/s]
 38%|███▊      | 4043/10707 [37:31<55:01,  2.02it/s]
 38%|███▊      | 4044/10707 [37:32<54:59,  2.02it/s]
 38%|███▊      | 4045/10707 [37:32<54:59,  2.02it/s]
 38%|███▊      | 4046/10707 [37:33<54:56,  2.02it/s]
 38%|███▊      | 4047/10707 [37:33<54:54,  2.02it/s]
 38%|███▊      | 4048/10707 [37:34<54:54,  2.02it/s]
 38%|███▊      | 4049/10707 [37:34<54:55,  2.02it/s]
 38%|███▊      | 4050/10707 [37:35<55:06,  2.01it/s]{'loss': 3.8183, 'grad_norm': 0.19633789360523224, 'learning_rate': 0.0007821408662867847, 'epoch': 0.38}
-
                                                    

 38%|███▊      | 4050/10707 [37:35<55:06,  2.01it/s]
 38%|███▊      | 4051/10707 [37:35<55:08,  2.01it/s]
 38%|███▊      | 4052/10707 [37:36<55:06,  2.01it/s]
 38%|███▊      | 4053/10707 [37:36<55:02,  2.01it/s]
 38%|███▊      | 4054/10707 [37:37<55:01,  2.02it/s]
 38%|███▊      | 4055/10707 [37:37<54:59,  2.02it/s]
 38%|███▊      | 4056/10707 [37:38<55:01,  2.01it/s]
 38%|███▊      | 4057/10707 [37:38<55:01,  2.01it/s]
 38%|███▊      | 4058/10707 [37:39<54:59,  2.02it/s]
 38%|███▊      | 4059/10707 [37:39<54:58,  2.02it/s]
 38%|███▊      | 4060/10707 [37:40<54:54,  2.02it/s]
 38%|███▊      | 4061/10707 [37:40<54:54,  2.02it/s]
 38%|███▊      | 4062/10707 [37:41<54:50,  2.02it/s]
 38%|███▊      | 4063/10707 [37:41<54:51,  2.02it/s]
 38%|███▊      | 4064/10707 [37:42<54:47,  2.02it/s]
 38%|███▊      | 4065/10707 [37:42<54:46,  2.02it/s]
 38%|███▊      | 4066/10707 [37:43<54:49,  2.02it/s]
 38%|███▊      | 4067/10707 [37:43<54:45,  2.02it/s]
 38%|███▊      | 4068/10707 [37:43<54:48,  2.02it/s]
 38%|███▊      | 4069/10707 [37:44<54:45,  2.02it/s]
 38%|███▊      | 4070/10707 [37:44<54:46,  2.02it/s]
 38%|███▊      | 4071/10707 [37:45<54:45,  2.02it/s]
 38%|███▊      | 4072/10707 [37:45<54:45,  2.02it/s]
 38%|███▊      | 4073/10707 [37:46<54:47,  2.02it/s]
 38%|███▊      | 4074/10707 [37:46<54:46,  2.02it/s]
 38%|███▊      | 4075/10707 [37:47<54:43,  2.02it/s]
                                                    
{'loss': 3.819, 'grad_norm': 0.19644653797149658, 'learning_rate': 0.0007787670090900327, 'epoch': 0.38}
-
 38%|███▊      | 4075/10707 [37:47<54:43,  2.02it/s]
 38%|███▊      | 4076/10707 [37:47<54:46,  2.02it/s]
 38%|███▊      | 4077/10707 [37:48<54:44,  2.02it/s]
 38%|███▊      | 4078/10707 [37:48<54:42,  2.02it/s]
 38%|███▊      | 4079/10707 [37:49<54:40,  2.02it/s]
 38%|███▊      | 4080/10707 [37:49<54:39,  2.02it/s]
 38%|███▊      | 4081/10707 [37:50<54:37,  2.02it/s]
 38%|███▊      | 4082/10707 [37:50<54:38,  2.02it/s]
 38%|███▊      | 4083/10707 [37:51<54:37,  2.02it/s]
 38%|███▊      | 4084/10707 [37:51<54:38,  2.02it/s]
 38%|███▊      | 4085/10707 [37:52<54:37,  2.02it/s]
 38%|███▊      | 4086/10707 [37:52<54:39,  2.02it/s]
 38%|███▊      | 4087/10707 [37:53<54:39,  2.02it/s]
 38%|███▊      | 4088/10707 [37:53<54:39,  2.02it/s]
 38%|███▊      | 4089/10707 [37:54<54:37,  2.02it/s]
 38%|███▊      | 4090/10707 [37:54<54:35,  2.02it/s]
 38%|███▊      | 4091/10707 [37:55<54:33,  2.02it/s]
 38%|███▊      | 4092/10707 [37:55<54:32,  2.02it/s]
 38%|███▊      | 4093/10707 [37:56<54:37,  2.02it/s]
 38%|███▊      | 4094/10707 [37:56<54:38,  2.02it/s]
 38%|███▊      | 4095/10707 [37:57<54:37,  2.02it/s]
 38%|███▊      | 4096/10707 [37:57<54:35,  2.02it/s]
 38%|███▊      | 4097/10707 [37:58<54:36,  2.02it/s]
 38%|███▊      | 4098/10707 [37:58<54:35,  2.02it/s]
 38%|███▊      | 4099/10707 [37:59<54:34,  2.02it/s]
 38%|███▊      | 4100/10707 [37:59<54:32,  2.02it/s]
                                                    {'loss': 3.809, 'grad_norm': 0.2135017216205597, 'learning_rate': 0.0007753746325683653, 'epoch': 0.38}
-
 38%|███▊      | 4100/10707 [37:59<54:32,  2.02it/s]
 38%|███▊      | 4101/10707 [38:00<54:35,  2.02it/s]
 38%|███▊      | 4102/10707 [38:00<54:33,  2.02it/s]
 38%|███▊      | 4103/10707 [38:01<54:32,  2.02it/s]
 38%|███▊      | 4104/10707 [38:01<54:30,  2.02it/s]
 38%|███▊      | 4105/10707 [38:02<54:27,  2.02it/s]
 38%|███▊      | 4106/10707 [38:02<54:39,  2.01it/s]
 38%|███▊      | 4107/10707 [38:03<54:39,  2.01it/s]
 38%|███▊      | 4108/10707 [38:03<54:36,  2.01it/s]
 38%|███▊      | 4109/10707 [38:04<54:34,  2.01it/s]
 38%|███▊      | 4110/10707 [38:04<54:30,  2.02it/s]
 38%|███▊      | 4111/10707 [38:05<54:31,  2.02it/s]
 38%|███▊      | 4112/10707 [38:05<54:27,  2.02it/s]
 38%|███▊      | 4113/10707 [38:06<54:31,  2.02it/s]
 38%|███▊      | 4114/10707 [38:06<54:27,  2.02it/s]
 38%|███▊      | 4115/10707 [38:07<54:25,  2.02it/s]
 38%|███▊      | 4116/10707 [38:07<54:24,  2.02it/s]
 38%|███▊      | 4117/10707 [38:08<54:25,  2.02it/s]
 38%|███▊      | 4118/10707 [38:08<54:24,  2.02it/s]
 38%|███▊      | 4119/10707 [38:09<54:23,  2.02it/s]
 38%|███▊      | 4120/10707 [38:09<54:21,  2.02it/s]
 38%|███▊      | 4121/10707 [38:10<54:24,  2.02it/s]
 38%|███▊      | 4122/10707 [38:10<54:21,  2.02it/s]
 39%|███▊      | 4123/10707 [38:11<54:19,  2.02it/s]
 39%|███▊      | 4124/10707 [38:11<54:19,  2.02it/s]
 39%|███▊      | 4125/10707 [38:12<54:19,  2.02it/s]{'loss': 3.8215, 'grad_norm': 0.19239623844623566, 'learning_rate': 0.0007719639620874856, 'epoch': 0.39}
-
                                                    

 39%|███▊      | 4125/10707 [38:12<54:19,  2.02it/s]
 39%|███▊      | 4126/10707 [38:12<54:23,  2.02it/s]
 39%|███▊      | 4127/10707 [38:13<54:21,  2.02it/s]
 39%|███▊      | 4128/10707 [38:13<54:21,  2.02it/s]
 39%|███▊      | 4129/10707 [38:14<54:18,  2.02it/s]
 39%|███▊      | 4130/10707 [38:14<54:18,  2.02it/s]
 39%|███▊      | 4131/10707 [38:15<54:16,  2.02it/s]
 39%|███▊      | 4132/10707 [38:15<54:16,  2.02it/s]
 39%|███▊      | 4133/10707 [38:16<54:16,  2.02it/s]
 39%|███▊      | 4134/10707 [38:16<54:15,  2.02it/s]
 39%|███▊      | 4135/10707 [38:17<54:16,  2.02it/s]
 39%|███▊      | 4136/10707 [38:17<54:12,  2.02it/s]
 39%|███▊      | 4137/10707 [38:18<54:11,  2.02it/s]
 39%|███▊      | 4138/10707 [38:18<54:15,  2.02it/s]
 39%|███▊      | 4139/10707 [38:19<54:15,  2.02it/s]
 39%|███▊      | 4140/10707 [38:19<54:14,  2.02it/s]
 39%|███▊      | 4141/10707 [38:20<54:13,  2.02it/s]
 39%|███▊      | 4142/10707 [38:20<54:08,  2.02it/s]
 39%|███▊      | 4143/10707 [38:21<54:10,  2.02it/s]
 39%|███▊      | 4144/10707 [38:21<54:08,  2.02it/s]
 39%|███▊      | 4145/10707 [38:22<54:06,  2.02it/s]
 39%|███▊      | 4146/10707 [38:22<54:03,  2.02it/s]
 39%|███▊      | 4147/10707 [38:23<54:05,  2.02it/s]
 39%|███▊      | 4148/10707 [38:23<54:12,  2.02it/s]
 39%|███▉      | 4149/10707 [38:24<54:09,  2.02it/s]
 39%|███▉      | 4150/10707 [38:24<54:09,  2.02it/s]{'loss': 3.8057, 'grad_norm': 0.21154125034809113, 'learning_rate': 0.0007685352242284193, 'epoch': 0.39}
-
                                                    

 39%|███▉      | 4150/10707 [38:24<54:09,  2.02it/s]
 39%|███▉      | 4151/10707 [38:25<54:12,  2.02it/s]
 39%|███▉      | 4152/10707 [38:25<54:20,  2.01it/s]
 39%|███▉      | 4153/10707 [38:26<54:18,  2.01it/s]
 39%|███▉      | 4154/10707 [38:26<54:14,  2.01it/s]
 39%|███▉      | 4155/10707 [38:27<54:13,  2.01it/s]
 39%|███▉      | 4156/10707 [38:27<54:09,  2.02it/s]
 39%|███▉      | 4157/10707 [38:28<54:09,  2.02it/s]
 39%|███▉      | 4158/10707 [38:28<54:15,  2.01it/s]
 39%|███▉      | 4159/10707 [38:29<54:15,  2.01it/s]
 39%|███▉      | 4160/10707 [38:29<54:11,  2.01it/s]
 39%|███▉      | 4161/10707 [38:30<54:08,  2.01it/s]
 39%|███▉      | 4162/10707 [38:30<54:04,  2.02it/s]
 39%|███▉      | 4163/10707 [38:31<54:04,  2.02it/s]
 39%|███▉      | 4164/10707 [38:31<54:02,  2.02it/s]
 39%|███▉      | 4165/10707 [38:32<54:03,  2.02it/s]
 39%|███▉      | 4166/10707 [38:32<53:59,  2.02it/s]
 39%|███▉      | 4167/10707 [38:33<54:00,  2.02it/s]
 39%|███▉      | 4168/10707 [38:33<53:55,  2.02it/s]
 39%|███▉      | 4169/10707 [38:34<53:56,  2.02it/s]
 39%|███▉      | 4170/10707 [38:34<53:53,  2.02it/s]
 39%|███▉      | 4171/10707 [38:35<53:55,  2.02it/s]
 39%|███▉      | 4172/10707 [38:35<53:52,  2.02it/s]
 39%|███▉      | 4173/10707 [38:36<53:56,  2.02it/s]
 39%|███▉      | 4174/10707 [38:36<53:52,  2.02it/s]
 39%|███▉      | 4175/10707 [38:37<53:53,  2.02it/s]
                                                    {'loss': 3.8022, 'grad_norm': 0.2079610526561737, 'learning_rate': 0.000765088646772462, 'epoch': 0.39}
-

 39%|███▉      | 4175/10707 [38:37<53:53,  2.02it/s]
 39%|███▉      | 4176/10707 [38:37<53:55,  2.02it/s]
 39%|███▉      | 4177/10707 [38:38<53:55,  2.02it/s]
 39%|███▉      | 4178/10707 [38:38<53:52,  2.02it/s]
 39%|███▉      | 4179/10707 [38:38<53:54,  2.02it/s]
 39%|███▉      | 4180/10707 [38:39<53:54,  2.02it/s]
 39%|███▉      | 4181/10707 [38:39<53:51,  2.02it/s]
 39%|███▉      | 4182/10707 [38:40<53:51,  2.02it/s]
 39%|███▉      | 4183/10707 [38:40<53:52,  2.02it/s]
 39%|███▉      | 4184/10707 [38:41<53:49,  2.02it/s]
 39%|███▉      | 4185/10707 [38:41<53:47,  2.02it/s]
 39%|███▉      | 4186/10707 [38:42<53:46,  2.02it/s]
 39%|███▉      | 4187/10707 [38:42<53:45,  2.02it/s]
 39%|███▉      | 4188/10707 [38:43<53:45,  2.02it/s]
 39%|███▉      | 4189/10707 [38:43<53:42,  2.02it/s]
 39%|███▉      | 4190/10707 [38:44<53:44,  2.02it/s]
 39%|███▉      | 4191/10707 [38:44<53:46,  2.02it/s]
 39%|███▉      | 4192/10707 [38:45<53:45,  2.02it/s]
 39%|███▉      | 4193/10707 [38:45<53:43,  2.02it/s]
 39%|███▉      | 4194/10707 [38:46<53:43,  2.02it/s]
 39%|███▉      | 4195/10707 [38:46<53:42,  2.02it/s]
 39%|███▉      | 4196/10707 [38:47<53:46,  2.02it/s]
 39%|███▉      | 4197/10707 [38:47<53:47,  2.02it/s]
 39%|███▉      | 4198/10707 [38:48<53:46,  2.02it/s]
 39%|███▉      | 4199/10707 [38:48<53:45,  2.02it/s]
 39%|███▉      | 4200/10707 [38:49<53:43,  2.02it/s]{'loss': 3.804, 'grad_norm': 0.20403452217578888, 'learning_rate': 0.0007616244586860465, 'epoch': 0.39}
-
                                                    

 39%|███▉      | 4200/10707 [38:49<53:43,  2.02it/s]
 39%|███▉      | 4201/10707 [38:49<53:47,  2.02it/s]
 39%|███▉      | 4202/10707 [38:50<53:49,  2.01it/s]
 39%|███▉      | 4203/10707 [38:50<53:58,  2.01it/s]
 39%|███▉      | 4204/10707 [38:51<53:50,  2.01it/s]
 39%|███▉      | 4205/10707 [38:51<53:49,  2.01it/s]
 39%|███▉      | 4206/10707 [38:52<53:44,  2.02it/s]
 39%|███▉      | 4207/10707 [38:52<53:45,  2.01it/s]
 39%|███▉      | 4208/10707 [38:53<53:40,  2.02it/s]
 39%|███▉      | 4209/10707 [38:53<53:49,  2.01it/s]
 39%|███▉      | 4210/10707 [38:54<53:47,  2.01it/s]
 39%|███▉      | 4211/10707 [38:54<53:44,  2.01it/s]
 39%|███▉      | 4212/10707 [38:55<53:41,  2.02it/s]
 39%|███▉      | 4213/10707 [38:55<53:37,  2.02it/s]
 39%|███▉      | 4214/10707 [38:56<53:38,  2.02it/s]
 39%|███▉      | 4215/10707 [38:56<53:37,  2.02it/s]
 39%|███▉      | 4216/10707 [38:57<53:39,  2.02it/s]
 39%|███▉      | 4217/10707 [38:57<53:37,  2.02it/s]
 39%|███▉      | 4218/10707 [38:58<53:39,  2.02it/s]
 39%|███▉      | 4219/10707 [38:58<53:36,  2.02it/s]
 39%|███▉      | 4220/10707 [38:59<53:36,  2.02it/s]
 39%|███▉      | 4221/10707 [38:59<53:34,  2.02it/s]
 39%|███▉      | 4222/10707 [39:00<53:32,  2.02it/s]
 39%|███▉      | 4223/10707 [39:00<53:38,  2.01it/s]
 39%|███▉      | 4224/10707 [39:01<53:32,  2.02it/s]
 39%|███▉      | 4225/10707 [39:01<53:32,  2.02it/s]{'loss': 3.8074, 'grad_norm': 0.21652792394161224, 'learning_rate': 0.0007581428901055332, 'epoch': 0.39}
-
                                                    

 39%|███▉      | 4225/10707 [39:01<53:32,  2.02it/s]
 39%|███▉      | 4226/10707 [39:02<53:36,  2.01it/s]
 39%|███▉      | 4227/10707 [39:02<53:37,  2.01it/s]
 39%|███▉      | 4228/10707 [39:03<53:38,  2.01it/s]
 39%|███▉      | 4229/10707 [39:03<53:31,  2.02it/s]
 40%|███▉      | 4230/10707 [39:04<53:31,  2.02it/s]
 40%|███▉      | 4231/10707 [39:04<53:29,  2.02it/s]
 40%|███▉      | 4232/10707 [39:05<53:28,  2.02it/s]
 40%|███▉      | 4233/10707 [39:05<53:27,  2.02it/s]
 40%|███▉      | 4234/10707 [39:06<53:30,  2.02it/s]
 40%|███▉      | 4235/10707 [39:06<53:25,  2.02it/s]
 40%|███▉      | 4236/10707 [39:07<53:31,  2.02it/s]
 40%|███▉      | 4237/10707 [39:07<53:27,  2.02it/s]
 40%|███▉      | 4238/10707 [39:08<53:24,  2.02it/s]
 40%|███▉      | 4239/10707 [39:08<53:20,  2.02it/s]
 40%|███▉      | 4240/10707 [39:09<53:22,  2.02it/s]
 40%|███▉      | 4241/10707 [39:09<53:19,  2.02it/s]
 40%|███▉      | 4242/10707 [39:10<53:18,  2.02it/s]
 40%|███▉      | 4243/10707 [39:10<53:15,  2.02it/s]
 40%|███▉      | 4244/10707 [39:11<53:16,  2.02it/s]
 40%|███▉      | 4245/10707 [39:11<53:13,  2.02it/s]
 40%|███▉      | 4246/10707 [39:12<53:19,  2.02it/s]
 40%|███▉      | 4247/10707 [39:12<53:16,  2.02it/s]
 40%|███▉      | 4248/10707 [39:13<53:16,  2.02it/s]
 40%|███▉      | 4249/10707 [39:13<53:15,  2.02it/s]
 40%|███▉      | 4250/10707 [39:14<53:17,  2.02it/s]{'loss': 3.8005, 'grad_norm': 0.19776679575443268, 'learning_rate': 0.0007546441723219197, 'epoch': 0.4}
                                                    
-
 40%|███▉      | 4250/10707 [39:14<53:17,  2.02it/s]
 40%|███▉      | 4251/10707 [39:14<53:16,  2.02it/s]
 40%|███▉      | 4252/10707 [39:15<53:16,  2.02it/s]
 40%|███▉      | 4253/10707 [39:15<53:21,  2.02it/s]
 40%|███▉      | 4254/10707 [39:16<53:20,  2.02it/s]
 40%|███▉      | 4255/10707 [39:16<53:18,  2.02it/s]
 40%|███▉      | 4256/10707 [39:17<53:18,  2.02it/s]
 40%|███▉      | 4257/10707 [39:17<53:17,  2.02it/s]
 40%|███▉      | 4258/10707 [39:18<53:15,  2.02it/s]
 40%|███▉      | 4259/10707 [39:18<53:15,  2.02it/s]
 40%|███▉      | 4260/10707 [39:19<53:16,  2.02it/s]
 40%|███▉      | 4261/10707 [39:19<53:14,  2.02it/s]
 40%|███▉      | 4262/10707 [39:20<53:14,  2.02it/s]
 40%|███▉      | 4263/10707 [39:20<53:13,  2.02it/s]
 40%|███▉      | 4264/10707 [39:21<53:12,  2.02it/s]
 40%|███▉      | 4265/10707 [39:21<53:11,  2.02it/s]
 40%|███▉      | 4266/10707 [39:22<53:13,  2.02it/s]
 40%|███▉      | 4267/10707 [39:22<53:06,  2.02it/s]
 40%|███▉      | 4268/10707 [39:23<53:07,  2.02it/s]
 40%|███▉      | 4269/10707 [39:23<53:05,  2.02it/s]
 40%|███▉      | 4270/10707 [39:24<53:05,  2.02it/s]
 40%|███▉      | 4271/10707 [39:24<53:05,  2.02it/s]
 40%|███▉      | 4272/10707 [39:25<53:03,  2.02it/s]
 40%|███▉      | 4273/10707 [39:25<53:03,  2.02it/s]
 40%|███▉      | 4274/10707 [39:26<53:04,  2.02it/s]
 40%|███▉      | 4275/10707 [39:26<53:01,  2.02it/s]{'loss': 3.8069, 'grad_norm': 0.2222370207309723, 'learning_rate': 0.0007511285377654771, 'epoch': 0.4}
                                                    
-
 40%|███▉      | 4275/10707 [39:26<53:01,  2.02it/s]
 40%|███▉      | 4276/10707 [39:27<53:08,  2.02it/s]
 40%|███▉      | 4277/10707 [39:27<53:05,  2.02it/s]
 40%|███▉      | 4278/10707 [39:28<53:06,  2.02it/s]
 40%|███▉      | 4279/10707 [39:28<53:04,  2.02it/s]
 40%|███▉      | 4280/10707 [39:29<53:03,  2.02it/s]
 40%|███▉      | 4281/10707 [39:29<53:01,  2.02it/s]
 40%|███▉      | 4282/10707 [39:30<53:00,  2.02it/s]
 40%|████      | 4283/10707 [39:30<52:59,  2.02it/s]
 40%|████      | 4284/10707 [39:31<52:58,  2.02it/s]
 40%|████      | 4285/10707 [39:31<52:55,  2.02it/s]
 40%|████      | 4286/10707 [39:32<52:57,  2.02it/s]
 40%|████      | 4287/10707 [39:32<52:54,  2.02it/s]
 40%|████      | 4288/10707 [39:33<52:56,  2.02it/s]
 40%|████      | 4289/10707 [39:33<52:53,  2.02it/s]
 40%|████      | 4290/10707 [39:33<52:57,  2.02it/s]
 40%|████      | 4291/10707 [39:34<52:54,  2.02it/s]
 40%|████      | 4292/10707 [39:34<52:53,  2.02it/s]
 40%|████      | 4293/10707 [39:35<52:53,  2.02it/s]
 40%|████      | 4294/10707 [39:35<52:53,  2.02it/s]
 40%|████      | 4295/10707 [39:36<52:50,  2.02it/s]
 40%|████      | 4296/10707 [39:36<52:52,  2.02it/s]
 40%|████      | 4297/10707 [39:37<52:51,  2.02it/s]
 40%|████      | 4298/10707 [39:37<52:52,  2.02it/s]
 40%|████      | 4299/10707 [39:38<52:53,  2.02it/s]
 40%|████      | 4300/10707 [39:38<52:50,  2.02it/s]{'loss': 3.7924, 'grad_norm': 0.1930372267961502, 'learning_rate': 0.0007475962199903079, 'epoch': 0.4}
-
                                                    

 40%|████      | 4300/10707 [39:38<52:50,  2.02it/s]
 40%|████      | 4301/10707 [39:39<52:56,  2.02it/s]
 40%|████      | 4302/10707 [39:39<52:50,  2.02it/s]
 40%|████      | 4303/10707 [39:40<52:51,  2.02it/s]
 40%|████      | 4304/10707 [39:40<52:47,  2.02it/s]
 40%|████      | 4305/10707 [39:41<52:47,  2.02it/s]
 40%|████      | 4306/10707 [39:41<52:48,  2.02it/s]
 40%|████      | 4307/10707 [39:42<52:51,  2.02it/s]
 40%|████      | 4308/10707 [39:42<52:47,  2.02it/s]
 40%|████      | 4309/10707 [39:43<52:49,  2.02it/s]
 40%|████      | 4310/10707 [39:43<52:46,  2.02it/s]
 40%|████      | 4311/10707 [39:44<52:47,  2.02it/s]
 40%|████      | 4312/10707 [39:44<52:46,  2.02it/s]
 40%|████      | 4313/10707 [39:45<52:47,  2.02it/s]
 40%|████      | 4314/10707 [39:45<52:48,  2.02it/s]
 40%|████      | 4315/10707 [39:46<52:44,  2.02it/s]
 40%|████      | 4316/10707 [39:46<52:45,  2.02it/s]
 40%|████      | 4317/10707 [39:47<52:45,  2.02it/s]
 40%|████      | 4318/10707 [39:47<52:45,  2.02it/s]
 40%|████      | 4319/10707 [39:48<52:44,  2.02it/s]
 40%|████      | 4320/10707 [39:48<52:44,  2.02it/s]
 40%|████      | 4321/10707 [39:49<52:41,  2.02it/s]
 40%|████      | 4322/10707 [39:49<52:43,  2.02it/s]
 40%|████      | 4323/10707 [39:50<52:41,  2.02it/s]
 40%|████      | 4324/10707 [39:50<52:41,  2.02it/s]
 40%|████      | 4325/10707 [39:51<52:39,  2.02it/s]
                                                    {'loss': 3.7995, 'grad_norm': 0.21055631339550018, 'learning_rate': 0.0007440474536588306, 'epoch': 0.4}
-
 40%|████      | 4325/10707 [39:51<52:39,  2.02it/s]
 40%|████      | 4326/10707 [39:51<52:43,  2.02it/s]
 40%|████      | 4327/10707 [39:52<52:42,  2.02it/s]
 40%|████      | 4328/10707 [39:52<52:41,  2.02it/s]
 40%|████      | 4329/10707 [39:53<52:41,  2.02it/s]
 40%|████      | 4330/10707 [39:53<52:39,  2.02it/s]
 40%|████      | 4331/10707 [39:54<52:39,  2.02it/s]
 40%|████      | 4332/10707 [39:54<52:37,  2.02it/s]
 40%|████      | 4333/10707 [39:55<52:40,  2.02it/s]
 40%|████      | 4334/10707 [39:55<52:39,  2.02it/s]
 40%|████      | 4335/10707 [39:56<52:38,  2.02it/s]
 40%|████      | 4336/10707 [39:56<52:38,  2.02it/s]
 41%|████      | 4337/10707 [39:57<52:39,  2.02it/s]
 41%|████      | 4338/10707 [39:57<52:34,  2.02it/s]
 41%|████      | 4339/10707 [39:58<52:35,  2.02it/s]
 41%|████      | 4340/10707 [39:58<52:35,  2.02it/s]
 41%|████      | 4341/10707 [39:59<52:36,  2.02it/s]
 41%|████      | 4342/10707 [39:59<52:32,  2.02it/s]
 41%|████      | 4343/10707 [40:00<52:33,  2.02it/s]
 41%|████      | 4344/10707 [40:00<52:31,  2.02it/s]
 41%|████      | 4345/10707 [40:01<52:32,  2.02it/s]
 41%|████      | 4346/10707 [40:01<52:32,  2.02it/s]
 41%|████      | 4347/10707 [40:02<52:30,  2.02it/s]
 41%|████      | 4348/10707 [40:02<52:28,  2.02it/s]
 41%|████      | 4349/10707 [40:03<52:26,  2.02it/s]
 41%|████      | 4350/10707 [40:03<52:26,  2.02it/s]{'loss': 3.7959, 'grad_norm': 0.22652867436408997, 'learning_rate': 0.0007404824745261909, 'epoch': 0.41}
-
                                                    

 41%|████      | 4350/10707 [40:03<52:26,  2.02it/s]
 41%|████      | 4351/10707 [40:04<52:29,  2.02it/s]
 41%|████      | 4352/10707 [40:04<52:29,  2.02it/s]
 41%|████      | 4353/10707 [40:05<52:26,  2.02it/s]
 41%|████      | 4354/10707 [40:05<52:24,  2.02it/s]
 41%|████      | 4355/10707 [40:06<52:24,  2.02it/s]
 41%|████      | 4356/10707 [40:06<52:25,  2.02it/s]
 41%|████      | 4357/10707 [40:07<52:21,  2.02it/s]
 41%|████      | 4358/10707 [40:07<52:24,  2.02it/s]
 41%|████      | 4359/10707 [40:08<52:24,  2.02it/s]
 41%|████      | 4360/10707 [40:08<52:22,  2.02it/s]
 41%|████      | 4361/10707 [40:09<52:20,  2.02it/s]
 41%|████      | 4362/10707 [40:09<52:19,  2.02it/s]
 41%|████      | 4363/10707 [40:10<52:22,  2.02it/s]
 41%|████      | 4364/10707 [40:10<52:21,  2.02it/s]
 41%|████      | 4365/10707 [40:11<52:23,  2.02it/s]
 41%|████      | 4366/10707 [40:11<52:22,  2.02it/s]
 41%|████      | 4367/10707 [40:12<52:20,  2.02it/s]
 41%|████      | 4368/10707 [40:12<52:20,  2.02it/s]
 41%|████      | 4369/10707 [40:13<52:19,  2.02it/s]
 41%|████      | 4370/10707 [40:13<52:19,  2.02it/s]
 41%|████      | 4371/10707 [40:14<52:17,  2.02it/s]
 41%|████      | 4372/10707 [40:14<52:16,  2.02it/s]
 41%|████      | 4373/10707 [40:15<52:13,  2.02it/s]
 41%|████      | 4374/10707 [40:15<52:15,  2.02it/s]
 41%|████      | 4375/10707 [40:16<52:13,  2.02it/s]{'loss': 3.7952, 'grad_norm': 0.1899864375591278, 'learning_rate': 0.0007369015194245985, 'epoch': 0.41}
-
                                                    

 41%|████      | 4375/10707 [40:16<52:13,  2.02it/s]
 41%|████      | 4376/10707 [40:16<52:20,  2.02it/s]
 41%|████      | 4377/10707 [40:17<52:19,  2.02it/s]
 41%|████      | 4378/10707 [40:17<52:19,  2.02it/s]
 41%|████      | 4379/10707 [40:18<52:20,  2.01it/s]
 41%|████      | 4380/10707 [40:18<52:17,  2.02it/s]
 41%|████      | 4381/10707 [40:19<52:17,  2.02it/s]
 41%|████      | 4382/10707 [40:19<52:16,  2.02it/s]
 41%|████      | 4383/10707 [40:20<52:19,  2.01it/s]
 41%|████      | 4384/10707 [40:20<52:13,  2.02it/s]
 41%|████      | 4385/10707 [40:21<52:13,  2.02it/s]
 41%|████      | 4386/10707 [40:21<52:13,  2.02it/s]
 41%|████      | 4387/10707 [40:22<52:12,  2.02it/s]
 41%|████      | 4388/10707 [40:22<52:09,  2.02it/s]
 41%|████      | 4389/10707 [40:23<52:11,  2.02it/s]
 41%|████      | 4390/10707 [40:23<52:12,  2.02it/s]
 41%|████      | 4391/10707 [40:24<52:13,  2.02it/s]
 41%|████      | 4392/10707 [40:24<52:07,  2.02it/s]
 41%|████      | 4393/10707 [40:25<52:08,  2.02it/s]
 41%|████      | 4394/10707 [40:25<52:03,  2.02it/s]
 41%|████      | 4395/10707 [40:26<52:08,  2.02it/s]
 41%|████      | 4396/10707 [40:26<52:03,  2.02it/s]
 41%|████      | 4397/10707 [40:26<52:06,  2.02it/s]
 41%|████      | 4398/10707 [40:27<52:07,  2.02it/s]
 41%|████      | 4399/10707 [40:27<52:07,  2.02it/s]
 41%|████      | 4400/10707 [40:28<52:05,  2.02it/s]{'loss': 3.7874, 'grad_norm': 0.21328960359096527, 'learning_rate': 0.000733304826247595, 'epoch': 0.41}
-
                                                    

 41%|████      | 4400/10707 [40:28<52:05,  2.02it/s]
 41%|████      | 4401/10707 [40:28<52:09,  2.02it/s]
 41%|████      | 4402/10707 [40:29<52:06,  2.02it/s]
 41%|████      | 4403/10707 [40:29<52:06,  2.02it/s]
 41%|████      | 4404/10707 [40:30<52:02,  2.02it/s]
 41%|████      | 4405/10707 [40:30<52:03,  2.02it/s]
 41%|████      | 4406/10707 [40:31<51:59,  2.02it/s]
 41%|████      | 4407/10707 [40:31<51:59,  2.02it/s]
 41%|████      | 4408/10707 [40:32<51:57,  2.02it/s]
 41%|████      | 4409/10707 [40:32<51:57,  2.02it/s]
 41%|████      | 4410/10707 [40:33<51:55,  2.02it/s]
 41%|████      | 4411/10707 [40:33<51:57,  2.02it/s]
 41%|████      | 4412/10707 [40:34<51:53,  2.02it/s]
 41%|████      | 4413/10707 [40:34<51:57,  2.02it/s]
 41%|████      | 4414/10707 [40:35<51:54,  2.02it/s]
 41%|████      | 4415/10707 [40:35<51:57,  2.02it/s]
 41%|████      | 4416/10707 [40:36<51:58,  2.02it/s]
 41%|████▏     | 4417/10707 [40:36<51:57,  2.02it/s]
 41%|████▏     | 4418/10707 [40:37<52:02,  2.01it/s]
 41%|████▏     | 4419/10707 [40:37<51:55,  2.02it/s]
 41%|████▏     | 4420/10707 [40:38<51:55,  2.02it/s]
 41%|████▏     | 4421/10707 [40:38<51:53,  2.02it/s]
 41%|████▏     | 4422/10707 [40:39<51:55,  2.02it/s]
 41%|████▏     | 4423/10707 [40:39<51:51,  2.02it/s]
 41%|████▏     | 4424/10707 [40:40<51:55,  2.02it/s]
 41%|████▏     | 4425/10707 [40:40<51:50,  2.02it/s]{'loss': 3.7828, 'grad_norm': 0.2060454785823822, 'learning_rate': 0.0007296926339342493, 'epoch': 0.41}
                                                    
-
 41%|████▏     | 4425/10707 [40:40<51:50,  2.02it/s]
 41%|████▏     | 4426/10707 [40:41<51:54,  2.02it/s]
 41%|████▏     | 4427/10707 [40:41<51:54,  2.02it/s]
 41%|████▏     | 4428/10707 [40:42<51:56,  2.02it/s]
 41%|████▏     | 4429/10707 [40:42<51:54,  2.02it/s]
 41%|████▏     | 4430/10707 [40:43<51:52,  2.02it/s]
 41%|████▏     | 4431/10707 [40:43<51:50,  2.02it/s]
 41%|████▏     | 4432/10707 [40:44<51:50,  2.02it/s]
 41%|████▏     | 4433/10707 [40:44<51:47,  2.02it/s]
 41%|████▏     | 4434/10707 [40:45<51:47,  2.02it/s]
 41%|████▏     | 4435/10707 [40:45<51:46,  2.02it/s]
 41%|████▏     | 4436/10707 [40:46<51:47,  2.02it/s]
 41%|████▏     | 4437/10707 [40:46<51:44,  2.02it/s]
 41%|████▏     | 4438/10707 [40:47<51:45,  2.02it/s]
 41%|████▏     | 4439/10707 [40:47<51:42,  2.02it/s]
 41%|████▏     | 4440/10707 [40:48<51:46,  2.02it/s]
 41%|████▏     | 4441/10707 [40:48<51:41,  2.02it/s]
 41%|████▏     | 4442/10707 [40:49<51:41,  2.02it/s]
 41%|████▏     | 4443/10707 [40:49<51:37,  2.02it/s]
 42%|████▏     | 4444/10707 [40:50<51:38,  2.02it/s]
 42%|████▏     | 4445/10707 [40:50<51:36,  2.02it/s]
 42%|████▏     | 4446/10707 [40:51<51:39,  2.02it/s]
 42%|████▏     | 4447/10707 [40:51<51:36,  2.02it/s]
 42%|████▏     | 4448/10707 [40:52<51:44,  2.02it/s]
 42%|████▏     | 4449/10707 [40:52<51:36,  2.02it/s]
 42%|████▏     | 4450/10707 [40:53<51:40,  2.02it/s]{'loss': 3.7866, 'grad_norm': 0.19736503064632416, 'learning_rate': 0.0007260651824532841, 'epoch': 0.42}
                                                    
-
 42%|████▏     | 4450/10707 [40:53<51:40,  2.02it/s]
 42%|████▏     | 4451/10707 [40:53<51:45,  2.01it/s]
 42%|████▏     | 4452/10707 [40:54<51:43,  2.02it/s]
 42%|████▏     | 4453/10707 [40:54<51:43,  2.02it/s]
 42%|████▏     | 4454/10707 [40:55<51:37,  2.02it/s]
 42%|████▏     | 4455/10707 [40:55<51:39,  2.02it/s]
 42%|████▏     | 4456/10707 [40:56<51:34,  2.02it/s]
 42%|████▏     | 4457/10707 [40:56<51:34,  2.02it/s]
 42%|████▏     | 4458/10707 [40:57<51:34,  2.02it/s]
 42%|████▏     | 4459/10707 [40:57<51:35,  2.02it/s]
 42%|████▏     | 4460/10707 [40:58<51:31,  2.02it/s]
 42%|████▏     | 4461/10707 [40:58<51:33,  2.02it/s]
 42%|████▏     | 4462/10707 [40:59<51:38,  2.02it/s]
 42%|████▏     | 4463/10707 [40:59<51:36,  2.02it/s]
 42%|████▏     | 4464/10707 [41:00<51:36,  2.02it/s]
 42%|████▏     | 4465/10707 [41:00<51:32,  2.02it/s]
 42%|████▏     | 4466/10707 [41:01<51:33,  2.02it/s]
 42%|████▏     | 4467/10707 [41:01<51:31,  2.02it/s]
 42%|████▏     | 4468/10707 [41:02<51:34,  2.02it/s]
 42%|████▏     | 4469/10707 [41:02<51:30,  2.02it/s]
 42%|████▏     | 4470/10707 [41:03<51:31,  2.02it/s]
 42%|████▏     | 4471/10707 [41:03<51:28,  2.02it/s]
 42%|████▏     | 4472/10707 [41:04<51:30,  2.02it/s]
 42%|████▏     | 4473/10707 [41:04<51:30,  2.02it/s]
 42%|████▏     | 4474/10707 [41:05<51:28,  2.02it/s]
 42%|████▏     | 4475/10707 [41:05<51:28,  2.02it/s]{'loss': 3.7855, 'grad_norm': 0.20592695474624634, 'learning_rate': 0.0007224227127871339, 'epoch': 0.42}
                                                    
-
 42%|████▏     | 4475/10707 [41:05<51:28,  2.02it/s]
 42%|████▏     | 4476/10707 [41:06<51:29,  2.02it/s]
 42%|████▏     | 4477/10707 [41:06<51:28,  2.02it/s]
 42%|████▏     | 4478/10707 [41:07<51:25,  2.02it/s]
 42%|████▏     | 4479/10707 [41:07<51:27,  2.02it/s]
 42%|████▏     | 4480/10707 [41:08<51:25,  2.02it/s]
 42%|████▏     | 4481/10707 [41:08<51:25,  2.02it/s]
 42%|████▏     | 4482/10707 [41:09<51:26,  2.02it/s]
 42%|████▏     | 4483/10707 [41:09<51:27,  2.02it/s]
 42%|████▏     | 4484/10707 [41:10<51:27,  2.02it/s]
 42%|████▏     | 4485/10707 [41:10<51:25,  2.02it/s]
 42%|████▏     | 4486/10707 [41:11<51:23,  2.02it/s]
 42%|████▏     | 4487/10707 [41:11<51:22,  2.02it/s]
 42%|████▏     | 4488/10707 [41:12<51:18,  2.02it/s]
 42%|████▏     | 4489/10707 [41:12<51:20,  2.02it/s]
 42%|████▏     | 4490/10707 [41:13<51:17,  2.02it/s]
 42%|████▏     | 4491/10707 [41:13<51:19,  2.02it/s]
 42%|████▏     | 4492/10707 [41:14<51:16,  2.02it/s]
 42%|���███▏     | 4493/10707 [41:14<51:16,  2.02it/s]
 42%|████▏     | 4494/10707 [41:15<51:13,  2.02it/s]
 42%|████▏     | 4495/10707 [41:15<51:17,  2.02it/s]
 42%|████▏     | 4496/10707 [41:16<51:12,  2.02it/s]
 42%|████▏     | 4497/10707 [41:16<51:12,  2.02it/s]
 42%|████▏     | 4498/10707 [41:17<51:10,  2.02it/s]
 42%|████▏     | 4499/10707 [41:17<51:13,  2.02it/s]
 42%|████▏     | 4500/10707 [41:18<51:11,  2.02it/s]{'loss': 3.7864, 'grad_norm': 0.20157848298549652, 'learning_rate': 0.0007187654669159368, 'epoch': 0.42}
-
                                                    

 42%|████▏     | 4500/10707 [41:18<51:11,  2.02it/s]
 42%|████▏     | 4501/10707 [41:18<51:16,  2.02it/s]
 42%|████▏     | 4502/10707 [41:19<51:15,  2.02it/s]
 42%|████▏     | 4503/10707 [41:19<51:13,  2.02it/s]
 42%|████▏     | 4504/10707 [41:20<51:14,  2.02it/s]
 42%|████▏     | 4505/10707 [41:20<51:11,  2.02it/s]
 42%|████▏     | 4506/10707 [41:21<51:14,  2.02it/s]
 42%|████▏     | 4507/10707 [41:21<51:09,  2.02it/s]
 42%|████▏     | 4508/10707 [41:21<51:10,  2.02it/s]
 42%|████▏     | 4509/10707 [41:22<51:05,  2.02it/s]
 42%|████▏     | 4510/10707 [41:22<51:06,  2.02it/s]
 42%|████▏     | 4511/10707 [41:23<51:06,  2.02it/s]
 42%|████▏     | 4512/10707 [41:23<51:05,  2.02it/s]
 42%|████▏     | 4513/10707 [41:24<51:02,  2.02it/s]
 42%|████▏     | 4514/10707 [41:24<51:04,  2.02it/s]
 42%|████▏     | 4515/10707 [41:25<51:03,  2.02it/s]
 42%|████▏     | 4516/10707 [41:25<51:05,  2.02it/s]
 42%|████▏     | 4517/10707 [41:26<51:01,  2.02it/s]
 42%|████▏     | 4518/10707 [41:26<51:03,  2.02it/s]
 42%|████▏     | 4519/10707 [41:27<50:59,  2.02it/s]
 42%|████▏     | 4520/10707 [41:27<51:03,  2.02it/s]
 42%|████▏     | 4521/10707 [41:28<51:01,  2.02it/s]
 42%|████▏     | 4522/10707 [41:28<51:04,  2.02it/s]
 42%|████▏     | 4523/10707 [41:29<51:04,  2.02it/s]
 42%|████▏     | 4524/10707 [41:29<51:05,  2.02it/s]
 42%|████▏     | 4525/10707 [41:30<51:04,  2.02it/s]{'loss': 3.7823, 'grad_norm': 0.19883131980895996, 'learning_rate': 0.0007150936878014578, 'epoch': 0.42}
                                                    
-
 42%|████▏     | 4525/10707 [41:30<51:04,  2.02it/s]
 42%|████▏     | 4526/10707 [41:30<51:07,  2.02it/s]
 42%|████▏     | 4527/10707 [41:31<51:04,  2.02it/s]
 42%|████▏     | 4528/10707 [41:31<51:01,  2.02it/s]
 42%|████▏     | 4529/10707 [41:32<51:00,  2.02it/s]
 42%|████▏     | 4530/10707 [41:32<51:02,  2.02it/s]
 42%|████▏     | 4531/10707 [41:33<51:01,  2.02it/s]
 42%|████▏     | 4532/10707 [41:33<51:01,  2.02it/s]
 42%|████▏     | 4533/10707 [41:34<50:59,  2.02it/s]
 42%|████▏     | 4534/10707 [41:34<50:59,  2.02it/s]
 42%|████▏     | 4535/10707 [41:35<50:55,  2.02it/s]
 42%|████▏     | 4536/10707 [41:35<50:57,  2.02it/s]
 42%|████▏     | 4537/10707 [41:36<50:55,  2.02it/s]
 42%|████▏     | 4538/10707 [41:36<50:55,  2.02it/s]
 42%|████▏     | 4539/10707 [41:37<50:53,  2.02it/s]
 42%|████▏     | 4540/10707 [41:37<50:56,  2.02it/s]
 42%|████▏     | 4541/10707 [41:38<50:52,  2.02it/s]
 42%|████▏     | 4542/10707 [41:38<50:55,  2.02it/s]
 42%|████▏     | 4543/10707 [41:39<50:49,  2.02it/s]
 42%|████▏     | 4544/10707 [41:39<50:56,  2.02it/s]
 42%|████▏     | 4545/10707 [41:40<50:51,  2.02it/s]
 42%|████▏     | 4546/10707 [41:40<50:53,  2.02it/s]
 42%|████▏     | 4547/10707 [41:41<50:48,  2.02it/s]
 42%|████▏     | 4548/10707 [41:41<50:48,  2.02it/s]
 42%|████▏     | 4549/10707 [41:42<50:45,  2.02it/s]
 42%|████▏     | 4550/10707 [41:42<50:46,  2.02it/s]{'loss': 3.7842, 'grad_norm': 0.19400477409362793, 'learning_rate': 0.0007114076193709487, 'epoch': 0.42}
-
                                                    

 42%|████▏     | 4550/10707 [41:42<50:46,  2.02it/s]
 43%|████▎     | 4551/10707 [41:43<50:50,  2.02it/s]
 43%|████▎     | 4552/10707 [41:43<50:49,  2.02it/s]
 43%|████▎     | 4553/10707 [41:44<50:47,  2.02it/s]
 43%|████▎     | 4554/10707 [41:44<50:50,  2.02it/s]
 43%|████▎     | 4555/10707 [41:45<50:48,  2.02it/s]
 43%|████▎     | 4556/10707 [41:45<50:46,  2.02it/s]
 43%|████▎     | 4557/10707 [41:46<50:46,  2.02it/s]
 43%|████▎     | 4558/10707 [41:46<50:44,  2.02it/s]
 43%|████▎     | 4559/10707 [41:47<50:45,  2.02it/s]
 43%|████▎     | 4560/10707 [41:47<50:43,  2.02it/s]
 43%|████▎     | 4561/10707 [41:48<50:44,  2.02it/s]
 43%|████▎     | 4562/10707 [41:48<50:43,  2.02it/s]
 43%|████▎     | 4563/10707 [41:49<50:46,  2.02it/s]
 43%|████▎     | 4564/10707 [41:49<50:44,  2.02it/s]
 43%|████▎     | 4565/10707 [41:50<50:44,  2.02it/s]
 43%|████▎     | 4566/10707 [41:50<50:44,  2.02it/s]
 43%|████▎     | 4567/10707 [41:51<50:42,  2.02it/s]
 43%|████▎     | 4568/10707 [41:51<50:46,  2.02it/s]
 43%|████▎     | 4569/10707 [41:52<50:41,  2.02it/s]
 43%|████▎     | 4570/10707 [41:52<50:41,  2.02it/s]
 43%|████▎     | 4571/10707 [41:53<50:40,  2.02it/s]
 43%|████▎     | 4572/10707 [41:53<50:40,  2.02it/s]
 43%|████▎     | 4573/10707 [41:54<50:34,  2.02it/s]
 43%|████▎     | 4574/10707 [41:54<50:36,  2.02it/s]
 43%|████▎     | 4575/10707 [41:55<50:33,  2.02it/s]
                                                    
{'loss': 3.78, 'grad_norm': 0.20181052386760712, 'learning_rate': 0.0007077075065009433, 'epoch': 0.43}
-
 43%|████▎     | 4575/10707 [41:55<50:33,  2.02it/s]
 43%|████▎     | 4576/10707 [41:55<50:42,  2.02it/s]
 43%|████▎     | 4577/10707 [41:56<50:40,  2.02it/s]
 43%|████▎     | 4578/10707 [41:56<50:37,  2.02it/s]
 43%|████▎     | 4579/10707 [41:57<50:36,  2.02it/s]
 43%|████▎     | 4580/10707 [41:57<50:35,  2.02it/s]
 43%|████▎     | 4581/10707 [41:58<50:35,  2.02it/s]
 43%|████▎     | 4582/10707 [41:58<50:31,  2.02it/s]
 43%|████▎     | 4583/10707 [41:59<50:34,  2.02it/s]
 43%|████▎     | 4584/10707 [41:59<50:32,  2.02it/s]
 43%|████▎     | 4585/10707 [42:00<50:34,  2.02it/s]
 43%|████▎     | 4586/10707 [42:00<50:33,  2.02it/s]
 43%|████▎     | 4587/10707 [42:01<50:33,  2.02it/s]
 43%|████▎     | 4588/10707 [42:01<50:31,  2.02it/s]
 43%|████▎     | 4589/10707 [42:02<50:30,  2.02it/s]
 43%|████▎     | 4590/10707 [42:02<50:29,  2.02it/s]
 43%|████▎     | 4591/10707 [42:03<50:29,  2.02it/s]
 43%|████▎     | 4592/10707 [42:03<50:29,  2.02it/s]
 43%|████▎     | 4593/10707 [42:04<50:26,  2.02it/s]
 43%|████▎     | 4594/10707 [42:04<50:27,  2.02it/s]
 43%|████▎     | 4595/10707 [42:05<50:26,  2.02it/s]
 43%|████▎     | 4596/10707 [42:05<50:29,  2.02it/s]
 43%|████▎     | 4597/10707 [42:06<50:28,  2.02it/s]
 43%|████▎     | 4598/10707 [42:06<50:30,  2.02it/s]
 43%|████▎     | 4599/10707 [42:07<50:27,  2.02it/s]
 43%|████▎     | 4600/10707 [42:07<50:27,  2.02it/s]{'loss': 3.7882, 'grad_norm': 0.22245673835277557, 'learning_rate': 0.0007039935950009894, 'epoch': 0.43}
-
                                                    

 43%|████▎     | 4600/10707 [42:07<50:27,  2.02it/s]
 43%|████▎     | 4601/10707 [42:08<50:31,  2.01it/s]
 43%|████▎     | 4602/10707 [42:08<50:29,  2.02it/s]
 43%|████▎     | 4603/10707 [42:09<50:28,  2.02it/s]
 43%|████▎     | 4604/10707 [42:09<50:25,  2.02it/s]
 43%|████▎     | 4605/10707 [42:10<50:24,  2.02it/s]
 43%|████▎     | 4606/10707 [42:10<50:23,  2.02it/s]
 43%|████▎     | 4607/10707 [42:11<50:20,  2.02it/s]
 43%|████▎     | 4608/10707 [42:11<50:20,  2.02it/s]
 43%|████▎     | 4609/10707 [42:12<50:17,  2.02it/s]
 43%|████▎     | 4610/10707 [42:12<50:17,  2.02it/s]
 43%|████▎     | 4611/10707 [42:13<50:14,  2.02it/s]
 43%|████▎     | 4612/10707 [42:13<50:17,  2.02it/s]
 43%|████▎     | 4613/10707 [42:14<50:16,  2.02it/s]
 43%|████▎     | 4614/10707 [42:14<50:18,  2.02it/s]
 43%|████▎     | 4615/10707 [42:14<50:15,  2.02it/s]
 43%|████▎     | 4616/10707 [42:15<50:16,  2.02it/s]
 43%|████▎     | 4617/10707 [42:15<50:11,  2.02it/s]
 43%|████▎     | 4618/10707 [42:16<50:13,  2.02it/s]
 43%|████▎     | 4619/10707 [42:16<50:09,  2.02it/s]
 43%|████▎     | 4620/10707 [42:17<50:12,  2.02it/s]
 43%|████▎     | 4621/10707 [42:17<50:15,  2.02it/s]
 43%|████▎     | 4622/10707 [42:18<50:15,  2.02it/s]
 43%|████▎     | 4623/10707 [42:18<50:15,  2.02it/s]
 43%|████▎     | 4624/10707 [42:19<50:15,  2.02it/s]
 43%|████▎     | 4625/10707 [42:19<50:16,  2.02it/s]{'loss': 3.781, 'grad_norm': 0.19251134991645813, 'learning_rate': 0.0007002661315973193, 'epoch': 0.43}
-
                                                    

 43%|████▎     | 4625/10707 [42:19<50:16,  2.02it/s]
 43%|████▎     | 4626/10707 [42:20<50:17,  2.02it/s]
 43%|████▎     | 4627/10707 [42:20<50:16,  2.02it/s]
 43%|████▎     | 4628/10707 [42:21<50:14,  2.02it/s]
 43%|████▎     | 4629/10707 [42:21<50:16,  2.02it/s]
 43%|████▎     | 4630/10707 [42:22<50:15,  2.02it/s]
 43%|████▎     | 4631/10707 [42:22<50:16,  2.01it/s]
 43%|████▎     | 4632/10707 [42:23<58:25,  1.73it/s]
 43%|████▎     | 4633/10707 [42:24<55:56,  1.81it/s]
 43%|████▎     | 4634/10707 [42:24<54:12,  1.87it/s]
 43%|████▎     | 4635/10707 [42:25<53:00,  1.91it/s]
 43%|████▎     | 4636/10707 [42:25<52:06,  1.94it/s]
 43%|████▎     | 4637/10707 [42:26<51:30,  1.96it/s]
 43%|████▎     | 4638/10707 [42:26<51:01,  1.98it/s]
 43%|████▎     | 4639/10707 [42:27<50:45,  1.99it/s]
 43%|████▎     | 4640/10707 [42:27<50:32,  2.00it/s]
 43%|████▎     | 4641/10707 [42:28<50:25,  2.01it/s]
 43%|████▎     | 4642/10707 [42:28<50:21,  2.01it/s]
 43%|████▎     | 4643/10707 [42:29<50:16,  2.01it/s]
 43%|████▎     | 4644/10707 [42:29<50:12,  2.01it/s]
 43%|████▎     | 4645/10707 [42:30<50:09,  2.01it/s]
 43%|████▎     | 4646/10707 [42:30<50:08,  2.01it/s]
 43%|████▎     | 4647/10707 [42:31<50:03,  2.02it/s]
 43%|████▎     | 4648/10707 [42:31<50:02,  2.02it/s]
 43%|████▎     | 4649/10707 [42:32<50:01,  2.02it/s]
 43%|████▎     | 4650/10707 [42:32<50:00,  2.02it/s]{'loss': 3.7792, 'grad_norm': 0.19299857318401337, 'learning_rate': 0.0006965253639164583, 'epoch': 0.43}
-
                                                    

 43%|████▎     | 4650/10707 [42:32<50:00,  2.02it/s]
 43%|████▎     | 4651/10707 [42:33<58:02,  1.74it/s]
 43%|████▎     | 4652/10707 [42:33<55:38,  1.81it/s]
 43%|████▎     | 4653/10707 [42:34<53:56,  1.87it/s]
 43%|████▎     | 4654/10707 [42:34<52:45,  1.91it/s]
 43%|████▎     | 4655/10707 [42:35<51:52,  1.94it/s]
 43%|████▎     | 4656/10707 [42:35<51:21,  1.96it/s]
 43%|████▎     | 4657/10707 [42:36<50:57,  1.98it/s]
 44%|████▎     | 4658/10707 [42:36<50:38,  1.99it/s]
 44%|████▎     | 4659/10707 [42:37<50:25,  2.00it/s]
 44%|████▎     | 4660/10707 [42:37<50:15,  2.01it/s]
 44%|████▎     | 4661/10707 [42:38<50:07,  2.01it/s]
 44%|████▎     | 4662/10707 [42:38<50:04,  2.01it/s]
 44%|████▎     | 4663/10707 [42:39<49:59,  2.01it/s]
 44%|████▎     | 4664/10707 [42:39<49:58,  2.02it/s]
 44%|████▎     | 4665/10707 [42:40<49:56,  2.02it/s]
 44%|████▎     | 4666/10707 [42:40<49:58,  2.01it/s]
 44%|████▎     | 4667/10707 [42:41<50:05,  2.01it/s]
 44%|████▎     | 4668/10707 [42:41<50:01,  2.01it/s]
 44%|████▎     | 4669/10707 [42:42<50:02,  2.01it/s]
 44%|████▎     | 4670/10707 [42:42<49:58,  2.01it/s]
 44%|████▎     | 4671/10707 [42:43<49:58,  2.01it/s]
 44%|████▎     | 4672/10707 [42:43<49:52,  2.02it/s]
 44%|████▎     | 4673/10707 [42:44<49:52,  2.02it/s]
 44%|████▎     | 4674/10707 [42:44<49:52,  2.02it/s]
 44%|████▎     | 4675/10707 [42:45<49:52,  2.02it/s]{'loss': 3.7762, 'grad_norm': 0.1870190054178238, 'learning_rate': 0.0006927715404687747, 'epoch': 0.44}
-
                                                    

 44%|████▎     | 4675/10707 [42:45<49:52,  2.02it/s]
 44%|████▎     | 4676/10707 [42:45<49:50,  2.02it/s]
 44%|████▎     | 4677/10707 [42:46<49:51,  2.02it/s]
 44%|████▎     | 4678/10707 [42:46<49:47,  2.02it/s]
 44%|████▎     | 4679/10707 [42:47<49:48,  2.02it/s]
 44%|████▎     | 4680/10707 [42:47<49:44,  2.02it/s]
 44%|████▎     | 4681/10707 [42:48<49:45,  2.02it/s]
 44%|████▎     | 4682/10707 [42:48<49:42,  2.02it/s]
 44%|████▎     | 4683/10707 [42:49<49:42,  2.02it/s]
 44%|████▎     | 4684/10707 [42:49<49:41,  2.02it/s]
 44%|████▍     | 4685/10707 [42:50<49:41,  2.02it/s]
 44%|████▍     | 4686/10707 [42:50<49:41,  2.02it/s]
 44%|████▍     | 4687/10707 [42:51<49:41,  2.02it/s]
 44%|████▍     | 4688/10707 [42:51<49:40,  2.02it/s]
 44%|████▍     | 4689/10707 [42:52<49:40,  2.02it/s]
 44%|████▍     | 4690/10707 [42:52<49:40,  2.02it/s]
 44%|████▍     | 4691/10707 [42:53<49:38,  2.02it/s]
 44%|████▍     | 4692/10707 [42:53<49:38,  2.02it/s]
 44%|████▍     | 4693/10707 [42:54<49:37,  2.02it/s]
 44%|████▍     | 4694/10707 [42:54<49:35,  2.02it/s]
 44%|████▍     | 4695/10707 [42:55<49:34,  2.02it/s]
 44%|████▍     | 4696/10707 [42:55<49:35,  2.02it/s]
 44%|████▍     | 4697/10707 [42:56<49:34,  2.02it/s]
 44%|████▍     | 4698/10707 [42:56<49:37,  2.02it/s]
 44%|████▍     | 4699/10707 [42:57<49:33,  2.02it/s]
 44%|████▍     | 4700/10707 [42:57<49:37,  2.02it/s]{'loss': 3.7733, 'grad_norm': 0.20529527962207794, 'learning_rate': 0.0006890049106319707, 'epoch': 0.44}
-
                                                    

 44%|████▍     | 4700/10707 [42:57<49:37,  2.02it/s]
 44%|████▍     | 4701/10707 [42:58<49:39,  2.02it/s]
 44%|████▍     | 4702/10707 [42:58<49:36,  2.02it/s]
 44%|████▍     | 4703/10707 [42:59<49:31,  2.02it/s]
 44%|████▍     | 4704/10707 [42:59<49:32,  2.02it/s]
 44%|████▍     | 4705/10707 [43:00<49:28,  2.02it/s]
 44%|████▍     | 4706/10707 [43:00<49:31,  2.02it/s]
 44%|████▍     | 4707/10707 [43:01<49:28,  2.02it/s]
 44%|████▍     | 4708/10707 [43:01<49:31,  2.02it/s]
 44%|████▍     | 4709/10707 [43:02<49:29,  2.02it/s]
 44%|████▍     | 4710/10707 [43:02<49:29,  2.02it/s]
 44%|████▍     | 4711/10707 [43:03<49:28,  2.02it/s]
 44%|████▍     | 4712/10707 [43:03<49:26,  2.02it/s]
 44%|████▍     | 4713/10707 [43:04<49:25,  2.02it/s]
 44%|████▍     | 4714/10707 [43:04<49:25,  2.02it/s]
 44%|████▍     | 4715/10707 [43:05<49:24,  2.02it/s]
 44%|████▍     | 4716/10707 [43:05<49:26,  2.02it/s]
 44%|████▍     | 4717/10707 [43:06<49:23,  2.02it/s]
 44%|████▍     | 4718/10707 [43:06<49:25,  2.02it/s]
 44%|████▍     | 4719/10707 [43:07<49:23,  2.02it/s]
 44%|████▍     | 4720/10707 [43:07<49:25,  2.02it/s]
 44%|████▍     | 4721/10707 [43:08<49:23,  2.02it/s]
 44%|████▍     | 4722/10707 [43:08<49:24,  2.02it/s]
 44%|████▍     | 4723/10707 [43:09<49:26,  2.02it/s]
 44%|████▍     | 4724/10707 [43:09<49:27,  2.02it/s]
 44%|████▍     | 4725/10707 [43:10<49:26,  2.02it/s]{'loss': 3.77, 'grad_norm': 0.19652652740478516, 'learning_rate': 0.0006852257246345145, 'epoch': 0.44}
-
                                                    

 44%|████▍     | 4725/10707 [43:10<49:26,  2.02it/s]
 44%|████▍     | 4726/10707 [43:10<49:27,  2.02it/s]
 44%|████▍     | 4727/10707 [43:11<49:26,  2.02it/s]
 44%|████▍     | 4728/10707 [43:11<49:25,  2.02it/s]
 44%|████▍     | 4729/10707 [43:12<49:22,  2.02it/s]
 44%|████▍     | 4730/10707 [43:12<49:21,  2.02it/s]
 44%|████▍     | 4731/10707 [43:13<49:22,  2.02it/s]
 44%|████▍     | 4732/10707 [43:13<49:19,  2.02it/s]
 44%|████▍     | 4733/10707 [43:13<49:15,  2.02it/s]
 44%|████▍     | 4734/10707 [43:14<49:14,  2.02it/s]
 44%|████▍     | 4735/10707 [43:14<49:15,  2.02it/s]
 44%|████▍     | 4736/10707 [43:15<49:14,  2.02it/s]
 44%|████▍     | 4737/10707 [43:15<49:13,  2.02it/s]
 44%|████▍     | 4738/10707 [43:16<49:14,  2.02it/s]
 44%|████▍     | 4739/10707 [43:16<49:14,  2.02it/s]
 44%|████▍     | 4740/10707 [43:17<49:13,  2.02it/s]
 44%|████▍     | 4741/10707 [43:17<49:14,  2.02it/s]
 44%|████▍     | 4742/10707 [43:18<49:14,  2.02it/s]
 44%|████▍     | 4743/10707 [43:18<49:14,  2.02it/s]
 44%|████▍     | 4744/10707 [43:19<49:15,  2.02it/s]
 44%|████▍     | 4745/10707 [43:19<49:13,  2.02it/s]
 44%|████▍     | 4746/10707 [43:20<49:11,  2.02it/s]
 44%|████▍     | 4747/10707 [43:20<49:10,  2.02it/s]
 44%|████▍     | 4748/10707 [43:21<49:09,  2.02it/s]
 44%|████▍     | 4749/10707 [43:21<49:09,  2.02it/s]
 44%|████▍     | 4750/10707 [43:22<49:07,  2.02it/s]
                                                    {'loss': 3.7724, 'grad_norm': 0.202806755900383, 'learning_rate': 0.0006814342335390184, 'epoch': 0.44}
-

 44%|████▍     | 4750/10707 [43:22<49:07,  2.02it/s]
 44%|████▍     | 4751/10707 [43:22<49:12,  2.02it/s]
 44%|████▍     | 4752/10707 [43:23<49:09,  2.02it/s]
 44%|████▍     | 4753/10707 [43:23<49:11,  2.02it/s]
 44%|████▍     | 4754/10707 [43:24<49:08,  2.02it/s]
 44%|████▍     | 4755/10707 [43:24<49:09,  2.02it/s]
 44%|████▍     | 4756/10707 [43:25<49:04,  2.02it/s]
 44%|████▍     | 4757/10707 [43:25<49:03,  2.02it/s]
 44%|████▍     | 4758/10707 [43:26<49:01,  2.02it/s]
 44%|████▍     | 4759/10707 [43:26<49:05,  2.02it/s]
 44%|████▍     | 4760/10707 [43:27<49:01,  2.02it/s]
 44%|████▍     | 4761/10707 [43:27<49:03,  2.02it/s]
 44%|████▍     | 4762/10707 [43:28<48:58,  2.02it/s]
 44%|████▍     | 4763/10707 [43:28<49:03,  2.02it/s]
 44%|████▍     | 4764/10707 [43:29<49:01,  2.02it/s]
 45%|████▍     | 4765/10707 [43:29<49:03,  2.02it/s]
 45%|████▍     | 4766/10707 [43:30<48:59,  2.02it/s]
 45%|████▍     | 4767/10707 [43:30<49:01,  2.02it/s]
 45%|████▍     | 4768/10707 [43:31<48:59,  2.02it/s]
 45%|████▍     | 4769/10707 [43:31<48:58,  2.02it/s]
 45%|████▍     | 4770/10707 [43:32<48:57,  2.02it/s]
 45%|████▍     | 4771/10707 [43:32<49:00,  2.02it/s]
 45%|████▍     | 4772/10707 [43:33<48:58,  2.02it/s]
 45%|████▍     | 4773/10707 [43:33<48:58,  2.02it/s]
 45%|████▍     | 4774/10707 [43:34<48:57,  2.02it/s]
 45%|████▍     | 4775/10707 [43:34<48:57,  2.02it/s]{'loss': 3.7734, 'grad_norm': 0.2084294557571411, 'learning_rate': 0.0006776306892255583, 'epoch': 0.45}
-
                                                    

 45%|████▍     | 4775/10707 [43:34<48:57,  2.02it/s]
 45%|████▍     | 4776/10707 [43:35<49:04,  2.01it/s]
 45%|████▍     | 4777/10707 [43:35<48:59,  2.02it/s]
 45%|████▍     | 4778/10707 [43:36<48:59,  2.02it/s]
 45%|████▍     | 4779/10707 [43:36<48:56,  2.02it/s]
 45%|████▍     | 4780/10707 [43:37<48:58,  2.02it/s]
 45%|████▍     | 4781/10707 [43:37<48:54,  2.02it/s]
 45%|████▍     | 4782/10707 [43:38<48:52,  2.02it/s]
 45%|████▍     | 4783/10707 [43:38<48:49,  2.02it/s]
 45%|████▍     | 4784/10707 [43:39<48:51,  2.02it/s]
 45%|████▍     | 4785/10707 [43:39<48:51,  2.02it/s]
 45%|████▍     | 4786/10707 [43:40<48:51,  2.02it/s]
 45%|████▍     | 4787/10707 [43:40<48:50,  2.02it/s]
 45%|████▍     | 4788/10707 [43:41<48:51,  2.02it/s]
 45%|████▍     | 4789/10707 [43:41<48:48,  2.02it/s]
 45%|████▍     | 4790/10707 [43:42<48:49,  2.02it/s]
 45%|████▍     | 4791/10707 [43:42<48:50,  2.02it/s]
 45%|████▍     | 4792/10707 [43:43<48:49,  2.02it/s]
 45%|████▍     | 4793/10707 [43:43<48:47,  2.02it/s]
 45%|████▍     | 4794/10707 [43:44<48:48,  2.02it/s]
 45%|████▍     | 4795/10707 [43:44<48:48,  2.02it/s]
 45%|████▍     | 4796/10707 [43:45<48:49,  2.02it/s]
 45%|████▍     | 4797/10707 [43:45<48:48,  2.02it/s]
 45%|████▍     | 4798/10707 [43:46<48:48,  2.02it/s]
 45%|████▍     | 4799/10707 [43:46<48:46,  2.02it/s]
 45%|████▍     | 4800/10707 [43:47<48:49,  2.02it/s]{'loss': 3.7668, 'grad_norm': 0.21428242325782776, 'learning_rate': 0.0006738153443749421, 'epoch': 0.45}
                                                    
-
 45%|████▍     | 4800/10707 [43:47<48:49,  2.02it/s]
 45%|████▍     | 4801/10707 [43:47<48:51,  2.01it/s]
 45%|████▍     | 4802/10707 [43:48<48:48,  2.02it/s]
 45%|████▍     | 4803/10707 [43:48<48:46,  2.02it/s]
 45%|████▍     | 4804/10707 [43:49<48:48,  2.02it/s]
 45%|████▍     | 4805/10707 [43:49<48:44,  2.02it/s]
 45%|████▍     | 4806/10707 [43:50<48:46,  2.02it/s]
 45%|████▍     | 4807/10707 [43:50<48:42,  2.02it/s]
 45%|████▍     | 4808/10707 [43:51<48:44,  2.02it/s]
 45%|████▍     | 4809/10707 [43:51<48:41,  2.02it/s]
 45%|████▍     | 4810/10707 [43:52<48:43,  2.02it/s]
 45%|████▍     | 4811/10707 [43:52<48:37,  2.02it/s]
 45%|████▍     | 4812/10707 [43:53<48:39,  2.02it/s]
 45%|████▍     | 4813/10707 [43:53<48:36,  2.02it/s]
 45%|████▍     | 4814/10707 [43:54<48:36,  2.02it/s]
 45%|████▍     | 4815/10707 [43:54<48:37,  2.02it/s]
 45%|████▍     | 4816/10707 [43:55<48:37,  2.02it/s]
 45%|████▍     | 4817/10707 [43:55<48:37,  2.02it/s]
 45%|████▍     | 4818/10707 [43:56<48:36,  2.02it/s]
 45%|████▌     | 4819/10707 [43:56<48:32,  2.02it/s]
 45%|████▌     | 4820/10707 [43:57<48:33,  2.02it/s]
 45%|████▌     | 4821/10707 [43:57<48:30,  2.02it/s]
 45%|████▌     | 4822/10707 [43:58<48:31,  2.02it/s]
 45%|████▌     | 4823/10707 [43:58<48:28,  2.02it/s]
 45%|████▌     | 4824/10707 [43:59<48:30,  2.02it/s]
 45%|████▌     | 4825/10707 [43:59<48:27,  2.02it/s]
                                                    {'loss': 3.7704, 'grad_norm': 0.2013889104127884, 'learning_rate': 0.0006699884524519219, 'epoch': 0.45}
-

 45%|████▌     | 4825/10707 [43:59<48:27,  2.02it/s]
 45%|████▌     | 4826/10707 [44:00<48:32,  2.02it/s]
 45%|████▌     | 4827/10707 [44:00<48:30,  2.02it/s]
 45%|████▌     | 4828/10707 [44:01<48:31,  2.02it/s]
 45%|████▌     | 4829/10707 [44:01<48:32,  2.02it/s]
 45%|████▌     | 4830/10707 [44:02<48:31,  2.02it/s]
 45%|████▌     | 4831/10707 [44:02<48:29,  2.02it/s]
 45%|████▌     | 4832/10707 [44:03<48:28,  2.02it/s]
 45%|████▌     | 4833/10707 [44:03<48:26,  2.02it/s]
 45%|████▌     | 4834/10707 [44:04<48:26,  2.02it/s]
 45%|████▌     | 4835/10707 [44:04<48:28,  2.02it/s]
 45%|████▌     | 4836/10707 [44:05<48:28,  2.02it/s]
 45%|████▌     | 4837/10707 [44:05<48:27,  2.02it/s]
 45%|████▌     | 4838/10707 [44:05<48:26,  2.02it/s]
 45%|████▌     | 4839/10707 [44:06<48:28,  2.02it/s]
 45%|████▌     | 4840/10707 [44:06<48:27,  2.02it/s]
 45%|████▌     | 4841/10707 [44:07<48:25,  2.02it/s]
 45%|████▌     | 4842/10707 [44:07<48:27,  2.02it/s]
 45%|████▌     | 4843/10707 [44:08<48:24,  2.02it/s]
 45%|████▌     | 4844/10707 [44:08<48:25,  2.02it/s]
 45%|████▌     | 4845/10707 [44:09<48:21,  2.02it/s]
 45%|████▌     | 4846/10707 [44:09<48:22,  2.02it/s]
 45%|████▌     | 4847/10707 [44:10<48:20,  2.02it/s]
 45%|████▌     | 4848/10707 [44:10<48:21,  2.02it/s]
 45%|████▌     | 4849/10707 [44:11<48:18,  2.02it/s]
 45%|████▌     | 4850/10707 [44:11<48:19,  2.02it/s]{'loss': 3.7684, 'grad_norm': 0.20233766734600067, 'learning_rate': 0.000666150267688357, 'epoch': 0.45}
-
                                                    

 45%|████▌     | 4850/10707 [44:11<48:19,  2.02it/s]
 45%|████▌     | 4851/10707 [44:12<48:21,  2.02it/s]
 45%|████▌     | 4852/10707 [44:12<48:19,  2.02it/s]
 45%|████▌     | 4853/10707 [44:13<48:18,  2.02it/s]
 45%|████▌     | 4854/10707 [44:13<48:15,  2.02it/s]
 45%|████▌     | 4855/10707 [44:14<48:16,  2.02it/s]
 45%|████▌     | 4856/10707 [44:14<48:19,  2.02it/s]
 45%|████▌     | 4857/10707 [44:15<48:19,  2.02it/s]
 45%|████▌     | 4858/10707 [44:15<48:17,  2.02it/s]
 45%|████▌     | 4859/10707 [44:16<48:17,  2.02it/s]
 45%|████▌     | 4860/10707 [44:16<48:17,  2.02it/s]
 45%|████▌     | 4861/10707 [44:17<48:18,  2.02it/s]
 45%|████▌     | 4862/10707 [44:17<48:13,  2.02it/s]
 45%|████▌     | 4863/10707 [44:18<48:13,  2.02it/s]
 45%|████▌     | 4864/10707 [44:18<48:11,  2.02it/s]
 45%|████▌     | 4865/10707 [44:19<48:11,  2.02it/s]
 45%|████▌     | 4866/10707 [44:19<48:09,  2.02it/s]
 45%|████▌     | 4867/10707 [44:20<48:09,  2.02it/s]
 45%|████▌     | 4868/10707 [44:20<48:08,  2.02it/s]
 45%|████▌     | 4869/10707 [44:21<48:10,  2.02it/s]
 45%|████▌     | 4870/10707 [44:21<48:10,  2.02it/s]
 45%|████▌     | 4871/10707 [44:22<48:13,  2.02it/s]
 46%|████▌     | 4872/10707 [44:22<48:11,  2.02it/s]
 46%|████▌     | 4873/10707 [44:23<48:10,  2.02it/s]
 46%|████▌     | 4874/10707 [44:23<48:08,  2.02it/s]
 46%|████▌     | 4875/10707 [44:24<48:09,  2.02it/s]{'loss': 3.7637, 'grad_norm': 0.20616765320301056, 'learning_rate': 0.0006623010450663232, 'epoch': 0.46}
-
                                                    

 46%|████▌     | 4875/10707 [44:24<48:09,  2.02it/s]
 46%|████▌     | 4876/10707 [44:24<48:09,  2.02it/s]
 46%|████▌     | 4877/10707 [44:25<48:09,  2.02it/s]
 46%|████▌     | 4878/10707 [44:25<48:05,  2.02it/s]
 46%|████▌     | 4879/10707 [44:26<48:06,  2.02it/s]
 46%|████▌     | 4880/10707 [44:26<48:02,  2.02it/s]
 46%|████▌     | 4881/10707 [44:27<48:03,  2.02it/s]
 46%|████▌     | 4882/10707 [44:27<48:04,  2.02it/s]
 46%|████▌     | 4883/10707 [44:28<48:03,  2.02it/s]
 46%|████▌     | 4884/10707 [44:28<48:02,  2.02it/s]
 46%|████▌     | 4885/10707 [44:29<48:02,  2.02it/s]
 46%|████▌     | 4886/10707 [44:29<47:57,  2.02it/s]
 46%|████▌     | 4887/10707 [44:30<47:58,  2.02it/s]
 46%|████▌     | 4888/10707 [44:30<47:57,  2.02it/s]
 46%|████▌     | 4889/10707 [44:31<47:59,  2.02it/s]
 46%|████▌     | 4890/10707 [44:31<47:57,  2.02it/s]
 46%|████▌     | 4891/10707 [44:32<47:59,  2.02it/s]
 46%|████▌     | 4892/10707 [44:32<48:01,  2.02it/s]
 46%|████▌     | 4893/10707 [44:33<48:01,  2.02it/s]
 46%|████▌     | 4894/10707 [44:33<48:03,  2.02it/s]
 46%|████▌     | 4895/10707 [44:34<48:00,  2.02it/s]
 46%|████▌     | 4896/10707 [44:34<48:00,  2.02it/s]
 46%|████▌     | 4897/10707 [44:35<47:58,  2.02it/s]
 46%|████▌     | 4898/10707 [44:35<47:57,  2.02it/s]
 46%|████▌     | 4899/10707 [44:36<47:56,  2.02it/s]
 46%|████▌     | 4900/10707 [44:36<47:55,  2.02it/s]{'loss': 3.7594, 'grad_norm': 0.21495549380779266, 'learning_rate': 0.0006584410403011746, 'epoch': 0.46}
-
                                                    

 46%|████▌     | 4900/10707 [44:36<47:55,  2.02it/s]
 46%|████▌     | 4901/10707 [44:37<48:00,  2.02it/s]
 46%|████▌     | 4902/10707 [44:37<47:57,  2.02it/s]
 46%|████▌     | 4903/10707 [44:38<47:57,  2.02it/s]
 46%|████▌     | 4904/10707 [44:38<47:56,  2.02it/s]
 46%|████▌     | 4905/10707 [44:39<47:55,  2.02it/s]
 46%|████▌     | 4906/10707 [44:39<47:55,  2.02it/s]
 46%|████▌     | 4907/10707 [44:40<47:51,  2.02it/s]
 46%|████▌     | 4908/10707 [44:40<47:52,  2.02it/s]
 46%|████▌     | 4909/10707 [44:41<47:50,  2.02it/s]
 46%|████▌     | 4910/10707 [44:41<47:54,  2.02it/s]
 46%|████▌     | 4911/10707 [44:42<47:48,  2.02it/s]
 46%|████▌     | 4912/10707 [44:42<47:48,  2.02it/s]
 46%|████▌     | 4913/10707 [44:43<47:47,  2.02it/s]
 46%|████▌     | 4914/10707 [44:43<47:50,  2.02it/s]
 46%|████▌     | 4915/10707 [44:44<47:48,  2.02it/s]
 46%|████▌     | 4916/10707 [44:44<47:47,  2.02it/s]
 46%|████▌     | 4917/10707 [44:45<47:44,  2.02it/s]
 46%|████▌     | 4918/10707 [44:45<47:45,  2.02it/s]
 46%|████▌     | 4919/10707 [44:46<47:41,  2.02it/s]
 46%|████▌     | 4920/10707 [44:46<47:42,  2.02it/s]
 46%|████▌     | 4921/10707 [44:47<47:41,  2.02it/s]
 46%|████▌     | 4922/10707 [44:47<47:44,  2.02it/s]
 46%|████▌     | 4923/10707 [44:48<47:40,  2.02it/s]
 46%|████▌     | 4924/10707 [44:48<47:40,  2.02it/s]
 46%|████▌     | 4925/10707 [44:49<47:41,  2.02it/s]{'loss': 3.7729, 'grad_norm': 0.20730841159820557, 'learning_rate': 0.0006545705098245551, 'epoch': 0.46}
-
                                                    

 46%|████▌     | 4925/10707 [44:49<47:41,  2.02it/s]
 46%|████▌     | 4926/10707 [44:49<47:46,  2.02it/s]
 46%|████▌     | 4927/10707 [44:50<47:42,  2.02it/s]
 46%|████▌     | 4928/10707 [44:50<47:44,  2.02it/s]
 46%|████▌     | 4929/10707 [44:51<47:46,  2.02it/s]
 46%|████▌     | 4930/10707 [44:51<47:44,  2.02it/s]
 46%|████▌     | 4931/10707 [44:52<47:45,  2.02it/s]
 46%|████▌     | 4932/10707 [44:52<47:43,  2.02it/s]
 46%|████▌     | 4933/10707 [44:53<47:44,  2.02it/s]
 46%|████▌     | 4934/10707 [44:53<47:40,  2.02it/s]
 46%|████▌     | 4935/10707 [44:54<47:42,  2.02it/s]
 46%|████▌     | 4936/10707 [44:54<47:38,  2.02it/s]
 46%|████▌     | 4937/10707 [44:55<47:39,  2.02it/s]
 46%|████▌     | 4938/10707 [44:55<47:38,  2.02it/s]
 46%|████▌     | 4939/10707 [44:56<47:39,  2.02it/s]
 46%|████▌     | 4940/10707 [44:56<47:35,  2.02it/s]
 46%|████▌     | 4941/10707 [44:57<47:36,  2.02it/s]
 46%|████▌     | 4942/10707 [44:57<47:35,  2.02it/s]
 46%|████▌     | 4943/10707 [44:57<47:34,  2.02it/s]
 46%|████▌     | 4944/10707 [44:58<47:33,  2.02it/s]
 46%|████▌     | 4945/10707 [44:58<47:31,  2.02it/s]
 46%|████▌     | 4946/10707 [44:59<47:33,  2.02it/s]
 46%|████▌     | 4947/10707 [44:59<47:32,  2.02it/s]
 46%|████▌     | 4948/10707 [45:00<47:34,  2.02it/s]
 46%|████▌     | 4949/10707 [45:00<47:31,  2.02it/s]
 46%|████▌     | 4950/10707 [45:01<47:30,  2.02it/s]
                                                    {'loss': 3.7496, 'grad_norm': 0.20756149291992188, 'learning_rate': 0.0006506897107673632, 'epoch': 0.46}
-
 46%|████▌     | 4950/10707 [45:01<47:30,  2.02it/s]
 46%|████▌     | 4951/10707 [45:01<47:30,  2.02it/s]
 46%|████▋     | 4952/10707 [45:02<47:30,  2.02it/s]
 46%|████▋     | 4953/10707 [45:02<47:27,  2.02it/s]
 46%|████▋     | 4954/10707 [45:03<47:29,  2.02it/s]
 46%|████▋     | 4955/10707 [45:03<47:29,  2.02it/s]
 46%|████▋     | 4956/10707 [45:04<47:27,  2.02it/s]
 46%|████▋     | 4957/10707 [45:04<47:28,  2.02it/s]
 46%|████▋     | 4958/10707 [45:05<47:25,  2.02it/s]
 46%|████▋     | 4959/10707 [45:05<47:26,  2.02it/s]
 46%|████▋     | 4960/10707 [45:06<47:24,  2.02it/s]
 46%|████▋     | 4961/10707 [45:06<47:29,  2.02it/s]
 46%|████▋     | 4962/10707 [45:07<47:25,  2.02it/s]
 46%|████▋     | 4963/10707 [45:07<47:27,  2.02it/s]
 46%|████▋     | 4964/10707 [45:08<47:22,  2.02it/s]
 46%|████▋     | 4965/10707 [45:08<47:23,  2.02it/s]
 46%|████▋     | 4966/10707 [45:09<47:21,  2.02it/s]
 46%|████▋     | 4967/10707 [45:09<47:22,  2.02it/s]
 46%|████▋     | 4968/10707 [45:10<47:19,  2.02it/s]
 46%|████▋     | 4969/10707 [45:10<47:22,  2.02it/s]
 46%|████▋     | 4970/10707 [45:11<47:23,  2.02it/s]
 46%|████▋     | 4971/10707 [45:11<47:23,  2.02it/s]
 46%|████▋     | 4972/10707 [45:12<47:22,  2.02it/s]
 46%|████▋     | 4973/10707 [45:12<47:22,  2.02it/s]
 46%|████▋     | 4974/10707 [45:13<47:21,  2.02it/s]
 46%|████▋     | 4975/10707 [45:13<47:21,  2.02it/s]{'loss': 3.7513, 'grad_norm': 0.20845744013786316, 'learning_rate': 0.0006467989009426699, 'epoch': 0.46}
-
                                                    

 46%|████▋     | 4975/10707 [45:13<47:21,  2.02it/s]
 46%|████▋     | 4976/10707 [45:14<47:25,  2.01it/s]
 46%|████▋     | 4977/10707 [45:14<47:23,  2.01it/s]
 46%|████▋     | 4978/10707 [45:15<47:17,  2.02it/s]
 47%|████▋     | 4979/10707 [45:15<47:18,  2.02it/s]
 47%|████▋     | 4980/10707 [45:16<47:12,  2.02it/s]
 47%|████▋     | 4981/10707 [45:16<47:13,  2.02it/s]
 47%|████▋     | 4982/10707 [45:17<47:11,  2.02it/s]
 47%|████▋     | 4983/10707 [45:17<47:12,  2.02it/s]
 47%|████▋     | 4984/10707 [45:18<47:09,  2.02it/s]
 47%|████▋     | 4985/10707 [45:18<47:10,  2.02it/s]
 47%|████▋     | 4986/10707 [45:19<47:07,  2.02it/s]
 47%|████▋     | 4987/10707 [45:19<47:10,  2.02it/s]
 47%|████▋     | 4988/10707 [45:20<47:08,  2.02it/s]
 47%|████▋     | 4989/10707 [45:20<47:12,  2.02it/s]
 47%|████▋     | 4990/10707 [45:21<47:07,  2.02it/s]
 47%|████▋     | 4991/10707 [45:21<47:08,  2.02it/s]
 47%|████▋     | 4992/10707 [45:22<47:07,  2.02it/s]
 47%|████▋     | 4993/10707 [45:22<47:08,  2.02it/s]
 47%|████▋     | 4994/10707 [45:23<47:06,  2.02it/s]
 47%|████▋     | 4995/10707 [45:23<47:06,  2.02it/s]
 47%|████▋     | 4996/10707 [45:24<47:04,  2.02it/s]
 47%|████▋     | 4997/10707 [45:24<47:04,  2.02it/s]
 47%|████▋     | 4998/10707 [45:25<47:04,  2.02it/s]
 47%|████▋     | 4999/10707 [45:25<47:05,  2.02it/s]
 47%|████▋     | 5000/10707 [45:26<47:04,  2.02it/s]{'loss': 3.7511, 'grad_norm': 0.20652948319911957, 'learning_rate': 0.0006428983388285909, 'epoch': 0.47}
-
                                                    

 47%|████▋     | 5000/10707 [45:26<47:04,  2.02it/s]
 47%|████▋     | 5001/10707 [45:26<47:06,  2.02it/s]
 47%|████▋     | 5002/10707 [45:27<47:07,  2.02it/s]
 47%|████▋     | 5003/10707 [45:27<47:07,  2.02it/s]
 47%|████▋     | 5004/10707 [45:28<47:04,  2.02it/s]
 47%|████▋     | 5005/10707 [45:28<47:02,  2.02it/s]
 47%|████▋     | 5006/10707 [45:29<47:04,  2.02it/s]
 47%|████▋     | 5007/10707 [45:29<46:59,  2.02it/s]
 47%|████▋     | 5008/10707 [45:30<47:04,  2.02it/s]
 47%|████▋     | 5009/10707 [45:30<46:59,  2.02it/s]
 47%|████▋     | 5010/10707 [45:31<46:57,  2.02it/s]
 47%|████▋     | 5011/10707 [45:31<46:57,  2.02it/s]
 47%|████▋     | 5012/10707 [45:32<46:58,  2.02it/s]
 47%|████▋     | 5013/10707 [45:32<46:57,  2.02it/s]
 47%|████▋     | 5014/10707 [45:33<46:58,  2.02it/s]
 47%|████▋     | 5015/10707 [45:33<46:56,  2.02it/s]
 47%|████▋     | 5016/10707 [45:34<46:56,  2.02it/s]
 47%|████▋     | 5017/10707 [45:34<46:54,  2.02it/s]
 47%|████▋     | 5018/10707 [45:35<46:54,  2.02it/s]
 47%|████▋     | 5019/10707 [45:35<46:56,  2.02it/s]
 47%|████▋     | 5020/10707 [45:36<46:53,  2.02it/s]
 47%|████▋     | 5021/10707 [45:36<46:55,  2.02it/s]
 47%|████▋     | 5022/10707 [45:37<46:54,  2.02it/s]
 47%|████▋     | 5023/10707 [45:37<46:56,  2.02it/s]
 47%|████▋     | 5024/10707 [45:38<46:52,  2.02it/s]
 47%|████▋     | 5025/10707 [45:38<46:52,  2.02it/s]{'loss': 3.7431, 'grad_norm': 0.19356027245521545, 'learning_rate': 0.0006389882835511162, 'epoch': 0.47}
-
                                                    

 47%|████▋     | 5025/10707 [45:38<46:52,  2.02it/s]
 47%|████▋     | 5026/10707 [45:39<46:53,  2.02it/s]
 47%|████▋     | 5027/10707 [45:39<46:53,  2.02it/s]
 47%|████▋     | 5028/10707 [45:40<46:50,  2.02it/s]
 47%|████▋     | 5029/10707 [45:40<46:51,  2.02it/s]
 47%|████▋     | 5030/10707 [45:41<46:52,  2.02it/s]
 47%|████▋     | 5031/10707 [45:41<46:50,  2.02it/s]
 47%|████▋     | 5032/10707 [45:42<46:51,  2.02it/s]
 47%|████▋     | 5033/10707 [45:42<46:50,  2.02it/s]
 47%|████▋     | 5034/10707 [45:43<46:50,  2.02it/s]
 47%|████▋     | 5035/10707 [45:43<46:49,  2.02it/s]
 47%|████▋     | 5036/10707 [45:44<46:50,  2.02it/s]
 47%|████▋     | 5037/10707 [45:44<46:47,  2.02it/s]
 47%|████▋     | 5038/10707 [45:45<46:46,  2.02it/s]
 47%|████▋     | 5039/10707 [45:45<46:44,  2.02it/s]
 47%|████▋     | 5040/10707 [45:46<46:43,  2.02it/s]
 47%|████▋     | 5041/10707 [45:46<46:44,  2.02it/s]
 47%|████▋     | 5042/10707 [45:47<46:43,  2.02it/s]
 47%|████▋     | 5043/10707 [45:47<46:45,  2.02it/s]
 47%|████▋     | 5044/10707 [45:48<46:45,  2.02it/s]
 47%|████▋     | 5045/10707 [45:48<46:42,  2.02it/s]
 47%|████▋     | 5046/10707 [45:48<46:40,  2.02it/s]
 47%|████▋     | 5047/10707 [45:49<46:40,  2.02it/s]
 47%|████▋     | 5048/10707 [45:49<46:38,  2.02it/s]
 47%|████▋     | 5049/10707 [45:50<46:39,  2.02it/s]
 47%|████▋     | 5050/10707 [45:50<46:37,  2.02it/s]
                                                    {'loss': 3.7552, 'grad_norm': 0.21887019276618958, 'learning_rate': 0.0006350689948668946, 'epoch': 0.47}
-
 47%|████▋     | 5050/10707 [45:50<46:37,  2.02it/s]
 47%|████▋     | 5051/10707 [45:51<46:42,  2.02it/s]
 47%|████▋     | 5052/10707 [45:51<46:44,  2.02it/s]
 47%|████▋     | 5053/10707 [45:52<46:41,  2.02it/s]
 47%|████▋     | 5054/10707 [45:52<46:40,  2.02it/s]
 47%|████▋     | 5055/10707 [45:53<46:40,  2.02it/s]
 47%|████▋     | 5056/10707 [45:53<46:39,  2.02it/s]
 47%|████▋     | 5057/10707 [45:54<46:38,  2.02it/s]
 47%|████▋     | 5058/10707 [45:54<46:36,  2.02it/s]
 47%|████▋     | 5059/10707 [45:55<46:36,  2.02it/s]
 47%|████▋     | 5060/10707 [45:55<46:34,  2.02it/s]
 47%|████▋     | 5061/10707 [45:56<46:35,  2.02it/s]
 47%|████▋     | 5062/10707 [45:56<46:34,  2.02it/s]
 47%|████▋     | 5063/10707 [45:57<46:35,  2.02it/s]
 47%|████▋     | 5064/10707 [45:57<46:33,  2.02it/s]
 47%|████▋     | 5065/10707 [45:58<46:34,  2.02it/s]
 47%|████▋     | 5066/10707 [45:58<46:34,  2.02it/s]
 47%|████▋     | 5067/10707 [45:59<46:33,  2.02it/s]
 47%|████▋     | 5068/10707 [45:59<46:33,  2.02it/s]
 47%|████▋     | 5069/10707 [46:00<46:31,  2.02it/s]
 47%|████▋     | 5070/10707 [46:00<46:33,  2.02it/s]
 47%|████▋     | 5071/10707 [46:01<46:32,  2.02it/s]
 47%|████▋     | 5072/10707 [46:01<46:33,  2.02it/s]
 47%|████▋     | 5073/10707 [46:02<46:30,  2.02it/s]
 47%|████▋     | 5074/10707 [46:02<46:30,  2.02it/s]
 47%|████▋     | 5075/10707 [46:03<46:26,  2.02it/s]{'loss': 3.7507, 'grad_norm': 0.21900996565818787, 'learning_rate': 0.000631140733145978, 'epoch': 0.47}
-
                                                    

 47%|████▋     | 5075/10707 [46:03<46:26,  2.02it/s]
 47%|████▋     | 5076/10707 [46:03<46:32,  2.02it/s]
 47%|████▋     | 5077/10707 [46:04<46:28,  2.02it/s]
 47%|████▋     | 5078/10707 [46:04<46:27,  2.02it/s]
 47%|████▋     | 5079/10707 [46:05<46:25,  2.02it/s]
 47%|████▋     | 5080/10707 [46:05<46:24,  2.02it/s]
 47%|████▋     | 5081/10707 [46:06<46:24,  2.02it/s]
 47%|████▋     | 5082/10707 [46:06<46:23,  2.02it/s]
 47%|████▋     | 5083/10707 [46:07<46:22,  2.02it/s]
 47%|████▋     | 5084/10707 [46:07<46:21,  2.02it/s]
 47%|████▋     | 5085/10707 [46:08<46:23,  2.02it/s]
 48%|████▊     | 5086/10707 [46:08<46:22,  2.02it/s]
 48%|████▊     | 5087/10707 [46:09<46:24,  2.02it/s]
 48%|████▊     | 5088/10707 [46:09<46:20,  2.02it/s]
 48%|████▊     | 5089/10707 [46:10<46:23,  2.02it/s]
 48%|████▊     | 5090/10707 [46:10<46:20,  2.02it/s]
 48%|████▊     | 5091/10707 [46:11<46:21,  2.02it/s]
 48%|████▊     | 5092/10707 [46:11<46:18,  2.02it/s]
 48%|████▊     | 5093/10707 [46:12<46:18,  2.02it/s]
 48%|████▊     | 5094/10707 [46:12<46:17,  2.02it/s]
 48%|████▊     | 5095/10707 [46:13<46:18,  2.02it/s]
 48%|████▊     | 5096/10707 [46:13<46:15,  2.02it/s]
 48%|████▊     | 5097/10707 [46:14<46:17,  2.02it/s]
 48%|████▊     | 5098/10707 [46:14<46:13,  2.02it/s]
 48%|████▊     | 5099/10707 [46:15<46:15,  2.02it/s]
 48%|████▊     | 5100/10707 [46:15<46:13,  2.02it/s]
                                                    {'loss': 3.7549, 'grad_norm': 0.19742192327976227, 'learning_rate': 0.0006272037593545237, 'epoch': 0.48}
-

 48%|████▊     | 5100/10707 [46:15<46:13,  2.02it/s]
 48%|████▊     | 5101/10707 [46:16<46:17,  2.02it/s]
 48%|████▊     | 5102/10707 [46:16<46:14,  2.02it/s]
 48%|████▊     | 5103/10707 [46:17<46:14,  2.02it/s]
 48%|████▊     | 5104/10707 [46:17<46:13,  2.02it/s]
 48%|████▊     | 5105/10707 [46:18<46:14,  2.02it/s]
 48%|████▊     | 5106/10707 [46:18<46:10,  2.02it/s]
 48%|████▊     | 5107/10707 [46:19<46:13,  2.02it/s]
 48%|████▊     | 5108/10707 [46:19<46:12,  2.02it/s]
 48%|████▊     | 5109/10707 [46:20<46:11,  2.02it/s]
 48%|████▊     | 5110/10707 [46:20<46:11,  2.02it/s]
 48%|████▊     | 5111/10707 [46:21<46:12,  2.02it/s]
 48%|████▊     | 5112/10707 [46:21<46:10,  2.02it/s]
 48%|████▊     | 5113/10707 [46:22<46:08,  2.02it/s]
 48%|████▊     | 5114/10707 [46:22<46:08,  2.02it/s]
 48%|████▊     | 5115/10707 [46:23<46:07,  2.02it/s]
 48%|████▊     | 5116/10707 [46:23<46:07,  2.02it/s]
 48%|████▊     | 5117/10707 [46:24<46:06,  2.02it/s]
 48%|████▊     | 5118/10707 [46:24<46:06,  2.02it/s]
 48%|████▊     | 5119/10707 [46:25<46:06,  2.02it/s]
 48%|████▊     | 5120/10707 [46:25<46:05,  2.02it/s]
 48%|████▊     | 5121/10707 [46:26<46:08,  2.02it/s]
 48%|████▊     | 5122/10707 [46:26<46:06,  2.02it/s]
 48%|████▊     | 5123/10707 [46:27<46:06,  2.02it/s]
 48%|████▊     | 5124/10707 [46:27<46:04,  2.02it/s]
 48%|████▊     | 5125/10707 [46:28<46:04,  2.02it/s]{'loss': 3.7494, 'grad_norm': 0.1959516704082489, 'learning_rate': 0.000623258335037458, 'epoch': 0.48}
-
                                                    

 48%|████▊     | 5125/10707 [46:28<46:04,  2.02it/s]
 48%|████▊     | 5126/10707 [46:28<46:06,  2.02it/s]
 48%|████▊     | 5127/10707 [46:29<46:06,  2.02it/s]
 48%|████▊     | 5128/10707 [46:29<46:04,  2.02it/s]
 48%|████▊     | 5129/10707 [46:30<46:07,  2.02it/s]
 48%|████▊     | 5130/10707 [46:30<46:05,  2.02it/s]
 48%|████▊     | 5131/10707 [46:31<46:02,  2.02it/s]
 48%|████��     | 5132/10707 [46:31<46:00,  2.02it/s]
 48%|████▊     | 5133/10707 [46:32<46:01,  2.02it/s]
 48%|████▊     | 5134/10707 [46:32<46:00,  2.02it/s]
 48%|████▊     | 5135/10707 [46:33<46:00,  2.02it/s]
 48%|████▊     | 5136/10707 [46:33<45:57,  2.02it/s]
 48%|████▊     | 5137/10707 [46:34<46:00,  2.02it/s]
 48%|████▊     | 5138/10707 [46:34<45:57,  2.02it/s]
 48%|████▊     | 5139/10707 [46:35<45:57,  2.02it/s]
 48%|████▊     | 5140/10707 [46:35<45:55,  2.02it/s]
 48%|████▊     | 5141/10707 [46:36<45:58,  2.02it/s]
 48%|████▊     | 5142/10707 [46:36<45:53,  2.02it/s]
 48%|████▊     | 5143/10707 [46:37<45:59,  2.02it/s]
 48%|████▊     | 5144/10707 [46:37<45:58,  2.02it/s]
 48%|████▊     | 5145/10707 [46:38<45:54,  2.02it/s]
 48%|████▊     | 5146/10707 [46:38<45:57,  2.02it/s]
 48%|████▊     | 5147/10707 [46:39<45:56,  2.02it/s]
 48%|████▊     | 5148/10707 [46:39<45:56,  2.02it/s]
 48%|████▊     | 5149/10707 [46:40<45:51,  2.02it/s]
 48%|████▊     | 5150/10707 [46:40<45:54,  2.02it/s]{'loss': 3.7453, 'grad_norm': 0.20721131563186646, 'learning_rate': 0.0006193047223011006, 'epoch': 0.48}
-
                                                    

 48%|████▊     | 5150/10707 [46:40<45:54,  2.02it/s]
 48%|████▊     | 5151/10707 [46:40<45:57,  2.01it/s]
 48%|████▊     | 5152/10707 [46:41<45:53,  2.02it/s]
 48%|████▊     | 5153/10707 [46:41<45:56,  2.01it/s]
 48%|████▊     | 5154/10707 [46:42<45:52,  2.02it/s]
 48%|████▊     | 5155/10707 [46:42<45:51,  2.02it/s]
 48%|████▊     | 5156/10707 [46:43<45:49,  2.02it/s]
 48%|████▊     | 5157/10707 [46:43<45:50,  2.02it/s]
 48%|████▊     | 5158/10707 [46:44<45:49,  2.02it/s]
 48%|████▊     | 5159/10707 [46:44<45:49,  2.02it/s]
 48%|████▊     | 5160/10707 [46:45<45:46,  2.02it/s]
 48%|████▊     | 5161/10707 [46:45<45:44,  2.02it/s]
 48%|████▊     | 5162/10707 [46:46<45:43,  2.02it/s]
 48%|████▊     | 5163/10707 [46:46<45:41,  2.02it/s]
 48%|████▊     | 5164/10707 [46:47<45:42,  2.02it/s]
 48%|████▊     | 5165/10707 [46:47<45:42,  2.02it/s]
 48%|████▊     | 5166/10707 [46:48<45:42,  2.02it/s]
 48%|████▊     | 5167/10707 [46:48<45:43,  2.02it/s]
 48%|████▊     | 5168/10707 [46:49<45:42,  2.02it/s]
 48%|████▊     | 5169/10707 [46:49<45:39,  2.02it/s]
 48%|████▊     | 5170/10707 [46:50<45:40,  2.02it/s]
 48%|████▊     | 5171/10707 [46:50<45:39,  2.02it/s]
 48%|████▊     | 5172/10707 [46:51<45:40,  2.02it/s]
 48%|████▊     | 5173/10707 [46:51<45:41,  2.02it/s]
 48%|████▊     | 5174/10707 [46:52<45:41,  2.02it/s]
 48%|████▊     | 5175/10707 [46:52<45:38,  2.02it/s]
                                                    
{'loss': 3.742, 'grad_norm': 0.21513019502162933, 'learning_rate': 0.0006153431837957528, 'epoch': 0.48}
-
 48%|████▊     | 5175/10707 [46:52<45:38,  2.02it/s]
 48%|████▊     | 5176/10707 [46:53<45:40,  2.02it/s]
 48%|████▊     | 5177/10707 [46:53<45:38,  2.02it/s]
 48%|████▊     | 5178/10707 [46:54<45:38,  2.02it/s]
 48%|████▊     | 5179/10707 [46:54<45:37,  2.02it/s]
 48%|████▊     | 5180/10707 [46:55<45:36,  2.02it/s]
 48%|████▊     | 5181/10707 [46:55<45:35,  2.02it/s]
 48%|████▊     | 5182/10707 [46:56<45:33,  2.02it/s]
 48%|████▊     | 5183/10707 [46:56<45:38,  2.02it/s]
 48%|████▊     | 5184/10707 [46:57<45:34,  2.02it/s]
 48%|████▊     | 5185/10707 [46:57<45:37,  2.02it/s]
 48%|████▊     | 5186/10707 [46:58<45:34,  2.02it/s]
 48%|████▊     | 5187/10707 [46:58<45:33,  2.02it/s]
 48%|████▊     | 5188/10707 [46:59<45:30,  2.02it/s]
 48%|████▊     | 5189/10707 [46:59<45:37,  2.02it/s]
 48%|████▊     | 5190/10707 [47:00<45:33,  2.02it/s]
 48%|████▊     | 5191/10707 [47:00<45:32,  2.02it/s]
 48%|████▊     | 5192/10707 [47:01<45:31,  2.02it/s]
 49%|████▊     | 5193/10707 [47:01<45:31,  2.02it/s]
 49%|████▊     | 5194/10707 [47:02<45:29,  2.02it/s]
 49%|████▊     | 5195/10707 [47:02<45:29,  2.02it/s]
 49%|████▊     | 5196/10707 [47:03<45:27,  2.02it/s]
 49%|████▊     | 5197/10707 [47:03<45:28,  2.02it/s]
 49%|████▊     | 5198/10707 [47:04<45:27,  2.02it/s]
 49%|████▊     | 5199/10707 [47:04<45:27,  2.02it/s]
 49%|████▊     | 5200/10707 [47:05<45:25,  2.02it/s]{'loss': 3.745, 'grad_norm': 0.1956460177898407, 'learning_rate': 0.0006113739826982479, 'epoch': 0.49}
-
                                                    

 49%|████▊     | 5200/10707 [47:05<45:25,  2.02it/s]
 49%|████▊     | 5201/10707 [47:05<45:28,  2.02it/s]
 49%|████▊     | 5202/10707 [47:06<45:27,  2.02it/s]
 49%|████▊     | 5203/10707 [47:06<45:26,  2.02it/s]
 49%|████▊     | 5204/10707 [47:07<45:26,  2.02it/s]
 49%|████▊     | 5205/10707 [47:07<45:24,  2.02it/s]
 49%|████▊     | 5206/10707 [47:08<45:26,  2.02it/s]
 49%|████▊     | 5207/10707 [47:08<45:23,  2.02it/s]
 49%|████▊     | 5208/10707 [47:09<45:20,  2.02it/s]
 49%|████▊     | 5209/10707 [47:09<45:21,  2.02it/s]
 49%|████▊     | 5210/10707 [47:10<45:20,  2.02it/s]
 49%|████▊     | 5211/10707 [47:10<45:22,  2.02it/s]
 49%|████▊     | 5212/10707 [47:11<45:21,  2.02it/s]
 49%|████▊     | 5213/10707 [47:11<45:22,  2.02it/s]
 49%|████▊     | 5214/10707 [47:12<45:19,  2.02it/s]
 49%|████▊     | 5215/10707 [47:12<45:26,  2.01it/s]
 49%|████▊     | 5216/10707 [47:13<45:24,  2.02it/s]
 49%|████▊     | 5217/10707 [47:13<45:23,  2.02it/s]
 49%|████▊     | 5218/10707 [47:14<45:20,  2.02it/s]
 49%|████▊     | 5219/10707 [47:14<45:22,  2.02it/s]
 49%|████▉     | 5220/10707 [47:15<45:17,  2.02it/s]
 49%|████▉     | 5221/10707 [47:15<45:19,  2.02it/s]
 49%|████▉     | 5222/10707 [47:16<45:16,  2.02it/s]
 49%|████▉     | 5223/10707 [47:16<45:17,  2.02it/s]
 49%|████▉     | 5224/10707 [47:17<45:15,  2.02it/s]
 49%|████▉     | 5225/10707 [47:17<45:14,  2.02it/s]{'loss': 3.7405, 'grad_norm': 0.21334019303321838, 'learning_rate': 0.0006073973826944683, 'epoch': 0.49}
-
                                                    

 49%|████▉     | 5225/10707 [47:17<45:14,  2.02it/s]
 49%|████▉     | 5226/10707 [47:18<45:21,  2.01it/s]
 49%|████▉     | 5227/10707 [47:18<45:16,  2.02it/s]
 49%|████▉     | 5228/10707 [47:19<45:17,  2.02it/s]
 49%|████▉     | 5229/10707 [47:19<45:14,  2.02it/s]
 49%|████▉     | 5230/10707 [47:20<45:13,  2.02it/s]
 49%|████▉     | 5231/10707 [47:20<45:10,  2.02it/s]
 49%|████▉     | 5232/10707 [47:21<45:11,  2.02it/s]
 49%|████▉     | 5233/10707 [47:21<45:09,  2.02it/s]
 49%|████▉     | 5234/10707 [47:22<45:11,  2.02it/s]
 49%|████▉     | 5235/10707 [47:22<45:09,  2.02it/s]
 49%|████▉     | 5236/10707 [47:23<45:12,  2.02it/s]
 49%|████▉     | 5237/10707 [47:23<45:08,  2.02it/s]
 49%|████▉     | 5238/10707 [47:24<45:08,  2.02it/s]
 49%|████▉     | 5239/10707 [47:24<45:06,  2.02it/s]
 49%|████▉     | 5240/10707 [47:25<45:07,  2.02it/s]
 49%|████▉     | 5241/10707 [47:25<45:04,  2.02it/s]
 49%|████▉     | 5242/10707 [47:26<45:07,  2.02it/s]
 49%|████▉     | 5243/10707 [47:26<45:05,  2.02it/s]
 49%|████▉     | 5244/10707 [47:27<45:07,  2.02it/s]
 49%|████▉     | 5245/10707 [47:27<45:05,  2.02it/s]
 49%|████▉     | 5246/10707 [47:28<45:05,  2.02it/s]
 49%|████▉     | 5247/10707 [47:28<45:04,  2.02it/s]
 49%|████▉     | 5248/10707 [47:29<45:05,  2.02it/s]
 49%|████▉     | 5249/10707 [47:29<45:05,  2.02it/s]
 49%|████▉     | 5250/10707 [47:30<45:05,  2.02it/s]{'loss': 3.7376, 'grad_norm': 0.20293313264846802, 'learning_rate': 0.000603413647961828, 'epoch': 0.49}
-
                                                    

 49%|████▉     | 5250/10707 [47:30<45:05,  2.02it/s]
 49%|████▉     | 5251/10707 [47:30<45:19,  2.01it/s]
 49%|████▉     | 5252/10707 [47:31<45:13,  2.01it/s]
 49%|████▉     | 5253/10707 [47:31<45:08,  2.01it/s]
 49%|████▉     | 5254/10707 [47:32<45:05,  2.02it/s]
 49%|████▉     | 5255/10707 [47:32<44:59,  2.02it/s]
 49%|████▉     | 5256/10707 [47:33<45:01,  2.02it/s]
 49%|████▉     | 5257/10707 [47:33<45:04,  2.02it/s]
 49%|████▉     | 5258/10707 [47:34<45:09,  2.01it/s]
 49%|████▉     | 5259/10707 [47:34<45:09,  2.01it/s]
 49%|████▉     | 5260/10707 [47:35<45:03,  2.01it/s]
 49%|████▉     | 5261/10707 [47:35<45:02,  2.02it/s]
 49%|████▉     | 5262/10707 [47:35<45:02,  2.02it/s]
 49%|████▉     | 5263/10707 [47:36<44:57,  2.02it/s]
 49%|████▉     | 5264/10707 [47:36<44:56,  2.02it/s]
 49%|████▉     | 5265/10707 [47:37<44:55,  2.02it/s]
 49%|████▉     | 5266/10707 [47:37<44:57,  2.02it/s]
 49%|████▉     | 5267/10707 [47:38<44:56,  2.02it/s]
 49%|████▉     | 5268/10707 [47:38<44:57,  2.02it/s]
 49%|████▉     | 5269/10707 [47:39<44:56,  2.02it/s]
 49%|████▉     | 5270/10707 [47:39<44:57,  2.02it/s]
 49%|████▉     | 5271/10707 [47:40<44:53,  2.02it/s]
 49%|████▉     | 5272/10707 [47:40<44:52,  2.02it/s]
 49%|████▉     | 5273/10707 [47:41<44:53,  2.02it/s]
 49%|████▉     | 5274/10707 [47:41<44:51,  2.02it/s]
 49%|████▉     | 5275/10707 [47:42<44:50,  2.02it/s]{'loss': 3.7437, 'grad_norm': 0.21730415523052216, 'learning_rate': 0.0005994230431517222, 'epoch': 0.49}
-
                                                    

 49%|████▉     | 5275/10707 [47:42<44:50,  2.02it/s]
 49%|████▉     | 5276/10707 [47:42<44:52,  2.02it/s]
 49%|████▉     | 5277/10707 [47:43<44:51,  2.02it/s]
 49%|████▉     | 5278/10707 [47:43<44:46,  2.02it/s]
 49%|████▉     | 5279/10707 [47:44<44:47,  2.02it/s]
 49%|████▉     | 5280/10707 [47:44<44:46,  2.02it/s]
 49%|████▉     | 5281/10707 [47:45<44:49,  2.02it/s]
 49%|████▉     | 5282/10707 [47:45<44:51,  2.02it/s]
 49%|████▉     | 5283/10707 [47:46<44:49,  2.02it/s]
 49%|████▉     | 5284/10707 [47:46<44:47,  2.02it/s]
 49%|████▉     | 5285/10707 [47:47<44:47,  2.02it/s]
 49%|████▉     | 5286/10707 [47:47<44:44,  2.02it/s]
 49%|████▉     | 5287/10707 [47:48<44:43,  2.02it/s]
 49%|████▉     | 5288/10707 [47:48<44:43,  2.02it/s]
 49%|████▉     | 5289/10707 [47:49<44:42,  2.02it/s]
 49%|████▉     | 5290/10707 [47:49<44:40,  2.02it/s]
 49%|████▉     | 5291/10707 [47:50<44:40,  2.02it/s]
 49%|████▉     | 5292/10707 [47:50<44:39,  2.02it/s]
 49%|████▉     | 5293/10707 [47:51<44:42,  2.02it/s]
 49%|████▉     | 5294/10707 [47:51<44:38,  2.02it/s]
 49%|████▉     | 5295/10707 [47:52<44:39,  2.02it/s]
 49%|████▉     | 5296/10707 [47:52<44:37,  2.02it/s]
 49%|████▉     | 5297/10707 [47:53<44:39,  2.02it/s]
 49%|████▉     | 5298/10707 [47:53<44:38,  2.02it/s]
 49%|████▉     | 5299/10707 [47:54<44:39,  2.02it/s]
 50%|████▉     | 5300/10707 [47:54<44:37,  2.02it/s]{'loss': 3.7321, 'grad_norm': 0.19508059322834015, 'learning_rate': 0.0005954258333719457, 'epoch': 0.49}
-
                                                    

 50%|████▉     | 5300/10707 [47:54<44:37,  2.02it/s]
 50%|████▉     | 5301/10707 [47:55<44:39,  2.02it/s]
 50%|████▉     | 5302/10707 [47:55<44:38,  2.02it/s]
 50%|████▉     | 5303/10707 [47:56<44:37,  2.02it/s]
 50%|████▉     | 5304/10707 [47:56<44:39,  2.02it/s]
 50%|████▉     | 5305/10707 [47:57<44:34,  2.02it/s]
 50%|████▉     | 5306/10707 [47:57<44:35,  2.02it/s]
 50%|████▉     | 5307/10707 [47:58<44:32,  2.02it/s]
 50%|████▉     | 5308/10707 [47:58<44:32,  2.02it/s]
 50%|████▉     | 5309/10707 [47:59<44:30,  2.02it/s]
 50%|████▉     | 5310/10707 [47:59<44:32,  2.02it/s]
 50%|████▉     | 5311/10707 [48:00<44:30,  2.02it/s]
 50%|████▉     | 5312/10707 [48:00<44:32,  2.02it/s]
 50%|████▉     | 5313/10707 [48:01<44:29,  2.02it/s]
 50%|████▉     | 5314/10707 [48:01<44:29,  2.02it/s]
 50%|████▉     | 5315/10707 [48:02<44:30,  2.02it/s]
 50%|████▉     | 5316/10707 [48:02<44:37,  2.01it/s]
 50%|████▉     | 5317/10707 [48:03<44:36,  2.01it/s]
 50%|████▉     | 5318/10707 [48:03<44:39,  2.01it/s]
 50%|████▉     | 5319/10707 [48:04<44:42,  2.01it/s]
 50%|████▉     | 5320/10707 [48:04<44:39,  2.01it/s]
 50%|████▉     | 5321/10707 [48:05<44:34,  2.01it/s]
 50%|████▉     | 5322/10707 [48:05<44:35,  2.01it/s]
 50%|████▉     | 5323/10707 [48:06<44:30,  2.02it/s]
 50%|████▉     | 5324/10707 [48:06<44:29,  2.02it/s]
 50%|████▉     | 5325/10707 [48:07<44:30,  2.02it/s]
                                                    
{'loss': 3.731, 'grad_norm': 0.19207492470741272, 'learning_rate': 0.0005914222841690814, 'epoch': 0.5}
-
 50%|████▉     | 5325/10707 [48:07<44:30,  2.02it/s]
 50%|████▉     | 5326/10707 [48:07<44:33,  2.01it/s]
 50%|████▉     | 5327/10707 [48:08<44:30,  2.01it/s]
 50%|████▉     | 5328/10707 [48:08<44:27,  2.02it/s]
 50%|████▉     | 5329/10707 [48:09<44:27,  2.02it/s]
 50%|████▉     | 5330/10707 [48:09<44:26,  2.02it/s]
 50%|████▉     | 5331/10707 [48:10<44:23,  2.02it/s]
 50%|████▉     | 5332/10707 [48:10<44:23,  2.02it/s]
 50%|████▉     | 5333/10707 [48:11<44:21,  2.02it/s]
 50%|████▉     | 5334/10707 [48:11<44:20,  2.02it/s]
 50%|████▉     | 5335/10707 [48:12<44:20,  2.02it/s]
 50%|████▉     | 5336/10707 [48:12<44:19,  2.02it/s]
 50%|████▉     | 5337/10707 [48:13<44:21,  2.02it/s]
 50%|████▉     | 5338/10707 [48:13<44:18,  2.02it/s]
 50%|████▉     | 5339/10707 [48:14<44:16,  2.02it/s]
 50%|████▉     | 5340/10707 [48:14<44:17,  2.02it/s]
 50%|████▉     | 5341/10707 [48:15<44:16,  2.02it/s]
 50%|████▉     | 5342/10707 [48:15<44:17,  2.02it/s]
 50%|████▉     | 5343/10707 [48:16<44:14,  2.02it/s]
 50%|████▉     | 5344/10707 [48:16<44:19,  2.02it/s]
 50%|████▉     | 5345/10707 [48:17<44:17,  2.02it/s]
 50%|████▉     | 5346/10707 [48:17<44:17,  2.02it/s]
 50%|████▉     | 5347/10707 [48:18<51:37,  1.73it/s]
 50%|████▉     | 5348/10707 [48:18<49:23,  1.81it/s]
 50%|████▉     | 5349/10707 [48:19<47:52,  1.87it/s]
 50%|████▉     | 5350/10707 [48:19<46:45,  1.91it/s]{'loss': 3.7259, 'grad_norm': 0.21463729441165924, 'learning_rate': 0.000587412661510859, 'epoch': 0.5}
-
                                                    

 50%|████▉     | 5350/10707 [48:19<46:45,  1.91it/s]
 50%|████▉     | 5351/10707 [48:20<46:00,  1.94it/s]
 50%|████▉     | 5352/10707 [48:20<45:25,  1.96it/s]
 50%|████▉     | 5353/10707 [48:21<45:03,  1.98it/s]
 50%|█████     | 5354/10707 [48:21<44:44,  1.99it/s]
 50%|█████     | 5355/10707 [48:22<44:39,  2.00it/s]
 50%|█████     | 5356/10707 [48:22<44:30,  2.00it/s]
 50%|█████     | 5357/10707 [48:23<44:23,  2.01it/s]
 50%|█████     | 5358/10707 [48:23<44:18,  2.01it/s]
 50%|█████     | 5359/10707 [48:24<44:15,  2.01it/s]
 50%|█████     | 5360/10707 [48:24<44:12,  2.02it/s]
 50%|█████     | 5361/10707 [48:25<44:09,  2.02it/s]
 50%|█████     | 5362/10707 [48:25<44:07,  2.02it/s]
 50%|█████     | 5363/10707 [48:26<44:06,  2.02it/s]
 50%|█████     | 5364/10707 [48:26<44:07,  2.02it/s]
 50%|█████     | 5365/10707 [48:27<44:05,  2.02it/s]
 50%|█████     | 5366/10707 [48:27<44:09,  2.02it/s]
 50%|█████     | 5367/10707 [48:28<44:07,  2.02it/s]
 50%|█████     | 5368/10707 [48:28<44:07,  2.02it/s]
 50%|█████     | 5369/10707 [48:29<44:07,  2.02it/s]
 50%|█████     | 5370/10707 [48:29<44:04,  2.02it/s]
 50%|█████     | 5371/10707 [48:30<51:03,  1.74it/s]
 50%|█████     | 5372/10707 [48:31<48:56,  1.82it/s]
 50%|█████     | 5373/10707 [48:31<47:28,  1.87it/s]
 50%|█████     | 5374/10707 [48:32<46:24,  1.92it/s]
 50%|█████     | 5375/10707 [48:32<45:46,  1.94it/s]{'loss': 3.7313, 'grad_norm': 0.21112915873527527, 'learning_rate': 0.0005833972317684858, 'epoch': 0.5}
                                                    
-
 50%|█████     | 5375/10707 [48:32<45:46,  1.94it/s]
 50%|█████     | 5376/10707 [48:33<45:15,  1.96it/s]
 50%|█████     | 5377/10707 [48:33<44:58,  1.98it/s]
 50%|█████     | 5378/10707 [48:34<44:49,  1.98it/s]
 50%|█████     | 5379/10707 [48:34<44:32,  1.99it/s]
 50%|█████     | 5380/10707 [48:35<44:24,  2.00it/s]
 50%|█████     | 5381/10707 [48:35<44:16,  2.01it/s]
 50%|█████     | 5382/10707 [48:36<44:10,  2.01it/s]
 50%|█████     | 5383/10707 [48:36<44:03,  2.01it/s]
 50%|█████     | 5384/10707 [48:36<44:03,  2.01it/s]
 50%|█████     | 5385/10707 [48:37<44:00,  2.02it/s]
 50%|█████     | 5386/10707 [48:37<44:01,  2.01it/s]
 50%|█████     | 5387/10707 [48:38<43:57,  2.02it/s]
 50%|█████     | 5388/10707 [48:38<43:55,  2.02it/s]
 50%|█████     | 5389/10707 [48:39<43:58,  2.02it/s]
 50%|█████     | 5390/10707 [48:39<43:55,  2.02it/s]
 50%|█████     | 5391/10707 [48:40<43:54,  2.02it/s]
 50%|█████     | 5392/10707 [48:40<43:54,  2.02it/s]
 50%|█████     | 5393/10707 [48:41<43:52,  2.02it/s]
 50%|█████     | 5394/10707 [48:41<43:49,  2.02it/s]
 50%|█████     | 5395/10707 [48:42<43:50,  2.02it/s]
 50%|█████     | 5396/10707 [48:42<43:47,  2.02it/s]
 50%|█████     | 5397/10707 [48:43<43:51,  2.02it/s]
 50%|█████     | 5398/10707 [48:43<43:47,  2.02it/s]
 50%|█████     | 5399/10707 [48:44<43:49,  2.02it/s]
 50%|█████     | 5400/10707 [48:44<43:48,  2.02it/s]{'loss': 3.7228, 'grad_norm': 0.2007475346326828, 'learning_rate': 0.0005793762616989512, 'epoch': 0.5}
-
                                                    

 50%|█████     | 5400/10707 [48:44<43:48,  2.02it/s]
 50%|█████     | 5401/10707 [48:45<43:51,  2.02it/s]
 50%|█████     | 5402/10707 [48:45<43:51,  2.02it/s]
 50%|█████     | 5403/10707 [48:46<43:48,  2.02it/s]
 50%|█████     | 5404/10707 [48:46<43:48,  2.02it/s]
 50%|█████     | 5405/10707 [48:47<43:47,  2.02it/s]
 50%|█████     | 5406/10707 [48:47<43:47,  2.02it/s]
 50%|█████     | 5407/10707 [48:48<43:45,  2.02it/s]
 51%|█████     | 5408/10707 [48:48<43:46,  2.02it/s]
 51%|█████     | 5409/10707 [48:49<43:43,  2.02it/s]
 51%|█████     | 5410/10707 [48:49<43:42,  2.02it/s]
 51%|█████     | 5411/10707 [48:50<43:44,  2.02it/s]
 51%|█████     | 5412/10707 [48:50<43:44,  2.02it/s]
 51%|█████     | 5413/10707 [48:51<43:42,  2.02it/s]
 51%|█████     | 5414/10707 [48:51<43:40,  2.02it/s]
 51%|█████     | 5415/10707 [48:52<43:50,  2.01it/s]
 51%|█████     | 5416/10707 [48:52<43:48,  2.01it/s]
 51%|█████     | 5417/10707 [48:53<43:52,  2.01it/s]
 51%|█████     | 5418/10707 [48:53<43:57,  2.01it/s]
 51%|█████     | 5419/10707 [48:54<43:59,  2.00it/s]
 51%|█████     | 5420/10707 [48:54<43:59,  2.00it/s]
 51%|█████     | 5421/10707 [48:55<43:56,  2.01it/s]
 51%|█████     | 5422/10707 [48:55<43:58,  2.00it/s]
 51%|█████     | 5423/10707 [48:56<43:59,  2.00it/s]
 51%|█████     | 5424/10707 [48:56<43:58,  2.00it/s]
 51%|█████     | 5425/10707 [48:57<44:02,  2.00it/s]{'loss': 3.7293, 'grad_norm': 0.21932096779346466, 'learning_rate': 0.0005753500184273051, 'epoch': 0.51}
-
                                                    

 51%|█████     | 5425/10707 [48:57<44:02,  2.00it/s]
 51%|█████     | 5426/10707 [48:57<44:00,  2.00it/s]
 51%|█████     | 5427/10707 [48:58<43:55,  2.00it/s]
 51%|█████     | 5428/10707 [48:58<43:53,  2.00it/s]
 51%|█████     | 5429/10707 [48:59<43:53,  2.00it/s]
 51%|█████     | 5430/10707 [48:59<43:50,  2.01it/s]
 51%|█████     | 5431/10707 [49:00<43:46,  2.01it/s]
 51%|█████     | 5432/10707 [49:00<43:47,  2.01it/s]
 51%|█████     | 5433/10707 [49:01<43:46,  2.01it/s]
 51%|█████     | 5434/10707 [49:01<43:41,  2.01it/s]
 51%|█████     | 5435/10707 [49:02<43:36,  2.01it/s]
 51%|█████     | 5436/10707 [49:02<43:38,  2.01it/s]
 51%|█████     | 5437/10707 [49:03<43:35,  2.01it/s]
 51%|█████     | 5438/10707 [49:03<43:41,  2.01it/s]
 51%|█████     | 5439/10707 [49:04<43:41,  2.01it/s]
 51%|█████     | 5440/10707 [49:04<43:34,  2.01it/s]
 51%|█████     | 5441/10707 [49:05<43:35,  2.01it/s]
 51%|█████     | 5442/10707 [49:05<43:31,  2.02it/s]
 51%|█████     | 5443/10707 [49:06<43:29,  2.02it/s]
 51%|█████     | 5444/10707 [49:06<43:26,  2.02it/s]
 51%|█████     | 5445/10707 [49:07<43:28,  2.02it/s]
 51%|█████     | 5446/10707 [49:07<43:25,  2.02it/s]
 51%|█████     | 5447/10707 [49:08<43:25,  2.02it/s]
 51%|█████     | 5448/10707 [49:08<43:24,  2.02it/s]
 51%|█████     | 5449/10707 [49:09<43:24,  2.02it/s]
 51%|█████     | 5450/10707 [49:09<43:23,  2.02it/s]{'loss': 3.731, 'grad_norm': 0.19414381682872772, 'learning_rate': 0.0005713187694289117, 'epoch': 0.51}
                                                    
-
 51%|█████     | 5450/10707 [49:09<43:23,  2.02it/s]
 51%|█████     | 5451/10707 [49:10<43:24,  2.02it/s]
 51%|█████     | 5452/10707 [49:10<43:21,  2.02it/s]
 51%|█████     | 5453/10707 [49:11<43:22,  2.02it/s]
 51%|█████     | 5454/10707 [49:11<43:22,  2.02it/s]
 51%|█████     | 5455/10707 [49:12<43:19,  2.02it/s]
 51%|█████     | 5456/10707 [49:12<43:21,  2.02it/s]
 51%|█████     | 5457/10707 [49:13<43:19,  2.02it/s]
 51%|█████     | 5458/10707 [49:13<43:20,  2.02it/s]
 51%|█████     | 5459/10707 [49:14<43:18,  2.02it/s]
 51%|█████     | 5460/10707 [49:14<43:19,  2.02it/s]
 51%|█████     | 5461/10707 [49:15<43:16,  2.02it/s]
 51%|█████     | 5462/10707 [49:15<43:18,  2.02it/s]
 51%|█████     | 5463/10707 [49:16<43:16,  2.02it/s]
 51%|█████     | 5464/10707 [49:16<43:19,  2.02it/s]
 51%|█████     | 5465/10707 [49:17<43:19,  2.02it/s]
 51%|█████     | 5466/10707 [49:17<43:20,  2.02it/s]
 51%|█████     | 5467/10707 [49:18<43:20,  2.01it/s]
 51%|█████     | 5468/10707 [49:18<43:16,  2.02it/s]
 51%|█████     | 5469/10707 [49:19<43:18,  2.02it/s]
 51%|█████     | 5470/10707 [49:19<43:14,  2.02it/s]
 51%|█████     | 5471/10707 [49:20<43:15,  2.02it/s]
 51%|█████     | 5472/10707 [49:20<43:12,  2.02it/s]
 51%|█████     | 5473/10707 [49:21<43:15,  2.02it/s]
 51%|█████     | 5474/10707 [49:21<43:12,  2.02it/s]
 51%|█████     | 5475/10707 [49:22<43:14,  2.02it/s]{'loss': 3.7257, 'grad_norm': 0.19901283085346222, 'learning_rate': 0.000567282782511681, 'epoch': 0.51}
-
                                                    

 51%|█████     | 5475/10707 [49:22<43:14,  2.02it/s]
 51%|█████     | 5476/10707 [49:22<43:14,  2.02it/s]
 51%|█████     | 5477/10707 [49:23<43:13,  2.02it/s]
 51%|█████     | 5478/10707 [49:23<43:10,  2.02it/s]
 51%|█████     | 5479/10707 [49:24<43:11,  2.02it/s]
 51%|█████     | 5480/10707 [49:24<43:08,  2.02it/s]
 51%|█████     | 5481/10707 [49:25<43:09,  2.02it/s]
 51%|█████     | 5482/10707 [49:25<43:12,  2.02it/s]
 51%|█████     | 5483/10707 [49:26<43:13,  2.01it/s]
 51%|█████     | 5484/10707 [49:26<43:12,  2.01it/s]
 51%|█████     | 5485/10707 [49:27<43:09,  2.02it/s]
 51%|█████     | 5486/10707 [49:27<43:07,  2.02it/s]
 51%|█████     | 5487/10707 [49:28<43:08,  2.02it/s]
 51%|█████▏    | 5488/10707 [49:28<43:06,  2.02it/s]
 51%|█████▏    | 5489/10707 [49:29<43:08,  2.02it/s]
 51%|█████▏    | 5490/10707 [49:29<43:03,  2.02it/s]
 51%|█████▏    | 5491/10707 [49:30<43:05,  2.02it/s]
 51%|█████▏    | 5492/10707 [49:30<43:00,  2.02it/s]
 51%|█████▏    | 5493/10707 [49:31<43:02,  2.02it/s]
 51%|█████▏    | 5494/10707 [49:31<43:01,  2.02it/s]
 51%|█████▏    | 5495/10707 [49:32<43:01,  2.02it/s]
 51%|█████▏    | 5496/10707 [49:32<42:59,  2.02it/s]
 51%|█████▏    | 5497/10707 [49:33<42:59,  2.02it/s]
 51%|█████▏    | 5498/10707 [49:33<43:03,  2.02it/s]
 51%|█████▏    | 5499/10707 [49:34<43:04,  2.01it/s]
 51%|█████▏    | 5500/10707 [49:34<43:04,  2.02it/s]{'loss': 3.7237, 'grad_norm': 0.20705921947956085, 'learning_rate': 0.0005632423257982766, 'epoch': 0.51}
-
                                                    

 51%|█████▏    | 5500/10707 [49:34<43:04,  2.02it/s]
 51%|█████▏    | 5501/10707 [49:35<43:03,  2.01it/s]
 51%|█████▏    | 5502/10707 [49:35<43:02,  2.02it/s]
 51%|█████▏    | 5503/10707 [49:36<42:59,  2.02it/s]
 51%|█████▏    | 5504/10707 [49:36<42:59,  2.02it/s]
 51%|█████▏    | 5505/10707 [49:37<43:00,  2.02it/s]
 51%|█████▏    | 5506/10707 [49:37<42:57,  2.02it/s]
 51%|█████▏    | 5507/10707 [49:38<42:59,  2.02it/s]
 51%|█████▏    | 5508/10707 [49:38<42:54,  2.02it/s]
 51%|█████▏    | 5509/10707 [49:39<42:56,  2.02it/s]
 51%|█████▏    | 5510/10707 [49:39<42:52,  2.02it/s]
 51%|█████▏    | 5511/10707 [49:40<42:54,  2.02it/s]
 51%|█████▏    | 5512/10707 [49:40<42:54,  2.02it/s]
 51%|█████▏    | 5513/10707 [49:40<42:55,  2.02it/s]
 51%|█████▏    | 5514/10707 [49:41<42:52,  2.02it/s]
 52%|█████▏    | 5515/10707 [49:41<42:50,  2.02it/s]
 52%|█████▏    | 5516/10707 [49:42<42:50,  2.02it/s]
 52%|█████▏    | 5517/10707 [49:42<42:49,  2.02it/s]
 52%|█████▏    | 5518/10707 [49:43<42:52,  2.02it/s]
 52%|█████▏    | 5519/10707 [49:43<42:51,  2.02it/s]
 52%|█████▏    | 5520/10707 [49:44<42:49,  2.02it/s]
 52%|█████▏    | 5521/10707 [49:44<42:48,  2.02it/s]
 52%|█████▏    | 5522/10707 [49:45<42:49,  2.02it/s]
 52%|█████▏    | 5523/10707 [49:45<42:47,  2.02it/s]
 52%|█████▏    | 5524/10707 [49:46<42:49,  2.02it/s]
 52%|█████▏    | 5525/10707 [49:46<42:46,  2.02it/s]
                                                    
{'loss': 3.7184, 'grad_norm': 0.19622330367565155, 'learning_rate': 0.0005591976677083044, 'epoch': 0.52}
-
 52%|█████▏    | 5525/10707 [49:46<42:46,  2.02it/s]
 52%|█████▏    | 5526/10707 [49:47<42:50,  2.02it/s]
 52%|█████▏    | 5527/10707 [49:47<42:53,  2.01it/s]
 52%|█████▏    | 5528/10707 [49:48<42:49,  2.02it/s]
 52%|█████▏    | 5529/10707 [49:48<42:49,  2.02it/s]
 52%|█████▏    | 5530/10707 [49:49<42:46,  2.02it/s]
 52%|█████▏    | 5531/10707 [49:49<42:45,  2.02it/s]
 52%|█████▏    | 5532/10707 [49:50<42:45,  2.02it/s]
 52%|█████▏    | 5533/10707 [49:50<42:44,  2.02it/s]
 52%|█████▏    | 5534/10707 [49:51<42:44,  2.02it/s]
 52%|█████▏    | 5535/10707 [49:51<42:42,  2.02it/s]
 52%|█████▏    | 5536/10707 [49:52<42:44,  2.02it/s]
 52%|█████▏    | 5537/10707 [49:52<42:43,  2.02it/s]
 52%|█████▏    | 5538/10707 [49:53<42:44,  2.02it/s]
 52%|█████▏    | 5539/10707 [49:53<42:41,  2.02it/s]
 52%|█████▏    | 5540/10707 [49:54<42:40,  2.02it/s]
 52%|█████▏    | 5541/10707 [49:54<42:40,  2.02it/s]
 52%|█████▏    | 5542/10707 [49:55<42:40,  2.02it/s]
 52%|█████▏    | 5543/10707 [49:55<42:39,  2.02it/s]
 52%|█████▏    | 5544/10707 [49:56<42:37,  2.02it/s]
 52%|█████▏    | 5545/10707 [49:56<42:39,  2.02it/s]
 52%|█████▏    | 5546/10707 [49:57<42:37,  2.02it/s]
 52%|█████▏    | 5547/10707 [49:57<42:37,  2.02it/s]
 52%|█████▏    | 5548/10707 [49:58<42:35,  2.02it/s]
 52%|█████▏    | 5549/10707 [49:58<42:36,  2.02it/s]
 52%|█████▏    | 5550/10707 [49:59<42:35,  2.02it/s]{'loss': 3.724, 'grad_norm': 0.1885838806629181, 'learning_rate': 0.0005551490769404803, 'epoch': 0.52}
-
                                                    

 52%|█████▏    | 5550/10707 [49:59<42:35,  2.02it/s]
 52%|█████▏    | 5551/10707 [49:59<42:37,  2.02it/s]
 52%|█████▏    | 5552/10707 [50:00<42:36,  2.02it/s]
 52%|█████▏    | 5553/10707 [50:00<42:34,  2.02it/s]
 52%|█████▏    | 5554/10707 [50:01<42:34,  2.02it/s]
 52%|█████▏    | 5555/10707 [50:01<42:34,  2.02it/s]
 52%|█████▏    | 5556/10707 [50:02<42:33,  2.02it/s]
 52%|█████▏    | 5557/10707 [50:02<42:33,  2.02it/s]
 52%|█████▏    | 5558/10707 [50:03<42:36,  2.01it/s]
 52%|█████▏    | 5559/10707 [50:03<42:43,  2.01it/s]
 52%|█████▏    | 5560/10707 [50:04<42:41,  2.01it/s]
 52%|█████▏    | 5561/10707 [50:04<42:38,  2.01it/s]
 52%|█████▏    | 5562/10707 [50:05<42:35,  2.01it/s]
 52%|█████▏    | 5563/10707 [50:05<42:33,  2.01it/s]
 52%|█████▏    | 5564/10707 [50:06<42:29,  2.02it/s]
 52%|█████▏    | 5565/10707 [50:06<42:29,  2.02it/s]
 52%|█████▏    | 5566/10707 [50:07<42:24,  2.02it/s]
 52%|█████▏    | 5567/10707 [50:07<42:26,  2.02it/s]
 52%|█████▏    | 5568/10707 [50:08<42:24,  2.02it/s]
 52%|█████▏    | 5569/10707 [50:08<42:25,  2.02it/s]
 52%|█████▏    | 5570/10707 [50:09<42:24,  2.02it/s]
 52%|█████▏    | 5571/10707 [50:09<42:26,  2.02it/s]
 52%|█████▏    | 5572/10707 [50:10<42:24,  2.02it/s]
 52%|█████▏    | 5573/10707 [50:10<42:23,  2.02it/s]
 52%|█████▏    | 5574/10707 [50:11<42:21,  2.02it/s]
 52%|█████▏    | 5575/10707 [50:11<42:23,  2.02it/s]{'loss': 3.7237, 'grad_norm': 0.197038933634758, 'learning_rate': 0.0005510968224547795, 'epoch': 0.52}
-
                                                    

 52%|█████▏    | 5575/10707 [50:11<42:23,  2.02it/s]
 52%|█████▏    | 5576/10707 [50:12<42:24,  2.02it/s]
 52%|█████▏    | 5577/10707 [50:12<42:22,  2.02it/s]
 52%|█████▏    | 5578/10707 [50:13<42:22,  2.02it/s]
 52%|█████▏    | 5579/10707 [50:13<42:19,  2.02it/s]
 52%|█████▏    | 5580/10707 [50:14<42:20,  2.02it/s]
 52%|█████▏    | 5581/10707 [50:14<42:17,  2.02it/s]
 52%|█████▏    | 5582/10707 [50:15<42:16,  2.02it/s]
 52%|█████▏    | 5583/10707 [50:15<42:15,  2.02it/s]
 52%|█████▏    | 5584/10707 [50:16<42:21,  2.02it/s]
 52%|█████▏    | 5585/10707 [50:16<42:20,  2.02it/s]
 52%|█████▏    | 5586/10707 [50:17<42:21,  2.01it/s]
 52%|█████▏    | 5587/10707 [50:17<42:20,  2.02it/s]
 52%|█████▏    | 5588/10707 [50:18<42:18,  2.02it/s]
 52%|█████▏    | 5589/10707 [50:18<42:17,  2.02it/s]
 52%|█████▏    | 5590/10707 [50:19<42:17,  2.02it/s]
 52%|█████▏    | 5591/10707 [50:19<42:17,  2.02it/s]
 52%|█████▏    | 5592/10707 [50:20<42:17,  2.02it/s]
 52%|█████▏    | 5593/10707 [50:20<42:18,  2.01it/s]
 52%|█████▏    | 5594/10707 [50:21<42:14,  2.02it/s]
 52%|█████▏    | 5595/10707 [50:21<42:14,  2.02it/s]
 52%|█████▏    | 5596/10707 [50:22<42:13,  2.02it/s]
 52%|█████▏    | 5597/10707 [50:22<42:11,  2.02it/s]
 52%|█████▏    | 5598/10707 [50:23<42:10,  2.02it/s]
 52%|█████▏    | 5599/10707 [50:23<42:10,  2.02it/s]
 52%|█████▏    | 5600/10707 [50:24<42:11,  2.02it/s]{'loss': 3.7211, 'grad_norm': 0.2043546885251999, 'learning_rate': 0.0005470411734545688, 'epoch': 0.52}
-
                                                    

 52%|█████▏    | 5600/10707 [50:24<42:11,  2.02it/s]
 52%|█████▏    | 5601/10707 [50:24<42:11,  2.02it/s]
 52%|█████▏    | 5602/10707 [50:25<42:11,  2.02it/s]
 52%|█████▏    | 5603/10707 [50:25<42:09,  2.02it/s]
 52%|█████▏    | 5604/10707 [50:26<42:09,  2.02it/s]
 52%|█████▏    | 5605/10707 [50:26<42:09,  2.02it/s]
 52%|█████▏    | 5606/10707 [50:27<42:10,  2.02it/s]
 52%|█████▏    | 5607/10707 [50:27<42:08,  2.02it/s]
 52%|█████▏    | 5608/10707 [50:28<42:07,  2.02it/s]
 52%|█████▏    | 5609/10707 [50:28<42:06,  2.02it/s]
 52%|█████▏    | 5610/10707 [50:29<42:06,  2.02it/s]
 52%|█████▏    | 5611/10707 [50:29<42:04,  2.02it/s]
 52%|█████▏    | 5612/10707 [50:30<42:05,  2.02it/s]
 52%|█████▏    | 5613/10707 [50:30<42:03,  2.02it/s]
 52%|█████▏    | 5614/10707 [50:31<42:04,  2.02it/s]
 52%|█████▏    | 5615/10707 [50:31<42:03,  2.02it/s]
 52%|█████▏    | 5616/10707 [50:32<42:03,  2.02it/s]
 52%|█████▏    | 5617/10707 [50:32<41:59,  2.02it/s]
 52%|█████▏    | 5618/10707 [50:33<42:02,  2.02it/s]
 52%|█████▏    | 5619/10707 [50:33<42:02,  2.02it/s]
 52%|█████▏    | 5620/10707 [50:34<42:08,  2.01it/s]
 52%|█████▏    | 5621/10707 [50:34<42:07,  2.01it/s]
 53%|█████▎    | 5622/10707 [50:35<42:04,  2.01it/s]
 53%|█████▎    | 5623/10707 [50:35<42:06,  2.01it/s]
 53%|█████▎    | 5624/10707 [50:36<42:02,  2.02it/s]
 53%|█████▎    | 5625/10707 [50:36<42:01,  2.02it/s]{'loss': 3.7252, 'grad_norm': 0.20586025714874268, 'learning_rate': 0.0005429823993687233, 'epoch': 0.53}
-
                                                    

 53%|█████▎    | 5625/10707 [50:36<42:01,  2.02it/s]
 53%|█████▎    | 5626/10707 [50:37<42:01,  2.01it/s]
 53%|█████▎    | 5627/10707 [50:37<42:03,  2.01it/s]
 53%|█████▎    | 5628/10707 [50:38<42:01,  2.01it/s]
 53%|█████▎    | 5629/10707 [50:38<42:00,  2.01it/s]
 53%|█████▎    | 5630/10707 [50:39<41:59,  2.02it/s]
 53%|█████▎    | 5631/10707 [50:39<41:57,  2.02it/s]
 53%|█████▎    | 5632/10707 [50:39<41:57,  2.02it/s]
 53%|█████▎    | 5633/10707 [50:40<41:56,  2.02it/s]
 53%|█████▎    | 5634/10707 [50:40<41:54,  2.02it/s]
 53%|█████▎    | 5635/10707 [50:41<41:52,  2.02it/s]
 53%|█████▎    | 5636/10707 [50:41<41:53,  2.02it/s]
 53%|█████▎    | 5637/10707 [50:42<41:54,  2.02it/s]
 53%|█████▎    | 5638/10707 [50:42<41:55,  2.01it/s]
 53%|█████▎    | 5639/10707 [50:43<41:53,  2.02it/s]
 53%|█████▎    | 5640/10707 [50:43<41:51,  2.02it/s]
 53%|█████▎    | 5641/10707 [50:44<41:49,  2.02it/s]
 53%|█████▎    | 5642/10707 [50:44<41:48,  2.02it/s]
 53%|█████▎    | 5643/10707 [50:45<41:47,  2.02it/s]
 53%|█████▎    | 5644/10707 [50:45<41:48,  2.02it/s]
 53%|█████▎    | 5645/10707 [50:46<41:47,  2.02it/s]
 53%|█████▎    | 5646/10707 [50:46<41:49,  2.02it/s]
 53%|█████▎    | 5647/10707 [50:47<41:50,  2.02it/s]
 53%|█████▎    | 5648/10707 [50:47<41:47,  2.02it/s]
 53%|█████▎    | 5649/10707 [50:48<41:49,  2.02it/s]
 53%|█████▎    | 5650/10707 [50:48<41:46,  2.02it/s]{'loss': 3.7145, 'grad_norm': 0.222206711769104, 'learning_rate': 0.0005389207698337263, 'epoch': 0.53}
-
                                                    

 53%|█████▎    | 5650/10707 [50:48<41:46,  2.02it/s]
 53%|█████▎    | 5651/10707 [50:49<41:59,  2.01it/s]
 53%|█████▎    | 5652/10707 [50:49<41:56,  2.01it/s]
 53%|█████▎    | 5653/10707 [50:50<41:51,  2.01it/s]
 53%|█████▎    | 5654/10707 [50:50<41:50,  2.01it/s]
 53%|█████▎    | 5655/10707 [50:51<41:48,  2.01it/s]
 53%|█████▎    | 5656/10707 [50:51<41:45,  2.02it/s]
 53%|█████▎    | 5657/10707 [50:52<41:46,  2.02it/s]
 53%|█████▎    | 5658/10707 [50:52<41:43,  2.02it/s]
 53%|█████▎    | 5659/10707 [50:53<41:43,  2.02it/s]
 53%|█████▎    | 5660/10707 [50:53<41:42,  2.02it/s]
 53%|█████▎    | 5661/10707 [50:54<41:41,  2.02it/s]
 53%|█████▎    | 5662/10707 [50:54<41:38,  2.02it/s]
 53%|█████▎    | 5663/10707 [50:55<41:39,  2.02it/s]
 53%|█████▎    | 5664/10707 [50:55<41:35,  2.02it/s]
 53%|█████▎    | 5665/10707 [50:56<41:38,  2.02it/s]
 53%|█████▎    | 5666/10707 [50:56<41:38,  2.02it/s]
 53%|█████▎    | 5667/10707 [50:57<41:37,  2.02it/s]
 53%|█████▎    | 5668/10707 [50:57<41:36,  2.02it/s]
 53%|█████▎    | 5669/10707 [50:58<41:35,  2.02it/s]
 53%|█████▎    | 5670/10707 [50:58<41:35,  2.02it/s]
 53%|█████▎    | 5671/10707 [50:59<41:33,  2.02it/s]
 53%|█████▎    | 5672/10707 [50:59<41:33,  2.02it/s]
 53%|█████▎    | 5673/10707 [51:00<41:30,  2.02it/s]
 53%|█████▎    | 5674/10707 [51:00<41:30,  2.02it/s]
 53%|█████▎    | 5675/10707 [51:01<41:29,  2.02it/s]{'loss': 3.715, 'grad_norm': 0.1892022341489792, 'learning_rate': 0.0005348565546757574, 'epoch': 0.53}
-
                                                    

 53%|█████▎    | 5675/10707 [51:01<41:29,  2.02it/s]
 53%|█████▎    | 5676/10707 [51:01<41:36,  2.02it/s]
 53%|█████▎    | 5677/10707 [51:02<41:32,  2.02it/s]
 53%|█████▎    | 5678/10707 [51:02<41:34,  2.02it/s]
 53%|█████▎    | 5679/10707 [51:03<41:32,  2.02it/s]
 53%|█████▎    | 5680/10707 [51:03<41:47,  2.00it/s]
 53%|█████▎    | 5681/10707 [51:04<41:41,  2.01it/s]
 53%|█████▎    | 5682/10707 [51:04<41:36,  2.01it/s]
 53%|█████▎    | 5683/10707 [51:05<41:31,  2.02it/s]
 53%|█████▎    | 5684/10707 [51:05<41:32,  2.02it/s]
 53%|█████▎    | 5685/10707 [51:06<41:31,  2.02it/s]
 53%|█████▎    | 5686/10707 [51:06<41:30,  2.02it/s]
 53%|█████▎    | 5687/10707 [51:07<41:27,  2.02it/s]
 53%|█████▎    | 5688/10707 [51:07<41:27,  2.02it/s]
 53%|█████▎    | 5689/10707 [51:08<41:25,  2.02it/s]
 53%|█████▎    | 5690/10707 [51:08<41:26,  2.02it/s]
 53%|█████▎    | 5691/10707 [51:09<41:24,  2.02it/s]
 53%|█████▎    | 5692/10707 [51:09<41:25,  2.02it/s]
 53%|█████▎    | 5693/10707 [51:10<41:22,  2.02it/s]
 53%|█████▎    | 5694/10707 [51:10<41:23,  2.02it/s]
 53%|█████▎    | 5695/10707 [51:11<41:22,  2.02it/s]
 53%|█████▎    | 5696/10707 [51:11<41:22,  2.02it/s]
 53%|█████▎    | 5697/10707 [51:12<41:26,  2.01it/s]
 53%|█████▎    | 5698/10707 [51:12<41:25,  2.02it/s]
 53%|█████▎    | 5699/10707 [51:13<41:24,  2.02it/s]
 53%|█████▎    | 5700/10707 [51:13<41:22,  2.02it/s]{'loss': 3.7211, 'grad_norm': 0.2014087289571762, 'learning_rate': 0.0005307900238927668, 'epoch': 0.53}
                                                    
-
 53%|█████▎    | 5700/10707 [51:13<41:22,  2.02it/s]
 53%|█████▎    | 5701/10707 [51:14<41:25,  2.01it/s]
 53%|█████▎    | 5702/10707 [51:14<41:24,  2.01it/s]
 53%|█████▎    | 5703/10707 [51:15<41:22,  2.02it/s]
 53%|█████▎    | 5704/10707 [51:15<41:20,  2.02it/s]
 53%|█████▎    | 5705/10707 [51:16<41:19,  2.02it/s]
 53%|█████▎    | 5706/10707 [51:16<41:16,  2.02it/s]
 53%|█████▎    | 5707/10707 [51:17<41:17,  2.02it/s]
 53%|█████▎    | 5708/10707 [51:17<41:14,  2.02it/s]
 53%|█████▎    | 5709/10707 [51:18<41:14,  2.02it/s]
 53%|█████▎    | 5710/10707 [51:18<41:12,  2.02it/s]
 53%|█████▎    | 5711/10707 [51:19<41:14,  2.02it/s]
 53%|█████▎    | 5712/10707 [51:19<41:12,  2.02it/s]
 53%|█████▎    | 5713/10707 [51:20<41:15,  2.02it/s]
 53%|█████▎    | 5714/10707 [51:20<41:13,  2.02it/s]
 53%|█████▎    | 5715/10707 [51:21<41:13,  2.02it/s]
 53%|█████▎    | 5716/10707 [51:21<41:11,  2.02it/s]
 53%|█████▎    | 5717/10707 [51:22<41:11,  2.02it/s]
 53%|█████▎    | 5718/10707 [51:22<41:11,  2.02it/s]
 53%|█████▎    | 5719/10707 [51:23<41:09,  2.02it/s]
 53%|█████▎    | 5720/10707 [51:23<41:08,  2.02it/s]
 53%|█████▎    | 5721/10707 [51:24<41:06,  2.02it/s]
 53%|█████▎    | 5722/10707 [51:24<41:05,  2.02it/s]
 53%|█████▎    | 5723/10707 [51:25<41:05,  2.02it/s]
 53%|█████▎    | 5724/10707 [51:25<41:04,  2.02it/s]
 53%|█████▎    | 5725/10707 [51:26<41:05,  2.02it/s]{'loss': 3.7178, 'grad_norm': 0.1980610191822052, 'learning_rate': 0.0005267214476365386, 'epoch': 0.53}
-
                                                    

 53%|█████▎    | 5725/10707 [51:26<41:05,  2.02it/s]
 53%|█████▎    | 5726/10707 [51:26<41:07,  2.02it/s]
 53%|█████▎    | 5727/10707 [51:27<41:07,  2.02it/s]
 53%|█████▎    | 5728/10707 [51:27<41:09,  2.02it/s]
 54%|█████▎    | 5729/10707 [51:28<41:06,  2.02it/s]
 54%|█████▎    | 5730/10707 [51:28<41:05,  2.02it/s]
 54%|█████▎    | 5731/10707 [51:29<41:03,  2.02it/s]
 54%|█████▎    | 5732/10707 [51:29<41:05,  2.02it/s]
 54%|█████▎    | 5733/10707 [51:30<41:04,  2.02it/s]
 54%|█████▎    | 5734/10707 [51:30<41:05,  2.02it/s]
 54%|█████▎    | 5735/10707 [51:31<41:01,  2.02it/s]
 54%|█████▎    | 5736/10707 [51:31<41:02,  2.02it/s]
 54%|█████▎    | 5737/10707 [51:32<41:00,  2.02it/s]
 54%|█████▎    | 5738/10707 [51:32<41:01,  2.02it/s]
 54%|█████▎    | 5739/10707 [51:33<41:00,  2.02it/s]
 54%|█████▎    | 5740/10707 [51:33<41:03,  2.02it/s]
 54%|█████▎    | 5741/10707 [51:34<41:12,  2.01it/s]
 54%|█████▎    | 5742/10707 [51:34<41:07,  2.01it/s]
 54%|█████▎    | 5743/10707 [51:35<41:06,  2.01it/s]
 54%|█████▎    | 5744/10707 [51:35<41:01,  2.02it/s]
 54%|█████▎    | 5745/10707 [51:36<40:59,  2.02it/s]
 54%|█████▎    | 5746/10707 [51:36<40:54,  2.02it/s]
 54%|█████▎    | 5747/10707 [51:36<40:56,  2.02it/s]
 54%|█████▎    | 5748/10707 [51:37<40:55,  2.02it/s]
 54%|█████▎    | 5749/10707 [51:37<40:57,  2.02it/s]
 54%|█████▎    | 5750/10707 [51:38<40:57,  2.02it/s]{'loss': 3.7145, 'grad_norm': 0.19791434705257416, 'learning_rate': 0.0005226510961947439, 'epoch': 0.54}
-
                                                    

 54%|█████▎    | 5750/10707 [51:38<40:57,  2.02it/s]
 54%|█████▎    | 5751/10707 [51:38<41:02,  2.01it/s]
 54%|█████▎    | 5752/10707 [51:39<41:03,  2.01it/s]
 54%|█████▎    | 5753/10707 [51:39<41:00,  2.01it/s]
 54%|█████▎    | 5754/10707 [51:40<40:59,  2.01it/s]
 54%|█████▎    | 5755/10707 [51:40<40:57,  2.02it/s]
 54%|█████▍    | 5756/10707 [51:41<40:51,  2.02it/s]
 54%|█████▍    | 5757/10707 [51:41<40:53,  2.02it/s]
 54%|█████▍    | 5758/10707 [51:42<40:53,  2.02it/s]
 54%|█████▍    | 5759/10707 [51:42<40:55,  2.02it/s]
 54%|█████▍    | 5760/10707 [51:43<40:51,  2.02it/s]
 54%|█████▍    | 5761/10707 [51:43<40:52,  2.02it/s]
 54%|█████▍    | 5762/10707 [51:44<40:48,  2.02it/s]
 54%|█████▍    | 5763/10707 [51:44<40:51,  2.02it/s]
 54%|█████▍    | 5764/10707 [51:45<40:47,  2.02it/s]
 54%|█████▍    | 5765/10707 [51:45<40:47,  2.02it/s]
 54%|█████▍    | 5766/10707 [51:46<40:48,  2.02it/s]
 54%|█████▍    | 5767/10707 [51:46<40:46,  2.02it/s]
 54%|█████▍    | 5768/10707 [51:47<40:45,  2.02it/s]
 54%|█████▍    | 5769/10707 [51:47<40:44,  2.02it/s]
 54%|█████▍    | 5770/10707 [51:48<40:44,  2.02it/s]
 54%|█████▍    | 5771/10707 [51:48<40:46,  2.02it/s]
 54%|█████▍    | 5772/10707 [51:49<40:45,  2.02it/s]
 54%|█████▍    | 5773/10707 [51:49<40:43,  2.02it/s]
 54%|█████▍    | 5774/10707 [51:50<40:42,  2.02it/s]
 54%|█████▍    | 5775/10707 [51:50<40:40,  2.02it/s]{'loss': 3.7089, 'grad_norm': 0.21259498596191406, 'learning_rate': 0.0005185792399729842, 'epoch': 0.54}
-
                                                    

 54%|█████▍    | 5775/10707 [51:50<40:40,  2.02it/s]
 54%|█████▍    | 5776/10707 [51:51<40:47,  2.02it/s]
 54%|█████▍    | 5777/10707 [51:51<40:44,  2.02it/s]
 54%|█████▍    | 5778/10707 [51:52<40:44,  2.02it/s]
 54%|█████▍    | 5779/10707 [51:52<40:42,  2.02it/s]
 54%|█████▍    | 5780/10707 [51:53<40:41,  2.02it/s]
 54%|█████▍    | 5781/10707 [51:53<40:45,  2.01it/s]
 54%|█████▍    | 5782/10707 [51:54<40:41,  2.02it/s]
 54%|█████▍    | 5783/10707 [51:54<40:40,  2.02it/s]
 54%|█████▍    | 5784/10707 [51:55<40:38,  2.02it/s]
 54%|█████▍    | 5785/10707 [51:55<40:38,  2.02it/s]
 54%|█████▍    | 5786/10707 [51:56<40:36,  2.02it/s]
 54%|█████▍    | 5787/10707 [51:56<40:37,  2.02it/s]
 54%|█████▍    | 5788/10707 [51:57<40:35,  2.02it/s]
 54%|█████▍    | 5789/10707 [51:57<40:35,  2.02it/s]
 54%|█████▍    | 5790/10707 [51:58<40:35,  2.02it/s]
 54%|█████▍    | 5791/10707 [51:58<40:33,  2.02it/s]
 54%|█████▍    | 5792/10707 [51:59<40:33,  2.02it/s]
 54%|█████▍    | 5793/10707 [51:59<40:32,  2.02it/s]
 54%|█████▍    | 5794/10707 [52:00<40:32,  2.02it/s]
 54%|█████▍    | 5795/10707 [52:00<40:30,  2.02it/s]
 54%|█████▍    | 5796/10707 [52:01<40:30,  2.02it/s]
 54%|█████▍    | 5797/10707 [52:01<40:31,  2.02it/s]
 54%|█████▍    | 5798/10707 [52:02<40:32,  2.02it/s]
 54%|█████▍    | 5799/10707 [52:02<40:29,  2.02it/s]
 54%|█████▍    | 5800/10707 [52:03<40:31,  2.02it/s]
                                                    
{'loss': 3.6984, 'grad_norm': 0.19928260147571564, 'learning_rate': 0.0005145061494768287, 'epoch': 0.54}
-
 54%|█████▍    | 5800/10707 [52:03<40:31,  2.02it/s]
 54%|█████▍    | 5801/10707 [52:03<40:39,  2.01it/s]
 54%|█████▍    | 5802/10707 [52:04<40:52,  2.00it/s]
 54%|█████▍    | 5803/10707 [52:04<40:44,  2.01it/s]
 54%|█████▍    | 5804/10707 [52:05<40:42,  2.01it/s]
 54%|█████▍    | 5805/10707 [52:05<40:36,  2.01it/s]
 54%|█████▍    | 5806/10707 [52:06<40:35,  2.01it/s]
 54%|█████▍    | 5807/10707 [52:06<40:33,  2.01it/s]
 54%|█████▍    | 5808/10707 [52:07<40:30,  2.02it/s]
 54%|█████▍    | 5809/10707 [52:07<40:30,  2.02it/s]
 54%|█████▍    | 5810/10707 [52:08<40:26,  2.02it/s]
 54%|█████▍    | 5811/10707 [52:08<40:26,  2.02it/s]
 54%|█████▍    | 5812/10707 [52:09<40:24,  2.02it/s]
 54%|█████▍    | 5813/10707 [52:09<40:25,  2.02it/s]
 54%|█████▍    | 5814/10707 [52:10<40:22,  2.02it/s]
 54%|█████▍    | 5815/10707 [52:10<40:21,  2.02it/s]
 54%|█████▍    | 5816/10707 [52:11<40:21,  2.02it/s]
 54%|█████▍    | 5817/10707 [52:11<40:21,  2.02it/s]
 54%|█████▍    | 5818/10707 [52:12<40:19,  2.02it/s]
 54%|█████▍    | 5819/10707 [52:12<40:27,  2.01it/s]
 54%|█████▍    | 5820/10707 [52:13<40:28,  2.01it/s]
 54%|█████▍    | 5821/10707 [52:13<40:25,  2.01it/s]
 54%|█████▍    | 5822/10707 [52:14<40:24,  2.01it/s]
 54%|█████▍    | 5823/10707 [52:14<40:26,  2.01it/s]
 54%|█████▍    | 5824/10707 [52:15<40:26,  2.01it/s]
 54%|█████▍    | 5825/10707 [52:15<40:23,  2.01it/s]{'loss': 3.7166, 'grad_norm': 0.20241673290729523, 'learning_rate': 0.0005104320952938426, 'epoch': 0.54}
                                                    
-
 54%|█████▍    | 5825/10707 [52:15<40:23,  2.01it/s]
 54%|█████▍    | 5826/10707 [52:16<40:25,  2.01it/s]
 54%|█████▍    | 5827/10707 [52:16<40:24,  2.01it/s]
 54%|█████▍    | 5828/10707 [52:17<40:25,  2.01it/s]
 54%|█████▍    | 5829/10707 [52:17<40:23,  2.01it/s]
 54%|█████▍    | 5830/10707 [52:18<40:20,  2.01it/s]
 54%|█████▍    | 5831/10707 [52:18<40:18,  2.02it/s]
 54%|█████▍    | 5832/10707 [52:19<40:16,  2.02it/s]
 54%|█████▍    | 5833/10707 [52:19<40:17,  2.02it/s]
 54%|█████▍    | 5834/10707 [52:20<40:16,  2.02it/s]
 54%|█████▍    | 5835/10707 [52:20<40:16,  2.02it/s]
 55%|█████▍    | 5836/10707 [52:21<40:11,  2.02it/s]
 55%|█████▍    | 5837/10707 [52:21<40:12,  2.02it/s]
 55%|█████▍    | 5838/10707 [52:22<40:11,  2.02it/s]
 55%|█████▍    | 5839/10707 [52:22<40:12,  2.02it/s]
 55%|█████▍    | 5840/10707 [52:23<40:12,  2.02it/s]
 55%|█████▍    | 5841/10707 [52:23<40:11,  2.02it/s]
 55%|█████▍    | 5842/10707 [52:24<40:07,  2.02it/s]
 55%|█████▍    | 5843/10707 [52:24<40:08,  2.02it/s]
 55%|█████▍    | 5844/10707 [52:25<40:07,  2.02it/s]
 55%|█████▍    | 5845/10707 [52:25<40:10,  2.02it/s]
 55%|█████▍    | 5846/10707 [52:26<40:09,  2.02it/s]
 55%|█████▍    | 5847/10707 [52:26<40:09,  2.02it/s]
 55%|█████▍    | 5848/10707 [52:27<40:11,  2.01it/s]
 55%|█████▍    | 5849/10707 [52:27<40:08,  2.02it/s]
 55%|█████▍    | 5850/10707 [52:28<40:09,  2.02it/s]
                                                    
{'loss': 3.7169, 'grad_norm': 0.2038862258195877, 'learning_rate': 0.0005063573480756123, 'epoch': 0.55}
-
 55%|█████▍    | 5850/10707 [52:28<40:09,  2.02it/s]
 55%|█████▍    | 5851/10707 [52:28<40:11,  2.01it/s]
 55%|█████▍    | 5852/10707 [52:29<40:08,  2.02it/s]
 55%|█████▍    | 5853/10707 [52:29<40:07,  2.02it/s]
 55%|█████▍    | 5854/10707 [52:30<40:08,  2.02it/s]
 55%|█████▍    | 5855/10707 [52:30<40:07,  2.02it/s]
 55%|█████▍    | 5856/10707 [52:31<40:03,  2.02it/s]
 55%|█████▍    | 5857/10707 [52:31<40:04,  2.02it/s]
 55%|█████▍    | 5858/10707 [52:32<40:01,  2.02it/s]
 55%|█████▍    | 5859/10707 [52:32<40:02,  2.02it/s]
 55%|█████▍    | 5860/10707 [52:33<40:01,  2.02it/s]
 55%|█████▍    | 5861/10707 [52:33<40:06,  2.01it/s]
 55%|█████▍    | 5862/10707 [52:34<40:12,  2.01it/s]
 55%|█████▍    | 5863/10707 [52:34<40:06,  2.01it/s]
 55%|█████▍    | 5864/10707 [52:35<40:04,  2.01it/s]
 55%|█████▍    | 5865/10707 [52:35<40:02,  2.02it/s]
 55%|█████▍    | 5866/10707 [52:36<40:01,  2.02it/s]
 55%|█████▍    | 5867/10707 [52:36<40:00,  2.02it/s]
 55%|█████▍    | 5868/10707 [52:37<39:59,  2.02it/s]
 55%|█████▍    | 5869/10707 [52:37<39:59,  2.02it/s]
 55%|█████▍    | 5870/10707 [52:37<39:57,  2.02it/s]
 55%|█████▍    | 5871/10707 [52:38<39:58,  2.02it/s]
 55%|█████▍    | 5872/10707 [52:38<39:56,  2.02it/s]
 55%|█████▍    | 5873/10707 [52:39<39:55,  2.02it/s]
 55%|█████▍    | 5874/10707 [52:39<39:52,  2.02it/s]
 55%|█████▍    | 5875/10707 [52:40<39:53,  2.02it/s]{'loss': 3.7096, 'grad_norm': 0.19629842042922974, 'learning_rate': 0.0005022821785197638, 'epoch': 0.55}
                                                    
-
 55%|█████▍    | 5875/10707 [52:40<39:53,  2.02it/s]
 55%|█████▍    | 5876/10707 [52:40<39:56,  2.02it/s]
 55%|█████▍    | 5877/10707 [52:41<39:57,  2.02it/s]
 55%|█████▍    | 5878/10707 [52:41<39:53,  2.02it/s]
 55%|█████▍    | 5879/10707 [52:42<39:52,  2.02it/s]
 55%|█████▍    | 5880/10707 [52:42<39:51,  2.02it/s]
 55%|█████▍    | 5881/10707 [52:43<39:51,  2.02it/s]
 55%|█████▍    | 5882/10707 [52:43<39:54,  2.02it/s]
 55%|█████▍    | 5883/10707 [52:44<39:51,  2.02it/s]
 55%|█████▍    | 5884/10707 [52:44<39:52,  2.02it/s]
 55%|█████▍    | 5885/10707 [52:45<39:49,  2.02it/s]
 55%|█████▍    | 5886/10707 [52:45<39:49,  2.02it/s]
 55%|█████▍    | 5887/10707 [52:46<39:50,  2.02it/s]
 55%|█████▍    | 5888/10707 [52:46<39:49,  2.02it/s]
 55%|█████▌    | 5889/10707 [52:47<39:46,  2.02it/s]
 55%|█████▌    | 5890/10707 [52:47<39:45,  2.02it/s]
 55%|█████▌    | 5891/10707 [52:48<39:46,  2.02it/s]
 55%|█████▌    | 5892/10707 [52:48<39:45,  2.02it/s]
 55%|█████▌    | 5893/10707 [52:49<39:47,  2.02it/s]
 55%|█████▌    | 5894/10707 [52:49<39:43,  2.02it/s]
 55%|█████▌    | 5895/10707 [52:50<39:43,  2.02it/s]
 55%|█████▌    | 5896/10707 [52:50<39:41,  2.02it/s]
 55%|█████▌    | 5897/10707 [52:51<39:43,  2.02it/s]
 55%|█████▌    | 5898/10707 [52:51<39:40,  2.02it/s]
 55%|█████▌    | 5899/10707 [52:52<39:41,  2.02it/s]
 55%|█████▌    | 5900/10707 [52:52<39:40,  2.02it/s]
                                                    
{'loss': 3.7141, 'grad_norm': 0.2124280482530594, 'learning_rate': 0.0004982068573519806, 'epoch': 0.55}
-
 55%|█████▌    | 5900/10707 [52:52<39:40,  2.02it/s]
 55%|█████▌    | 5901/10707 [52:53<39:44,  2.02it/s]
 55%|█████▌    | 5902/10707 [52:53<39:41,  2.02it/s]
 55%|█████▌    | 5903/10707 [52:54<39:41,  2.02it/s]
 55%|█████▌    | 5904/10707 [52:54<39:39,  2.02it/s]
 55%|█████▌    | 5905/10707 [52:55<39:37,  2.02it/s]
 55%|█████▌    | 5906/10707 [52:55<39:36,  2.02it/s]
 55%|█████▌    | 5907/10707 [52:56<39:37,  2.02it/s]
 55%|█████▌    | 5908/10707 [52:56<39:37,  2.02it/s]
 55%|█████▌    | 5909/10707 [52:57<39:35,  2.02it/s]
 55%|█████▌    | 5910/10707 [52:57<39:35,  2.02it/s]
 55%|█████▌    | 5911/10707 [52:58<39:33,  2.02it/s]
 55%|█████▌    | 5912/10707 [52:58<39:35,  2.02it/s]
 55%|█████▌    | 5913/10707 [52:59<39:31,  2.02it/s]
 55%|█████▌    | 5914/10707 [52:59<39:33,  2.02it/s]
 55%|█████▌    | 5915/10707 [53:00<39:31,  2.02it/s]
 55%|█████▌    | 5916/10707 [53:00<39:34,  2.02it/s]
 55%|█████▌    | 5917/10707 [53:01<39:30,  2.02it/s]
 55%|█████▌    | 5918/10707 [53:01<39:32,  2.02it/s]
 55%|█████▌    | 5919/10707 [53:02<39:30,  2.02it/s]
 55%|█████▌    | 5920/10707 [53:02<39:32,  2.02it/s]
 55%|█████▌    | 5921/10707 [53:03<39:35,  2.01it/s]
 55%|█████▌    | 5922/10707 [53:03<39:39,  2.01it/s]
 55%|█████▌    | 5923/10707 [53:04<39:39,  2.01it/s]
 55%|█████▌    | 5924/10707 [53:04<39:37,  2.01it/s]
 55%|█████▌    | 5925/10707 [53:05<39:32,  2.02it/s]{'loss': 3.7021, 'grad_norm': 0.20711711049079895, 'learning_rate': 0.0004941316553080188, 'epoch': 0.55}
-
                                                    

 55%|█████▌    | 5925/10707 [53:05<39:32,  2.02it/s]
 55%|█████▌    | 5926/10707 [53:05<39:36,  2.01it/s]
 55%|█████▌    | 5927/10707 [53:06<39:33,  2.01it/s]
 55%|█████▌    | 5928/10707 [53:06<39:32,  2.01it/s]
 55%|█████▌    | 5929/10707 [53:07<39:28,  2.02it/s]
 55%|█████▌    | 5930/10707 [53:07<39:28,  2.02it/s]
 55%|█████▌    | 5931/10707 [53:08<39:27,  2.02it/s]
 55%|█████▌    | 5932/10707 [53:08<39:26,  2.02it/s]
 55%|█████▌    | 5933/10707 [53:09<39:27,  2.02it/s]
 55%|█████▌    | 5934/10707 [53:09<39:23,  2.02it/s]
 55%|█████▌    | 5935/10707 [53:10<39:25,  2.02it/s]
 55%|█████▌    | 5936/10707 [53:10<39:23,  2.02it/s]
 55%|█████▌    | 5937/10707 [53:11<39:23,  2.02it/s]
 55%|█████▌    | 5938/10707 [53:11<39:21,  2.02it/s]
 55%|█████▌    | 5939/10707 [53:12<39:22,  2.02it/s]
 55%|█████▌    | 5940/10707 [53:12<39:19,  2.02it/s]
 55%|█████▌    | 5941/10707 [53:13<39:21,  2.02it/s]
 55%|█████▌    | 5942/10707 [53:13<39:16,  2.02it/s]
 56%|█████▌    | 5943/10707 [53:14<39:18,  2.02it/s]
 56%|█████▌    | 5944/10707 [53:14<39:15,  2.02it/s]
 56%|█████▌    | 5945/10707 [53:15<39:17,  2.02it/s]
 56%|█████▌    | 5946/10707 [53:15<39:14,  2.02it/s]
 56%|█████▌    | 5947/10707 [53:16<39:16,  2.02it/s]
 56%|█████▌    | 5948/10707 [53:16<39:14,  2.02it/s]
 56%|█████▌    | 5949/10707 [53:17<39:17,  2.02it/s]
 56%|█████▌    | 5950/10707 [53:17<39:15,  2.02it/s]{'loss': 3.7013, 'grad_norm': 0.19911937415599823, 'learning_rate': 0.0004900568431157198, 'epoch': 0.56}
-
                                                    

 56%|█████▌    | 5950/10707 [53:17<39:15,  2.02it/s]
 56%|█████▌    | 5951/10707 [53:18<39:19,  2.02it/s]
 56%|█████▌    | 5952/10707 [53:18<39:21,  2.01it/s]
 56%|█████▌    | 5953/10707 [53:19<39:17,  2.02it/s]
 56%|█████▌    | 5954/10707 [53:19<39:16,  2.02it/s]
 56%|█████▌    | 5955/10707 [53:20<39:14,  2.02it/s]
 56%|█████▌    | 5956/10707 [53:20<39:12,  2.02it/s]
 56%|█████▌    | 5957/10707 [53:21<39:10,  2.02it/s]
 56%|█████▌    | 5958/10707 [53:21<39:11,  2.02it/s]
 56%|█████▌    | 5959/10707 [53:22<39:10,  2.02it/s]
 56%|█████▌    | 5960/10707 [53:22<39:09,  2.02it/s]
 56%|█████▌    | 5961/10707 [53:23<39:07,  2.02it/s]
 56%|█████▌    | 5962/10707 [53:23<39:08,  2.02it/s]
 56%|█████▌    | 5963/10707 [53:24<39:07,  2.02it/s]
 56%|█████▌    | 5964/10707 [53:24<39:08,  2.02it/s]
 56%|█████▌    | 5965/10707 [53:25<39:09,  2.02it/s]
 56%|█████▌    | 5966/10707 [53:25<39:09,  2.02it/s]
 56%|█████▌    | 5967/10707 [53:26<39:08,  2.02it/s]
 56%|█████▌    | 5968/10707 [53:26<39:07,  2.02it/s]
 56%|█████▌    | 5969/10707 [53:27<39:07,  2.02it/s]
 56%|█████▌    | 5970/10707 [53:27<39:06,  2.02it/s]
 56%|█████▌    | 5971/10707 [53:28<39:06,  2.02it/s]
 56%|█████▌    | 5972/10707 [53:28<39:07,  2.02it/s]
 56%|█████▌    | 5973/10707 [53:29<39:05,  2.02it/s]
 56%|█████▌    | 5974/10707 [53:29<39:06,  2.02it/s]
 56%|█████▌    | 5975/10707 [53:30<39:05,  2.02it/s]{'loss': 3.7016, 'grad_norm': 0.2025490701198578, 'learning_rate': 0.0004859826914770267, 'epoch': 0.56}
-
                                                    

 56%|█████▌    | 5975/10707 [53:30<39:05,  2.02it/s]
 56%|█████▌    | 5976/10707 [53:30<39:07,  2.02it/s]
 56%|█████▌    | 5977/10707 [53:31<39:05,  2.02it/s]
 56%|█████▌    | 5978/10707 [53:31<39:03,  2.02it/s]
 56%|█████▌    | 5979/10707 [53:32<39:03,  2.02it/s]
 56%|█████▌    | 5980/10707 [53:32<39:00,  2.02it/s]
 56%|█████▌    | 5981/10707 [53:32<39:00,  2.02it/s]
 56%|█████▌    | 5982/10707 [53:33<39:20,  2.00it/s]
 56%|█████▌    | 5983/10707 [53:34<39:19,  2.00it/s]
 56%|█████▌    | 5984/10707 [53:34<39:11,  2.01it/s]
 56%|█████▌    | 5985/10707 [53:34<39:09,  2.01it/s]
 56%|█████▌    | 5986/10707 [53:35<39:04,  2.01it/s]
 56%|█████▌    | 5987/10707 [53:35<39:02,  2.02it/s]
 56%|█████▌    | 5988/10707 [53:36<38:58,  2.02it/s]
 56%|█████▌    | 5989/10707 [53:36<38:59,  2.02it/s]
 56%|█████▌    | 5990/10707 [53:37<38:56,  2.02it/s]
 56%|█████▌    | 5991/10707 [53:37<38:57,  2.02it/s]
 56%|█████▌    | 5992/10707 [53:38<38:55,  2.02it/s]
 56%|█████▌    | 5993/10707 [53:38<38:57,  2.02it/s]
 56%|█████▌    | 5994/10707 [53:39<38:54,  2.02it/s]
 56%|█████▌    | 5995/10707 [53:39<38:57,  2.02it/s]
 56%|█████▌    | 5996/10707 [53:40<38:57,  2.02it/s]
 56%|█████▌    | 5997/10707 [53:40<38:54,  2.02it/s]
 56%|█████▌    | 5998/10707 [53:41<38:55,  2.02it/s]
 56%|█████▌    | 5999/10707 [53:41<38:51,  2.02it/s]
 56%|█████▌    | 6000/10707 [53:42<38:52,  2.02it/s]{'loss': 3.6961, 'grad_norm': 0.1965174674987793, 'learning_rate': 0.00048190947104999934, 'epoch': 0.56}
-
                                                    

 56%|█████▌    | 6000/10707 [53:42<38:52,  2.02it/s]
 56%|█████▌    | 6001/10707 [53:42<38:53,  2.02it/s]
 56%|█████▌    | 6002/10707 [53:43<38:54,  2.02it/s]
 56%|█████▌    | 6003/10707 [53:43<38:53,  2.02it/s]
 56%|█████▌    | 6004/10707 [53:44<38:52,  2.02it/s]
 56%|█████▌    | 6005/10707 [53:44<38:51,  2.02it/s]
 56%|█████▌    | 6006/10707 [53:45<38:50,  2.02it/s]
 56%|█████▌    | 6007/10707 [53:45<38:50,  2.02it/s]
 56%|█████▌    | 6008/10707 [53:46<38:49,  2.02it/s]
 56%|█████▌    | 6009/10707 [53:46<38:50,  2.02it/s]
 56%|█████▌    | 6010/10707 [53:47<38:47,  2.02it/s]
 56%|█████▌    | 6011/10707 [53:47<38:48,  2.02it/s]
 56%|█████▌    | 6012/10707 [53:48<38:44,  2.02it/s]
 56%|█████▌    | 6013/10707 [53:48<38:45,  2.02it/s]
 56%|█████▌    | 6014/10707 [53:49<38:44,  2.02it/s]
 56%|█████▌    | 6015/10707 [53:49<38:45,  2.02it/s]
 56%|█████▌    | 6016/10707 [53:50<38:41,  2.02it/s]
 56%|█████▌    | 6017/10707 [53:50<38:43,  2.02it/s]
 56%|█████▌    | 6018/10707 [53:51<38:40,  2.02it/s]
 56%|█████▌    | 6019/10707 [53:51<38:43,  2.02it/s]
 56%|█████▌    | 6020/10707 [53:52<38:40,  2.02it/s]
 56%|█████▌    | 6021/10707 [53:52<38:42,  2.02it/s]
 56%|█████▌    | 6022/10707 [53:53<38:39,  2.02it/s]
 56%|█████▋    | 6023/10707 [53:53<38:39,  2.02it/s]
 56%|█████▋    | 6024/10707 [53:54<38:39,  2.02it/s]
 56%|█████▋    | 6025/10707 [53:54<38:40,  2.02it/s]{'loss': 3.6892, 'grad_norm': 0.20315933227539062, 'learning_rate': 0.0004778374524308354, 'epoch': 0.56}
                                                    
-
 56%|█████▋    | 6025/10707 [53:54<38:40,  2.02it/s]
 56%|█████▋    | 6026/10707 [53:55<38:43,  2.01it/s]
 56%|█████▋    | 6027/10707 [53:55<38:42,  2.02it/s]
 56%|█████▋    | 6028/10707 [53:56<38:39,  2.02it/s]
 56%|█████▋    | 6029/10707 [53:56<38:40,  2.02it/s]
 56%|█████▋    | 6030/10707 [53:57<38:38,  2.02it/s]
 56%|█████▋    | 6031/10707 [53:57<38:38,  2.02it/s]
 56%|█████▋    | 6032/10707 [53:58<38:36,  2.02it/s]
 56%|█████▋    | 6033/10707 [53:58<38:38,  2.02it/s]
 56%|█████▋    | 6034/10707 [53:59<38:38,  2.02it/s]
 56%|█████▋    | 6035/10707 [53:59<38:37,  2.02it/s]
 56%|█████▋    | 6036/10707 [54:00<38:32,  2.02it/s]
 56%|█████▋    | 6037/10707 [54:00<38:34,  2.02it/s]
 56%|█████▋    | 6038/10707 [54:01<38:31,  2.02it/s]
 56%|█████▋    | 6039/10707 [54:01<38:32,  2.02it/s]
 56%|█████▋    | 6040/10707 [54:02<38:31,  2.02it/s]
 56%|█████▋    | 6041/10707 [54:02<38:31,  2.02it/s]
 56%|█████▋    | 6042/10707 [54:03<38:30,  2.02it/s]
 56%|█████▋    | 6043/10707 [54:03<38:35,  2.01it/s]
 56%|█████▋    | 6044/10707 [54:04<38:44,  2.01it/s]
 56%|█████▋    | 6045/10707 [54:04<38:42,  2.01it/s]
 56%|█████▋    | 6046/10707 [54:05<38:36,  2.01it/s]
 56%|█████▋    | 6047/10707 [54:05<38:35,  2.01it/s]
 56%|█████▋    | 6048/10707 [54:06<38:28,  2.02it/s]
 56%|█████▋    | 6049/10707 [54:06<38:30,  2.02it/s]
 57%|█████▋    | 6050/10707 [54:07<38:26,  2.02it/s]{'loss': 3.6969, 'grad_norm': 0.21122252941131592, 'learning_rate': 0.0004737669061358921, 'epoch': 0.57}
-
                                                    

 57%|█████▋    | 6050/10707 [54:07<38:26,  2.02it/s]
 57%|█████▋    | 6051/10707 [54:07<38:29,  2.02it/s]
 57%|█████▋    | 6052/10707 [54:08<38:27,  2.02it/s]
 57%|█████▋    | 6053/10707 [54:08<38:26,  2.02it/s]
 57%|█████▋    | 6054/10707 [54:09<38:26,  2.02it/s]
 57%|█████▋    | 6055/10707 [54:09<38:24,  2.02it/s]
 57%|█████▋    | 6056/10707 [54:10<38:26,  2.02it/s]
 57%|█████▋    | 6057/10707 [54:10<38:25,  2.02it/s]
 57%|█████▋    | 6058/10707 [54:11<38:25,  2.02it/s]
 57%|█████▋    | 6059/10707 [54:11<38:23,  2.02it/s]
 57%|█████▋    | 6060/10707 [54:12<38:24,  2.02it/s]
 57%|█████▋    | 6061/10707 [54:12<38:23,  2.02it/s]
 57%|█████▋    | 6062/10707 [54:13<38:21,  2.02it/s]
 57%|█████▋    | 6063/10707 [54:13<38:23,  2.02it/s]
 57%|█████▋    | 6064/10707 [54:14<38:21,  2.02it/s]
 57%|█████▋    | 6065/10707 [54:14<38:21,  2.02it/s]
 57%|█████▋    | 6066/10707 [54:15<38:19,  2.02it/s]
 57%|█████▋    | 6067/10707 [54:15<44:39,  1.73it/s]
 57%|█████▋    | 6068/10707 [54:16<42:44,  1.81it/s]
 57%|█████▋    | 6069/10707 [54:16<41:22,  1.87it/s]
 57%|█████▋    | 6070/10707 [54:17<40:27,  1.91it/s]
 57%|█████▋    | 6071/10707 [54:17<39:50,  1.94it/s]
 57%|█████▋    | 6072/10707 [54:18<39:19,  1.96it/s]
 57%|█████▋    | 6073/10707 [54:18<39:00,  1.98it/s]
 57%|█████▋    | 6074/10707 [54:19<38:45,  1.99it/s]
 57%|█████▋    | 6075/10707 [54:19<38:37,  2.00it/s]{'loss': 3.6988, 'grad_norm': 0.19844235479831696, 'learning_rate': 0.00046969810258371594, 'epoch': 0.57}
-
                                                    

 57%|█████▋    | 6075/10707 [54:19<38:37,  2.00it/s]
 57%|█████▋    | 6076/10707 [54:20<38:30,  2.00it/s]
 57%|█████▋    | 6077/10707 [54:20<38:27,  2.01it/s]
 57%|█████▋    | 6078/10707 [54:21<38:21,  2.01it/s]
 57%|█████▋    | 6079/10707 [54:21<38:18,  2.01it/s]
 57%|█████▋    | 6080/10707 [54:22<38:15,  2.02it/s]
 57%|█████▋    | 6081/10707 [54:22<38:12,  2.02it/s]
 57%|█████▋    | 6082/10707 [54:23<38:12,  2.02it/s]
 57%|█████▋    | 6083/10707 [54:23<38:12,  2.02it/s]
 57%|█████▋    | 6084/10707 [54:24<38:08,  2.02it/s]
 57%|█████▋    | 6085/10707 [54:24<38:09,  2.02it/s]
 57%|█████▋    | 6086/10707 [54:25<38:09,  2.02it/s]
 57%|█████▋    | 6087/10707 [54:25<38:06,  2.02it/s]
 57%|█████▋    | 6088/10707 [54:26<38:06,  2.02it/s]
 57%|█████▋    | 6089/10707 [54:26<38:07,  2.02it/s]
 57%|█████▋    | 6090/10707 [54:27<38:11,  2.02it/s]
 57%|█████▋    | 6091/10707 [54:27<38:08,  2.02it/s]
 57%|█████▋    | 6092/10707 [54:28<44:06,  1.74it/s]
 57%|█████▋    | 6093/10707 [54:29<42:16,  1.82it/s]
 57%|█████▋    | 6094/10707 [54:29<41:03,  1.87it/s]
 57%|█████▋    | 6095/10707 [54:30<40:05,  1.92it/s]
 57%|█████▋    | 6096/10707 [54:30<39:28,  1.95it/s]
 57%|█████▋    | 6097/10707 [54:31<39:03,  1.97it/s]
 57%|█████▋    | 6098/10707 [54:31<38:45,  1.98it/s]
 57%|█████▋    | 6099/10707 [54:32<38:33,  1.99it/s]
 57%|█████▋    | 6100/10707 [54:32<38:24,  2.00it/s]{'loss': 3.6933, 'grad_norm': 0.1913803219795227, 'learning_rate': 0.0004656313120770776, 'epoch': 0.57}
-
                                                    

 57%|█████▋    | 6100/10707 [54:32<38:24,  2.00it/s]
 57%|█████▋    | 6101/10707 [54:33<38:19,  2.00it/s]
 57%|█████▋    | 6102/10707 [54:33<38:18,  2.00it/s]
 57%|█████▋    | 6103/10707 [54:34<38:17,  2.00it/s]
 57%|█████▋    | 6104/10707 [54:34<38:13,  2.01it/s]
 57%|█████▋    | 6105/10707 [54:35<38:07,  2.01it/s]
 57%|█████▋    | 6106/10707 [54:35<38:04,  2.01it/s]
 57%|█████▋    | 6107/10707 [54:35<38:02,  2.02it/s]
 57%|█████▋    | 6108/10707 [54:36<38:00,  2.02it/s]
 57%|█████▋    | 6109/10707 [54:36<38:01,  2.02it/s]
 57%|█████▋    | 6110/10707 [54:37<37:58,  2.02it/s]
 57%|█████▋    | 6111/10707 [54:37<37:56,  2.02it/s]
 57%|█████▋    | 6112/10707 [54:38<37:57,  2.02it/s]
 57%|█████▋    | 6113/10707 [54:38<37:56,  2.02it/s]
 57%|█████▋    | 6114/10707 [54:39<37:53,  2.02it/s]
 57%|█████▋    | 6115/10707 [54:39<37:54,  2.02it/s]
 57%|█████▋    | 6116/10707 [54:40<37:54,  2.02it/s]
 57%|█████▋    | 6117/10707 [54:40<37:54,  2.02it/s]
 57%|█████▋    | 6118/10707 [54:41<37:51,  2.02it/s]
 57%|█████▋    | 6119/10707 [54:41<37:52,  2.02it/s]
 57%|█████▋    | 6120/10707 [54:42<37:49,  2.02it/s]
 57%|█████▋    | 6121/10707 [54:42<37:49,  2.02it/s]
 57%|█████▋    | 6122/10707 [54:43<37:47,  2.02it/s]
 57%|█████▋    | 6123/10707 [54:43<37:50,  2.02it/s]
 57%|█████▋    | 6124/10707 [54:44<37:47,  2.02it/s]
 57%|█████▋    | 6125/10707 [54:44<37:49,  2.02it/s]{'loss': 3.688, 'grad_norm': 0.20418143272399902, 'learning_rate': 0.0004615668047850157, 'epoch': 0.57}
-
                                                    

 57%|█████▋    | 6125/10707 [54:44<37:49,  2.02it/s]
 57%|█████▋    | 6126/10707 [54:45<37:51,  2.02it/s]
 57%|█████▋    | 6127/10707 [54:45<37:49,  2.02it/s]
 57%|█████▋    | 6128/10707 [54:46<37:47,  2.02it/s]
 57%|█████▋    | 6129/10707 [54:46<37:46,  2.02it/s]
 57%|█████▋    | 6130/10707 [54:47<37:48,  2.02it/s]
 57%|█████▋    | 6131/10707 [54:47<37:45,  2.02it/s]
 57%|█████▋    | 6132/10707 [54:48<37:45,  2.02it/s]
 57%|█████▋    | 6133/10707 [54:48<37:43,  2.02it/s]
 57%|█████▋    | 6134/10707 [54:49<37:44,  2.02it/s]
 57%|█████▋    | 6135/10707 [54:49<37:42,  2.02it/s]
 57%|█████▋    | 6136/10707 [54:50<37:44,  2.02it/s]
 57%|█████▋    | 6137/10707 [54:50<37:42,  2.02it/s]
 57%|█████▋    | 6138/10707 [54:51<37:41,  2.02it/s]
 57%|█████▋    | 6139/10707 [54:51<37:39,  2.02it/s]
 57%|█████▋    | 6140/10707 [54:52<37:39,  2.02it/s]
 57%|█████▋    | 6141/10707 [54:52<37:37,  2.02it/s]
 57%|█████▋    | 6142/10707 [54:53<37:40,  2.02it/s]
 57%|█████▋    | 6143/10707 [54:53<37:39,  2.02it/s]
 57%|█████▋    | 6144/10707 [54:54<37:39,  2.02it/s]
 57%|█████▋    | 6145/10707 [54:54<37:38,  2.02it/s]
 57%|█████▋    | 6146/10707 [54:55<37:39,  2.02it/s]
 57%|█████▋    | 6147/10707 [54:55<37:37,  2.02it/s]
 57%|█████▋    | 6148/10707 [54:56<37:37,  2.02it/s]
 57%|█████▋    | 6149/10707 [54:56<37:33,  2.02it/s]
 57%|█████▋    | 6150/10707 [54:57<37:38,  2.02it/s]{'loss': 3.6902, 'grad_norm': 0.19882845878601074, 'learning_rate': 0.0004575048507248879, 'epoch': 0.57}
-
                                                    

 57%|█████▋    | 6150/10707 [54:57<37:38,  2.02it/s]
 57%|█████▋    | 6151/10707 [54:57<37:39,  2.02it/s]
 57%|█████▋    | 6152/10707 [54:58<37:37,  2.02it/s]
 57%|█████▋    | 6153/10707 [54:58<37:35,  2.02it/s]
 57%|█████▋    | 6154/10707 [54:59<37:35,  2.02it/s]
 57%|█████▋    | 6155/10707 [54:59<37:32,  2.02it/s]
 57%|█████▋    | 6156/10707 [55:00<37:34,  2.02it/s]
 58%|█████▊    | 6157/10707 [55:00<37:34,  2.02it/s]
 58%|█████▊    | 6158/10707 [55:01<37:33,  2.02it/s]
 58%|█████▊    | 6159/10707 [55:01<37:33,  2.02it/s]
 58%|█████▊    | 6160/10707 [55:02<37:32,  2.02it/s]
 58%|█████▊    | 6161/10707 [55:02<37:32,  2.02it/s]
 58%|█████▊    | 6162/10707 [55:03<37:31,  2.02it/s]
 58%|█████▊    | 6163/10707 [55:03<37:36,  2.01it/s]
 58%|█████▊    | 6164/10707 [55:04<37:37,  2.01it/s]
 58%|█████▊    | 6165/10707 [55:04<37:34,  2.01it/s]
 58%|█████▊    | 6166/10707 [55:05<37:34,  2.01it/s]
 58%|█████▊    | 6167/10707 [55:05<37:32,  2.02it/s]
 58%|█████▊    | 6168/10707 [55:06<37:32,  2.02it/s]
 58%|█████▊    | 6169/10707 [55:06<37:28,  2.02it/s]
 58%|█████▊    | 6170/10707 [55:07<37:28,  2.02it/s]
 58%|█████▊    | 6171/10707 [55:07<37:25,  2.02it/s]
 58%|█████▊    | 6172/10707 [55:08<37:27,  2.02it/s]
 58%|█████▊    | 6173/10707 [55:08<37:26,  2.02it/s]
 58%|█████▊    | 6174/10707 [55:09<37:26,  2.02it/s]
 58%|█████▊    | 6175/10707 [55:09<37:26,  2.02it/s]{'loss': 3.6958, 'grad_norm': 0.19925130903720856, 'learning_rate': 0.0004534457197444325, 'epoch': 0.58}
-
                                                    

 58%|█████▊    | 6175/10707 [55:09<37:26,  2.02it/s]
 58%|█████▊    | 6176/10707 [55:10<37:26,  2.02it/s]
 58%|█████▊    | 6177/10707 [55:10<37:26,  2.02it/s]
 58%|█████▊    | 6178/10707 [55:11<37:23,  2.02it/s]
 58%|█████▊    | 6179/10707 [55:11<37:23,  2.02it/s]
 58%|█████▊    | 6180/10707 [55:12<37:21,  2.02it/s]
 58%|█████▊    | 6181/10707 [55:12<37:24,  2.02it/s]
 58%|█████▊    | 6182/10707 [55:13<37:22,  2.02it/s]
 58%|█████▊    | 6183/10707 [55:13<37:23,  2.02it/s]
 58%|█████▊    | 6184/10707 [55:14<37:23,  2.02it/s]
 58%|█████▊    | 6185/10707 [55:14<37:22,  2.02it/s]
 58%|█████▊    | 6186/10707 [55:15<37:25,  2.01it/s]
 58%|█████▊    | 6187/10707 [55:15<37:23,  2.02it/s]
 58%|█████▊    | 6188/10707 [55:16<37:24,  2.01it/s]
 58%|█████▊    | 6189/10707 [55:16<37:19,  2.02it/s]
 58%|█████▊    | 6190/10707 [55:17<37:20,  2.02it/s]
 58%|█████▊    | 6191/10707 [55:17<37:17,  2.02it/s]
 58%|█████▊    | 6192/10707 [55:18<37:17,  2.02it/s]
 58%|█████▊    | 6193/10707 [55:18<37:16,  2.02it/s]
 58%|█████▊    | 6194/10707 [55:19<37:18,  2.02it/s]
 58%|█████▊    | 6195/10707 [55:19<37:17,  2.02it/s]
 58%|█████▊    | 6196/10707 [55:20<37:15,  2.02it/s]
 58%|█████▊    | 6197/10707 [55:20<37:16,  2.02it/s]
 58%|█████▊    | 6198/10707 [55:21<37:19,  2.01it/s]
 58%|█████▊    | 6199/10707 [55:21<37:16,  2.02it/s]
 58%|█████▊    | 6200/10707 [55:22<37:16,  2.02it/s]
                                                    
{'loss': 3.702, 'grad_norm': 0.2019549310207367, 'learning_rate': 0.00044938968150384274, 'epoch': 0.58}
-
 58%|█████▊    | 6200/10707 [55:22<37:16,  2.02it/s]
 58%|█████▊    | 6201/10707 [55:22<37:15,  2.02it/s]
 58%|█████▊    | 6202/10707 [55:23<37:16,  2.01it/s]
 58%|█████▊    | 6203/10707 [55:23<37:15,  2.02it/s]
 58%|█████▊    | 6204/10707 [55:24<37:14,  2.02it/s]
 58%|█████▊    | 6205/10707 [55:24<37:12,  2.02it/s]
 58%|█████▊    | 6206/10707 [55:25<37:09,  2.02it/s]
 58%|█████▊    | 6207/10707 [55:25<37:10,  2.02it/s]
 58%|█████▊    | 6208/10707 [55:26<37:08,  2.02it/s]
 58%|█████▊    | 6209/10707 [55:26<37:09,  2.02it/s]
 58%|█████▊    | 6210/10707 [55:27<37:07,  2.02it/s]
 58%|█████▊    | 6211/10707 [55:27<37:06,  2.02it/s]
 58%|█████▊    | 6212/10707 [55:28<37:04,  2.02it/s]
 58%|█████▊    | 6213/10707 [55:28<37:04,  2.02it/s]
 58%|█████▊    | 6214/10707 [55:29<37:02,  2.02it/s]
 58%|█████▊    | 6215/10707 [55:29<37:03,  2.02it/s]
 58%|█████▊    | 6216/10707 [55:30<37:03,  2.02it/s]
 58%|█████▊    | 6217/10707 [55:30<37:03,  2.02it/s]
 58%|█████▊    | 6218/10707 [55:30<37:02,  2.02it/s]
 58%|█████▊    | 6219/10707 [55:31<37:00,  2.02it/s]
 58%|█████▊    | 6220/10707 [55:31<37:01,  2.02it/s]
 58%|█████▊    | 6221/10707 [55:32<37:01,  2.02it/s]
 58%|█████▊    | 6222/10707 [55:32<37:00,  2.02it/s]
 58%|█████▊    | 6223/10707 [55:33<37:02,  2.02it/s]
 58%|█████▊    | 6224/10707 [55:33<37:07,  2.01it/s]
 58%|█████▊    | 6225/10707 [55:34<37:05,  2.01it/s]{'loss': 3.6854, 'grad_norm': 0.19646720588207245, 'learning_rate': 0.00044533700545785197, 'epoch': 0.58}
-
                                                    

 58%|█████▊    | 6225/10707 [55:34<37:05,  2.01it/s]
 58%|█████▊    | 6226/10707 [55:34<37:10,  2.01it/s]
 58%|█████▊    | 6227/10707 [55:35<37:06,  2.01it/s]
 58%|█████▊    | 6228/10707 [55:35<37:04,  2.01it/s]
 58%|█████▊    | 6229/10707 [55:36<37:01,  2.02it/s]
 58%|█████▊    | 6230/10707 [55:36<37:02,  2.01it/s]
 58%|█████▊    | 6231/10707 [55:37<36:58,  2.02it/s]
 58%|█████▊    | 6232/10707 [55:37<36:58,  2.02it/s]
 58%|█████▊    | 6233/10707 [55:38<36:55,  2.02it/s]
 58%|█████▊    | 6234/10707 [55:38<36:57,  2.02it/s]
 58%|█████▊    | 6235/10707 [55:39<36:57,  2.02it/s]
 58%|█████▊    | 6236/10707 [55:39<36:56,  2.02it/s]
 58%|█████▊    | 6237/10707 [55:40<36:54,  2.02it/s]
 58%|█████▊    | 6238/10707 [55:40<36:55,  2.02it/s]
 58%|█████▊    | 6239/10707 [55:41<36:53,  2.02it/s]
 58%|█████▊    | 6240/10707 [55:41<36:53,  2.02it/s]
 58%|█████▊    | 6241/10707 [55:42<36:53,  2.02it/s]
 58%|█████▊    | 6242/10707 [55:42<36:50,  2.02it/s]
 58%|█████▊    | 6243/10707 [55:43<36:52,  2.02it/s]
 58%|█████▊    | 6244/10707 [55:43<36:51,  2.02it/s]
 58%|█████▊    | 6245/10707 [55:44<36:50,  2.02it/s]
 58%|█████▊    | 6246/10707 [55:44<36:51,  2.02it/s]
 58%|█████▊    | 6247/10707 [55:45<36:50,  2.02it/s]
 58%|█████▊    | 6248/10707 [55:45<36:51,  2.02it/s]
 58%|█████▊    | 6249/10707 [55:46<36:50,  2.02it/s]
 58%|█████▊    | 6250/10707 [55:46<36:49,  2.02it/s]{'loss': 3.6874, 'grad_norm': 0.19415268301963806, 'learning_rate': 0.00044128796083783264, 'epoch': 0.58}
-
                                                    

 58%|█████▊    | 6250/10707 [55:46<36:49,  2.02it/s]
 58%|█████▊    | 6251/10707 [55:47<36:52,  2.01it/s]
 58%|█████▊    | 6252/10707 [55:47<36:48,  2.02it/s]
 58%|█████▊    | 6253/10707 [55:48<36:48,  2.02it/s]
 58%|█████▊    | 6254/10707 [55:48<36:46,  2.02it/s]
 58%|█████▊    | 6255/10707 [55:49<36:46,  2.02it/s]
 58%|█████▊    | 6256/10707 [55:49<36:46,  2.02it/s]
 58%|█████▊    | 6257/10707 [55:50<36:44,  2.02it/s]
 58%|█████▊    | 6258/10707 [55:50<36:41,  2.02it/s]
 58%|█████▊    | 6259/10707 [55:51<36:43,  2.02it/s]
 58%|█████▊    | 6260/10707 [55:51<36:40,  2.02it/s]
 58%|█████▊    | 6261/10707 [55:52<36:48,  2.01it/s]
 58%|█████▊    | 6262/10707 [55:52<36:48,  2.01it/s]
 58%|█████▊    | 6263/10707 [55:53<36:43,  2.02it/s]
 59%|█████▊    | 6264/10707 [55:53<36:43,  2.02it/s]
 59%|█████▊    | 6265/10707 [55:54<36:41,  2.02it/s]
 59%|█████▊    | 6266/10707 [55:54<36:41,  2.02it/s]
 59%|█████▊    | 6267/10707 [55:55<36:37,  2.02it/s]
 59%|█████▊    | 6268/10707 [55:55<36:38,  2.02it/s]
 59%|█████▊    | 6269/10707 [55:56<36:36,  2.02it/s]
 59%|█████▊    | 6270/10707 [55:56<36:39,  2.02it/s]
 59%|█████▊    | 6271/10707 [55:57<36:36,  2.02it/s]
 59%|█████▊    | 6272/10707 [55:57<36:36,  2.02it/s]
 59%|█████▊    | 6273/10707 [55:58<36:32,  2.02it/s]
 59%|█████▊    | 6274/10707 [55:58<36:33,  2.02it/s]
 59%|█████▊    | 6275/10707 [55:59<36:31,  2.02it/s]{'loss': 3.6854, 'grad_norm': 0.2042492926120758, 'learning_rate': 0.0004372428166339106, 'epoch': 0.59}
-
                                                    

 59%|█████▊    | 6275/10707 [55:59<36:31,  2.02it/s]
 59%|█████▊    | 6276/10707 [55:59<36:33,  2.02it/s]
 59%|█████▊    | 6277/10707 [56:00<36:35,  2.02it/s]
 59%|█████▊    | 6278/10707 [56:00<36:34,  2.02it/s]
 59%|█████▊    | 6279/10707 [56:01<36:34,  2.02it/s]
 59%|█████▊    | 6280/10707 [56:01<36:32,  2.02it/s]
 59%|█████▊    | 6281/10707 [56:02<36:31,  2.02it/s]
 59%|█████▊    | 6282/10707 [56:02<36:30,  2.02it/s]
 59%|█████▊    | 6283/10707 [56:03<36:32,  2.02it/s]
 59%|█████▊    | 6284/10707 [56:03<36:36,  2.01it/s]
 59%|█████▊    | 6285/10707 [56:04<36:36,  2.01it/s]
 59%|█████▊    | 6286/10707 [56:04<36:34,  2.01it/s]
 59%|█████▊    | 6287/10707 [56:05<36:31,  2.02it/s]
 59%|█████▊    | 6288/10707 [56:05<36:27,  2.02it/s]
 59%|█████▊    | 6289/10707 [56:06<36:27,  2.02it/s]
 59%|█████▊    | 6290/10707 [56:06<36:27,  2.02it/s]
 59%|█████▉    | 6291/10707 [56:07<36:27,  2.02it/s]
 59%|█████▉    | 6292/10707 [56:07<36:25,  2.02it/s]
 59%|█████▉    | 6293/10707 [56:08<36:27,  2.02it/s]
 59%|█████▉    | 6294/10707 [56:08<36:23,  2.02it/s]
 59%|█████▉    | 6295/10707 [56:09<36:22,  2.02it/s]
 59%|█████▉    | 6296/10707 [56:09<36:23,  2.02it/s]
 59%|█████▉    | 6297/10707 [56:10<36:22,  2.02it/s]
 59%|█████▉    | 6298/10707 [56:10<36:23,  2.02it/s]
 59%|█████▉    | 6299/10707 [56:11<36:20,  2.02it/s]
 59%|█████▉    | 6300/10707 [56:11<36:22,  2.02it/s]{'loss': 3.6792, 'grad_norm': 0.2014482468366623, 'learning_rate': 0.000433201841577096, 'epoch': 0.59}
-
                                                    

 59%|█████▉    | 6300/10707 [56:11<36:22,  2.02it/s]
 59%|█████▉    | 6301/10707 [56:12<36:22,  2.02it/s]
 59%|█████▉    | 6302/10707 [56:12<36:23,  2.02it/s]
 59%|█████▉    | 6303/10707 [56:13<36:19,  2.02it/s]
 59%|█████▉    | 6304/10707 [56:13<36:20,  2.02it/s]
 59%|█████▉    | 6305/10707 [56:14<36:18,  2.02it/s]
 59%|█████▉    | 6306/10707 [56:14<36:19,  2.02it/s]
 59%|█████▉    | 6307/10707 [56:15<36:18,  2.02it/s]
 59%|█████▉    | 6308/10707 [56:15<36:19,  2.02it/s]
 59%|█████▉    | 6309/10707 [56:16<36:19,  2.02it/s]
 59%|█████▉    | 6310/10707 [56:16<36:18,  2.02it/s]
 59%|█████▉    | 6311/10707 [56:17<36:15,  2.02it/s]
 59%|█████▉    | 6312/10707 [56:17<36:15,  2.02it/s]
 59%|█████▉    | 6313/10707 [56:18<36:17,  2.02it/s]
 59%|█████▉    | 6314/10707 [56:18<36:16,  2.02it/s]
 59%|█████▉    | 6315/10707 [56:19<36:18,  2.02it/s]
 59%|█████▉    | 6316/10707 [56:19<36:13,  2.02it/s]
 59%|█████▉    | 6317/10707 [56:20<36:13,  2.02it/s]
 59%|█████▉    | 6318/10707 [56:20<36:11,  2.02it/s]
 59%|█████▉    | 6319/10707 [56:21<36:16,  2.02it/s]
 59%|█████▉    | 6320/10707 [56:21<36:14,  2.02it/s]
 59%|█████▉    | 6321/10707 [56:22<36:14,  2.02it/s]
 59%|█████▉    | 6322/10707 [56:22<36:14,  2.02it/s]
 59%|█████▉    | 6323/10707 [56:23<36:12,  2.02it/s]
 59%|█████▉    | 6324/10707 [56:23<36:11,  2.02it/s]
 59%|█████▉    | 6325/10707 [56:24<36:12,  2.02it/s]{'loss': 3.6927, 'grad_norm': 0.20235152542591095, 'learning_rate': 0.00042916530412142917, 'epoch': 0.59}
-
                                                    

 59%|█████▉    | 6325/10707 [56:24<36:12,  2.02it/s]
 59%|█████▉    | 6326/10707 [56:24<36:14,  2.01it/s]
 59%|█████▉    | 6327/10707 [56:25<36:11,  2.02it/s]
 59%|█████▉    | 6328/10707 [56:25<36:10,  2.02it/s]
 59%|█████▉    | 6329/10707 [56:26<36:11,  2.02it/s]
 59%|█████▉    | 6330/10707 [56:26<36:09,  2.02it/s]
 59%|█████▉    | 6331/10707 [56:26<36:08,  2.02it/s]
 59%|█████▉    | 6332/10707 [56:27<36:07,  2.02it/s]
 59%|█████▉    | 6333/10707 [56:27<36:09,  2.02it/s]
 59%|█████▉    | 6334/10707 [56:28<36:07,  2.02it/s]
 59%|█████▉    | 6335/10707 [56:28<36:09,  2.02it/s]
 59%|█████▉    | 6336/10707 [56:29<36:05,  2.02it/s]
 59%|█████▉    | 6337/10707 [56:29<36:05,  2.02it/s]
 59%|█████▉    | 6338/10707 [56:30<36:03,  2.02it/s]
 59%|█████▉    | 6339/10707 [56:30<36:03,  2.02it/s]
 59%|█████▉    | 6340/10707 [56:31<36:04,  2.02it/s]
 59%|█████▉    | 6341/10707 [56:31<36:01,  2.02it/s]
 59%|█████▉    | 6342/10707 [56:32<36:02,  2.02it/s]
 59%|█████▉    | 6343/10707 [56:32<36:01,  2.02it/s]
 59%|█████▉    | 6344/10707 [56:33<36:11,  2.01it/s]
 59%|█████▉    | 6345/10707 [56:33<36:09,  2.01it/s]
 59%|█████▉    | 6346/10707 [56:34<36:05,  2.01it/s]
 59%|█████▉    | 6347/10707 [56:34<36:03,  2.02it/s]
 59%|█████▉    | 6348/10707 [56:35<35:58,  2.02it/s]
 59%|█████▉    | 6349/10707 [56:35<35:58,  2.02it/s]
 59%|█████▉    | 6350/10707 [56:36<35:57,  2.02it/s]{'loss': 3.6781, 'grad_norm': 0.21678656339645386, 'learning_rate': 0.00042513347242614834, 'epoch': 0.59}
-
                                                    

 59%|█████▉    | 6350/10707 [56:36<35:57,  2.02it/s]
 59%|█████▉    | 6351/10707 [56:36<36:01,  2.01it/s]
 59%|█████▉    | 6352/10707 [56:37<35:56,  2.02it/s]
 59%|█████▉    | 6353/10707 [56:37<35:58,  2.02it/s]
 59%|█████▉    | 6354/10707 [56:38<35:56,  2.02it/s]
 59%|█████▉    | 6355/10707 [56:38<35:57,  2.02it/s]
 59%|█████▉    | 6356/10707 [56:39<35:56,  2.02it/s]
 59%|█████▉    | 6357/10707 [56:39<35:55,  2.02it/s]
 59%|█████▉    | 6358/10707 [56:40<35:55,  2.02it/s]
 59%|█████▉    | 6359/10707 [56:40<35:54,  2.02it/s]
 59%|█████▉    | 6360/10707 [56:41<35:59,  2.01it/s]
 59%|█████▉    | 6361/10707 [56:41<35:55,  2.02it/s]
 59%|█████▉    | 6362/10707 [56:42<35:57,  2.01it/s]
 59%|█████▉    | 6363/10707 [56:42<35:54,  2.02it/s]
 59%|█████▉    | 6364/10707 [56:43<35:53,  2.02it/s]
 59%|█████▉    | 6365/10707 [56:43<35:54,  2.02it/s]
 59%|█████▉    | 6366/10707 [56:44<35:51,  2.02it/s]
 59%|█████▉    | 6367/10707 [56:44<35:53,  2.02it/s]
 59%|█████▉    | 6368/10707 [56:45<35:50,  2.02it/s]
 59%|█████▉    | 6369/10707 [56:45<35:53,  2.01it/s]
 59%|█████▉    | 6370/10707 [56:46<35:50,  2.02it/s]
 60%|█████▉    | 6371/10707 [56:46<35:48,  2.02it/s]
 60%|█████▉    | 6372/10707 [56:47<35:50,  2.02it/s]
 60%|█████▉    | 6373/10707 [56:47<35:50,  2.02it/s]
 60%|█████▉    | 6374/10707 [56:48<35:47,  2.02it/s]
 60%|█████▉    | 6375/10707 [56:48<35:47,  2.02it/s]{'loss': 3.6893, 'grad_norm': 0.20518457889556885, 'learning_rate': 0.00042110661433787346, 'epoch': 0.6}
-
                                                    

 60%|█████▉    | 6375/10707 [56:48<35:47,  2.02it/s]
 60%|█████▉    | 6376/10707 [56:49<35:49,  2.01it/s]
 60%|█████▉    | 6377/10707 [56:49<35:49,  2.01it/s]
 60%|█████▉    | 6378/10707 [56:50<35:48,  2.01it/s]
 60%|█████▉    | 6379/10707 [56:50<35:47,  2.01it/s]
 60%|█████▉    | 6380/10707 [56:51<35:47,  2.02it/s]
 60%|█████▉    | 6381/10707 [56:51<35:44,  2.02it/s]
 60%|█████▉    | 6382/10707 [56:52<35:45,  2.02it/s]
 60%|█████▉    | 6383/10707 [56:52<35:41,  2.02it/s]
 60%|█████▉    | 6384/10707 [56:53<35:41,  2.02it/s]
 60%|█████▉    | 6385/10707 [56:53<35:40,  2.02it/s]
 60%|█████▉    | 6386/10707 [56:54<35:41,  2.02it/s]
 60%|█████▉    | 6387/10707 [56:54<35:39,  2.02it/s]
 60%|��████▉    | 6388/10707 [56:55<35:40,  2.02it/s]
 60%|█████▉    | 6389/10707 [56:55<35:36,  2.02it/s]
 60%|█████▉    | 6390/10707 [56:56<35:38,  2.02it/s]
 60%|█████▉    | 6391/10707 [56:56<35:33,  2.02it/s]
 60%|█████▉    | 6392/10707 [56:57<35:36,  2.02it/s]
 60%|█████▉    | 6393/10707 [56:57<35:36,  2.02it/s]
 60%|█████▉    | 6394/10707 [56:58<35:36,  2.02it/s]
 60%|█████▉    | 6395/10707 [56:58<35:33,  2.02it/s]
 60%|█████▉    | 6396/10707 [56:59<35:36,  2.02it/s]
 60%|█████▉    | 6397/10707 [56:59<35:35,  2.02it/s]
 60%|█████▉    | 6398/10707 [57:00<35:34,  2.02it/s]
 60%|█████▉    | 6399/10707 [57:00<35:31,  2.02it/s]
 60%|█████▉    | 6400/10707 [57:01<35:31,  2.02it/s]
                                                    
{'loss': 3.6777, 'grad_norm': 0.1949084997177124, 'learning_rate': 0.00041708499737281305, 'epoch': 0.6}
-
 60%|█████▉    | 6400/10707 [57:01<35:31,  2.02it/s]
 60%|█████▉    | 6401/10707 [57:01<35:37,  2.01it/s]
 60%|█████▉    | 6402/10707 [57:02<35:33,  2.02it/s]
 60%|█████▉    | 6403/10707 [57:02<35:35,  2.02it/s]
 60%|█████▉    | 6404/10707 [57:03<35:30,  2.02it/s]
 60%|█████▉    | 6405/10707 [57:03<35:37,  2.01it/s]
 60%|█████▉    | 6406/10707 [57:04<35:40,  2.01it/s]
 60%|█████▉    | 6407/10707 [57:04<35:35,  2.01it/s]
 60%|█████▉    | 6408/10707 [57:05<35:33,  2.01it/s]
 60%|█████▉    | 6409/10707 [57:05<35:30,  2.02it/s]
 60%|█████▉    | 6410/10707 [57:06<35:29,  2.02it/s]
 60%|█████▉    | 6411/10707 [57:06<35:25,  2.02it/s]
 60%|█████▉    | 6412/10707 [57:07<35:26,  2.02it/s]
 60%|█████▉    | 6413/10707 [57:07<35:23,  2.02it/s]
 60%|█████▉    | 6414/10707 [57:08<35:25,  2.02it/s]
 60%|█████▉    | 6415/10707 [57:08<35:24,  2.02it/s]
 60%|█████▉    | 6416/10707 [57:09<35:25,  2.02it/s]
 60%|█████▉    | 6417/10707 [57:09<35:23,  2.02it/s]
 60%|█████▉    | 6418/10707 [57:10<35:24,  2.02it/s]
 60%|█████▉    | 6419/10707 [57:10<35:21,  2.02it/s]
 60%|█████▉    | 6420/10707 [57:11<35:22,  2.02it/s]
 60%|█████▉    | 6421/10707 [57:11<35:20,  2.02it/s]
 60%|█████▉    | 6422/10707 [57:12<35:20,  2.02it/s]
 60%|█████▉    | 6423/10707 [57:12<35:18,  2.02it/s]
 60%|█████▉    | 6424/10707 [57:13<35:20,  2.02it/s]
 60%|██████    | 6425/10707 [57:13<35:19,  2.02it/s]
                                                    
{'loss': 3.6776, 'grad_norm': 0.212349995970726, 'learning_rate': 0.00041306888869899183, 'epoch': 0.6}
-
 60%|██████    | 6425/10707 [57:13<35:19,  2.02it/s]
 60%|██████    | 6426/10707 [57:14<35:22,  2.02it/s]
 60%|██████    | 6427/10707 [57:14<35:19,  2.02it/s]
 60%|██████    | 6428/10707 [57:15<35:20,  2.02it/s]
 60%|██████    | 6429/10707 [57:15<35:19,  2.02it/s]
 60%|██████    | 6430/10707 [57:16<35:20,  2.02it/s]
 60%|██████    | 6431/10707 [57:16<35:18,  2.02it/s]
 60%|██████    | 6432/10707 [57:17<35:19,  2.02it/s]
 60%|██████    | 6433/10707 [57:17<35:19,  2.02it/s]
 60%|██████    | 6434/10707 [57:18<35:18,  2.02it/s]
 60%|██████    | 6435/10707 [57:18<35:19,  2.02it/s]
 60%|██████    | 6436/10707 [57:19<35:18,  2.02it/s]
 60%|██████    | 6437/10707 [57:19<35:15,  2.02it/s]
 60%|██████    | 6438/10707 [57:20<35:16,  2.02it/s]
 60%|██████    | 6439/10707 [57:20<35:14,  2.02it/s]
 60%|██████    | 6440/10707 [57:21<35:13,  2.02it/s]
 60%|██████    | 6441/10707 [57:21<35:12,  2.02it/s]
 60%|██████    | 6442/10707 [57:22<35:13,  2.02it/s]
 60%|██████    | 6443/10707 [57:22<35:10,  2.02it/s]
 60%|██████    | 6444/10707 [57:22<35:10,  2.02it/s]
 60%|██████    | 6445/10707 [57:23<35:09,  2.02it/s]
 60%|██████    | 6446/10707 [57:23<35:11,  2.02it/s]
 60%|██████    | 6447/10707 [57:24<35:09,  2.02it/s]
 60%|██████    | 6448/10707 [57:24<35:10,  2.02it/s]
 60%|██████    | 6449/10707 [57:25<35:07,  2.02it/s]
 60%|██████    | 6450/10707 [57:25<35:09,  2.02it/s]
                                                    {'loss': 3.6833, 'grad_norm': 0.1929328590631485, 'learning_rate': 0.00040905855511850304, 'epoch': 0.6}
-
 60%|██████    | 6450/10707 [57:25<35:09,  2.02it/s]
 60%|██████    | 6451/10707 [57:26<35:13,  2.01it/s]
 60%|██████    | 6452/10707 [57:26<35:10,  2.02it/s]
 60%|██████    | 6453/10707 [57:27<35:08,  2.02it/s]
 60%|██████    | 6454/10707 [57:27<35:05,  2.02it/s]
 60%|██████    | 6455/10707 [57:28<35:05,  2.02it/s]
 60%|██████    | 6456/10707 [57:28<35:04,  2.02it/s]
 60%|██████    | 6457/10707 [57:29<35:05,  2.02it/s]
 60%|██████    | 6458/10707 [57:29<35:03,  2.02it/s]
 60%|██████    | 6459/10707 [57:30<35:04,  2.02it/s]
 60%|██████    | 6460/10707 [57:30<35:03,  2.02it/s]
 60%|██████    | 6461/10707 [57:31<35:02,  2.02it/s]
 60%|██████    | 6462/10707 [57:31<35:02,  2.02it/s]
 60%|██████    | 6463/10707 [57:32<35:01,  2.02it/s]
 60%|██████    | 6464/10707 [57:32<35:02,  2.02it/s]
 60%|██████    | 6465/10707 [57:33<35:01,  2.02it/s]
 60%|██████    | 6466/10707 [57:33<35:01,  2.02it/s]
 60%|██████    | 6467/10707 [57:34<34:59,  2.02it/s]
 60%|██████    | 6468/10707 [57:34<35:00,  2.02it/s]
 60%|██████    | 6469/10707 [57:35<34:58,  2.02it/s]
 60%|██████    | 6470/10707 [57:35<34:59,  2.02it/s]
 60%|██████    | 6471/10707 [57:36<34:56,  2.02it/s]
 60%|██████    | 6472/10707 [57:36<34:57,  2.02it/s]
 60%|██████    | 6473/10707 [57:37<34:54,  2.02it/s]
 60%|██████    | 6474/10707 [57:37<34:55,  2.02it/s]
 60%|██████    | 6475/10707 [57:38<34:53,  2.02it/s]{'loss': 3.6802, 'grad_norm': 0.1952439695596695, 'learning_rate': 0.0004050542630497826, 'epoch': 0.6}
-
                                                    

 60%|██████    | 6475/10707 [57:38<34:53,  2.02it/s]
 60%|██████    | 6476/10707 [57:38<34:58,  2.02it/s]
 60%|██████    | 6477/10707 [57:39<34:56,  2.02it/s]
 61%|██████    | 6478/10707 [57:39<34:56,  2.02it/s]
 61%|██████    | 6479/10707 [57:40<34:53,  2.02it/s]
 61%|██████    | 6480/10707 [57:40<34:54,  2.02it/s]
 61%|██████    | 6481/10707 [57:41<34:54,  2.02it/s]
 61%|██████    | 6482/10707 [57:41<34:52,  2.02it/s]
 61%|██████    | 6483/10707 [57:42<34:51,  2.02it/s]
 61%|██████    | 6484/10707 [57:42<34:50,  2.02it/s]
 61%|██████    | 6485/10707 [57:43<34:50,  2.02it/s]
 61%|██████    | 6486/10707 [57:43<34:49,  2.02it/s]
 61%|██████    | 6487/10707 [57:44<34:48,  2.02it/s]
 61%|██████    | 6488/10707 [57:44<34:47,  2.02it/s]
 61%|██████    | 6489/10707 [57:45<34:47,  2.02it/s]
 61%|██████    | 6490/10707 [57:45<34:45,  2.02it/s]
 61%|██████    | 6491/10707 [57:46<34:44,  2.02it/s]
 61%|██████    | 6492/10707 [57:46<34:44,  2.02it/s]
 61%|██████    | 6493/10707 [57:47<34:43,  2.02it/s]
 61%|██████    | 6494/10707 [57:47<34:44,  2.02it/s]
 61%|██████    | 6495/10707 [57:48<34:43,  2.02it/s]
 61%|██████    | 6496/10707 [57:48<34:43,  2.02it/s]
 61%|██████    | 6497/10707 [57:49<34:42,  2.02it/s]
 61%|██████    | 6498/10707 [57:49<34:43,  2.02it/s]
 61%|██████    | 6499/10707 [57:50<34:42,  2.02it/s]
 61%|██████    | 6500/10707 [57:50<34:41,  2.02it/s]{'loss': 3.6759, 'grad_norm': 0.20619872212409973, 'learning_rate': 0.00040105627850991083, 'epoch': 0.61}
-
                                                    

 61%|██████    | 6500/10707 [57:50<34:41,  2.02it/s]
 61%|██████    | 6501/10707 [57:51<34:45,  2.02it/s]
 61%|██████    | 6502/10707 [57:51<34:44,  2.02it/s]
 61%|██████    | 6503/10707 [57:52<34:44,  2.02it/s]
 61%|██████    | 6504/10707 [57:52<34:42,  2.02it/s]
 61%|██████    | 6505/10707 [57:53<34:42,  2.02it/s]
 61%|██████    | 6506/10707 [57:53<34:41,  2.02it/s]
 61%|██████    | 6507/10707 [57:54<34:42,  2.02it/s]
 61%|██████    | 6508/10707 [57:54<34:39,  2.02it/s]
 61%|██████    | 6509/10707 [57:55<34:39,  2.02it/s]
 61%|██████    | 6510/10707 [57:55<34:36,  2.02it/s]
 61%|██████    | 6511/10707 [57:56<34:36,  2.02it/s]
 61%|██████    | 6512/10707 [57:56<34:34,  2.02it/s]
 61%|██████    | 6513/10707 [57:57<34:36,  2.02it/s]
 61%|██████    | 6514/10707 [57:57<34:34,  2.02it/s]
 61%|██████    | 6515/10707 [57:58<34:34,  2.02it/s]
 61%|██████    | 6516/10707 [57:58<34:33,  2.02it/s]
 61%|██████    | 6517/10707 [57:59<34:36,  2.02it/s]
 61%|██████    | 6518/10707 [57:59<34:33,  2.02it/s]
 61%|██████    | 6519/10707 [58:00<34:35,  2.02it/s]
 61%|██████    | 6520/10707 [58:00<34:33,  2.02it/s]
 61%|██████    | 6521/10707 [58:01<34:34,  2.02it/s]
 61%|██████    | 6522/10707 [58:01<34:34,  2.02it/s]
 61%|██████    | 6523/10707 [58:02<34:33,  2.02it/s]
 61%|██████    | 6524/10707 [58:02<34:34,  2.02it/s]
 61%|██████    | 6525/10707 [58:03<34:31,  2.02it/s]{'loss': 3.6766, 'grad_norm': 0.1919352114200592, 'learning_rate': 0.0003970648670969401, 'epoch': 0.61}
-
                                                    

 61%|██████    | 6525/10707 [58:03<34:31,  2.02it/s]
 61%|██████    | 6526/10707 [58:03<34:34,  2.02it/s]
 61%|██████    | 6527/10707 [58:04<34:30,  2.02it/s]
 61%|██████    | 6528/10707 [58:04<34:31,  2.02it/s]
 61%|██████    | 6529/10707 [58:05<34:28,  2.02it/s]
 61%|██████    | 6530/10707 [58:05<34:29,  2.02it/s]
 61%|██████    | 6531/10707 [58:06<34:28,  2.02it/s]
 61%|██████    | 6532/10707 [58:06<34:28,  2.02it/s]
 61%|██████    | 6533/10707 [58:07<34:27,  2.02it/s]
 61%|██████    | 6534/10707 [58:07<34:27,  2.02it/s]
 61%|██████    | 6535/10707 [58:08<34:27,  2.02it/s]
 61%|██████    | 6536/10707 [58:08<34:27,  2.02it/s]
 61%|██████    | 6537/10707 [58:09<34:30,  2.01it/s]
 61%|██████    | 6538/10707 [58:09<34:26,  2.02it/s]
 61%|██████    | 6539/10707 [58:10<34:25,  2.02it/s]
 61%|██████    | 6540/10707 [58:10<34:22,  2.02it/s]
 61%|██████    | 6541/10707 [58:11<34:23,  2.02it/s]
 61%|██████    | 6542/10707 [58:11<34:22,  2.02it/s]
 61%|██████    | 6543/10707 [58:12<34:24,  2.02it/s]
 61%|██████    | 6544/10707 [58:12<34:22,  2.02it/s]
 61%|██████    | 6545/10707 [58:13<34:22,  2.02it/s]
 61%|██████    | 6546/10707 [58:13<34:20,  2.02it/s]
 61%|██████    | 6547/10707 [58:14<34:20,  2.02it/s]
 61%|██████    | 6548/10707 [58:14<34:20,  2.02it/s]
 61%|██████    | 6549/10707 [58:15<34:20,  2.02it/s]
 61%|██████    | 6550/10707 [58:15<34:17,  2.02it/s]{'loss': 3.6632, 'grad_norm': 0.18991342186927795, 'learning_rate': 0.00039308029397225076, 'epoch': 0.61}
-
                                                    

 61%|██████    | 6550/10707 [58:15<34:17,  2.02it/s]
 61%|██████    | 6551/10707 [58:15<34:20,  2.02it/s]
 61%|██████    | 6552/10707 [58:16<34:19,  2.02it/s]
 61%|██████    | 6553/10707 [58:16<34:17,  2.02it/s]
 61%|██████    | 6554/10707 [58:17<34:16,  2.02it/s]
 61%|██████    | 6555/10707 [58:17<34:15,  2.02it/s]
 61%|██████    | 6556/10707 [58:18<34:15,  2.02it/s]
 61%|██████    | 6557/10707 [58:18<34:14,  2.02it/s]
 61%|██████    | 6558/10707 [58:19<34:14,  2.02it/s]
 61%|██████▏   | 6559/10707 [58:19<34:13,  2.02it/s]
 61%|██████▏   | 6560/10707 [58:20<34:14,  2.02it/s]
 61%|██████▏   | 6561/10707 [58:20<34:12,  2.02it/s]
 61%|██████▏   | 6562/10707 [58:21<34:13,  2.02it/s]
 61%|██████▏   | 6563/10707 [58:21<34:12,  2.02it/s]
 61%|██████▏   | 6564/10707 [58:22<34:12,  2.02it/s]
 61%|██████▏   | 6565/10707 [58:22<34:10,  2.02it/s]
 61%|██████▏   | 6566/10707 [58:23<34:10,  2.02it/s]
 61%|██████▏   | 6567/10707 [58:23<34:12,  2.02it/s]
 61%|██████▏   | 6568/10707 [58:24<34:09,  2.02it/s]
 61%|██████▏   | 6569/10707 [58:24<34:10,  2.02it/s]
 61%|██████▏   | 6570/10707 [58:25<34:08,  2.02it/s]
 61%|██████▏   | 6571/10707 [58:25<34:10,  2.02it/s]
 61%|██████▏   | 6572/10707 [58:26<34:07,  2.02it/s]
 61%|██████▏   | 6573/10707 [58:26<34:07,  2.02it/s]
 61%|██████▏   | 6574/10707 [58:27<34:08,  2.02it/s]
 61%|██████▏   | 6575/10707 [58:27<34:07,  2.02it/s]{'loss': 3.6736, 'grad_norm': 0.2033180147409439, 'learning_rate': 0.00038910282384293474, 'epoch': 0.61}
                                                    
-
 61%|██████▏   | 6575/10707 [58:27<34:07,  2.02it/s]
 61%|██████▏   | 6576/10707 [58:28<34:08,  2.02it/s]
 61%|██████▏   | 6577/10707 [58:28<34:08,  2.02it/s]
 61%|██████▏   | 6578/10707 [58:29<34:08,  2.02it/s]
 61%|██████▏   | 6579/10707 [58:29<34:06,  2.02it/s]
 61%|██████▏   | 6580/10707 [58:30<34:05,  2.02it/s]
 61%|██████▏   | 6581/10707 [58:30<34:04,  2.02it/s]
 61%|██████▏   | 6582/10707 [58:31<34:02,  2.02it/s]
 61%|██████▏   | 6583/10707 [58:31<34:00,  2.02it/s]
 61%|██████▏   | 6584/10707 [58:32<34:00,  2.02it/s]
 62%|██████▏   | 6585/10707 [58:32<33:58,  2.02it/s]
 62%|██████▏   | 6586/10707 [58:33<33:59,  2.02it/s]
 62%|██████▏   | 6587/10707 [58:33<33:59,  2.02it/s]
 62%|██████▏   | 6588/10707 [58:34<33:59,  2.02it/s]
 62%|██████▏   | 6589/10707 [58:34<33:57,  2.02it/s]
 62%|██████▏   | 6590/10707 [58:35<33:56,  2.02it/s]
 62%|██████▏   | 6591/10707 [58:35<33:56,  2.02it/s]
 62%|██████▏   | 6592/10707 [58:36<33:54,  2.02it/s]
 62%|██████▏   | 6593/10707 [58:36<33:55,  2.02it/s]
 62%|██████▏   | 6594/10707 [58:37<33:53,  2.02it/s]
 62%|██████▏   | 6595/10707 [58:37<33:53,  2.02it/s]
 62%|██████▏   | 6596/10707 [58:38<33:52,  2.02it/s]
 62%|██████▏   | 6597/10707 [58:38<33:53,  2.02it/s]
 62%|██████▏   | 6598/10707 [58:39<33:53,  2.02it/s]
 62%|██████▏   | 6599/10707 [58:39<33:54,  2.02it/s]
 62%|██████▏   | 6600/10707 [58:40<33:51,  2.02it/s]{'loss': 3.6679, 'grad_norm': 0.2001279592514038, 'learning_rate': 0.00038513272094421074, 'epoch': 0.62}
-
                                                    

 62%|██████▏   | 6600/10707 [58:40<33:51,  2.02it/s]
 62%|██████▏   | 6601/10707 [58:40<33:54,  2.02it/s]
 62%|██████▏   | 6602/10707 [58:41<33:55,  2.02it/s]
 62%|██████▏   | 6603/10707 [58:41<33:51,  2.02it/s]
 62%|██████▏   | 6604/10707 [58:42<33:50,  2.02it/s]
 62%|██████▏   | 6605/10707 [58:42<33:49,  2.02it/s]
 62%|██████▏   | 6606/10707 [58:43<33:50,  2.02it/s]
 62%|██████▏   | 6607/10707 [58:43<33:50,  2.02it/s]
 62%|██████▏   | 6608/10707 [58:44<33:48,  2.02it/s]
 62%|██████▏   | 6609/10707 [58:44<33:49,  2.02it/s]
 62%|██████▏   | 6610/10707 [58:45<33:49,  2.02it/s]
 62%|██████▏   | 6611/10707 [58:45<33:48,  2.02it/s]
 62%|██████▏   | 6612/10707 [58:46<33:48,  2.02it/s]
 62%|██████▏   | 6613/10707 [58:46<33:46,  2.02it/s]
 62%|██████▏   | 6614/10707 [58:47<33:46,  2.02it/s]
 62%|██████▏   | 6615/10707 [58:47<33:43,  2.02it/s]
 62%|██████▏   | 6616/10707 [58:48<33:43,  2.02it/s]
 62%|██████▏   | 6617/10707 [58:48<33:43,  2.02it/s]
 62%|██████▏   | 6618/10707 [58:49<33:43,  2.02it/s]
 62%|██████▏   | 6619/10707 [58:49<33:43,  2.02it/s]
 62%|██████▏   | 6620/10707 [58:50<33:44,  2.02it/s]
 62%|██████▏   | 6621/10707 [58:50<33:42,  2.02it/s]
 62%|██████▏   | 6622/10707 [58:51<33:43,  2.02it/s]
 62%|██████▏   | 6623/10707 [58:51<33:40,  2.02it/s]
 62%|██████▏   | 6624/10707 [58:52<33:41,  2.02it/s]
 62%|██████▏   | 6625/10707 [58:52<33:38,  2.02it/s]
                                                    
{'loss': 3.667, 'grad_norm': 0.20763695240020752, 'learning_rate': 0.00038117024902187057, 'epoch': 0.62}
-
 62%|██████▏   | 6625/10707 [58:52<33:38,  2.02it/s]
 62%|██████▏   | 6626/10707 [58:53<33:42,  2.02it/s]
 62%|██████▏   | 6627/10707 [58:53<33:40,  2.02it/s]
 62%|██████▏   | 6628/10707 [58:54<33:40,  2.02it/s]
 62%|██████▏   | 6629/10707 [58:54<33:39,  2.02it/s]
 62%|██████▏   | 6630/10707 [58:55<33:38,  2.02it/s]
 62%|██████▏   | 6631/10707 [58:55<33:38,  2.02it/s]
 62%|██████▏   | 6632/10707 [58:56<33:37,  2.02it/s]
 62%|██████▏   | 6633/10707 [58:56<33:39,  2.02it/s]
 62%|██████▏   | 6634/10707 [58:57<33:37,  2.02it/s]
 62%|██████▏   | 6635/10707 [58:57<33:38,  2.02it/s]
 62%|██████▏   | 6636/10707 [58:58<33:34,  2.02it/s]
 62%|██████▏   | 6637/10707 [58:58<33:34,  2.02it/s]
 62%|██████▏   | 6638/10707 [58:59<33:31,  2.02it/s]
 62%|██████▏   | 6639/10707 [58:59<33:34,  2.02it/s]
 62%|██████▏   | 6640/10707 [59:00<33:32,  2.02it/s]
 62%|██████▏   | 6641/10707 [59:00<33:34,  2.02it/s]
 62%|██████▏   | 6642/10707 [59:01<33:33,  2.02it/s]
 62%|██████▏   | 6643/10707 [59:01<33:33,  2.02it/s]
 62%|██████▏   | 6644/10707 [59:02<33:30,  2.02it/s]
 62%|██████▏   | 6645/10707 [59:02<33:29,  2.02it/s]
 62%|██████▏   | 6646/10707 [59:03<33:29,  2.02it/s]
 62%|██████▏   | 6647/10707 [59:03<33:27,  2.02it/s]
 62%|██████▏   | 6648/10707 [59:04<33:28,  2.02it/s]
 62%|██████▏   | 6649/10707 [59:04<33:29,  2.02it/s]
 62%|██████▏   | 6650/10707 [59:05<33:27,  2.02it/s]{'loss': 3.6678, 'grad_norm': 0.20027126371860504, 'learning_rate': 0.0003772156713147578, 'epoch': 0.62}
-
                                                    

 62%|██████▏   | 6650/10707 [59:05<33:27,  2.02it/s]
 62%|██████▏   | 6651/10707 [59:05<33:30,  2.02it/s]
 62%|██████▏   | 6652/10707 [59:06<33:30,  2.02it/s]
 62%|██████▏   | 6653/10707 [59:06<33:29,  2.02it/s]
 62%|██████▏   | 6654/10707 [59:06<33:29,  2.02it/s]
 62%|██████▏   | 6655/10707 [59:07<33:29,  2.02it/s]
 62%|██████▏   | 6656/10707 [59:07<33:26,  2.02it/s]
 62%|██████▏   | 6657/10707 [59:08<33:26,  2.02it/s]
 62%|██████▏   | 6658/10707 [59:08<33:24,  2.02it/s]
 62%|██████▏   | 6659/10707 [59:09<33:26,  2.02it/s]
 62%|██████▏   | 6660/10707 [59:09<33:21,  2.02it/s]
 62%|██████▏   | 6661/10707 [59:10<33:24,  2.02it/s]
 62%|██████▏   | 6662/10707 [59:10<33:21,  2.02it/s]
 62%|██████▏   | 6663/10707 [59:11<33:22,  2.02it/s]
 62%|██████▏   | 6664/10707 [59:11<33:20,  2.02it/s]
 62%|██████▏   | 6665/10707 [59:12<33:22,  2.02it/s]
 62%|██████▏   | 6666/10707 [59:12<33:19,  2.02it/s]
 62%|██████▏   | 6667/10707 [59:13<33:21,  2.02it/s]
 62%|██████▏   | 6668/10707 [59:13<33:18,  2.02it/s]
 62%|██████▏   | 6669/10707 [59:14<33:19,  2.02it/s]
 62%|██████▏   | 6670/10707 [59:14<33:16,  2.02it/s]
 62%|██████▏   | 6671/10707 [59:15<33:17,  2.02it/s]
 62%|██████▏   | 6672/10707 [59:15<33:16,  2.02it/s]
 62%|██████▏   | 6673/10707 [59:16<33:17,  2.02it/s]
 62%|██████▏   | 6674/10707 [59:16<33:17,  2.02it/s]
 62%|██████▏   | 6675/10707 [59:17<33:16,  2.02it/s]{'loss': 3.6631, 'grad_norm': 0.19815798103809357, 'learning_rate': 0.00037326925053727893, 'epoch': 0.62}
                                                    
-
 62%|██████▏   | 6675/10707 [59:17<33:16,  2.02it/s]
 62%|██████▏   | 6676/10707 [59:17<33:21,  2.01it/s]
 62%|██████▏   | 6677/10707 [59:18<33:16,  2.02it/s]
 62%|██████▏   | 6678/10707 [59:18<33:17,  2.02it/s]
 62%|██████▏   | 6679/10707 [59:19<33:13,  2.02it/s]
 62%|██████▏   | 6680/10707 [59:19<33:15,  2.02it/s]
 62%|██████▏   | 6681/10707 [59:20<33:12,  2.02it/s]
 62%|██████▏   | 6682/10707 [59:20<33:12,  2.02it/s]
 62%|██████▏   | 6683/10707 [59:21<33:10,  2.02it/s]
 62%|██████▏   | 6684/10707 [59:21<33:11,  2.02it/s]
 62%|██████▏   | 6685/10707 [59:22<33:10,  2.02it/s]
 62%|██████▏   | 6686/10707 [59:22<33:11,  2.02it/s]
 62%|██████▏   | 6687/10707 [59:23<33:10,  2.02it/s]
 62%|██████▏   | 6688/10707 [59:23<33:10,  2.02it/s]
 62%|██████▏   | 6689/10707 [59:24<33:09,  2.02it/s]
 62%|██████▏   | 6690/10707 [59:24<33:09,  2.02it/s]
 62%|██████▏   | 6691/10707 [59:25<33:06,  2.02it/s]
 63%|██████▎   | 6692/10707 [59:25<33:08,  2.02it/s]
 63%|██████▎   | 6693/10707 [59:26<33:06,  2.02it/s]
 63%|██████▎   | 6694/10707 [59:26<33:08,  2.02it/s]
 63%|██████▎   | 6695/10707 [59:27<33:05,  2.02it/s]
 63%|██████▎   | 6696/10707 [59:27<33:06,  2.02it/s]
 63%|██████▎   | 6697/10707 [59:28<33:09,  2.02it/s]
 63%|██████▎   | 6698/10707 [59:28<33:07,  2.02it/s]
 63%|██████▎   | 6699/10707 [59:29<33:05,  2.02it/s]
 63%|██████▎   | 6700/10707 [59:29<33:04,  2.02it/s]{'loss': 3.6699, 'grad_norm': 0.19983115792274475, 'learning_rate': 0.00036933124886195197, 'epoch': 0.63}
-
                                                    

 63%|██████▎   | 6700/10707 [59:29<33:04,  2.02it/s]
 63%|██████▎   | 6701/10707 [59:30<33:05,  2.02it/s]
 63%|██████▎   | 6702/10707 [59:30<33:06,  2.02it/s]
 63%|██████▎   | 6703/10707 [59:31<33:04,  2.02it/s]
 63%|██████▎   | 6704/10707 [59:31<33:04,  2.02it/s]
 63%|██████▎   | 6705/10707 [59:32<33:01,  2.02it/s]
 63%|██████▎   | 6706/10707 [59:32<33:02,  2.02it/s]
 63%|██████▎   | 6707/10707 [59:33<33:01,  2.02it/s]
 63%|██████▎   | 6708/10707 [59:33<33:00,  2.02it/s]
 63%|██████▎   | 6709/10707 [59:34<33:01,  2.02it/s]
 63%|██████▎   | 6710/10707 [59:34<33:01,  2.02it/s]
 63%|██████▎   | 6711/10707 [59:35<32:59,  2.02it/s]
 63%|██████▎   | 6712/10707 [59:35<33:01,  2.02it/s]
 63%|██████▎   | 6713/10707 [59:36<32:59,  2.02it/s]
 63%|██████▎   | 6714/10707 [59:36<32:59,  2.02it/s]
 63%|██████▎   | 6715/10707 [59:37<32:57,  2.02it/s]
 63%|██████▎   | 6716/10707 [59:37<32:57,  2.02it/s]
 63%|██████▎   | 6717/10707 [59:38<32:56,  2.02it/s]
 63%|██████▎   | 6718/10707 [59:38<32:55,  2.02it/s]
 63%|██████▎   | 6719/10707 [59:39<32:52,  2.02it/s]
 63%|██████▎   | 6720/10707 [59:39<32:53,  2.02it/s]
 63%|██████▎   | 6721/10707 [59:40<32:50,  2.02it/s]
 63%|██████▎   | 6722/10707 [59:40<32:51,  2.02it/s]
 63%|██████▎   | 6723/10707 [59:41<32:51,  2.02it/s]
 63%|██████▎   | 6724/10707 [59:41<32:52,  2.02it/s]
 63%|██████▎   | 6725/10707 [59:42<32:50,  2.02it/s]{'loss': 3.6578, 'grad_norm': 0.20259660482406616, 'learning_rate': 0.00036540192790198805, 'epoch': 0.63}
-
                                                    

 63%|██████▎   | 6725/10707 [59:42<32:50,  2.02it/s]
 63%|██████▎   | 6726/10707 [59:42<32:53,  2.02it/s]
 63%|██████▎   | 6727/10707 [59:43<32:51,  2.02it/s]
 63%|██████▎   | 6728/10707 [59:43<32:51,  2.02it/s]
 63%|██████▎   | 6729/10707 [59:44<32:50,  2.02it/s]
 63%|██████▎   | 6730/10707 [59:44<32:52,  2.02it/s]
 63%|██████▎   | 6731/10707 [59:45<32:50,  2.02it/s]
 63%|██████▎   | 6732/10707 [59:45<32:49,  2.02it/s]
 63%|██████▎   | 6733/10707 [59:46<32:50,  2.02it/s]
 63%|██████▎   | 6734/10707 [59:46<32:48,  2.02it/s]
 63%|██████▎   | 6735/10707 [59:47<32:47,  2.02it/s]
 63%|██████▎   | 6736/10707 [59:47<32:45,  2.02it/s]
 63%|██████▎   | 6737/10707 [59:48<32:46,  2.02it/s]
 63%|██████▎   | 6738/10707 [59:48<32:47,  2.02it/s]
 63%|██████▎   | 6739/10707 [59:49<32:46,  2.02it/s]
 63%|██████▎   | 6740/10707 [59:49<32:45,  2.02it/s]
 63%|██████▎   | 6741/10707 [59:50<32:44,  2.02it/s]
 63%|██████▎   | 6742/10707 [59:50<32:44,  2.02it/s]
 63%|██████▎   | 6743/10707 [59:51<32:42,  2.02it/s]
 63%|██████▎   | 6744/10707 [59:51<32:42,  2.02it/s]
 63%|██████▎   | 6745/10707 [59:52<32:41,  2.02it/s]
 63%|██████▎   | 6746/10707 [59:52<32:40,  2.02it/s]
 63%|██████▎   | 6747/10707 [59:53<32:39,  2.02it/s]
 63%|██████▎   | 6748/10707 [59:53<32:40,  2.02it/s]
 63%|██████▎   | 6749/10707 [59:54<32:38,  2.02it/s]
 63%|██████▎   | 6750/10707 [59:54<32:38,  2.02it/s]{'loss': 3.6665, 'grad_norm': 0.20548219978809357, 'learning_rate': 0.00036148154869391363, 'epoch': 0.63}
-
                                                    

 63%|██████▎   | 6750/10707 [59:54<32:38,  2.02it/s]
 63%|██████▎   | 6751/10707 [59:55<32:37,  2.02it/s]
 63%|██████▎   | 6752/10707 [59:55<32:38,  2.02it/s]
 63%|██████▎   | 6753/10707 [59:56<32:36,  2.02it/s]
 63%|██████▎   | 6754/10707 [59:56<32:37,  2.02it/s]
 63%|██████▎   | 6755/10707 [59:57<32:37,  2.02it/s]
 63%|██████▎   | 6756/10707 [59:57<32:37,  2.02it/s]
 63%|██████▎   | 6757/10707 [59:58<32:33,  2.02it/s]
 63%|██████▎   | 6758/10707 [59:58<32:35,  2.02it/s]
 63%|██████▎   | 6759/10707 [59:59<32:42,  2.01it/s]
 63%|██████▎   | 6760/10707 [59:59<32:38,  2.02it/s]
 63%|██████▎   | 6761/10707 [59:59<32:38,  2.01it/s]
 63%|██████▎   | 6762/10707 [1:00:00<32:34,  2.02it/s]
 63%|██████▎   | 6763/10707 [1:00:00<32:33,  2.02it/s]
 63%|██████▎   | 6764/10707 [1:00:01<32:31,  2.02it/s]
 63%|██████▎   | 6765/10707 [1:00:01<32:31,  2.02it/s]
 63%|██████▎   | 6766/10707 [1:00:02<32:30,  2.02it/s]
 63%|██████▎   | 6767/10707 [1:00:02<32:30,  2.02it/s]
 63%|██████▎   | 6768/10707 [1:00:03<32:27,  2.02it/s]
 63%|██████▎   | 6769/10707 [1:00:03<32:28,  2.02it/s]
 63%|██████▎   | 6770/10707 [1:00:04<32:27,  2.02it/s]
 63%|██████▎   | 6771/10707 [1:00:04<32:29,  2.02it/s]
 63%|██████▎   | 6772/10707 [1:00:05<32:30,  2.02it/s]
 63%|██████▎   | 6773/10707 [1:00:05<32:30,  2.02it/s]
 63%|██████▎   | 6774/10707 [1:00:06<32:29,  2.02it/s]
 63%|██████▎   | 6775/10707 [1:00:06<32:27,  2.02it/s]{'loss': 3.666, 'grad_norm': 0.2012614607810974, 'learning_rate': 0.0003575703716802268, 'epoch': 0.63}
                                                      
-
 63%|██████▎   | 6775/10707 [1:00:06<32:27,  2.02it/s]
 63%|██████▎   | 6776/10707 [1:00:07<32:28,  2.02it/s]
 63%|██████▎   | 6777/10707 [1:00:07<32:27,  2.02it/s]
 63%|██████▎   | 6778/10707 [1:00:08<32:25,  2.02it/s]
 63%|██████▎   | 6779/10707 [1:00:08<32:24,  2.02it/s]
 63%|██████▎   | 6780/10707 [1:00:09<32:24,  2.02it/s]
 63%|██████▎   | 6781/10707 [1:00:09<32:24,  2.02it/s]
 63%|██████▎   | 6782/10707 [1:00:10<37:35,  1.74it/s]
 63%|██████▎   | 6783/10707 [1:00:11<36:01,  1.82it/s]
 63%|██████▎   | 6784/10707 [1:00:11<34:57,  1.87it/s]
 63%|██████▎   | 6785/10707 [1:00:12<34:12,  1.91it/s]
 63%|██████▎   | 6786/10707 [1:00:12<33:39,  1.94it/s]
 63%|██████▎   | 6787/10707 [1:00:13<33:13,  1.97it/s]
 63%|██████▎   | 6788/10707 [1:00:13<32:57,  1.98it/s]
 63%|██████▎   | 6789/10707 [1:00:14<32:45,  1.99it/s]
 63%|██████▎   | 6790/10707 [1:00:14<32:37,  2.00it/s]
 63%|██████▎   | 6791/10707 [1:00:15<32:30,  2.01it/s]
 63%|██████▎   | 6792/10707 [1:00:15<32:27,  2.01it/s]
 63%|██████▎   | 6793/10707 [1:00:16<32:23,  2.01it/s]
 63%|██████▎   | 6794/10707 [1:00:16<32:22,  2.01it/s]
 63%|██████▎   | 6795/10707 [1:00:17<32:17,  2.02it/s]
 63%|██████▎   | 6796/10707 [1:00:17<32:19,  2.02it/s]
 63%|██████▎   | 6797/10707 [1:00:18<32:15,  2.02it/s]
 63%|██████▎   | 6798/10707 [1:00:18<32:16,  2.02it/s]
 64%|██████▎   | 6799/10707 [1:00:19<32:15,  2.02it/s]
 64%|██████▎   | 6800/10707 [1:00:19<32:16,  2.02it/s]{'loss': 3.6701, 'grad_norm': 0.20090706646442413, 'learning_rate': 0.000353668656692097, 'epoch': 0.64}
-
                                                      

 64%|██████▎   | 6800/10707 [1:00:19<32:16,  2.02it/s]
 64%|██████▎   | 6801/10707 [1:00:20<32:17,  2.02it/s]
 64%|██████▎   | 6802/10707 [1:00:20<32:17,  2.02it/s]
 64%|██████▎   | 6803/10707 [1:00:21<32:15,  2.02it/s]
 64%|██████▎   | 6804/10707 [1:00:21<32:14,  2.02it/s]
 64%|██████▎   | 6805/10707 [1:00:22<32:13,  2.02it/s]
 64%|██████▎   | 6806/10707 [1:00:22<32:11,  2.02it/s]
 64%|██████▎   | 6807/10707 [1:00:23<32:12,  2.02it/s]
 64%|██████▎   | 6808/10707 [1:00:23<32:10,  2.02it/s]
 64%|██████▎   | 6809/10707 [1:00:24<32:11,  2.02it/s]
 64%|██████▎   | 6810/10707 [1:00:24<32:08,  2.02it/s]
 64%|██████▎   | 6811/10707 [1:00:25<32:09,  2.02it/s]
 64%|██████▎   | 6812/10707 [1:00:25<37:17,  1.74it/s]
 64%|██████▎   | 6813/10707 [1:00:26<35:46,  1.81it/s]
 64%|██████▎   | 6814/10707 [1:00:26<34:40,  1.87it/s]
 64%|██████▎   | 6815/10707 [1:00:27<33:54,  1.91it/s]
 64%|██████▎   | 6816/10707 [1:00:27<33:22,  1.94it/s]
 64%|██████▎   | 6817/10707 [1:00:28<32:58,  1.97it/s]
 64%|██████▎   | 6818/10707 [1:00:28<32:41,  1.98it/s]
 64%|██████▎   | 6819/10707 [1:00:29<32:30,  1.99it/s]
 64%|██████▎   | 6820/10707 [1:00:29<32:21,  2.00it/s]
 64%|██████▎   | 6821/10707 [1:00:30<32:15,  2.01it/s]
 64%|██████▎   | 6822/10707 [1:00:30<32:10,  2.01it/s]
 64%|██████▎   | 6823/10707 [1:00:31<32:07,  2.01it/s]
 64%|██████▎   | 6824/10707 [1:00:31<32:02,  2.02it/s]
 64%|██████▎   | 6825/10707 [1:00:32<32:02,  2.02it/s]{'loss': 3.6586, 'grad_norm': 0.19795367121696472, 'learning_rate': 0.0003497766629321025, 'epoch': 0.64}
-
                                                      

 64%|██████▎   | 6825/10707 [1:00:32<32:02,  2.02it/s]
 64%|██████▍   | 6826/10707 [1:00:32<32:11,  2.01it/s]
 64%|██████▍   | 6827/10707 [1:00:33<32:06,  2.01it/s]
 64%|██████▍   | 6828/10707 [1:00:33<32:06,  2.01it/s]
 64%|██████▍   | 6829/10707 [1:00:34<32:02,  2.02it/s]
 64%|██████▍   | 6830/10707 [1:00:34<32:01,  2.02it/s]
 64%|██████▍   | 6831/10707 [1:00:35<31:58,  2.02it/s]
 64%|██████▍   | 6832/10707 [1:00:35<31:59,  2.02it/s]
 64%|██████▍   | 6833/10707 [1:00:36<31:56,  2.02it/s]
 64%|██████▍   | 6834/10707 [1:00:36<31:57,  2.02it/s]
 64%|██████▍   | 6835/10707 [1:00:37<31:55,  2.02it/s]
 64%|██████▍   | 6836/10707 [1:00:37<31:54,  2.02it/s]
 64%|██████▍   | 6837/10707 [1:00:38<31:54,  2.02it/s]
 64%|██████▍   | 6838/10707 [1:00:38<31:55,  2.02it/s]
 64%|██████▍   | 6839/10707 [1:00:39<31:53,  2.02it/s]
 64%|██████▍   | 6840/10707 [1:00:39<31:52,  2.02it/s]
 64%|██████▍   | 6841/10707 [1:00:40<31:52,  2.02it/s]
 64%|██████▍   | 6842/10707 [1:00:40<31:52,  2.02it/s]
 64%|██████▍   | 6843/10707 [1:00:41<31:49,  2.02it/s]
 64%|██████▍   | 6844/10707 [1:00:41<31:50,  2.02it/s]
 64%|██████▍   | 6845/10707 [1:00:42<31:49,  2.02it/s]
 64%|██████▍   | 6846/10707 [1:00:42<31:48,  2.02it/s]
 64%|██████▍   | 6847/10707 [1:00:43<31:47,  2.02it/s]
 64%|██████▍   | 6848/10707 [1:00:43<31:49,  2.02it/s]
 64%|██████▍   | 6849/10707 [1:00:44<31:46,  2.02it/s]
 64%|██████▍   | 6850/10707 [1:00:44<31:47,  2.02it/s]{'loss': 3.6634, 'grad_norm': 0.208807572722435, 'learning_rate': 0.00034589464895701206, 'epoch': 0.64}
-
                                                      

 64%|██████▍   | 6850/10707 [1:00:44<31:47,  2.02it/s]
 64%|██████▍   | 6851/10707 [1:00:45<31:48,  2.02it/s]
 64%|██████▍   | 6852/10707 [1:00:45<31:48,  2.02it/s]
 64%|██████▍   | 6853/10707 [1:00:46<31:46,  2.02it/s]
 64%|██████▍   | 6854/10707 [1:00:46<31:47,  2.02it/s]
 64%|██████▍   | 6855/10707 [1:00:47<31:44,  2.02it/s]
 64%|██████▍   | 6856/10707 [1:00:47<31:45,  2.02it/s]
 64%|██████▍   | 6857/10707 [1:00:48<31:43,  2.02it/s]
 64%|██████▍   | 6858/10707 [1:00:48<31:44,  2.02it/s]
 64%|██████▍   | 6859/10707 [1:00:49<31:43,  2.02it/s]
 64%|██████▍   | 6860/10707 [1:00:49<31:44,  2.02it/s]
 64%|██████▍   | 6861/10707 [1:00:50<31:43,  2.02it/s]
 64%|██████▍   | 6862/10707 [1:00:50<31:42,  2.02it/s]
 64%|██████▍   | 6863/10707 [1:00:51<31:39,  2.02it/s]
 64%|██████▍   | 6864/10707 [1:00:51<31:40,  2.02it/s]
 64%|██████▍   | 6865/10707 [1:00:52<31:38,  2.02it/s]
 64%|██████▍   | 6866/10707 [1:00:52<31:41,  2.02it/s]
 64%|██████▍   | 6867/10707 [1:00:52<31:40,  2.02it/s]
 64%|██████▍   | 6868/10707 [1:00:53<31:39,  2.02it/s]
 64%|██████▍   | 6869/10707 [1:00:53<31:39,  2.02it/s]
 64%|██████▍   | 6870/10707 [1:00:54<31:38,  2.02it/s]
 64%|██████▍   | 6871/10707 [1:00:54<31:37,  2.02it/s]
 64%|██████▍   | 6872/10707 [1:00:55<31:37,  2.02it/s]
 64%|██████▍   | 6873/10707 [1:00:55<31:35,  2.02it/s]
 64%|██████▍   | 6874/10707 [1:00:56<31:36,  2.02it/s]
 64%|██████▍   | 6875/10707 [1:00:56<31:35,  2.02it/s]{'loss': 3.6552, 'grad_norm': 0.20368102192878723, 'learning_rate': 0.0003420228726606072, 'epoch': 0.64}
-
                                                      

 64%|██████▍   | 6875/10707 [1:00:56<31:35,  2.02it/s]
 64%|██████▍   | 6876/10707 [1:00:57<31:36,  2.02it/s]
 64%|██████▍   | 6877/10707 [1:00:57<31:36,  2.02it/s]
 64%|██████▍   | 6878/10707 [1:00:58<31:33,  2.02it/s]
 64%|██████▍   | 6879/10707 [1:00:58<31:34,  2.02it/s]
 64%|██████▍   | 6880/10707 [1:00:59<31:32,  2.02it/s]
 64%|██████▍   | 6881/10707 [1:00:59<31:33,  2.02it/s]
 64%|██████▍   | 6882/10707 [1:01:00<31:30,  2.02it/s]
 64%|██████▍   | 6883/10707 [1:01:00<31:30,  2.02it/s]
 64%|██████▍   | 6884/10707 [1:01:01<31:30,  2.02it/s]
 64%|██████▍   | 6885/10707 [1:01:01<31:30,  2.02it/s]
 64%|██████▍   | 6886/10707 [1:01:02<31:30,  2.02it/s]
 64%|██████▍   | 6887/10707 [1:01:02<31:29,  2.02it/s]
 64%|██████▍   | 6888/10707 [1:01:03<31:29,  2.02it/s]
 64%|██████▍   | 6889/10707 [1:01:03<31:29,  2.02it/s]
 64%|██████▍   | 6890/10707 [1:01:04<31:28,  2.02it/s]
 64%|██████▍   | 6891/10707 [1:01:04<31:28,  2.02it/s]
 64%|██████▍   | 6892/10707 [1:01:05<31:27,  2.02it/s]
 64%|██████▍   | 6893/10707 [1:01:05<31:26,  2.02it/s]
 64%|██████▍   | 6894/10707 [1:01:06<31:26,  2.02it/s]
 64%|██████▍   | 6895/10707 [1:01:06<31:25,  2.02it/s]
 64%|██████▍   | 6896/10707 [1:01:07<31:25,  2.02it/s]
 64%|██████▍   | 6897/10707 [1:01:07<31:25,  2.02it/s]
 64%|██████▍   | 6898/10707 [1:01:08<31:25,  2.02it/s]
 64%|██████▍   | 6899/10707 [1:01:08<31:24,  2.02it/s]
 64%|██████▍   | 6900/10707 [1:01:09<31:25,  2.02it/s]{'loss': 3.6664, 'grad_norm': 0.2008521556854248, 'learning_rate': 0.00033816159125654954, 'epoch': 0.64}
-
                                                      

 64%|██████▍   | 6900/10707 [1:01:09<31:25,  2.02it/s]
 64%|██████▍   | 6901/10707 [1:01:09<31:28,  2.02it/s]
 64%|██████▍   | 6902/10707 [1:01:10<31:26,  2.02it/s]
 64%|██████▍   | 6903/10707 [1:01:10<31:25,  2.02it/s]
 64%|██████▍   | 6904/10707 [1:01:11<31:22,  2.02it/s]
 64%|██████▍   | 6905/10707 [1:01:11<31:22,  2.02it/s]
 64%|██████▍   | 6906/10707 [1:01:12<31:20,  2.02it/s]
 65%|██████▍   | 6907/10707 [1:01:12<31:21,  2.02it/s]
 65%|██████▍   | 6908/10707 [1:01:13<31:20,  2.02it/s]
 65%|██████▍   | 6909/10707 [1:01:13<31:19,  2.02it/s]
 65%|██████▍   | 6910/10707 [1:01:14<31:19,  2.02it/s]
 65%|██████▍   | 6911/10707 [1:01:14<31:21,  2.02it/s]
 65%|██████▍   | 6912/10707 [1:01:15<31:18,  2.02it/s]
 65%|██████▍   | 6913/10707 [1:01:15<31:20,  2.02it/s]
 65%|██████▍   | 6914/10707 [1:01:16<31:16,  2.02it/s]
 65%|██████▍   | 6915/10707 [1:01:16<31:16,  2.02it/s]
 65%|██████▍   | 6916/10707 [1:01:17<31:13,  2.02it/s]
 65%|██████▍   | 6917/10707 [1:01:17<31:13,  2.02it/s]
 65%|██████▍   | 6918/10707 [1:01:18<31:14,  2.02it/s]
 65%|██████▍   | 6919/10707 [1:01:18<31:15,  2.02it/s]
 65%|██████▍   | 6920/10707 [1:01:19<31:12,  2.02it/s]
 65%|██████▍   | 6921/10707 [1:01:19<31:11,  2.02it/s]
 65%|██████▍   | 6922/10707 [1:01:20<31:10,  2.02it/s]
 65%|██████▍   | 6923/10707 [1:01:20<31:08,  2.02it/s]
 65%|██████▍   | 6924/10707 [1:01:21<31:10,  2.02it/s]
 65%|██████▍   | 6925/10707 [1:01:21<31:08,  2.02it/s]{'loss': 3.6601, 'grad_norm': 0.2025429606437683, 'learning_rate': 0.0003343110612612942, 'epoch': 0.65}
-
                                                      

 65%|██████▍   | 6925/10707 [1:01:21<31:08,  2.02it/s]
 65%|██████▍   | 6926/10707 [1:01:22<31:12,  2.02it/s]
 65%|██████▍   | 6927/10707 [1:01:22<31:09,  2.02it/s]
 65%|██████▍   | 6928/10707 [1:01:23<31:12,  2.02it/s]
 65%|██████▍   | 6929/10707 [1:01:23<31:11,  2.02it/s]
 65%|██████▍   | 6930/10707 [1:01:24<31:10,  2.02it/s]
 65%|██████▍   | 6931/10707 [1:01:24<31:09,  2.02it/s]
 65%|██████▍   | 6932/10707 [1:01:25<31:08,  2.02it/s]
 65%|██████▍   | 6933/10707 [1:01:25<31:08,  2.02it/s]
 65%|██████▍   | 6934/10707 [1:01:26<31:08,  2.02it/s]
 65%|██████▍   | 6935/10707 [1:01:26<31:07,  2.02it/s]
 65%|██████▍   | 6936/10707 [1:01:27<31:06,  2.02it/s]
 65%|██████▍   | 6937/10707 [1:01:27<31:04,  2.02it/s]
 65%|██████▍   | 6938/10707 [1:01:28<31:06,  2.02it/s]
 65%|██████▍   | 6939/10707 [1:01:28<31:08,  2.02it/s]
 65%|██████▍   | 6940/10707 [1:01:29<31:06,  2.02it/s]
 65%|██████▍   | 6941/10707 [1:01:29<31:04,  2.02it/s]
 65%|██████▍   | 6942/10707 [1:01:30<31:03,  2.02it/s]
 65%|██████▍   | 6943/10707 [1:01:30<31:03,  2.02it/s]
 65%|██████▍   | 6944/10707 [1:01:31<31:02,  2.02it/s]
 65%|██████▍   | 6945/10707 [1:01:31<31:03,  2.02it/s]
 65%|██████▍   | 6946/10707 [1:01:32<31:02,  2.02it/s]
 65%|██████▍   | 6947/10707 [1:01:32<31:03,  2.02it/s]
 65%|██████▍   | 6948/10707 [1:01:33<31:01,  2.02it/s]
 65%|██████▍   | 6949/10707 [1:01:33<31:03,  2.02it/s]
 65%|██████▍   | 6950/10707 [1:01:34<31:02,  2.02it/s]{'loss': 3.6604, 'grad_norm': 0.1972043216228485, 'learning_rate': 0.00033047153847704777, 'epoch': 0.65}
-
                                                      

 65%|██████▍   | 6950/10707 [1:01:34<31:02,  2.02it/s]
 65%|██████▍   | 6951/10707 [1:01:34<31:02,  2.02it/s]
 65%|██████▍   | 6952/10707 [1:01:35<31:00,  2.02it/s]
 65%|██████▍   | 6953/10707 [1:01:35<30:59,  2.02it/s]
 65%|██████▍   | 6954/10707 [1:01:36<30:59,  2.02it/s]
 65%|██████▍   | 6955/10707 [1:01:36<30:56,  2.02it/s]
 65%|██████▍   | 6956/10707 [1:01:37<30:56,  2.02it/s]
 65%|██████▍   | 6957/10707 [1:01:37<30:55,  2.02it/s]
 65%|██████▍   | 6958/10707 [1:01:38<30:55,  2.02it/s]
 65%|██████▍   | 6959/10707 [1:01:38<30:53,  2.02it/s]
 65%|██████▌   | 6960/10707 [1:01:39<30:55,  2.02it/s]
 65%|██████▌   | 6961/10707 [1:01:39<30:54,  2.02it/s]
 65%|██████▌   | 6962/10707 [1:01:40<30:53,  2.02it/s]
 65%|██████▌   | 6963/10707 [1:01:40<30:54,  2.02it/s]
 65%|██████▌   | 6964/10707 [1:01:41<30:54,  2.02it/s]
 65%|██████▌   | 6965/10707 [1:01:41<30:51,  2.02it/s]
 65%|██████▌   | 6966/10707 [1:01:42<30:51,  2.02it/s]
 65%|██████▌   | 6967/10707 [1:01:42<30:51,  2.02it/s]
 65%|██████▌   | 6968/10707 [1:01:42<30:50,  2.02it/s]
 65%|██████▌   | 6969/10707 [1:01:43<30:50,  2.02it/s]
 65%|██████▌   | 6970/10707 [1:01:43<30:50,  2.02it/s]
 65%|██████▌   | 6971/10707 [1:01:44<30:51,  2.02it/s]
 65%|██████▌   | 6972/10707 [1:01:44<30:50,  2.02it/s]
 65%|██████▌   | 6973/10707 [1:01:45<30:50,  2.02it/s]
 65%|██████▌   | 6974/10707 [1:01:45<30:48,  2.02it/s]
 65%|██████▌   | 6975/10707 [1:01:46<30:49,  2.02it/s]{'loss': 3.6606, 'grad_norm': 0.20020616054534912, 'learning_rate': 0.00032664327797477525, 'epoch': 0.65}
-
                                                      

 65%|██████▌   | 6975/10707 [1:01:46<30:49,  2.02it/s]
 65%|██████▌   | 6976/10707 [1:01:46<30:49,  2.02it/s]
 65%|██████▌   | 6977/10707 [1:01:47<30:49,  2.02it/s]
 65%|██████▌   | 6978/10707 [1:01:47<30:45,  2.02it/s]
 65%|██████▌   | 6979/10707 [1:01:48<30:46,  2.02it/s]
 65%|██████▌   | 6980/10707 [1:01:48<30:43,  2.02it/s]
 65%|██████▌   | 6981/10707 [1:01:49<30:44,  2.02it/s]
 65%|██████▌   | 6982/10707 [1:01:49<30:42,  2.02it/s]
 65%|██████▌   | 6983/10707 [1:01:50<30:43,  2.02it/s]
 65%|██████▌   | 6984/10707 [1:01:50<30:43,  2.02it/s]
 65%|██████▌   | 6985/10707 [1:01:51<30:43,  2.02it/s]
 65%|██████▌   | 6986/10707 [1:01:51<30:42,  2.02it/s]
 65%|██████▌   | 6987/10707 [1:01:52<30:40,  2.02it/s]
 65%|██████▌   | 6988/10707 [1:01:52<30:39,  2.02it/s]
 65%|██████▌   | 6989/10707 [1:01:53<30:39,  2.02it/s]
 65%|██████▌   | 6990/10707 [1:01:53<30:38,  2.02it/s]
 65%|██████▌   | 6991/10707 [1:01:54<30:38,  2.02it/s]
 65%|██���███▌   | 6992/10707 [1:01:54<30:37,  2.02it/s]
 65%|██████▌   | 6993/10707 [1:01:55<30:37,  2.02it/s]
 65%|██████▌   | 6994/10707 [1:01:55<30:39,  2.02it/s]
 65%|██████▌   | 6995/10707 [1:01:56<30:39,  2.02it/s]
 65%|██████▌   | 6996/10707 [1:01:56<30:36,  2.02it/s]
 65%|██████▌   | 6997/10707 [1:01:57<30:38,  2.02it/s]
 65%|██████▌   | 6998/10707 [1:01:57<30:38,  2.02it/s]
 65%|██████▌   | 6999/10707 [1:01:58<30:37,  2.02it/s]
 65%|██████▌   | 7000/10707 [1:01:58<30:36,  2.02it/s]{'loss': 3.653, 'grad_norm': 0.2046518623828888, 'learning_rate': 0.00032282653407725417, 'epoch': 0.65}
-
                                                      

 65%|██████▌   | 7000/10707 [1:01:58<30:36,  2.02it/s]
 65%|██████▌   | 7001/10707 [1:01:59<30:37,  2.02it/s]
 65%|██████▌   | 7002/10707 [1:01:59<30:37,  2.02it/s]
 65%|██████▌   | 7003/10707 [1:02:00<30:35,  2.02it/s]
 65%|██████▌   | 7004/10707 [1:02:00<30:33,  2.02it/s]
 65%|██████▌   | 7005/10707 [1:02:01<30:31,  2.02it/s]
 65%|██████▌   | 7006/10707 [1:02:01<30:33,  2.02it/s]
 65%|██████▌   | 7007/10707 [1:02:02<30:31,  2.02it/s]
 65%|██████▌   | 7008/10707 [1:02:02<30:31,  2.02it/s]
 65%|██████▌   | 7009/10707 [1:02:03<30:30,  2.02it/s]
 65%|██████▌   | 7010/10707 [1:02:03<30:29,  2.02it/s]
 65%|██████▌   | 7011/10707 [1:02:04<30:28,  2.02it/s]
 65%|██████▌   | 7012/10707 [1:02:04<30:28,  2.02it/s]
 65%|██████▌   | 7013/10707 [1:02:05<30:26,  2.02it/s]
 66%|██████▌   | 7014/10707 [1:02:05<30:29,  2.02it/s]
 66%|██████▌   | 7015/10707 [1:02:06<30:27,  2.02it/s]
 66%|██████▌   | 7016/10707 [1:02:06<30:26,  2.02it/s]
 66%|██████▌   | 7017/10707 [1:02:07<30:25,  2.02it/s]
 66%|██████▌   | 7018/10707 [1:02:07<30:25,  2.02it/s]
 66%|██████▌   | 7019/10707 [1:02:08<30:24,  2.02it/s]
 66%|██████▌   | 7020/10707 [1:02:08<30:25,  2.02it/s]
 66%|██████▌   | 7021/10707 [1:02:09<30:24,  2.02it/s]
 66%|██████▌   | 7022/10707 [1:02:09<30:24,  2.02it/s]
 66%|██████▌   | 7023/10707 [1:02:10<30:21,  2.02it/s]
 66%|██████▌   | 7024/10707 [1:02:10<30:23,  2.02it/s]
 66%|██████▌   | 7025/10707 [1:02:11<30:20,  2.02it/s]{'loss': 3.654, 'grad_norm': 0.19894011318683624, 'learning_rate': 0.00031902156034218, 'epoch': 0.66}
-
                                                      

 66%|██████▌   | 7025/10707 [1:02:11<30:20,  2.02it/s]
 66%|██████▌   | 7026/10707 [1:02:11<30:25,  2.02it/s]
 66%|██████▌   | 7027/10707 [1:02:12<30:24,  2.02it/s]
 66%|██████▌   | 7028/10707 [1:02:12<30:23,  2.02it/s]
 66%|██████▌   | 7029/10707 [1:02:13<30:23,  2.02it/s]
 66%|██████▌   | 7030/10707 [1:02:13<30:20,  2.02it/s]
 66%|██████▌   | 7031/10707 [1:02:14<30:21,  2.02it/s]
 66%|██████▌   | 7032/10707 [1:02:14<30:18,  2.02it/s]
 66%|██████▌   | 7033/10707 [1:02:15<30:18,  2.02it/s]
 66%|██████▌   | 7034/10707 [1:02:15<30:16,  2.02it/s]
 66%|██████▌   | 7035/10707 [1:02:16<30:17,  2.02it/s]
 66%|██████▌   | 7036/10707 [1:02:16<30:17,  2.02it/s]
 66%|██████▌   | 7037/10707 [1:02:17<30:18,  2.02it/s]
 66%|██████▌   | 7038/10707 [1:02:17<30:15,  2.02it/s]
 66%|██████▌   | 7039/10707 [1:02:18<30:15,  2.02it/s]
 66%|██████▌   | 7040/10707 [1:02:18<30:14,  2.02it/s]
 66%|██████▌   | 7041/10707 [1:02:19<30:14,  2.02it/s]
 66%|██████▌   | 7042/10707 [1:02:19<30:11,  2.02it/s]
 66%|██████▌   | 7043/10707 [1:02:20<30:12,  2.02it/s]
 66%|██████▌   | 7044/10707 [1:02:20<30:11,  2.02it/s]
 66%|██████▌   | 7045/10707 [1:02:21<30:11,  2.02it/s]
 66%|██████▌   | 7046/10707 [1:02:21<30:09,  2.02it/s]
 66%|██████▌   | 7047/10707 [1:02:22<30:12,  2.02it/s]
 66%|██████▌   | 7048/10707 [1:02:22<30:09,  2.02it/s]
 66%|██████▌   | 7049/10707 [1:02:23<30:11,  2.02it/s]
 66%|██████▌   | 7050/10707 [1:02:23<30:08,  2.02it/s]
                                                      {'loss': 3.6535, 'grad_norm': 0.20083139836788177, 'learning_rate': 0.0003152286095453207, 'epoch': 0.66}
-
 66%|██████▌   | 7050/10707 [1:02:23<30:08,  2.02it/s]
 66%|██████▌   | 7051/10707 [1:02:24<30:11,  2.02it/s]
 66%|██████▌   | 7052/10707 [1:02:24<30:10,  2.02it/s]
 66%|██████▌   | 7053/10707 [1:02:25<30:09,  2.02it/s]
 66%|██████▌   | 7054/10707 [1:02:25<30:08,  2.02it/s]
 66%|██████▌   | 7055/10707 [1:02:26<30:09,  2.02it/s]
 66%|██████���   | 7056/10707 [1:02:26<30:08,  2.02it/s]
 66%|██████▌   | 7057/10707 [1:02:27<30:09,  2.02it/s]
 66%|██████▌   | 7058/10707 [1:02:27<30:08,  2.02it/s]
 66%|██████▌   | 7059/10707 [1:02:28<30:07,  2.02it/s]
 66%|██████▌   | 7060/10707 [1:02:28<30:05,  2.02it/s]
 66%|██████▌   | 7061/10707 [1:02:29<30:05,  2.02it/s]
 66%|██████▌   | 7062/10707 [1:02:29<30:04,  2.02it/s]
 66%|██████▌   | 7063/10707 [1:02:30<30:03,  2.02it/s]
 66%|██████▌   | 7064/10707 [1:02:30<30:02,  2.02it/s]
 66%|██████▌   | 7065/10707 [1:02:31<30:02,  2.02it/s]
 66%|██████▌   | 7066/10707 [1:02:31<30:00,  2.02it/s]
 66%|██████▌   | 7067/10707 [1:02:32<30:02,  2.02it/s]
 66%|██████▌   | 7068/10707 [1:02:32<30:00,  2.02it/s]
 66%|██████▌   | 7069/10707 [1:02:32<30:01,  2.02it/s]
 66%|██████▌   | 7070/10707 [1:02:33<29:59,  2.02it/s]
 66%|██████▌   | 7071/10707 [1:02:33<30:05,  2.01it/s]
 66%|██████▌   | 7072/10707 [1:02:34<30:02,  2.02it/s]
 66%|██████▌   | 7073/10707 [1:02:34<30:01,  2.02it/s]
 66%|██████▌   | 7074/10707 [1:02:35<30:01,  2.02it/s]
 66%|██████▌   | 7075/10707 [1:02:35<29:58,  2.02it/s]{'loss': 3.6534, 'grad_norm': 0.20796798169612885, 'learning_rate': 0.0003114479336637254, 'epoch': 0.66}
-
                                                      

 66%|██████▌   | 7075/10707 [1:02:35<29:58,  2.02it/s]
 66%|██████▌   | 7076/10707 [1:02:36<30:00,  2.02it/s]
 66%|██████▌   | 7077/10707 [1:02:36<29:59,  2.02it/s]
 66%|██████▌   | 7078/10707 [1:02:37<30:00,  2.02it/s]
 66%|██████▌   | 7079/10707 [1:02:37<29:58,  2.02it/s]
 66%|██████▌   | 7080/10707 [1:02:38<29:58,  2.02it/s]
 66%|██████▌   | 7081/10707 [1:02:38<29:56,  2.02it/s]
 66%|██████▌   | 7082/10707 [1:02:39<29:56,  2.02it/s]
 66%|██████▌   | 7083/10707 [1:02:39<29:56,  2.02it/s]
 66%|██████▌   | 7084/10707 [1:02:40<29:54,  2.02it/s]
 66%|██████▌   | 7085/10707 [1:02:40<29:55,  2.02it/s]
 66%|██████▌   | 7086/10707 [1:02:41<29:52,  2.02it/s]
 66%|██████▌   | 7087/10707 [1:02:41<29:53,  2.02it/s]
 66%|██████▌   | 7088/10707 [1:02:42<29:50,  2.02it/s]
 66%|██████▌   | 7089/10707 [1:02:42<29:51,  2.02it/s]
 66%|██████▌   | 7090/10707 [1:02:43<29:50,  2.02it/s]
 66%|██████▌   | 7091/10707 [1:02:43<29:51,  2.02it/s]
 66%|██████▌   | 7092/10707 [1:02:44<29:49,  2.02it/s]
 66%|██████▌   | 7093/10707 [1:02:44<29:51,  2.02it/s]
 66%|██████▋   | 7094/10707 [1:02:45<29:48,  2.02it/s]
 66%|██████▋   | 7095/10707 [1:02:45<29:49,  2.02it/s]
 66%|██████▋   | 7096/10707 [1:02:46<29:46,  2.02it/s]
 66%|██████▋   | 7097/10707 [1:02:46<29:48,  2.02it/s]
 66%|██████▋   | 7098/10707 [1:02:47<29:44,  2.02it/s]
 66%|██████▋   | 7099/10707 [1:02:47<29:44,  2.02it/s]
 66%|██████▋   | 7100/10707 [1:02:48<29:44,  2.02it/s]{'loss': 3.6489, 'grad_norm': 0.20896080136299133, 'learning_rate': 0.000307679783858983, 'epoch': 0.66}
                                                      
-
 66%|██████▋   | 7100/10707 [1:02:48<29:44,  2.02it/s]
 66%|██████▋   | 7101/10707 [1:02:48<29:47,  2.02it/s]
 66%|██████▋   | 7102/10707 [1:02:49<29:45,  2.02it/s]
 66%|██████▋   | 7103/10707 [1:02:49<29:46,  2.02it/s]
 66%|██████▋   | 7104/10707 [1:02:50<29:44,  2.02it/s]
 66%|██████▋   | 7105/10707 [1:02:50<29:44,  2.02it/s]
 66%|██████▋   | 7106/10707 [1:02:51<29:41,  2.02it/s]
 66%|██████▋   | 7107/10707 [1:02:51<29:42,  2.02it/s]
 66%|██████▋   | 7108/10707 [1:02:52<29:39,  2.02it/s]
 66%|██████▋   | 7109/10707 [1:02:52<29:39,  2.02it/s]
 66%|██████▋   | 7110/10707 [1:02:53<29:39,  2.02it/s]
 66%|██████▋   | 7111/10707 [1:02:53<29:40,  2.02it/s]
 66%|██████▋   | 7112/10707 [1:02:54<29:38,  2.02it/s]
 66%|██████▋   | 7113/10707 [1:02:54<29:38,  2.02it/s]
 66%|██████▋   | 7114/10707 [1:02:55<29:38,  2.02it/s]
 66%|██████▋   | 7115/10707 [1:02:55<29:37,  2.02it/s]
 66%|██████▋   | 7116/10707 [1:02:56<29:38,  2.02it/s]
 66%|██████▋   | 7117/10707 [1:02:56<29:36,  2.02it/s]
 66%|██████▋   | 7118/10707 [1:02:57<29:37,  2.02it/s]
 66%|██████▋   | 7119/10707 [1:02:57<29:34,  2.02it/s]
 66%|██████▋   | 7120/10707 [1:02:58<29:33,  2.02it/s]
 67%|██████▋   | 7121/10707 [1:02:58<29:32,  2.02it/s]
 67%|██████▋   | 7122/10707 [1:02:59<29:37,  2.02it/s]
 67%|██████▋   | 7123/10707 [1:02:59<29:33,  2.02it/s]
 67%|██████▋   | 7124/10707 [1:03:00<29:35,  2.02it/s]
 67%|██████▋   | 7125/10707 [1:03:00<29:32,  2.02it/s]{'loss': 3.6523, 'grad_norm': 0.19364063441753387, 'learning_rate': 0.0003039244104605384, 'epoch': 0.67}
-
                                                      

 67%|██████▋   | 7125/10707 [1:03:00<29:32,  2.02it/s]
 67%|██████▋   | 7126/10707 [1:03:01<29:39,  2.01it/s]
 67%|██████▋   | 7127/10707 [1:03:01<29:36,  2.02it/s]
 67%|██████▋   | 7128/10707 [1:03:02<29:32,  2.02it/s]
 67%|██████▋   | 7129/10707 [1:03:02<29:33,  2.02it/s]
 67%|██████▋   | 7130/10707 [1:03:03<29:30,  2.02it/s]
 67%|██████▋   | 7131/10707 [1:03:03<29:30,  2.02it/s]
 67%|██████▋   | 7132/10707 [1:03:04<29:28,  2.02it/s]
 67%|██████▋   | 7133/10707 [1:03:04<29:28,  2.02it/s]
 67%|██████▋   | 7134/10707 [1:03:05<29:26,  2.02it/s]
 67%|██████▋   | 7135/10707 [1:03:05<29:29,  2.02it/s]
 67%|██████▋   | 7136/10707 [1:03:06<29:27,  2.02it/s]
 67%|██████▋   | 7137/10707 [1:03:06<29:27,  2.02it/s]
 67%|██████▋   | 7138/10707 [1:03:07<29:27,  2.02it/s]
 67%|██████▋   | 7139/10707 [1:03:07<29:26,  2.02it/s]
 67%|██████▋   | 7140/10707 [1:03:08<29:25,  2.02it/s]
 67%|██████▋   | 7141/10707 [1:03:08<29:24,  2.02it/s]
 67%|██████▋   | 7142/10707 [1:03:09<29:23,  2.02it/s]
 67%|██████▋   | 7143/10707 [1:03:09<29:24,  2.02it/s]
 67%|██████▋   | 7144/10707 [1:03:10<29:23,  2.02it/s]
 67%|██████▋   | 7145/10707 [1:03:10<29:22,  2.02it/s]
 67%|██████▋   | 7146/10707 [1:03:11<29:23,  2.02it/s]
 67%|██████▋   | 7147/10707 [1:03:11<29:20,  2.02it/s]
 67%|██████▋   | 7148/10707 [1:03:12<29:23,  2.02it/s]
 67%|██████▋   | 7149/10707 [1:03:12<29:20,  2.02it/s]
 67%|██████▋   | 7150/10707 [1:03:13<29:20,  2.02it/s]{'loss': 3.6434, 'grad_norm': 0.2033149003982544, 'learning_rate': 0.000300182062949061, 'epoch': 0.67}
-
                                                      

 67%|██████▋   | 7150/10707 [1:03:13<29:20,  2.02it/s]
 67%|██████▋   | 7151/10707 [1:03:13<29:21,  2.02it/s]
 67%|██████▋   | 7152/10707 [1:03:14<29:22,  2.02it/s]
 67%|██████▋   | 7153/10707 [1:03:14<29:20,  2.02it/s]
 67%|██████▋   | 7154/10707 [1:03:15<29:20,  2.02it/s]
 67%|██████▋   | 7155/10707 [1:03:15<29:18,  2.02it/s]
 67%|██████▋   | 7156/10707 [1:03:16<29:18,  2.02it/s]
 67%|██████▋   | 7157/10707 [1:03:16<29:17,  2.02it/s]
 67%|██████▋   | 7158/10707 [1:03:17<29:17,  2.02it/s]
 67%|██████▋   | 7159/10707 [1:03:17<29:15,  2.02it/s]
 67%|██████▋   | 7160/10707 [1:03:18<29:16,  2.02it/s]
 67%|██████▋   | 7161/10707 [1:03:18<29:13,  2.02it/s]
 67%|██████▋   | 7162/10707 [1:03:19<29:14,  2.02it/s]
 67%|██████▋   | 7163/10707 [1:03:19<29:12,  2.02it/s]
 67%|██████▋   | 7164/10707 [1:03:20<29:12,  2.02it/s]
 67%|██████▋   | 7165/10707 [1:03:20<29:12,  2.02it/s]
 67%|██████▋   | 7166/10707 [1:03:21<29:13,  2.02it/s]
 67%|██████▋   | 7167/10707 [1:03:21<29:11,  2.02it/s]
 67%|██████▋   | 7168/10707 [1:03:22<29:11,  2.02it/s]
 67%|██████▋   | 7169/10707 [1:03:22<29:10,  2.02it/s]
 67%|██████▋   | 7170/10707 [1:03:23<29:10,  2.02it/s]
 67%|██████▋   | 7171/10707 [1:03:23<29:10,  2.02it/s]
 67%|██████▋   | 7172/10707 [1:03:23<29:09,  2.02it/s]
 67%|██████▋   | 7173/10707 [1:03:24<29:09,  2.02it/s]
 67%|██████▋   | 7174/10707 [1:03:24<29:09,  2.02it/s]
 67%|██████▋   | 7175/10707 [1:03:25<29:11,  2.02it/s]{'loss': 3.6513, 'grad_norm': 0.20410624146461487, 'learning_rate': 0.0002964529899398725, 'epoch': 0.67}
-
                                                      

 67%|██████▋   | 7175/10707 [1:03:25<29:11,  2.02it/s]
 67%|██████▋   | 7176/10707 [1:03:25<29:11,  2.02it/s]
 67%|██████▋   | 7177/10707 [1:03:26<29:10,  2.02it/s]
 67%|██████▋   | 7178/10707 [1:03:26<29:09,  2.02it/s]
 67%|██████▋   | 7179/10707 [1:03:27<29:08,  2.02it/s]
 67%|██████▋   | 7180/10707 [1:03:27<29:07,  2.02it/s]
 67%|██████▋   | 7181/10707 [1:03:28<29:06,  2.02it/s]
 67%|██████▋   | 7182/10707 [1:03:28<29:05,  2.02it/s]
 67%|██████▋   | 7183/10707 [1:03:29<29:06,  2.02it/s]
 67%|██████▋   | 7184/10707 [1:03:29<29:07,  2.02it/s]
 67%|██████▋   | 7185/10707 [1:03:30<29:04,  2.02it/s]
 67%|██████▋   | 7186/10707 [1:03:30<29:04,  2.02it/s]
 67%|██████▋   | 7187/10707 [1:03:31<29:03,  2.02it/s]
 67%|██████▋   | 7188/10707 [1:03:31<29:03,  2.02it/s]
 67%|██████▋   | 7189/10707 [1:03:32<29:01,  2.02it/s]
 67%|██████▋   | 7190/10707 [1:03:32<29:02,  2.02it/s]
 67%|██████▋   | 7191/10707 [1:03:33<29:00,  2.02it/s]
 67%|██████▋   | 7192/10707 [1:03:33<29:00,  2.02it/s]
 67%|██████▋   | 7193/10707 [1:03:34<28:58,  2.02it/s]
 67%|██████▋   | 7194/10707 [1:03:34<28:59,  2.02it/s]
 67%|██████▋   | 7195/10707 [1:03:35<28:57,  2.02it/s]
 67%|██████▋   | 7196/10707 [1:03:35<28:58,  2.02it/s]
 67%|██████▋   | 7197/10707 [1:03:36<28:56,  2.02it/s]
 67%|██████▋   | 7198/10707 [1:03:36<28:56,  2.02it/s]
 67%|██████▋   | 7199/10707 [1:03:37<28:56,  2.02it/s]
 67%|██████▋   | 7200/10707 [1:03:37<28:54,  2.02it/s]
                                                      
{'loss': 3.6453, 'grad_norm': 0.20454050600528717, 'learning_rate': 0.000292737439166429, 'epoch': 0.67}
-
 67%|██████▋   | 7200/10707 [1:03:37<28:54,  2.02it/s]
 67%|██████▋   | 7201/10707 [1:03:38<28:58,  2.02it/s]
 67%|██████▋   | 7202/10707 [1:03:38<28:58,  2.02it/s]
 67%|██████▋   | 7203/10707 [1:03:39<28:56,  2.02it/s]
 67%|██████▋   | 7204/10707 [1:03:39<28:56,  2.02it/s]
 67%|██████▋   | 7205/10707 [1:03:40<28:54,  2.02it/s]
 67%|██████▋   | 7206/10707 [1:03:40<28:56,  2.02it/s]
 67%|██████▋   | 7207/10707 [1:03:41<28:53,  2.02it/s]
 67%|██████▋   | 7208/10707 [1:03:41<28:53,  2.02it/s]
 67%|██████▋   | 7209/10707 [1:03:42<28:50,  2.02it/s]
 67%|██████▋   | 7210/10707 [1:03:42<28:51,  2.02it/s]
 67%|██████▋   | 7211/10707 [1:03:43<28:49,  2.02it/s]
 67%|██████▋   | 7212/10707 [1:03:43<28:50,  2.02it/s]
 67%|██████▋   | 7213/10707 [1:03:44<28:49,  2.02it/s]
 67%|██████▋   | 7214/10707 [1:03:44<28:51,  2.02it/s]
 67%|██████▋   | 7215/10707 [1:03:45<28:48,  2.02it/s]
 67%|██████▋   | 7216/10707 [1:03:45<28:49,  2.02it/s]
 67%|██████▋   | 7217/10707 [1:03:46<28:46,  2.02it/s]
 67%|██████▋   | 7218/10707 [1:03:46<28:48,  2.02it/s]
 67%|██████▋   | 7219/10707 [1:03:47<28:46,  2.02it/s]
 67%|██████▋   | 7220/10707 [1:03:47<28:47,  2.02it/s]
 67%|██████▋   | 7221/10707 [1:03:48<28:46,  2.02it/s]
 67%|██████▋   | 7222/10707 [1:03:48<28:45,  2.02it/s]
 67%|██████▋   | 7223/10707 [1:03:49<28:44,  2.02it/s]
 67%|██████▋   | 7224/10707 [1:03:49<28:44,  2.02it/s]
 67%|██████▋   | 7225/10707 [1:03:50<28:44,  2.02it/s]{'loss': 3.6487, 'grad_norm': 0.2020494043827057, 'learning_rate': 0.00028903565746386464, 'epoch': 0.67}
                                                      
-
 67%|██████▋   | 7225/10707 [1:03:50<28:44,  2.02it/s]
 67%|██████▋   | 7226/10707 [1:03:50<28:44,  2.02it/s]
 67%|██████▋   | 7227/10707 [1:03:51<28:43,  2.02it/s]
 68%|██████▊   | 7228/10707 [1:03:51<28:42,  2.02it/s]
 68%|██████▊   | 7229/10707 [1:03:52<28:41,  2.02it/s]
 68%|██████▊   | 7230/10707 [1:03:52<28:40,  2.02it/s]
 68%|██████▊   | 7231/10707 [1:03:53<28:41,  2.02it/s]
 68%|██████▊   | 7232/10707 [1:03:53<28:39,  2.02it/s]
 68%|██████▊   | 7233/10707 [1:03:54<28:42,  2.02it/s]
 68%|██████▊   | 7234/10707 [1:03:54<28:40,  2.02it/s]
 68%|██████▊   | 7235/10707 [1:03:55<28:40,  2.02it/s]
 68%|██████▊   | 7236/10707 [1:03:55<28:37,  2.02it/s]
 68%|██████▊   | 7237/10707 [1:03:56<28:37,  2.02it/s]
 68%|██████▊   | 7238/10707 [1:03:56<28:35,  2.02it/s]
 68%|██████▊   | 7239/10707 [1:03:57<28:39,  2.02it/s]
 68%|██████▊   | 7240/10707 [1:03:57<28:35,  2.02it/s]
 68%|██████▊   | 7241/10707 [1:03:58<28:35,  2.02it/s]
 68%|██████▊   | 7242/10707 [1:03:58<28:34,  2.02it/s]
 68%|██████▊   | 7243/10707 [1:03:59<28:35,  2.02it/s]
 68%|██████▊   | 7244/10707 [1:03:59<28:34,  2.02it/s]
 68%|██████▊   | 7245/10707 [1:04:00<28:34,  2.02it/s]
 68%|██████▊   | 7246/10707 [1:04:00<28:33,  2.02it/s]
 68%|██████▊   | 7247/10707 [1:04:01<28:31,  2.02it/s]
 68%|██████▊   | 7248/10707 [1:04:01<28:31,  2.02it/s]
 68%|██████▊   | 7249/10707 [1:04:02<28:30,  2.02it/s]
 68%|██████▊   | 7250/10707 [1:04:02<28:34,  2.02it/s]{'loss': 3.6403, 'grad_norm': 0.20050588250160217, 'learning_rate': 0.0002853478907525922, 'epoch': 0.68}
-
                                                      

 68%|██████▊   | 7250/10707 [1:04:02<28:34,  2.02it/s]
 68%|██████▊   | 7251/10707 [1:04:03<28:35,  2.01it/s]
 68%|██████▊   | 7252/10707 [1:04:03<28:34,  2.02it/s]
 68%|██████▊   | 7253/10707 [1:04:04<28:31,  2.02it/s]
 68%|██████▊   | 7254/10707 [1:04:04<28:32,  2.02it/s]
 68%|██████▊   | 7255/10707 [1:04:05<28:30,  2.02it/s]
 68%|██████▊   | 7256/10707 [1:04:05<28:30,  2.02it/s]
 68%|██████▊   | 7257/10707 [1:04:06<28:27,  2.02it/s]
 68%|██████▊   | 7258/10707 [1:04:06<28:27,  2.02it/s]
 68%|██████▊   | 7259/10707 [1:04:07<28:25,  2.02it/s]
 68%|██████▊   | 7260/10707 [1:04:07<28:25,  2.02it/s]
 68%|██████▊   | 7261/10707 [1:04:08<28:24,  2.02it/s]
 68%|██████▊   | 7262/10707 [1:04:08<28:25,  2.02it/s]
 68%|██████▊   | 7263/10707 [1:04:09<28:23,  2.02it/s]
 68%|██████▊   | 7264/10707 [1:04:09<28:23,  2.02it/s]
 68%|██████▊   | 7265/10707 [1:04:10<28:21,  2.02it/s]
 68%|██████▊   | 7266/10707 [1:04:10<28:24,  2.02it/s]
 68%|██████▊   | 7267/10707 [1:04:11<28:21,  2.02it/s]
 68%|██████▊   | 7268/10707 [1:04:11<28:21,  2.02it/s]
 68%|██████▊   | 7269/10707 [1:04:12<28:20,  2.02it/s]
 68%|██████▊   | 7270/10707 [1:04:12<28:20,  2.02it/s]
 68%|██████▊   | 7271/10707 [1:04:13<28:21,  2.02it/s]
 68%|██████▊   | 7272/10707 [1:04:13<28:20,  2.02it/s]
 68%|██████▊   | 7273/10707 [1:04:14<28:21,  2.02it/s]
 68%|██████▊   | 7274/10707 [1:04:14<28:20,  2.02it/s]
 68%|██████▊   | 7275/10707 [1:04:15<28:19,  2.02it/s]{'loss': 3.6393, 'grad_norm': 0.1978965401649475, 'learning_rate': 0.000281674384021968, 'epoch': 0.68}
-
                                                      

 68%|██████▊   | 7275/10707 [1:04:15<28:19,  2.02it/s]
 68%|██████▊   | 7276/10707 [1:04:15<28:21,  2.02it/s]
 68%|██████▊   | 7277/10707 [1:04:15<28:20,  2.02it/s]
 68%|██████▊   | 7278/10707 [1:04:16<28:19,  2.02it/s]
 68%|██████▊   | 7279/10707 [1:04:16<28:19,  2.02it/s]
 68%|██████▊   | 7280/10707 [1:04:17<28:19,  2.02it/s]
 68%|██████▊   | 7281/10707 [1:04:17<28:17,  2.02it/s]
 68%|██████▊   | 7282/10707 [1:04:18<28:18,  2.02it/s]
 68%|██████▊   | 7283/10707 [1:04:18<28:16,  2.02it/s]
 68%|██████▊   | 7284/10707 [1:04:19<28:16,  2.02it/s]
 68%|██████▊   | 7285/10707 [1:04:19<28:14,  2.02it/s]
 68%|██████▊   | 7286/10707 [1:04:20<28:13,  2.02it/s]
 68%|██████▊   | 7287/10707 [1:04:20<28:13,  2.02it/s]
 68%|██████▊   | 7288/10707 [1:04:21<28:12,  2.02it/s]
 68%|██████▊   | 7289/10707 [1:04:21<28:11,  2.02it/s]
 68%|██████▊   | 7290/10707 [1:04:22<28:12,  2.02it/s]
 68%|██████▊   | 7291/10707 [1:04:22<28:09,  2.02it/s]
 68%|██████▊   | 7292/10707 [1:04:23<28:11,  2.02it/s]
 68%|██████▊   | 7293/10707 [1:04:23<28:09,  2.02it/s]
 68%|██████▊   | 7294/10707 [1:04:24<28:10,  2.02it/s]
 68%|██████▊   | 7295/10707 [1:04:24<28:08,  2.02it/s]
 68%|██████▊   | 7296/10707 [1:04:25<28:09,  2.02it/s]
 68%|██████▊   | 7297/10707 [1:04:25<28:06,  2.02it/s]
 68%|██████▊   | 7298/10707 [1:04:26<28:06,  2.02it/s]
 68%|██████▊   | 7299/10707 [1:04:26<28:03,  2.02it/s]
 68%|██████▊   | 7300/10707 [1:04:27<28:05,  2.02it/s]{'loss': 3.6481, 'grad_norm': 0.20991620421409607, 'learning_rate': 0.00027801538131401444, 'epoch': 0.68}
-
                                                      

 68%|██████▊   | 7300/10707 [1:04:27<28:05,  2.02it/s]
 68%|██████▊   | 7301/10707 [1:04:27<28:04,  2.02it/s]
 68%|██████▊   | 7302/10707 [1:04:28<28:06,  2.02it/s]
 68%|██████▊   | 7303/10707 [1:04:28<28:03,  2.02it/s]
 68%|██████▊   | 7304/10707 [1:04:29<28:05,  2.02it/s]
 68%|██████▊   | 7305/10707 [1:04:29<28:03,  2.02it/s]
 68%|██████▊   | 7306/10707 [1:04:30<28:03,  2.02it/s]
 68%|██████▊   | 7307/10707 [1:04:30<28:01,  2.02it/s]
 68%|██████▊   | 7308/10707 [1:04:31<28:01,  2.02it/s]
 68%|██████▊   | 7309/10707 [1:04:31<28:00,  2.02it/s]
 68%|██████▊   | 7310/10707 [1:04:32<28:01,  2.02it/s]
 68%|██████▊   | 7311/10707 [1:04:32<27:58,  2.02it/s]
 68%|██████▊   | 7312/10707 [1:04:33<27:58,  2.02it/s]
 68%|██████▊   | 7313/10707 [1:04:33<27:56,  2.02it/s]
 68%|██████▊   | 7314/10707 [1:04:34<27:58,  2.02it/s]
 68%|██████▊   | 7315/10707 [1:04:34<27:57,  2.02it/s]
 68%|██████▊   | 7316/10707 [1:04:35<27:58,  2.02it/s]
 68%|██████▊   | 7317/10707 [1:04:35<27:56,  2.02it/s]
 68%|██████▊   | 7318/10707 [1:04:36<27:57,  2.02it/s]
 68%|██████▊   | 7319/10707 [1:04:36<27:55,  2.02it/s]
 68%|██████▊   | 7320/10707 [1:04:37<27:55,  2.02it/s]
 68%|██████▊   | 7321/10707 [1:04:37<27:54,  2.02it/s]
 68%|██████▊   | 7322/10707 [1:04:38<27:55,  2.02it/s]
 68%|██████▊   | 7323/10707 [1:04:38<27:55,  2.02it/s]
 68%|██████▊   | 7324/10707 [1:04:39<27:55,  2.02it/s]
 68%|██████▊   | 7325/10707 [1:04:39<27:56,  2.02it/s]{'loss': 3.6438, 'grad_norm': 0.20783311128616333, 'learning_rate': 0.0002743711257072087, 'epoch': 0.68}
-
                                                      

 68%|██████▊   | 7325/10707 [1:04:39<27:56,  2.02it/s]
 68%|██████▊   | 7326/10707 [1:04:40<27:57,  2.02it/s]
 68%|██████▊   | 7327/10707 [1:04:40<27:56,  2.02it/s]
 68%|██████▊   | 7328/10707 [1:04:41<27:55,  2.02it/s]
 68%|██████▊   | 7329/10707 [1:04:41<27:53,  2.02it/s]
 68%|██████▊   | 7330/10707 [1:04:42<27:51,  2.02it/s]
 68%|██████▊   | 7331/10707 [1:04:42<27:50,  2.02it/s]
 68%|██████▊   | 7332/10707 [1:04:43<27:51,  2.02it/s]
 68%|██████▊   | 7333/10707 [1:04:43<27:50,  2.02it/s]
 68%|██████▊   | 7334/10707 [1:04:44<27:50,  2.02it/s]
 69%|██████▊   | 7335/10707 [1:04:44<27:48,  2.02it/s]
 69%|██████▊   | 7336/10707 [1:04:45<27:47,  2.02it/s]
 69%|██████▊   | 7337/10707 [1:04:45<27:45,  2.02it/s]
 69%|██████▊   | 7338/10707 [1:04:46<27:48,  2.02it/s]
 69%|██████▊   | 7339/10707 [1:04:46<27:46,  2.02it/s]
 69%|██████▊   | 7340/10707 [1:04:47<27:46,  2.02it/s]
 69%|██████▊   | 7341/10707 [1:04:47<27:44,  2.02it/s]
 69%|██████▊   | 7342/10707 [1:04:48<27:43,  2.02it/s]
 69%|██████▊   | 7343/10707 [1:04:48<27:42,  2.02it/s]
 69%|██████▊   | 7344/10707 [1:04:49<27:44,  2.02it/s]
 69%|██████▊   | 7345/10707 [1:04:49<27:42,  2.02it/s]
 69%|██████▊   | 7346/10707 [1:04:50<27:42,  2.02it/s]
 69%|██████▊   | 7347/10707 [1:04:50<27:41,  2.02it/s]
 69%|██████▊   | 7348/10707 [1:04:51<27:42,  2.02it/s]
 69%|██████▊   | 7349/10707 [1:04:51<27:40,  2.02it/s]
 69%|██████▊   | 7350/10707 [1:04:52<27:41,  2.02it/s]{'loss': 3.6453, 'grad_norm': 0.2012365311384201, 'learning_rate': 0.0002707418593003331, 'epoch': 0.69}
-
                                                      

 69%|██████▊   | 7350/10707 [1:04:52<27:41,  2.02it/s]
 69%|██████▊   | 7351/10707 [1:04:52<27:40,  2.02it/s]
 69%|██████▊   | 7352/10707 [1:04:53<27:40,  2.02it/s]
 69%|██████▊   | 7353/10707 [1:04:53<27:38,  2.02it/s]
 69%|██████▊   | 7354/10707 [1:04:54<27:39,  2.02it/s]
 69%|██████▊   | 7355/10707 [1:04:54<27:39,  2.02it/s]
 69%|██████▊   | 7356/10707 [1:04:55<27:39,  2.02it/s]
 69%|██████▊   | 7357/10707 [1:04:55<27:37,  2.02it/s]
 69%|██████▊   | 7358/10707 [1:04:56<27:38,  2.02it/s]
 69%|██████▊   | 7359/10707 [1:04:56<27:37,  2.02it/s]
 69%|██████▊   | 7360/10707 [1:04:57<27:37,  2.02it/s]
 69%|██████▊   | 7361/10707 [1:04:57<27:37,  2.02it/s]
 69%|██████▉   | 7362/10707 [1:04:58<27:37,  2.02it/s]
 69%|██████▉   | 7363/10707 [1:04:58<27:37,  2.02it/s]
 69%|██████▉   | 7364/10707 [1:04:59<27:36,  2.02it/s]
 69%|██████▉   | 7365/10707 [1:04:59<27:36,  2.02it/s]
 69%|██████▉   | 7366/10707 [1:05:00<27:33,  2.02it/s]
 69%|██████▉   | 7367/10707 [1:05:00<27:36,  2.02it/s]
 69%|██████▉   | 7368/10707 [1:05:01<27:33,  2.02it/s]
 69%|██████▉   | 7369/10707 [1:05:01<27:34,  2.02it/s]
 69%|██████▉   | 7370/10707 [1:05:02<27:31,  2.02it/s]
 69%|██████▉   | 7371/10707 [1:05:02<27:32,  2.02it/s]
 69%|██████▉   | 7372/10707 [1:05:03<27:31,  2.02it/s]
 69%|██████▉   | 7373/10707 [1:05:03<27:31,  2.02it/s]
 69%|██████▉   | 7374/10707 [1:05:04<27:29,  2.02it/s]
 69%|██████▉   | 7375/10707 [1:05:04<27:30,  2.02it/s]
                                                      {'loss': 3.6418, 'grad_norm': 0.20302054286003113, 'learning_rate': 0.0002671278231963943, 'epoch': 0.69}
-
 69%|██████▉   | 7375/10707 [1:05:04<27:30,  2.02it/s]
 69%|██████▉   | 7376/10707 [1:05:05<27:28,  2.02it/s]
 69%|██████▉   | 7377/10707 [1:05:05<27:31,  2.02it/s]
 69%|██████▉   | 7378/10707 [1:05:05<27:28,  2.02it/s]
 69%|██████▉   | 7379/10707 [1:05:06<27:28,  2.02it/s]
 69%|██████▉   | 7380/10707 [1:05:06<27:27,  2.02it/s]
 69%|██████▉   | 7381/10707 [1:05:07<27:27,  2.02it/s]
 69%|██████▉   | 7382/10707 [1:05:07<27:26,  2.02it/s]
 69%|██████▉   | 7383/10707 [1:05:08<27:25,  2.02it/s]
 69%|██████▉   | 7384/10707 [1:05:08<27:24,  2.02it/s]
 69%|██████▉   | 7385/10707 [1:05:09<27:23,  2.02it/s]
 69%|██████▉   | 7386/10707 [1:05:09<27:22,  2.02it/s]
 69%|██████▉   | 7387/10707 [1:05:10<27:22,  2.02it/s]
 69%|██████▉   | 7388/10707 [1:05:10<27:22,  2.02it/s]
 69%|██████▉   | 7389/10707 [1:05:11<27:21,  2.02it/s]
 69%|██████▉   | 7390/10707 [1:05:11<27:22,  2.02it/s]
 69%|██████▉   | 7391/10707 [1:05:12<27:21,  2.02it/s]
 69%|██████▉   | 7392/10707 [1:05:12<27:22,  2.02it/s]
 69%|██████▉   | 7393/10707 [1:05:13<27:21,  2.02it/s]
 69%|██████▉   | 7394/10707 [1:05:13<27:21,  2.02it/s]
 69%|██████▉   | 7395/10707 [1:05:14<27:18,  2.02it/s]
 69%|██████▉   | 7396/10707 [1:05:14<27:19,  2.02it/s]
 69%|██████▉   | 7397/10707 [1:05:15<27:17,  2.02it/s]
 69%|██████▉   | 7398/10707 [1:05:15<27:19,  2.02it/s]
 69%|██████▉   | 7399/10707 [1:05:16<27:16,  2.02it/s]
 69%|██████▉   | 7400/10707 [1:05:16<27:17,  2.02it/s]{'loss': 3.6403, 'grad_norm': 0.19571241736412048, 'learning_rate': 0.00026352925748660314, 'epoch': 0.69}
-
                                                      

 69%|██████▉   | 7400/10707 [1:05:16<27:17,  2.02it/s]
 69%|██████▉   | 7401/10707 [1:05:17<27:18,  2.02it/s]
 69%|██████▉   | 7402/10707 [1:05:17<27:17,  2.02it/s]
 69%|██████▉   | 7403/10707 [1:05:18<27:14,  2.02it/s]
 69%|██████▉   | 7404/10707 [1:05:18<27:14,  2.02it/s]
 69%|██████▉   | 7405/10707 [1:05:19<27:13,  2.02it/s]
 69%|██████▉   | 7406/10707 [1:05:19<27:16,  2.02it/s]
 69%|██████▉   | 7407/10707 [1:05:20<27:13,  2.02it/s]
 69%|██████▉   | 7408/10707 [1:05:20<27:13,  2.02it/s]
 69%|██████▉   | 7409/10707 [1:05:21<27:13,  2.02it/s]
 69%|██████▉   | 7410/10707 [1:05:21<27:13,  2.02it/s]
 69%|██████▉   | 7411/10707 [1:05:22<27:13,  2.02it/s]
 69%|██████▉   | 7412/10707 [1:05:22<27:12,  2.02it/s]
 69%|██████▉   | 7413/10707 [1:05:23<27:10,  2.02it/s]
 69%|██████▉   | 7414/10707 [1:05:23<27:10,  2.02it/s]
 69%|██████▉   | 7415/10707 [1:05:24<27:11,  2.02it/s]
 69%|██████▉   | 7416/10707 [1:05:24<27:10,  2.02it/s]
 69%|██████▉   | 7417/10707 [1:05:25<27:09,  2.02it/s]
 69%|██████▉   | 7418/10707 [1:05:25<27:09,  2.02it/s]
 69%|██████▉   | 7419/10707 [1:05:26<27:08,  2.02it/s]
 69%|██████▉   | 7420/10707 [1:05:26<27:08,  2.02it/s]
 69%|██████▉   | 7421/10707 [1:05:27<27:07,  2.02it/s]
 69%|██████▉   | 7422/10707 [1:05:27<27:07,  2.02it/s]
 69%|██████▉   | 7423/10707 [1:05:28<27:06,  2.02it/s]
 69%|██████▉   | 7424/10707 [1:05:28<27:06,  2.02it/s]
 69%|██████▉   | 7425/10707 [1:05:29<27:04,  2.02it/s]{'loss': 3.6366, 'grad_norm': 0.20488198101520538, 'learning_rate': 0.0002599464012344264, 'epoch': 0.69}
-
                                                      

 69%|██████▉   | 7425/10707 [1:05:29<27:04,  2.02it/s]
 69%|██████▉   | 7426/10707 [1:05:29<27:07,  2.02it/s]
 69%|██████▉   | 7427/10707 [1:05:30<27:03,  2.02it/s]
 69%|██████▉   | 7428/10707 [1:05:30<27:03,  2.02it/s]
 69%|██████▉   | 7429/10707 [1:05:31<27:01,  2.02it/s]
 69%|██████▉   | 7430/10707 [1:05:31<27:04,  2.02it/s]
 69%|██████▉   | 7431/10707 [1:05:32<27:02,  2.02it/s]
 69%|██████▉   | 7432/10707 [1:05:32<27:02,  2.02it/s]
 69%|██████▉   | 7433/10707 [1:05:33<27:01,  2.02it/s]
 69%|██████▉   | 7434/10707 [1:05:33<27:01,  2.02it/s]
 69%|██████▉   | 7435/10707 [1:05:34<26:59,  2.02it/s]
 69%|██████▉   | 7436/10707 [1:05:34<27:00,  2.02it/s]
 69%|██████▉   | 7437/10707 [1:05:35<26:59,  2.02it/s]
 69%|██████▉   | 7438/10707 [1:05:35<26:58,  2.02it/s]
 69%|██████▉   | 7439/10707 [1:05:36<26:59,  2.02it/s]
 69%|██████▉   | 7440/10707 [1:05:36<26:58,  2.02it/s]
 69%|██████▉   | 7441/10707 [1:05:37<26:58,  2.02it/s]
 70%|██████▉   | 7442/10707 [1:05:37<26:56,  2.02it/s]
 70%|██████▉   | 7443/10707 [1:05:38<26:56,  2.02it/s]
 70%|██████▉   | 7444/10707 [1:05:38<26:54,  2.02it/s]
 70%|██████▉   | 7445/10707 [1:05:39<26:54,  2.02it/s]
 70%|██████▉   | 7446/10707 [1:05:39<26:53,  2.02it/s]
 70%|██████▉   | 7447/10707 [1:05:40<26:52,  2.02it/s]
 70%|██████▉   | 7448/10707 [1:05:40<26:51,  2.02it/s]
 70%|██████▉   | 7449/10707 [1:05:41<26:52,  2.02it/s]
 70%|██████▉   | 7450/10707 [1:05:41<26:51,  2.02it/s]{'loss': 3.6361, 'grad_norm': 0.19595849514007568, 'learning_rate': 0.00025637949245970425, 'epoch': 0.7}
-
                                                      

 70%|██████▉   | 7450/10707 [1:05:41<26:51,  2.02it/s]
 70%|██████▉   | 7451/10707 [1:05:42<26:54,  2.02it/s]
 70%|██████▉   | 7452/10707 [1:05:42<26:52,  2.02it/s]
 70%|██████▉   | 7453/10707 [1:05:43<26:51,  2.02it/s]
 70%|██████▉   | 7454/10707 [1:05:43<26:50,  2.02it/s]
 70%|██████▉   | 7455/10707 [1:05:44<26:51,  2.02it/s]
 70%|██████▉   | 7456/10707 [1:05:44<26:50,  2.02it/s]
 70%|██████▉   | 7457/10707 [1:05:45<26:51,  2.02it/s]
 70%|██████▉   | 7458/10707 [1:05:45<26:48,  2.02it/s]
 70%|██████▉   | 7459/10707 [1:05:46<26:48,  2.02it/s]
 70%|██████▉   | 7460/10707 [1:05:46<26:48,  2.02it/s]
 70%|██████▉   | 7461/10707 [1:05:47<26:48,  2.02it/s]
 70%|██████▉   | 7462/10707 [1:05:47<26:45,  2.02it/s]
 70%|██████▉   | 7463/10707 [1:05:48<26:47,  2.02it/s]
 70%|██████▉   | 7464/10707 [1:05:48<26:44,  2.02it/s]
 70%|██████▉   | 7465/10707 [1:05:49<26:44,  2.02it/s]
 70%|██████▉   | 7466/10707 [1:05:49<26:42,  2.02it/s]
 70%|██████▉   | 7467/10707 [1:05:50<26:43,  2.02it/s]
 70%|██████▉   | 7468/10707 [1:05:50<26:41,  2.02it/s]
 70%|██████▉   | 7469/10707 [1:05:51<26:42,  2.02it/s]
 70%|██████▉   | 7470/10707 [1:05:51<26:40,  2.02it/s]
 70%|██████▉   | 7471/10707 [1:05:52<26:41,  2.02it/s]
 70%|██████▉   | 7472/10707 [1:05:52<26:39,  2.02it/s]
 70%|██████▉   | 7473/10707 [1:05:53<26:41,  2.02it/s]
 70%|██████▉   | 7474/10707 [1:05:53<26:39,  2.02it/s]
 70%|██████▉   | 7475/10707 [1:05:54<26:41,  2.02it/s]{'loss': 3.6349, 'grad_norm': 0.19910091161727905, 'learning_rate': 0.0002528287681228394, 'epoch': 0.7}
                                                      
-
 70%|██████▉   | 7475/10707 [1:05:54<26:41,  2.02it/s]
 70%|██████▉   | 7476/10707 [1:05:54<26:42,  2.02it/s]
 70%|██████▉   | 7477/10707 [1:05:55<26:41,  2.02it/s]
 70%|██████▉   | 7478/10707 [1:05:55<26:40,  2.02it/s]
 70%|██████▉   | 7479/10707 [1:05:56<26:38,  2.02it/s]
 70%|██████▉   | 7480/10707 [1:05:56<26:39,  2.02it/s]
 70%|██████▉   | 7481/10707 [1:05:56<26:38,  2.02it/s]
 70%|██████▉   | 7482/10707 [1:05:57<26:37,  2.02it/s]
 70%|██████▉   | 7483/10707 [1:05:57<26:36,  2.02it/s]
 70%|██████▉   | 7484/10707 [1:05:58<26:36,  2.02it/s]
 70%|██████▉   | 7485/10707 [1:05:58<26:33,  2.02it/s]
 70%|██████▉   | 7486/10707 [1:05:59<26:36,  2.02it/s]
 70%|██████▉   | 7487/10707 [1:05:59<26:34,  2.02it/s]
 70%|██████▉   | 7488/10707 [1:06:00<26:33,  2.02it/s]
 70%|██████▉   | 7489/10707 [1:06:00<26:32,  2.02it/s]
 70%|██████▉   | 7490/10707 [1:06:01<26:32,  2.02it/s]
 70%|██████▉   | 7491/10707 [1:06:01<26:30,  2.02it/s]
 70%|██████▉   | 7492/10707 [1:06:02<26:30,  2.02it/s]
 70%|██████▉   | 7493/10707 [1:06:02<26:29,  2.02it/s]
 70%|██████▉   | 7494/10707 [1:06:03<26:30,  2.02it/s]
 70%|███████   | 7495/10707 [1:06:03<26:28,  2.02it/s]
 70%|███████   | 7496/10707 [1:06:04<26:29,  2.02it/s]
 70%|███████   | 7497/10707 [1:06:04<26:28,  2.02it/s]
 70%|███████   | 7498/10707 [1:06:05<26:28,  2.02it/s]
 70%|███████   | 7499/10707 [1:06:05<26:27,  2.02it/s]
 70%|███████   | 7500/10707 [1:06:06<26:27,  2.02it/s]{'loss': 3.623, 'grad_norm': 0.1997041553258896, 'learning_rate': 0.00024929446410905317, 'epoch': 0.7}
                                                      
-
 70%|███████   | 7500/10707 [1:06:06<26:27,  2.02it/s]
 70%|███████   | 7501/10707 [1:06:06<26:30,  2.02it/s]
 70%|███████   | 7502/10707 [1:06:07<30:45,  1.74it/s]
 70%|███████   | 7503/10707 [1:06:08<29:27,  1.81it/s]
 70%|███████   | 7504/10707 [1:06:08<28:32,  1.87it/s]
 70%|███████   | 7505/10707 [1:06:09<27:54,  1.91it/s]
 70%|███████   | 7506/10707 [1:06:09<27:28,  1.94it/s]
 70%|███████   | 7507/10707 [1:06:10<27:09,  1.96it/s]
 70%|███████   | 7508/10707 [1:06:10<26:54,  1.98it/s]
 70%|███████   | 7509/10707 [1:06:11<26:45,  1.99it/s]
 70%|███████   | 7510/10707 [1:06:11<26:36,  2.00it/s]
 70%|███████   | 7511/10707 [1:06:12<26:31,  2.01it/s]
 70%|███████   | 7512/10707 [1:06:12<26:27,  2.01it/s]
 70%|███████   | 7513/10707 [1:06:13<26:26,  2.01it/s]
 70%|███████   | 7514/10707 [1:06:13<26:23,  2.02it/s]
 70%|██████��   | 7515/10707 [1:06:14<26:22,  2.02it/s]
 70%|███████   | 7516/10707 [1:06:14<26:21,  2.02it/s]
 70%|███████   | 7517/10707 [1:06:15<26:20,  2.02it/s]
 70%|███████   | 7518/10707 [1:06:15<26:20,  2.02it/s]
 70%|███████   | 7519/10707 [1:06:16<26:18,  2.02it/s]
 70%|███████   | 7520/10707 [1:06:16<26:18,  2.02it/s]
 70%|███████   | 7521/10707 [1:06:17<26:16,  2.02it/s]
 70%|███████   | 7522/10707 [1:06:17<26:16,  2.02it/s]
 70%|███████   | 7523/10707 [1:06:18<26:15,  2.02it/s]
 70%|███████   | 7524/10707 [1:06:18<26:15,  2.02it/s]
 70%|███████   | 7525/10707 [1:06:19<26:14,  2.02it/s]
                                                      
{'loss': 3.6309, 'grad_norm': 0.1981370747089386, 'learning_rate': 0.0002457768152127155, 'epoch': 0.7}
-
 70%|███████   | 7525/10707 [1:06:19<26:14,  2.02it/s]
 70%|███████   | 7526/10707 [1:06:19<26:16,  2.02it/s]
 70%|███████   | 7527/10707 [1:06:20<26:16,  2.02it/s]
 70%|███████   | 7528/10707 [1:06:20<26:14,  2.02it/s]
 70%|███████   | 7529/10707 [1:06:21<26:13,  2.02it/s]
 70%|███████   | 7530/10707 [1:06:21<26:13,  2.02it/s]
 70%|███████   | 7531/10707 [1:06:22<26:15,  2.02it/s]
 70%|███████   | 7532/10707 [1:06:22<26:17,  2.01it/s]
 70%|███████   | 7533/10707 [1:06:23<30:42,  1.72it/s]
 70%|███████   | 7534/10707 [1:06:23<29:20,  1.80it/s]
 70%|███████   | 7535/10707 [1:06:24<28:23,  1.86it/s]
 70%|███████   | 7536/10707 [1:06:24<27:44,  1.91it/s]
 70%|███████   | 7537/10707 [1:06:25<27:14,  1.94it/s]
 70%|███████   | 7538/10707 [1:06:25<26:54,  1.96it/s]
 70%|███████   | 7539/10707 [1:06:26<26:39,  1.98it/s]
 70%|███████   | 7540/10707 [1:06:26<26:30,  1.99it/s]
 70%|███████   | 7541/10707 [1:06:27<26:23,  2.00it/s]
 70%|███████   | 7542/10707 [1:06:27<26:19,  2.00it/s]
 70%|███████   | 7543/10707 [1:06:28<26:15,  2.01it/s]
 70%|███████   | 7544/10707 [1:06:28<26:10,  2.01it/s]
 70%|███████   | 7545/10707 [1:06:29<26:08,  2.02it/s]
 70%|███████   | 7546/10707 [1:06:29<26:05,  2.02it/s]
 70%|███████   | 7547/10707 [1:06:30<26:05,  2.02it/s]
 70%|███████   | 7548/10707 [1:06:30<26:05,  2.02it/s]
 71%|███████   | 7549/10707 [1:06:31<26:04,  2.02it/s]
 71%|███████   | 7550/10707 [1:06:31<26:02,  2.02it/s]{'loss': 3.6375, 'grad_norm': 0.2013542354106903, 'learning_rate': 0.00024227605512174794, 'epoch': 0.71}
-
                                                      

 71%|███████   | 7550/10707 [1:06:31<26:02,  2.02it/s]
 71%|███████   | 7551/10707 [1:06:32<26:03,  2.02it/s]
 71%|███████   | 7552/10707 [1:06:32<26:01,  2.02it/s]
 71%|███████   | 7553/10707 [1:06:33<26:02,  2.02it/s]
 71%|███████   | 7554/10707 [1:06:33<25:59,  2.02it/s]
 71%|███████   | 7555/10707 [1:06:34<26:00,  2.02it/s]
 71%|███████   | 7556/10707 [1:06:34<25:58,  2.02it/s]
 71%|███████   | 7557/10707 [1:06:35<26:00,  2.02it/s]
 71%|███████   | 7558/10707 [1:06:35<25:59,  2.02it/s]
 71%|███████   | 7559/10707 [1:06:36<25:58,  2.02it/s]
 71%|███████   | 7560/10707 [1:06:36<25:56,  2.02it/s]
 71%|███████   | 7561/10707 [1:06:37<25:57,  2.02it/s]
 71%|███████   | 7562/10707 [1:06:37<25:55,  2.02it/s]
 71%|███████   | 7563/10707 [1:06:38<25:55,  2.02it/s]
 71%|███████   | 7564/10707 [1:06:38<25:54,  2.02it/s]
 71%|███████   | 7565/10707 [1:06:39<25:55,  2.02it/s]
 71%|███████   | 7566/10707 [1:06:39<25:52,  2.02it/s]
 71%|███████   | 7567/10707 [1:06:40<25:55,  2.02it/s]
 71%|███████   | 7568/10707 [1:06:40<25:54,  2.02it/s]
 71%|███████   | 7569/10707 [1:06:41<25:53,  2.02it/s]
 71%|███████   | 7570/10707 [1:06:41<25:53,  2.02it/s]
 71%|███████   | 7571/10707 [1:06:42<25:53,  2.02it/s]
 71%|███████   | 7572/10707 [1:06:42<25:52,  2.02it/s]
 71%|███████   | 7573/10707 [1:06:43<25:52,  2.02it/s]
 71%|███████   | 7574/10707 [1:06:43<25:51,  2.02it/s]
 71%|███████   | 7575/10707 [1:06:44<25:50,  2.02it/s]
                                                      
{'loss': 3.6353, 'grad_norm': 0.19786237180233002, 'learning_rate': 0.00023879241640209782, 'epoch': 0.71}
-
 71%|███████   | 7575/10707 [1:06:44<25:50,  2.02it/s]
 71%|███████   | 7576/10707 [1:06:44<25:52,  2.02it/s]
 71%|███████   | 7577/10707 [1:06:45<25:50,  2.02it/s]
 71%|███████   | 7578/10707 [1:06:45<25:50,  2.02it/s]
 71%|███████   | 7579/10707 [1:06:46<25:47,  2.02it/s]
 71%|███████   | 7580/10707 [1:06:46<25:50,  2.02it/s]
 71%|███████   | 7581/10707 [1:06:47<25:47,  2.02it/s]
 71%|███████   | 7582/10707 [1:06:47<25:47,  2.02it/s]
 71%|███████   | 7583/10707 [1:06:48<25:45,  2.02it/s]
 71%|███████   | 7584/10707 [1:06:48<25:46,  2.02it/s]
 71%|███████   | 7585/10707 [1:06:49<25:45,  2.02it/s]
 71%|███████   | 7586/10707 [1:06:49<25:45,  2.02it/s]
 71%|███████   | 7587/10707 [1:06:50<25:43,  2.02it/s]
 71%|███████   | 7588/10707 [1:06:50<25:42,  2.02it/s]
 71%|███████   | 7589/10707 [1:06:51<25:41,  2.02it/s]
 71%|███████   | 7590/10707 [1:06:51<25:41,  2.02it/s]
 71%|███████   | 7591/10707 [1:06:52<25:41,  2.02it/s]
 71%|███████   | 7592/10707 [1:06:52<25:40,  2.02it/s]
 71%|███████   | 7593/10707 [1:06:52<25:40,  2.02it/s]
 71%|███████   | 7594/10707 [1:06:53<25:39,  2.02it/s]
 71%|███████   | 7595/10707 [1:06:53<25:39,  2.02it/s]
 71%|███████   | 7596/10707 [1:06:54<25:38,  2.02it/s]
 71%|███████   | 7597/10707 [1:06:54<25:37,  2.02it/s]
 71%|███████   | 7598/10707 [1:06:55<25:38,  2.02it/s]
 71%|███████   | 7599/10707 [1:06:55<25:38,  2.02it/s]
 71%|███████   | 7600/10707 [1:06:56<25:36,  2.02it/s]{'loss': 3.6255, 'grad_norm': 0.20742152631282806, 'learning_rate': 0.0002353261304822887, 'epoch': 0.71}
-
                                                      

 71%|███████   | 7600/10707 [1:06:56<25:36,  2.02it/s]
 71%|███████   | 7601/10707 [1:06:56<25:40,  2.02it/s]
 71%|███████   | 7602/10707 [1:06:57<25:39,  2.02it/s]
 71%|███████   | 7603/10707 [1:06:57<25:39,  2.02it/s]
 71%|███████   | 7604/10707 [1:06:58<25:37,  2.02it/s]
 71%|███████   | 7605/10707 [1:06:58<25:36,  2.02it/s]
 71%|███████   | 7606/10707 [1:06:59<25:34,  2.02it/s]
 71%|███████   | 7607/10707 [1:06:59<25:33,  2.02it/s]
 71%|███████   | 7608/10707 [1:07:00<25:30,  2.02it/s]
 71%|███████   | 7609/10707 [1:07:00<25:32,  2.02it/s]
 71%|███████   | 7610/10707 [1:07:01<25:30,  2.02it/s]
 71%|███████   | 7611/10707 [1:07:01<25:31,  2.02it/s]
 71%|███████   | 7612/10707 [1:07:02<25:29,  2.02it/s]
 71%|███████   | 7613/10707 [1:07:02<25:31,  2.02it/s]
 71%|███████   | 7614/10707 [1:07:03<25:29,  2.02it/s]
 71%|███████   | 7615/10707 [1:07:03<25:30,  2.02it/s]
 71%|███████   | 7616/10707 [1:07:04<25:29,  2.02it/s]
 71%|███████   | 7617/10707 [1:07:04<25:28,  2.02it/s]
 71%|███████   | 7618/10707 [1:07:05<25:27,  2.02it/s]
 71%|███████   | 7619/10707 [1:07:05<25:26,  2.02it/s]
 71%|███████   | 7620/10707 [1:07:06<25:25,  2.02it/s]
 71%|███████   | 7621/10707 [1:07:06<25:26,  2.02it/s]
 71%|███████   | 7622/10707 [1:07:07<25:25,  2.02it/s]
 71%|███████   | 7623/10707 [1:07:07<25:27,  2.02it/s]
 71%|███████   | 7624/10707 [1:07:08<25:25,  2.02it/s]
 71%|███████   | 7625/10707 [1:07:08<25:24,  2.02it/s]{'loss': 3.6301, 'grad_norm': 0.20106655359268188, 'learning_rate': 0.0002318774276380458, 'epoch': 0.71}
-
                                                      

 71%|███████   | 7625/10707 [1:07:08<25:24,  2.02it/s]
 71%|███████   | 7626/10707 [1:07:09<25:26,  2.02it/s]
 71%|███████   | 7627/10707 [1:07:09<25:26,  2.02it/s]
 71%|███████   | 7628/10707 [1:07:10<25:25,  2.02it/s]
 71%|███████▏  | 7629/10707 [1:07:10<25:25,  2.02it/s]
 71%|███████▏  | 7630/10707 [1:07:11<25:23,  2.02it/s]
 71%|███████▏  | 7631/10707 [1:07:11<25:24,  2.02it/s]
 71%|███████▏  | 7632/10707 [1:07:12<25:23,  2.02it/s]
 71%|███████▏  | 7633/10707 [1:07:12<25:21,  2.02it/s]
 71%|███████▏  | 7634/10707 [1:07:13<25:20,  2.02it/s]
 71%|███████▏  | 7635/10707 [1:07:13<25:20,  2.02it/s]
 71%|███████▏  | 7636/10707 [1:07:14<25:18,  2.02it/s]
 71%|███████▏  | 7637/10707 [1:07:14<25:18,  2.02it/s]
 71%|███████▏  | 7638/10707 [1:07:15<25:17,  2.02it/s]
 71%|███████▏  | 7639/10707 [1:07:15<25:17,  2.02it/s]
 71%|███████▏  | 7640/10707 [1:07:16<25:16,  2.02it/s]
 71%|███████▏  | 7641/10707 [1:07:16<25:16,  2.02it/s]
 71%|███████▏  | 7642/10707 [1:07:17<25:15,  2.02it/s]
 71%|███████▏  | 7643/10707 [1:07:17<25:15,  2.02it/s]
 71%|███████▏  | 7644/10707 [1:07:18<25:15,  2.02it/s]
 71%|███████▏  | 7645/10707 [1:07:18<25:15,  2.02it/s]
 71%|███████▏  | 7646/10707 [1:07:19<25:15,  2.02it/s]
 71%|███████▏  | 7647/10707 [1:07:19<25:14,  2.02it/s]
 71%|███████▏  | 7648/10707 [1:07:20<25:14,  2.02it/s]
 71%|███████▏  | 7649/10707 [1:07:20<25:12,  2.02it/s]
 71%|███████▏  | 7650/10707 [1:07:21<25:11,  2.02it/s]
                                                      
{'loss': 3.6231, 'grad_norm': 0.1948511153459549, 'learning_rate': 0.0002284465369769983, 'epoch': 0.71}
-
 71%|███████▏  | 7650/10707 [1:07:21<25:11,  2.02it/s]
 71%|███████▏  | 7651/10707 [1:07:21<25:12,  2.02it/s]
 71%|███████▏  | 7652/10707 [1:07:22<25:13,  2.02it/s]
 71%|███████▏  | 7653/10707 [1:07:22<25:12,  2.02it/s]
 71%|███████▏  | 7654/10707 [1:07:23<25:12,  2.02it/s]
 71%|███████▏  | 7655/10707 [1:07:23<25:10,  2.02it/s]
 72%|███████▏  | 7656/10707 [1:07:24<25:09,  2.02it/s]
 72%|███████▏  | 7657/10707 [1:07:24<25:10,  2.02it/s]
 72%|███████▏  | 7658/10707 [1:07:25<25:10,  2.02it/s]
 72%|███████▏  | 7659/10707 [1:07:25<25:10,  2.02it/s]
 72%|███████▏  | 7660/10707 [1:07:26<25:10,  2.02it/s]
 72%|███████▏  | 7661/10707 [1:07:26<25:08,  2.02it/s]
 72%|███████▏  | 7662/10707 [1:07:27<25:06,  2.02it/s]
 72%|███████▏  | 7663/10707 [1:07:27<25:07,  2.02it/s]
 72%|███████▏  | 7664/10707 [1:07:28<25:06,  2.02it/s]
 72%|███████▏  | 7665/10707 [1:07:28<25:05,  2.02it/s]
 72%|███████▏  | 7666/10707 [1:07:29<25:05,  2.02it/s]
 72%|███████▏  | 7667/10707 [1:07:29<25:05,  2.02it/s]
 72%|███████▏  | 7668/10707 [1:07:30<25:03,  2.02it/s]
 72%|███████▏  | 7669/10707 [1:07:30<25:03,  2.02it/s]
 72%|███████▏  | 7670/10707 [1:07:31<25:02,  2.02it/s]
 72%|███████▏  | 7671/10707 [1:07:31<25:02,  2.02it/s]
 72%|███████▏  | 7672/10707 [1:07:32<25:01,  2.02it/s]
 72%|███████▏  | 7673/10707 [1:07:32<25:01,  2.02it/s]
 72%|███████▏  | 7674/10707 [1:07:33<25:01,  2.02it/s]
 72%|███████▏  | 7675/10707 [1:07:33<25:01,  2.02it/s]{'loss': 3.6256, 'grad_norm': 0.1953551173210144, 'learning_rate': 0.00022503368642345863, 'epoch': 0.72}
-
                                                      

 72%|███████▏  | 7675/10707 [1:07:33<25:01,  2.02it/s]
 72%|███████▏  | 7676/10707 [1:07:34<25:04,  2.01it/s]
 72%|███████▏  | 7677/10707 [1:07:34<25:03,  2.02it/s]
 72%|███████▏  | 7678/10707 [1:07:35<25:02,  2.02it/s]
 72%|███████▏  | 7679/10707 [1:07:35<25:00,  2.02it/s]
 72%|███████▏  | 7680/10707 [1:07:36<24:59,  2.02it/s]
 72%|███████▏  | 7681/10707 [1:07:36<24:58,  2.02it/s]
 72%|███████▏  | 7682/10707 [1:07:37<24:58,  2.02it/s]
 72%|███████▏  | 7683/10707 [1:07:37<24:57,  2.02it/s]
 72%|███████▏  | 7684/10707 [1:07:38<24:56,  2.02it/s]
 72%|███████▏  | 7685/10707 [1:07:38<24:55,  2.02it/s]
 72%|███████▏  | 7686/10707 [1:07:39<24:54,  2.02it/s]
 72%|███████▏  | 7687/10707 [1:07:39<24:54,  2.02it/s]
 72%|███████▏  | 7688/10707 [1:07:40<24:53,  2.02it/s]
 72%|███████▏  | 7689/10707 [1:07:40<24:53,  2.02it/s]
 72%|███████▏  | 7690/10707 [1:07:41<24:53,  2.02it/s]
 72%|███████▏  | 7691/10707 [1:07:41<24:51,  2.02it/s]
 72%|███████▏  | 7692/10707 [1:07:42<24:51,  2.02it/s]
 72%|███████▏  | 7693/10707 [1:07:42<24:49,  2.02it/s]
 72%|███████▏  | 7694/10707 [1:07:42<24:50,  2.02it/s]
 72%|███████▏  | 7695/10707 [1:07:43<24:49,  2.02it/s]
 72%|███████▏  | 7696/10707 [1:07:43<24:49,  2.02it/s]
 72%|███████▏  | 7697/10707 [1:07:44<24:50,  2.02it/s]
 72%|███████▏  | 7698/10707 [1:07:44<24:49,  2.02it/s]
 72%|███████▏  | 7699/10707 [1:07:45<24:48,  2.02it/s]
 72%|███████▏  | 7700/10707 [1:07:45<24:48,  2.02it/s]{'loss': 3.6142, 'grad_norm': 0.2011794000864029, 'learning_rate': 0.00022163910270328114, 'epoch': 0.72}
-
                                                      

 72%|███████▏  | 7700/10707 [1:07:45<24:48,  2.02it/s]
 72%|███████▏  | 7701/10707 [1:07:46<24:50,  2.02it/s]
 72%|███████▏  | 7702/10707 [1:07:46<24:49,  2.02it/s]
 72%|███████▏  | 7703/10707 [1:07:47<24:49,  2.02it/s]
 72%|███████▏  | 7704/10707 [1:07:47<24:48,  2.02it/s]
 72%|███████▏  | 7705/10707 [1:07:48<24:49,  2.02it/s]
 72%|███████▏  | 7706/10707 [1:07:48<24:45,  2.02it/s]
 72%|███████▏  | 7707/10707 [1:07:49<24:45,  2.02it/s]
 72%|███████▏  | 7708/10707 [1:07:49<24:43,  2.02it/s]
 72%|███████▏  | 7709/10707 [1:07:50<24:44,  2.02it/s]
 72%|███████▏  | 7710/10707 [1:07:50<24:41,  2.02it/s]
 72%|███████▏  | 7711/10707 [1:07:51<24:42,  2.02it/s]
 72%|███████▏  | 7712/10707 [1:07:51<24:40,  2.02it/s]
 72%|███████▏  | 7713/10707 [1:07:52<24:41,  2.02it/s]
 72%|███████▏  | 7714/10707 [1:07:52<24:39,  2.02it/s]
 72%|███████▏  | 7715/10707 [1:07:53<24:39,  2.02it/s]
 72%|███████▏  | 7716/10707 [1:07:53<24:37,  2.02it/s]
 72%|███████▏  | 7717/10707 [1:07:54<24:39,  2.02it/s]
 72%|███████▏  | 7718/10707 [1:07:54<24:38,  2.02it/s]
 72%|███████▏  | 7719/10707 [1:07:55<24:40,  2.02it/s]
 72%|███████▏  | 7720/10707 [1:07:55<24:39,  2.02it/s]
 72%|███████▏  | 7721/10707 [1:07:56<24:37,  2.02it/s]
 72%|███████▏  | 7722/10707 [1:07:56<24:36,  2.02it/s]
 72%|███████▏  | 7723/10707 [1:07:57<24:36,  2.02it/s]
 72%|███████▏  | 7724/10707 [1:07:57<24:34,  2.02it/s]
 72%|███████▏  | 7725/10707 [1:07:58<24:34,  2.02it/s]{'loss': 3.6287, 'grad_norm': 0.1961234211921692, 'learning_rate': 0.00021826301132879978, 'epoch': 0.72}
-
                                                      

 72%|███████▏  | 7725/10707 [1:07:58<24:34,  2.02it/s]
 72%|███████▏  | 7726/10707 [1:07:58<24:36,  2.02it/s]
 72%|███████▏  | 7727/10707 [1:07:59<24:35,  2.02it/s]
 72%|███████▏  | 7728/10707 [1:07:59<24:34,  2.02it/s]
 72%|███████▏  | 7729/10707 [1:08:00<24:34,  2.02it/s]
 72%|███████▏  | 7730/10707 [1:08:00<24:35,  2.02it/s]
 72%|███████▏  | 7731/10707 [1:08:01<24:33,  2.02it/s]
 72%|███████▏  | 7732/10707 [1:08:01<24:33,  2.02it/s]
 72%|███████▏  | 7733/10707 [1:08:02<24:31,  2.02it/s]
 72%|███████▏  | 7734/10707 [1:08:02<24:32,  2.02it/s]
 72%|███████▏  | 7735/10707 [1:08:03<24:30,  2.02it/s]
 72%|███████▏  | 7736/10707 [1:08:03<24:30,  2.02it/s]
 72%|███████▏  | 7737/10707 [1:08:04<24:29,  2.02it/s]
 72%|███████▏  | 7738/10707 [1:08:04<24:29,  2.02it/s]
 72%|███████▏  | 7739/10707 [1:08:05<24:28,  2.02it/s]
 72%|███████▏  | 7740/10707 [1:08:05<24:29,  2.02it/s]
 72%|███████▏  | 7741/10707 [1:08:06<24:29,  2.02it/s]
 72%|███████▏  | 7742/10707 [1:08:06<24:28,  2.02it/s]
 72%|███████▏  | 7743/10707 [1:08:07<24:28,  2.02it/s]
 72%|███████▏  | 7744/10707 [1:08:07<24:26,  2.02it/s]
 72%|███████▏  | 7745/10707 [1:08:08<24:25,  2.02it/s]
 72%|███████▏  | 7746/10707 [1:08:08<24:24,  2.02it/s]
 72%|███████▏  | 7747/10707 [1:08:09<24:24,  2.02it/s]
 72%|███████▏  | 7748/10707 [1:08:09<24:24,  2.02it/s]
 72%|███████▏  | 7749/10707 [1:08:10<24:23,  2.02it/s]
 72%|███████▏  | 7750/10707 [1:08:10<24:22,  2.02it/s]{'loss': 3.6308, 'grad_norm': 0.20930619537830353, 'learning_rate': 0.00021490563658384688, 'epoch': 0.72}
-
                                                      

 72%|███████▏  | 7750/10707 [1:08:10<24:22,  2.02it/s]
 72%|███████▏  | 7751/10707 [1:08:11<24:25,  2.02it/s]
 72%|███████▏  | 7752/10707 [1:08:11<24:24,  2.02it/s]
 72%|███████▏  | 7753/10707 [1:08:12<24:23,  2.02it/s]
 72%|███████▏  | 7754/10707 [1:08:12<24:20,  2.02it/s]
 72%|███████▏  | 7755/10707 [1:08:13<24:21,  2.02it/s]
 72%|███████▏  | 7756/10707 [1:08:13<24:19,  2.02it/s]
 72%|███████▏  | 7757/10707 [1:08:14<24:19,  2.02it/s]
 72%|███████▏  | 7758/10707 [1:08:14<24:18,  2.02it/s]
 72%|███████▏  | 7759/10707 [1:08:15<24:19,  2.02it/s]
 72%|███████▏  | 7760/10707 [1:08:15<24:17,  2.02it/s]
 72%|███████▏  | 7761/10707 [1:08:16<24:18,  2.02it/s]
 72%|███████▏  | 7762/10707 [1:08:16<24:17,  2.02it/s]
 73%|███████▎  | 7763/10707 [1:08:17<24:17,  2.02it/s]
 73%|███████▎  | 7764/10707 [1:08:17<24:15,  2.02it/s]
 73%|███████▎  | 7765/10707 [1:08:18<24:15,  2.02it/s]
 73%|███████▎  | 7766/10707 [1:08:18<24:13,  2.02it/s]
 73%|███████▎  | 7767/10707 [1:08:19<24:14,  2.02it/s]
 73%|███████▎  | 7768/10707 [1:08:19<24:12,  2.02it/s]
 73%|███████▎  | 7769/10707 [1:08:20<24:12,  2.02it/s]
 73%|███████▎  | 7770/10707 [1:08:20<24:12,  2.02it/s]
 73%|███████▎  | 7771/10707 [1:08:21<24:12,  2.02it/s]
 73%|███████▎  | 7772/10707 [1:08:21<24:12,  2.02it/s]
 73%|███████▎  | 7773/10707 [1:08:22<24:13,  2.02it/s]
 73%|███���███▎  | 7774/10707 [1:08:22<24:11,  2.02it/s]
 73%|███████▎  | 7775/10707 [1:08:23<24:11,  2.02it/s]
                                                      
{'loss': 3.6251, 'grad_norm': 0.20498313009738922, 'learning_rate': 0.0002115672015088531, 'epoch': 0.73}
-
 73%|███████▎  | 7775/10707 [1:08:23<24:11,  2.02it/s]
 73%|███████▎  | 7776/10707 [1:08:23<24:12,  2.02it/s]
 73%|███████▎  | 7777/10707 [1:08:24<24:11,  2.02it/s]
 73%|███████▎  | 7778/10707 [1:08:24<24:09,  2.02it/s]
 73%|███████▎  | 7779/10707 [1:08:25<24:08,  2.02it/s]
 73%|███████▎  | 7780/10707 [1:08:25<24:06,  2.02it/s]
 73%|███████▎  | 7781/10707 [1:08:26<24:07,  2.02it/s]
 73%|███████▎  | 7782/10707 [1:08:26<24:05,  2.02it/s]
 73%|███████▎  | 7783/10707 [1:08:27<24:05,  2.02it/s]
 73%|███████▎  | 7784/10707 [1:08:27<24:05,  2.02it/s]
 73%|███████▎  | 7785/10707 [1:08:28<24:06,  2.02it/s]
 73%|███████▎  | 7786/10707 [1:08:28<24:05,  2.02it/s]
 73%|███████▎  | 7787/10707 [1:08:29<24:05,  2.02it/s]
 73%|███████▎  | 7788/10707 [1:08:29<24:04,  2.02it/s]
 73%|███████▎  | 7789/10707 [1:08:30<24:03,  2.02it/s]
 73%|███████▎  | 7790/10707 [1:08:30<24:03,  2.02it/s]
 73%|███████▎  | 7791/10707 [1:08:30<24:03,  2.02it/s]
 73%|███████▎  | 7792/10707 [1:08:31<24:03,  2.02it/s]
 73%|███████▎  | 7793/10707 [1:08:31<24:02,  2.02it/s]
 73%|███████▎  | 7794/10707 [1:08:32<24:01,  2.02it/s]
 73%|███████▎  | 7795/10707 [1:08:32<23:59,  2.02it/s]
 73%|███████▎  | 7796/10707 [1:08:33<24:00,  2.02it/s]
 73%|███████▎  | 7797/10707 [1:08:33<23:59,  2.02it/s]
 73%|███████▎  | 7798/10707 [1:08:34<23:58,  2.02it/s]
 73%|███████▎  | 7799/10707 [1:08:34<23:59,  2.02it/s]
 73%|███████▎  | 7800/10707 [1:08:35<23:59,  2.02it/s]{'loss': 3.6116, 'grad_norm': 0.2022058516740799, 'learning_rate': 0.00020824792788603025, 'epoch': 0.73}
-
                                                      

 73%|███████▎  | 7800/10707 [1:08:35<23:59,  2.02it/s]
 73%|███████▎  | 7801/10707 [1:08:35<24:00,  2.02it/s]
 73%|███████▎  | 7802/10707 [1:08:36<24:00,  2.02it/s]
 73%|███████▎  | 7803/10707 [1:08:36<23:57,  2.02it/s]
 73%|███████▎  | 7804/10707 [1:08:37<23:57,  2.02it/s]
 73%|███████▎  | 7805/10707 [1:08:37<23:56,  2.02it/s]
 73%|███████▎  | 7806/10707 [1:08:38<23:55,  2.02it/s]
 73%|███████▎  | 7807/10707 [1:08:38<23:54,  2.02it/s]
 73%|███████▎  | 7808/10707 [1:08:39<23:54,  2.02it/s]
 73%|███████▎  | 7809/10707 [1:08:39<23:52,  2.02it/s]
 73%|███████▎  | 7810/10707 [1:08:40<23:53,  2.02it/s]
 73%|███████▎  | 7811/10707 [1:08:40<23:52,  2.02it/s]
 73%|███████▎  | 7812/10707 [1:08:41<23:54,  2.02it/s]
 73%|███████▎  | 7813/10707 [1:08:41<23:52,  2.02it/s]
 73%|███████▎  | 7814/10707 [1:08:42<23:53,  2.02it/s]
 73%|███████▎  | 7815/10707 [1:08:42<23:52,  2.02it/s]
 73%|███████▎  | 7816/10707 [1:08:43<23:53,  2.02it/s]
 73%|███████▎  | 7817/10707 [1:08:43<23:51,  2.02it/s]
 73%|███████▎  | 7818/10707 [1:08:44<23:51,  2.02it/s]
 73%|███████▎  | 7819/10707 [1:08:44<23:50,  2.02it/s]
 73%|███████▎  | 7820/10707 [1:08:45<23:50,  2.02it/s]
 73%|███████▎  | 7821/10707 [1:08:45<23:49,  2.02it/s]
 73%|███████▎  | 7822/10707 [1:08:46<23:49,  2.02it/s]
 73%|███████▎  | 7823/10707 [1:08:46<23:48,  2.02it/s]
 73%|███████▎  | 7824/10707 [1:08:47<23:47,  2.02it/s]
 73%|███████▎  | 7825/10707 [1:08:47<23:46,  2.02it/s]
                                                      
{'loss': 3.6304, 'grad_norm': 0.20006491243839264, 'learning_rate': 0.00020494803622463775, 'epoch': 0.73}
-
 73%|███████▎  | 7825/10707 [1:08:47<23:46,  2.02it/s]
 73%|███████▎  | 7826/10707 [1:08:48<23:47,  2.02it/s]
 73%|███████▎  | 7827/10707 [1:08:48<23:47,  2.02it/s]
 73%|███████▎  | 7828/10707 [1:08:49<23:47,  2.02it/s]
 73%|███████▎  | 7829/10707 [1:08:49<23:44,  2.02it/s]
 73%|███████▎  | 7830/10707 [1:08:50<23:43,  2.02it/s]
 73%|███████▎  | 7831/10707 [1:08:50<23:42,  2.02it/s]
 73%|███████▎  | 7832/10707 [1:08:51<23:42,  2.02it/s]
 73%|███████▎  | 7833/10707 [1:08:51<23:41,  2.02it/s]
 73%|███████▎  | 7834/10707 [1:08:52<23:41,  2.02it/s]
 73%|███████▎  | 7835/10707 [1:08:52<23:40,  2.02it/s]
 73%|███████▎  | 7836/10707 [1:08:53<23:39,  2.02it/s]
 73%|███████▎  | 7837/10707 [1:08:53<23:38,  2.02it/s]
 73%|███████▎  | 7838/10707 [1:08:54<23:39,  2.02it/s]
 73%|███████▎  | 7839/10707 [1:08:54<23:39,  2.02it/s]
 73%|███████▎  | 7840/10707 [1:08:55<23:39,  2.02it/s]
 73%|███████▎  | 7841/10707 [1:08:55<23:38,  2.02it/s]
 73%|███████▎  | 7842/10707 [1:08:56<23:39,  2.02it/s]
 73%|███████▎  | 7843/10707 [1:08:56<23:37,  2.02it/s]
 73%|███████▎  | 7844/10707 [1:08:57<23:37,  2.02it/s]
 73%|███████▎  | 7845/10707 [1:08:57<23:36,  2.02it/s]
 73%|███████▎  | 7846/10707 [1:08:58<23:36,  2.02it/s]
 73%|███████▎  | 7847/10707 [1:08:58<23:35,  2.02it/s]
 73%|███████▎  | 7848/10707 [1:08:59<23:35,  2.02it/s]
 73%|███████▎  | 7849/10707 [1:08:59<23:34,  2.02it/s]
 73%|███████▎  | 7850/10707 [1:09:00<23:33,  2.02it/s]{'loss': 3.6259, 'grad_norm': 0.197265625, 'learning_rate': 0.00020166774574633328, 'epoch': 0.73}
-
                                                      

 73%|███████▎  | 7850/10707 [1:09:00<23:33,  2.02it/s]
 73%|███████▎  | 7851/10707 [1:09:00<23:35,  2.02it/s]
 73%|███████▎  | 7852/10707 [1:09:01<23:34,  2.02it/s]
 73%|███████▎  | 7853/10707 [1:09:01<23:34,  2.02it/s]
 73%|███████▎  | 7854/10707 [1:09:02<23:33,  2.02it/s]
 73%|███████▎  | 7855/10707 [1:09:02<23:34,  2.02it/s]
 73%|███████▎  | 7856/10707 [1:09:03<23:32,  2.02it/s]
 73%|███████▎  | 7857/10707 [1:09:03<23:31,  2.02it/s]
 73%|███████▎  | 7858/10707 [1:09:04<23:32,  2.02it/s]
 73%|███████▎  | 7859/10707 [1:09:04<23:31,  2.02it/s]
 73%|███████▎  | 7860/10707 [1:09:05<23:29,  2.02it/s]
 73%|███████▎  | 7861/10707 [1:09:05<23:28,  2.02it/s]
 73%|███████▎  | 7862/10707 [1:09:06<23:29,  2.02it/s]
 73%|███████▎  | 7863/10707 [1:09:06<23:27,  2.02it/s]
 73%|███████▎  | 7864/10707 [1:09:07<23:27,  2.02it/s]
 73%|███████▎  | 7865/10707 [1:09:07<23:25,  2.02it/s]
 73%|███████▎  | 7866/10707 [1:09:08<23:26,  2.02it/s]
 73%|███████▎  | 7867/10707 [1:09:08<23:25,  2.02it/s]
 73%|███████▎  | 7868/10707 [1:09:09<23:25,  2.02it/s]
 73%|███████▎  | 7869/10707 [1:09:09<23:23,  2.02it/s]
 74%|███████▎  | 7870/10707 [1:09:10<23:23,  2.02it/s]
 74%|███████▎  | 7871/10707 [1:09:10<23:23,  2.02it/s]
 74%|███████▎  | 7872/10707 [1:09:11<23:22,  2.02it/s]
 74%|███████▎  | 7873/10707 [1:09:11<23:21,  2.02it/s]
 74%|███████▎  | 7874/10707 [1:09:12<23:21,  2.02it/s]
 74%|███████▎  | 7875/10707 [1:09:12<23:20,  2.02it/s]
                                                      
{'loss': 3.6189, 'grad_norm': 0.19722211360931396, 'learning_rate': 0.0001984072743706095, 'epoch': 0.74}
-
 74%|███████▎  | 7875/10707 [1:09:12<23:20,  2.02it/s]
 74%|███████▎  | 7876/10707 [1:09:13<23:22,  2.02it/s]
 74%|███████▎  | 7877/10707 [1:09:13<23:23,  2.02it/s]
 74%|███████▎  | 7878/10707 [1:09:14<23:21,  2.02it/s]
 74%|███████▎  | 7879/10707 [1:09:14<23:20,  2.02it/s]
 74%|███████▎  | 7880/10707 [1:09:15<23:19,  2.02it/s]
 74%|███████▎  | 7881/10707 [1:09:15<23:18,  2.02it/s]
 74%|███████▎  | 7882/10707 [1:09:16<23:19,  2.02it/s]
 74%|███████▎  | 7883/10707 [1:09:16<23:18,  2.02it/s]
 74%|███████▎  | 7884/10707 [1:09:17<23:17,  2.02it/s]
 74%|███████▎  | 7885/10707 [1:09:17<23:17,  2.02it/s]
 74%|███████▎  | 7886/10707 [1:09:18<23:16,  2.02it/s]
 74%|███████▎  | 7887/10707 [1:09:18<23:16,  2.02it/s]
 74%|███████▎  | 7888/10707 [1:09:19<23:16,  2.02it/s]
 74%|███████▎  | 7889/10707 [1:09:19<23:15,  2.02it/s]
 74%|███████▎  | 7890/10707 [1:09:20<23:15,  2.02it/s]
 74%|███████▎  | 7891/10707 [1:09:20<23:13,  2.02it/s]
 74%|███████▎  | 7892/10707 [1:09:21<23:13,  2.02it/s]
 74%|███████▎  | 7893/10707 [1:09:21<23:12,  2.02it/s]
 74%|███████▎  | 7894/10707 [1:09:21<23:12,  2.02it/s]
 74%|███████▎  | 7895/10707 [1:09:22<23:12,  2.02it/s]
 74%|███████▎  | 7896/10707 [1:09:22<23:11,  2.02it/s]
 74%|███████▍  | 7897/10707 [1:09:23<23:11,  2.02it/s]
 74%|███████▍  | 7898/10707 [1:09:23<23:10,  2.02it/s]
 74%|███████▍  | 7899/10707 [1:09:24<23:10,  2.02it/s]
 74%|███████▍  | 7900/10707 [1:09:24<23:10,  2.02it/s]
                                                      
{'loss': 3.6171, 'grad_norm': 0.19725504517555237, 'learning_rate': 0.0001951668387003169, 'epoch': 0.74}
-
 74%|███████▍  | 7900/10707 [1:09:24<23:10,  2.02it/s]
 74%|███████▍  | 7901/10707 [1:09:25<23:12,  2.02it/s]
 74%|███████▍  | 7902/10707 [1:09:25<23:11,  2.02it/s]
 74%|███████▍  | 7903/10707 [1:09:26<23:11,  2.02it/s]
 74%|███████▍  | 7904/10707 [1:09:26<23:09,  2.02it/s]
 74%|███████▍  | 7905/10707 [1:09:27<23:08,  2.02it/s]
 74%|███████▍  | 7906/10707 [1:09:27<23:07,  2.02it/s]
 74%|███████▍  | 7907/10707 [1:09:28<23:08,  2.02it/s]
 74%|███████▍  | 7908/10707 [1:09:28<23:05,  2.02it/s]
 74%|███████▍  | 7909/10707 [1:09:29<23:05,  2.02it/s]
 74%|███████▍  | 7910/10707 [1:09:29<23:05,  2.02it/s]
 74%|███████▍  | 7911/10707 [1:09:30<23:05,  2.02it/s]
 74%|███████▍  | 7912/10707 [1:09:30<23:03,  2.02it/s]
 74%|███████▍  | 7913/10707 [1:09:31<23:03,  2.02it/s]
 74%|███████▍  | 7914/10707 [1:09:31<23:01,  2.02it/s]
 74%|███████▍  | 7915/10707 [1:09:32<23:02,  2.02it/s]
 74%|███████▍  | 7916/10707 [1:09:32<23:02,  2.02it/s]
 74%|███████▍  | 7917/10707 [1:09:33<23:02,  2.02it/s]
 74%|███████▍  | 7918/10707 [1:09:33<22:59,  2.02it/s]
 74%|███████▍  | 7919/10707 [1:09:34<23:00,  2.02it/s]
 74%|███████▍  | 7920/10707 [1:09:34<22:59,  2.02it/s]
 74%|███████▍  | 7921/10707 [1:09:35<23:00,  2.02it/s]
 74%|███████▍  | 7922/10707 [1:09:35<22:58,  2.02it/s]
 74%|███████▍  | 7923/10707 [1:09:36<22:58,  2.02it/s]
 74%|███████▍  | 7924/10707 [1:09:36<22:57,  2.02it/s]
 74%|███████▍  | 7925/10707 [1:09:37<22:57,  2.02it/s]{'loss': 3.6185, 'grad_norm': 0.1997210681438446, 'learning_rate': 0.0001919466540072741, 'epoch': 0.74}
-
                                                      

 74%|███████▍  | 7925/10707 [1:09:37<22:57,  2.02it/s]
 74%|███████▍  | 7926/10707 [1:09:37<22:58,  2.02it/s]
 74%|███████▍  | 7927/10707 [1:09:38<22:57,  2.02it/s]
 74%|███████▍  | 7928/10707 [1:09:38<22:56,  2.02it/s]
 74%|███████▍  | 7929/10707 [1:09:39<22:56,  2.02it/s]
 74%|███████▍  | 7930/10707 [1:09:39<22:56,  2.02it/s]
 74%|███████▍  | 7931/10707 [1:09:40<22:53,  2.02it/s]
 74%|███████▍  | 7932/10707 [1:09:40<22:54,  2.02it/s]
 74%|███████▍  | 7933/10707 [1:09:41<22:52,  2.02it/s]
 74%|███████▍  | 7934/10707 [1:09:41<22:53,  2.02it/s]
 74%|███████▍  | 7935/10707 [1:09:42<22:51,  2.02it/s]
 74%|███████▍  | 7936/10707 [1:09:42<22:51,  2.02it/s]
 74%|███████▍  | 7937/10707 [1:09:43<22:49,  2.02it/s]
 74%|███████▍  | 7938/10707 [1:09:43<22:51,  2.02it/s]
 74%|███████▍  | 7939/10707 [1:09:44<22:49,  2.02it/s]
 74%|███████▍  | 7940/10707 [1:09:44<22:50,  2.02it/s]
 74%|███████▍  | 7941/10707 [1:09:45<22:48,  2.02it/s]
 74%|███████▍  | 7942/10707 [1:09:45<22:49,  2.02it/s]
 74%|███████▍  | 7943/10707 [1:09:46<22:47,  2.02it/s]
 74%|███████▍  | 7944/10707 [1:09:46<22:47,  2.02it/s]
 74%|███████▍  | 7945/10707 [1:09:47<22:46,  2.02it/s]
 74%|███████▍  | 7946/10707 [1:09:47<22:46,  2.02it/s]
 74%|███████▍  | 7947/10707 [1:09:48<22:44,  2.02it/s]
 74%|███████▍  | 7948/10707 [1:09:48<22:45,  2.02it/s]
 74%|███████▍  | 7949/10707 [1:09:49<22:45,  2.02it/s]
 74%|███████▍  | 7950/10707 [1:09:49<22:46,  2.02it/s]{'loss': 3.6236, 'grad_norm': 0.1970883309841156, 'learning_rate': 0.00018874693421796728, 'epoch': 0.74}
                                                      
-
 74%|███████▍  | 7950/10707 [1:09:49<22:46,  2.02it/s]
 74%|███████▍  | 7951/10707 [1:09:50<22:47,  2.01it/s]
 74%|███████▍  | 7952/10707 [1:09:50<22:46,  2.02it/s]
 74%|███████▍  | 7953/10707 [1:09:51<22:46,  2.02it/s]
 74%|███████▍  | 7954/10707 [1:09:51<22:43,  2.02it/s]
 74%|███████▍  | 7955/10707 [1:09:52<22:43,  2.02it/s]
 74%|███████▍  | 7956/10707 [1:09:52<22:43,  2.02it/s]
 74%|███████▍  | 7957/10707 [1:09:53<22:42,  2.02it/s]
 74%|███████▍  | 7958/10707 [1:09:53<22:41,  2.02it/s]
 74%|███████▍  | 7959/10707 [1:09:54<22:41,  2.02it/s]
 74%|███████▍  | 7960/10707 [1:09:54<22:41,  2.02it/s]
 74%|███████▍  | 7961/10707 [1:09:55<22:40,  2.02it/s]
 74%|███████▍  | 7962/10707 [1:09:55<22:40,  2.02it/s]
 74%|███████▍  | 7963/10707 [1:09:56<22:39,  2.02it/s]
 74%|███████▍  | 7964/10707 [1:09:56<22:39,  2.02it/s]
 74%|███████▍  | 7965/10707 [1:09:57<22:37,  2.02it/s]
 74%|███████▍  | 7966/10707 [1:09:57<22:37,  2.02it/s]
 74%|███████▍  | 7967/10707 [1:09:58<22:35,  2.02it/s]
 74%|███████▍  | 7968/10707 [1:09:58<22:35,  2.02it/s]
 74%|███████▍  | 7969/10707 [1:09:59<22:34,  2.02it/s]
 74%|███████▍  | 7970/10707 [1:09:59<22:35,  2.02it/s]
 74%|███████▍  | 7971/10707 [1:10:00<22:33,  2.02it/s]
 74%|███████▍  | 7972/10707 [1:10:00<22:34,  2.02it/s]
 74%|███████▍  | 7973/10707 [1:10:01<22:32,  2.02it/s]
 74%|███████▍  | 7974/10707 [1:10:01<22:32,  2.02it/s]
 74%|███████▍  | 7975/10707 [1:10:02<22:30,  2.02it/s]{'loss': 3.6146, 'grad_norm': 0.200435072183609, 'learning_rate': 0.00018556789189933692, 'epoch': 0.74}
                                                      
-
 74%|███████▍  | 7975/10707 [1:10:02<22:30,  2.02it/s]
 74%|███████▍  | 7976/10707 [1:10:02<22:32,  2.02it/s]
 75%|███████▍  | 7977/10707 [1:10:03<22:30,  2.02it/s]
 75%|███████▍  | 7978/10707 [1:10:03<22:31,  2.02it/s]
 75%|███████▍  | 7979/10707 [1:10:04<22:29,  2.02it/s]
 75%|███████▍  | 7980/10707 [1:10:04<22:31,  2.02it/s]
 75%|███████▍  | 7981/10707 [1:10:05<22:29,  2.02it/s]
 75%|███████▍  | 7982/10707 [1:10:05<22:29,  2.02it/s]
 75%|███████▍  | 7983/10707 [1:10:06<22:29,  2.02it/s]
 75%|███████▍  | 7984/10707 [1:10:06<22:28,  2.02it/s]
 75%|███████▍  | 7985/10707 [1:10:07<22:28,  2.02it/s]
 75%|███████▍  | 7986/10707 [1:10:07<22:27,  2.02it/s]
 75%|███████▍  | 7987/10707 [1:10:08<22:27,  2.02it/s]
 75%|███████▍  | 7988/10707 [1:10:08<22:26,  2.02it/s]
 75%|███████▍  | 7989/10707 [1:10:09<22:25,  2.02it/s]
 75%|███████▍  | 7990/10707 [1:10:09<22:24,  2.02it/s]
 75%|███████▍  | 7991/10707 [1:10:10<22:23,  2.02it/s]
 75%|███████▍  | 7992/10707 [1:10:10<22:26,  2.02it/s]
 75%|███████▍  | 7993/10707 [1:10:11<22:25,  2.02it/s]
 75%|███████▍  | 7994/10707 [1:10:11<22:24,  2.02it/s]
 75%|███████▍  | 7995/10707 [1:10:12<22:23,  2.02it/s]
 75%|███████▍  | 7996/10707 [1:10:12<22:24,  2.02it/s]
 75%|███████▍  | 7997/10707 [1:10:13<22:23,  2.02it/s]
 75%|███████▍  | 7998/10707 [1:10:13<22:23,  2.02it/s]
 75%|███████▍  | 7999/10707 [1:10:13<22:21,  2.02it/s]
 75%|███████▍  | 8000/10707 [1:10:14<22:21,  2.02it/s]{'loss': 3.6187, 'grad_norm': 0.1972166895866394, 'learning_rate': 0.00018240973824465862, 'epoch': 0.75}
-
                                                      

 75%|███████▍  | 8000/10707 [1:10:14<22:21,  2.02it/s]
 75%|███████▍  | 8001/10707 [1:10:14<22:22,  2.02it/s]
 75%|███████▍  | 8002/10707 [1:10:15<22:20,  2.02it/s]
 75%|███████▍  | 8003/10707 [1:10:15<22:20,  2.02it/s]
 75%|███████▍  | 8004/10707 [1:10:16<22:18,  2.02it/s]
 75%|███████▍  | 8005/10707 [1:10:16<22:19,  2.02it/s]
 75%|███████▍  | 8006/10707 [1:10:17<22:16,  2.02it/s]
 75%|███████▍  | 8007/10707 [1:10:17<22:17,  2.02it/s]
 75%|███████▍  | 8008/10707 [1:10:18<22:15,  2.02it/s]
 75%|███████▍  | 8009/10707 [1:10:18<22:16,  2.02it/s]
 75%|███████▍  | 8010/10707 [1:10:19<22:14,  2.02it/s]
 75%|███████▍  | 8011/10707 [1:10:19<22:14,  2.02it/s]
 75%|███████▍  | 8012/10707 [1:10:20<22:12,  2.02it/s]
 75%|███████▍  | 8013/10707 [1:10:20<22:12,  2.02it/s]
 75%|███████▍  | 8014/10707 [1:10:21<22:10,  2.02it/s]
 75%|███████▍  | 8015/10707 [1:10:21<22:12,  2.02it/s]
 75%|███████▍  | 8016/10707 [1:10:22<22:11,  2.02it/s]
 75%|███████▍  | 8017/10707 [1:10:22<22:12,  2.02it/s]
 75%|███████▍  | 8018/10707 [1:10:23<22:10,  2.02it/s]
 75%|███████▍  | 8019/10707 [1:10:23<22:10,  2.02it/s]
 75%|███████▍  | 8020/10707 [1:10:24<22:09,  2.02it/s]
 75%|███████▍  | 8021/10707 [1:10:24<22:09,  2.02it/s]
 75%|███████▍  | 8022/10707 [1:10:25<22:08,  2.02it/s]
 75%|███████▍  | 8023/10707 [1:10:25<22:08,  2.02it/s]
 75%|███████▍  | 8024/10707 [1:10:26<22:07,  2.02it/s]
 75%|███████▍  | 8025/10707 [1:10:26<22:07,  2.02it/s]{'loss': 3.6164, 'grad_norm': 0.19571930170059204, 'learning_rate': 0.00017927268305951088, 'epoch': 0.75}
-
                                                      

 75%|███████▍  | 8025/10707 [1:10:26<22:07,  2.02it/s]
 75%|███████▍  | 8026/10707 [1:10:27<22:09,  2.02it/s]
 75%|███████▍  | 8027/10707 [1:10:27<22:09,  2.02it/s]
 75%|███████▍  | 8028/10707 [1:10:28<22:09,  2.02it/s]
 75%|███████▍  | 8029/10707 [1:10:28<22:07,  2.02it/s]
 75%|███████▍  | 8030/10707 [1:10:29<22:06,  2.02it/s]
 75%|███████▌  | 8031/10707 [1:10:29<22:04,  2.02it/s]
 75%|███████▌  | 8032/10707 [1:10:30<22:03,  2.02it/s]
 75%|███████▌  | 8033/10707 [1:10:30<22:03,  2.02it/s]
 75%|███████▌  | 8034/10707 [1:10:31<22:05,  2.02it/s]
 75%|███████▌  | 8035/10707 [1:10:31<22:03,  2.02it/s]
 75%|███████▌  | 8036/10707 [1:10:32<22:02,  2.02it/s]
 75%|███████▌  | 8037/10707 [1:10:32<22:00,  2.02it/s]
 75%|███████▌  | 8038/10707 [1:10:33<22:00,  2.02it/s]
 75%|███████▌  | 8039/10707 [1:10:33<22:00,  2.02it/s]
 75%|███████▌  | 8040/10707 [1:10:34<22:00,  2.02it/s]
 75%|███████▌  | 8041/10707 [1:10:34<21:59,  2.02it/s]
 75%|███████▌  | 8042/10707 [1:10:35<21:59,  2.02it/s]
 75%|███████▌  | 8043/10707 [1:10:35<21:58,  2.02it/s]
 75%|███████▌  | 8044/10707 [1:10:36<21:58,  2.02it/s]
 75%|███████▌  | 8045/10707 [1:10:36<21:57,  2.02it/s]
 75%|███████▌  | 8046/10707 [1:10:37<21:59,  2.02it/s]
 75%|███████▌  | 8047/10707 [1:10:37<21:58,  2.02it/s]
 75%|███████▌  | 8048/10707 [1:10:38<21:58,  2.02it/s]
 75%|███████▌  | 8049/10707 [1:10:38<21:57,  2.02it/s]
 75%|███████▌  | 8050/10707 [1:10:39<21:56,  2.02it/s]{'loss': 3.6186, 'grad_norm': 0.2024877369403839, 'learning_rate': 0.00017615693474783828, 'epoch': 0.75}
-
                                                      

 75%|███████▌  | 8050/10707 [1:10:39<21:56,  2.02it/s]
 75%|███████▌  | 8051/10707 [1:10:39<21:57,  2.02it/s]
 75%|███████▌  | 8052/10707 [1:10:40<21:55,  2.02it/s]
 75%|███████▌  | 8053/10707 [1:10:40<21:55,  2.02it/s]
 75%|███████▌  | 8054/10707 [1:10:41<21:53,  2.02it/s]
 75%|███████▌  | 8055/10707 [1:10:41<21:52,  2.02it/s]
 75%|███████▌  | 8056/10707 [1:10:42<21:52,  2.02it/s]
 75%|███████▌  | 8057/10707 [1:10:42<21:53,  2.02it/s]
 75%|███████▌  | 8058/10707 [1:10:43<21:52,  2.02it/s]
 75%|███████▌  | 8059/10707 [1:10:43<21:52,  2.02it/s]
 75%|███████▌  | 8060/10707 [1:10:44<21:51,  2.02it/s]
 75%|███████▌  | 8061/10707 [1:10:44<21:50,  2.02it/s]
 75%|███████▌  | 8062/10707 [1:10:45<21:50,  2.02it/s]
 75%|███████▌  | 8063/10707 [1:10:45<21:48,  2.02it/s]
 75%|███████▌  | 8064/10707 [1:10:46<21:47,  2.02it/s]
 75%|███████▌  | 8065/10707 [1:10:46<21:47,  2.02it/s]
 75%|███████▌  | 8066/10707 [1:10:47<21:45,  2.02it/s]
 75%|███████▌  | 8067/10707 [1:10:47<21:45,  2.02it/s]
 75%|███████▌  | 8068/10707 [1:10:48<21:44,  2.02it/s]
 75%|███████▌  | 8069/10707 [1:10:48<21:45,  2.02it/s]
 75%|███████▌  | 8070/10707 [1:10:49<21:45,  2.02it/s]
 75%|███████▌  | 8071/10707 [1:10:49<21:45,  2.02it/s]
 75%|███████▌  | 8072/10707 [1:10:50<21:44,  2.02it/s]
 75%|███████▌  | 8073/10707 [1:10:50<21:44,  2.02it/s]
 75%|███████▌  | 8074/10707 [1:10:51<21:44,  2.02it/s]
 75%|███████▌  | 8075/10707 [1:10:51<21:43,  2.02it/s]{'loss': 3.6102, 'grad_norm': 0.19916808605194092, 'learning_rate': 0.00017306270029810556, 'epoch': 0.75}
-
                                                      

 75%|███████▌  | 8075/10707 [1:10:51<21:43,  2.02it/s]
 75%|███████▌  | 8076/10707 [1:10:52<21:44,  2.02it/s]
 75%|███████▌  | 8077/10707 [1:10:52<21:43,  2.02it/s]
 75%|███████▌  | 8078/10707 [1:10:53<21:43,  2.02it/s]
 75%|███████▌  | 8079/10707 [1:10:53<21:41,  2.02it/s]
 75%|███████▌  | 8080/10707 [1:10:54<21:40,  2.02it/s]
 75%|███████▌  | 8081/10707 [1:10:54<21:38,  2.02it/s]
 75%|███████▌  | 8082/10707 [1:10:55<21:38,  2.02it/s]
 75%|███████▌  | 8083/10707 [1:10:55<21:38,  2.02it/s]
 76%|███████▌  | 8084/10707 [1:10:56<21:38,  2.02it/s]
 76%|███████▌  | 8085/10707 [1:10:56<21:36,  2.02it/s]
 76%|███████▌  | 8086/10707 [1:10:57<21:36,  2.02it/s]
 76%|███████▌  | 8087/10707 [1:10:57<21:35,  2.02it/s]
 76%|███████▌  | 8088/10707 [1:10:58<21:35,  2.02it/s]
 76%|███████▌  | 8089/10707 [1:10:58<21:36,  2.02it/s]
 76%|███████▌  | 8090/10707 [1:10:59<21:35,  2.02it/s]
 76%|███████▌  | 8091/10707 [1:10:59<21:34,  2.02it/s]
 76%|███████▌  | 8092/10707 [1:11:00<21:34,  2.02it/s]
 76%|███████▌  | 8093/10707 [1:11:00<21:33,  2.02it/s]
 76%|███████▌  | 8094/10707 [1:11:01<21:32,  2.02it/s]
 76%|███████▌  | 8095/10707 [1:11:01<21:35,  2.02it/s]
 76%|███████▌  | 8096/10707 [1:11:02<21:36,  2.01it/s]
 76%|███████▌  | 8097/10707 [1:11:02<21:34,  2.02it/s]
 76%|███████▌  | 8098/10707 [1:11:03<21:33,  2.02it/s]
 76%|███████▌  | 8099/10707 [1:11:03<21:32,  2.02it/s]
 76%|███████▌  | 8100/10707 [1:11:04<21:33,  2.01it/s]{'loss': 3.6096, 'grad_norm': 0.20199918746948242, 'learning_rate': 0.00016999018526954774, 'epoch': 0.76}
-
                                                      

 76%|███████▌  | 8100/10707 [1:11:04<21:33,  2.01it/s]
 76%|███████▌  | 8101/10707 [1:11:04<21:35,  2.01it/s]
 76%|███████▌  | 8102/10707 [1:11:05<21:33,  2.01it/s]
 76%|███████▌  | 8103/10707 [1:11:05<21:32,  2.01it/s]
 76%|███████▌  | 8104/10707 [1:11:05<21:31,  2.02it/s]
 76%|███████▌  | 8105/10707 [1:11:06<21:29,  2.02it/s]
 76%|███████▌  | 8106/10707 [1:11:06<21:28,  2.02it/s]
 76%|███████▌  | 8107/10707 [1:11:07<21:28,  2.02it/s]
 76%|███████▌  | 8108/10707 [1:11:07<21:26,  2.02it/s]
 76%|███████▌  | 8109/10707 [1:11:08<21:26,  2.02it/s]
 76%|███████▌  | 8110/10707 [1:11:08<21:25,  2.02it/s]
 76%|███████▌  | 8111/10707 [1:11:09<21:24,  2.02it/s]
 76%|███████▌  | 8112/10707 [1:11:09<21:24,  2.02it/s]
 76%|███████▌  | 8113/10707 [1:11:10<21:25,  2.02it/s]
 76%|███████▌  | 8114/10707 [1:11:10<21:25,  2.02it/s]
 76%|███████▌  | 8115/10707 [1:11:11<21:24,  2.02it/s]
 76%|███████▌  | 8116/10707 [1:11:11<21:23,  2.02it/s]
 76%|███████▌  | 8117/10707 [1:11:12<21:22,  2.02it/s]
 76%|███████▌  | 8118/10707 [1:11:12<21:22,  2.02it/s]
 76%|███████▌  | 8119/10707 [1:11:13<21:21,  2.02it/s]
 76%|███████▌  | 8120/10707 [1:11:13<21:20,  2.02it/s]
 76%|███████▌  | 8121/10707 [1:11:14<21:19,  2.02it/s]
 76%|███████▌  | 8122/10707 [1:11:14<21:19,  2.02it/s]
 76%|███████▌  | 8123/10707 [1:11:15<21:18,  2.02it/s]
 76%|███████▌  | 8124/10707 [1:11:15<21:17,  2.02it/s]
 76%|███████▌  | 8125/10707 [1:11:16<21:17,  2.02it/s]{'loss': 3.6102, 'grad_norm': 0.19730238616466522, 'learning_rate': 0.00016693959377851438, 'epoch': 0.76}
-
                                                      

 76%|███████▌  | 8125/10707 [1:11:16<21:17,  2.02it/s]
 76%|███████▌  | 8126/10707 [1:11:16<21:19,  2.02it/s]
 76%|███████▌  | 8127/10707 [1:11:17<21:18,  2.02it/s]
 76%|███████▌  | 8128/10707 [1:11:17<21:18,  2.02it/s]
 76%|███████▌  | 8129/10707 [1:11:18<21:18,  2.02it/s]
 76%|███████▌  | 8130/10707 [1:11:18<21:17,  2.02it/s]
 76%|███████▌  | 8131/10707 [1:11:19<21:16,  2.02it/s]
 76%|███████▌  | 8132/10707 [1:11:19<21:15,  2.02it/s]
 76%|███████▌  | 8133/10707 [1:11:20<21:14,  2.02it/s]
 76%|███████▌  | 8134/10707 [1:11:20<21:14,  2.02it/s]
 76%|███████▌  | 8135/10707 [1:11:21<21:15,  2.02it/s]
 76%|███████▌  | 8136/10707 [1:11:21<21:12,  2.02it/s]
 76%|███████▌  | 8137/10707 [1:11:22<21:12,  2.02it/s]
 76%|███████▌  | 8138/10707 [1:11:22<21:10,  2.02it/s]
 76%|███████▌  | 8139/10707 [1:11:23<21:12,  2.02it/s]
 76%|███████▌  | 8140/10707 [1:11:23<21:10,  2.02it/s]
 76%|███████▌  | 8141/10707 [1:11:24<21:10,  2.02it/s]
 76%|███████▌  | 8142/10707 [1:11:24<21:09,  2.02it/s]
 76%|███████▌  | 8143/10707 [1:11:25<21:10,  2.02it/s]
 76%|███████▌  | 8144/10707 [1:11:25<21:07,  2.02it/s]
 76%|███████▌  | 8145/10707 [1:11:26<21:07,  2.02it/s]
 76%|███████▌  | 8146/10707 [1:11:26<21:06,  2.02it/s]
 76%|███████▌  | 8147/10707 [1:11:27<21:07,  2.02it/s]
 76%|███████▌  | 8148/10707 [1:11:27<21:05,  2.02it/s]
 76%|███████▌  | 8149/10707 [1:11:28<21:06,  2.02it/s]
 76%|███████▌  | 8150/10707 [1:11:28<21:04,  2.02it/s]{'loss': 3.6206, 'grad_norm': 0.20802269876003265, 'learning_rate': 0.00016391112848490786, 'epoch': 0.76}
-
                                                      

 76%|███████▌  | 8150/10707 [1:11:28<21:04,  2.02it/s]
 76%|███████▌  | 8151/10707 [1:11:29<21:06,  2.02it/s]
 76%|███████▌  | 8152/10707 [1:11:29<21:04,  2.02it/s]
 76%|███████▌  | 8153/10707 [1:11:30<21:04,  2.02it/s]
 76%|███████▌  | 8154/10707 [1:11:30<21:02,  2.02it/s]
 76%|███████▌  | 8155/10707 [1:11:31<21:02,  2.02it/s]
 76%|███████▌  | 8156/10707 [1:11:31<21:01,  2.02it/s]
 76%|███████▌  | 8157/10707 [1:11:32<21:01,  2.02it/s]
 76%|███████▌  | 8158/10707 [1:11:32<21:00,  2.02it/s]
 76%|███████▌  | 8159/10707 [1:11:33<21:00,  2.02it/s]
 76%|███████▌  | 8160/10707 [1:11:33<21:02,  2.02it/s]
 76%|███████▌  | 8161/10707 [1:11:34<21:03,  2.02it/s]
 76%|███████▌  | 8162/10707 [1:11:34<21:02,  2.02it/s]
 76%|███████▌  | 8163/10707 [1:11:35<21:00,  2.02it/s]
 76%|███████▌  | 8164/10707 [1:11:35<20:58,  2.02it/s]
 76%|███████▋  | 8165/10707 [1:11:36<20:57,  2.02it/s]
 76%|███████▋  | 8166/10707 [1:11:36<20:57,  2.02it/s]
 76%|███████▋  | 8167/10707 [1:11:37<20:57,  2.02it/s]
 76%|███████▋  | 8168/10707 [1:11:37<20:56,  2.02it/s]
 76%|███████▋  | 8169/10707 [1:11:38<20:56,  2.02it/s]
 76%|███████▋  | 8170/10707 [1:11:38<20:55,  2.02it/s]
 76%|███████▋  | 8171/10707 [1:11:39<20:55,  2.02it/s]
 76%|███████▋  | 8172/10707 [1:11:39<20:56,  2.02it/s]
 76%|███████▋  | 8173/10707 [1:11:40<20:55,  2.02it/s]
 76%|███████▋  | 8174/10707 [1:11:40<20:55,  2.02it/s]
 76%|███████▋  | 8175/10707 [1:11:41<20:56,  2.02it/s]{'loss': 3.6129, 'grad_norm': 0.20088517665863037, 'learning_rate': 0.00016090499057872193, 'epoch': 0.76}
                                                      
-
 76%|███████▋  | 8175/10707 [1:11:41<20:56,  2.02it/s]
 76%|███████▋  | 8176/10707 [1:11:41<20:56,  2.01it/s]
 76%|███████▋  | 8177/10707 [1:11:42<20:55,  2.02it/s]
 76%|███████▋  | 8178/10707 [1:11:42<20:53,  2.02it/s]
 76%|███████▋  | 8179/10707 [1:11:43<20:52,  2.02it/s]
 76%|███████▋  | 8180/10707 [1:11:43<20:50,  2.02it/s]
 76%|███████▋  | 8181/10707 [1:11:44<20:51,  2.02it/s]
 76%|███████▋  | 8182/10707 [1:11:44<20:50,  2.02it/s]
 76%|███████▋  | 8183/10707 [1:11:45<20:51,  2.02it/s]
 76%|███████▋  | 8184/10707 [1:11:45<20:50,  2.02it/s]
 76%|███████▋  | 8185/10707 [1:11:46<20:50,  2.02it/s]
 76%|███████▋  | 8186/10707 [1:11:46<20:49,  2.02it/s]
 76%|███████▋  | 8187/10707 [1:11:47<20:48,  2.02it/s]
 76%|███████▋  | 8188/10707 [1:11:47<20:48,  2.02it/s]
 76%|███████▋  | 8189/10707 [1:11:48<20:47,  2.02it/s]
 76%|███████▋  | 8190/10707 [1:11:48<20:46,  2.02it/s]
 77%|███████▋  | 8191/10707 [1:11:49<20:45,  2.02it/s]
 77%|███████▋  | 8192/10707 [1:11:49<20:44,  2.02it/s]
 77%|███████▋  | 8193/10707 [1:11:50<20:44,  2.02it/s]
 77%|███████▋  | 8194/10707 [1:11:50<20:44,  2.02it/s]
 77%|███████▋  | 8195/10707 [1:11:51<20:44,  2.02it/s]
 77%|███████▋  | 8196/10707 [1:11:51<20:44,  2.02it/s]
 77%|███████▋  | 8197/10707 [1:11:52<20:43,  2.02it/s]
 77%|███████▋  | 8198/10707 [1:11:52<20:42,  2.02it/s]
 77%|███████▋  | 8199/10707 [1:11:53<20:41,  2.02it/s]
 77%|███████▋  | 8200/10707 [1:11:53<20:41,  2.02it/s]{'loss': 3.6068, 'grad_norm': 0.20051276683807373, 'learning_rate': 0.00015792137976667508, 'epoch': 0.77}
-
                                                      

 77%|███████▋  | 8200/10707 [1:11:53<20:41,  2.02it/s]
 77%|███████▋  | 8201/10707 [1:11:54<20:42,  2.02it/s]
 77%|███████▋  | 8202/10707 [1:11:54<20:40,  2.02it/s]
 77%|███████▋  | 8203/10707 [1:11:55<20:40,  2.02it/s]
 77%|███████▋  | 8204/10707 [1:11:55<20:40,  2.02it/s]
 77%|███████▋  | 8205/10707 [1:11:56<20:40,  2.02it/s]
 77%|███████▋  | 8206/10707 [1:11:56<20:39,  2.02it/s]
 77%|███████▋  | 8207/10707 [1:11:57<20:38,  2.02it/s]
 77%|███████▋  | 8208/10707 [1:11:57<20:37,  2.02it/s]
 77%|███████▋  | 8209/10707 [1:11:57<20:36,  2.02it/s]
 77%|███████▋  | 8210/10707 [1:11:58<20:36,  2.02it/s]
 77%|███████▋  | 8211/10707 [1:11:58<20:35,  2.02it/s]
 77%|███████▋  | 8212/10707 [1:11:59<20:34,  2.02it/s]
 77%|███████▋  | 8213/10707 [1:11:59<20:33,  2.02it/s]
 77%|███████▋  | 8214/10707 [1:12:00<20:32,  2.02it/s]
 77%|███████▋  | 8215/10707 [1:12:00<20:31,  2.02it/s]
 77%|███████▋  | 8216/10707 [1:12:01<20:32,  2.02it/s]
 77%|███████▋  | 8217/10707 [1:12:02<23:48,  1.74it/s]
 77%|███████▋  | 8218/10707 [1:12:02<22:50,  1.82it/s]
 77%|███████▋  | 8219/10707 [1:12:03<22:08,  1.87it/s]
 77%|███████▋  | 8220/10707 [1:12:03<21:41,  1.91it/s]
 77%|███████▋  | 8221/10707 [1:12:04<21:21,  1.94it/s]
 77%|███████▋  | 8222/10707 [1:12:04<21:05,  1.96it/s]
 77%|███████▋  | 8223/10707 [1:12:05<20:53,  1.98it/s]
 77%|███████▋  | 8224/10707 [1:12:05<20:45,  1.99it/s]
 77%|███████▋  | 8225/10707 [1:12:06<20:40,  2.00it/s]
                                                      
{'loss': 3.6053, 'grad_norm': 0.20060594379901886, 'learning_rate': 0.0001549604942589441, 'epoch': 0.77}
-
 77%|███████▋  | 8225/10707 [1:12:06<20:40,  2.00it/s]
 77%|███████▋  | 8226/10707 [1:12:06<20:39,  2.00it/s]
 77%|███████▋  | 8227/10707 [1:12:07<20:34,  2.01it/s]
 77%|███████▋  | 8228/10707 [1:12:07<20:32,  2.01it/s]
 77%|███████▋  | 8229/10707 [1:12:08<20:29,  2.01it/s]
 77%|███████▋  | 8230/10707 [1:12:08<20:28,  2.02it/s]
 77%|███████▋  | 8231/10707 [1:12:09<20:26,  2.02it/s]
 77%|███████▋  | 8232/10707 [1:12:09<20:26,  2.02it/s]
 77%|███████▋  | 8233/10707 [1:12:10<20:24,  2.02it/s]
 77%|███████▋  | 8234/10707 [1:12:10<20:24,  2.02it/s]
 77%|███████▋  | 8235/10707 [1:12:11<20:22,  2.02it/s]
 77%|███████▋  | 8236/10707 [1:12:11<20:22,  2.02it/s]
 77%|███████▋  | 8237/10707 [1:12:12<20:22,  2.02it/s]
 77%|███████▋  | 8238/10707 [1:12:12<20:23,  2.02it/s]
 77%|███████▋  | 8239/10707 [1:12:13<20:21,  2.02it/s]
 77%|███████▋  | 8240/10707 [1:12:13<20:22,  2.02it/s]
 77%|███████▋  | 8241/10707 [1:12:14<20:20,  2.02it/s]
 77%|███████▋  | 8242/10707 [1:12:14<20:20,  2.02it/s]
 77%|███████▋  | 8243/10707 [1:12:15<20:19,  2.02it/s]
 77%|███████▋  | 8244/10707 [1:12:15<20:19,  2.02it/s]
 77%|███████▋  | 8245/10707 [1:12:16<20:17,  2.02it/s]
 77%|███████▋  | 8246/10707 [1:12:16<20:17,  2.02it/s]
 77%|███████▋  | 8247/10707 [1:12:17<20:16,  2.02it/s]
 77%|███████▋  | 8248/10707 [1:12:17<20:16,  2.02it/s]
 77%|███████▋  | 8249/10707 [1:12:18<20:16,  2.02it/s]
 77%|███████▋  | 8250/10707 [1:12:18<20:16,  2.02it/s]{'loss': 3.6084, 'grad_norm': 0.2028883993625641, 'learning_rate': 0.0001520225307559951, 'epoch': 0.77}
-
                                                      

 77%|███████▋  | 8250/10707 [1:12:18<20:16,  2.02it/s]
 77%|███████▋  | 8251/10707 [1:12:19<20:17,  2.02it/s]
 77%|███████▋  | 8252/10707 [1:12:19<20:15,  2.02it/s]
 77%|███████▋  | 8253/10707 [1:12:20<23:27,  1.74it/s]
 77%|███████▋  | 8254/10707 [1:12:20<22:28,  1.82it/s]
 77%|███████▋  | 8255/10707 [1:12:21<21:46,  1.88it/s]
 77%|███████▋  | 8256/10707 [1:12:21<21:18,  1.92it/s]
 77%|███████▋  | 8257/10707 [1:12:22<20:57,  1.95it/s]
 77%|███████▋  | 8258/10707 [1:12:22<20:44,  1.97it/s]
 77%|███████▋  | 8259/10707 [1:12:23<20:31,  1.99it/s]
 77%|███████▋  | 8260/10707 [1:12:23<20:26,  2.00it/s]
 77%|███████▋  | 8261/10707 [1:12:24<20:19,  2.01it/s]
 77%|███████▋  | 8262/10707 [1:12:24<20:17,  2.01it/s]
 77%|███████▋  | 8263/10707 [1:12:25<20:13,  2.01it/s]
 77%|███████▋  | 8264/10707 [1:12:25<20:11,  2.02it/s]
 77%|███████▋  | 8265/10707 [1:12:26<20:09,  2.02it/s]
 77%|███████▋  | 8266/10707 [1:12:26<20:08,  2.02it/s]
 77%|███████▋  | 8267/10707 [1:12:27<20:06,  2.02it/s]
 77%|███████▋  | 8268/10707 [1:12:27<20:07,  2.02it/s]
 77%|███████▋  | 8269/10707 [1:12:28<20:05,  2.02it/s]
 77%|███████▋  | 8270/10707 [1:12:28<20:05,  2.02it/s]
 77%|███████▋  | 8271/10707 [1:12:29<20:05,  2.02it/s]
 77%|███████▋  | 8272/10707 [1:12:29<20:06,  2.02it/s]
 77%|███████▋  | 8273/10707 [1:12:30<20:04,  2.02it/s]
 77%|███████▋  | 8274/10707 [1:12:30<20:04,  2.02it/s]
 77%|███████▋  | 8275/10707 [1:12:31<20:03,  2.02it/s]
                                                      {'loss': 3.6055, 'grad_norm': 0.2071843147277832, 'learning_rate': 0.00014910768443551764, 'epoch': 0.77}
-

 77%|███████▋  | 8275/10707 [1:12:31<20:03,  2.02it/s]
 77%|███████▋  | 8276/10707 [1:12:31<20:05,  2.02it/s]
 77%|███████▋  | 8277/10707 [1:12:32<20:03,  2.02it/s]
 77%|███████▋  | 8278/10707 [1:12:32<20:03,  2.02it/s]
 77%|███████▋  | 8279/10707 [1:12:33<20:02,  2.02it/s]
 77%|███████▋  | 8280/10707 [1:12:33<20:06,  2.01it/s]
 77%|███████▋  | 8281/10707 [1:12:34<20:06,  2.01it/s]
 77%|███████▋  | 8282/10707 [1:12:34<20:02,  2.02it/s]
 77%|███████▋  | 8283/10707 [1:12:35<20:02,  2.02it/s]
 77%|███████▋  | 8284/10707 [1:12:35<20:00,  2.02it/s]
 77%|███████▋  | 8285/10707 [1:12:36<20:00,  2.02it/s]
 77%|███████▋  | 8286/10707 [1:12:36<19:59,  2.02it/s]
 77%|███████▋  | 8287/10707 [1:12:37<19:59,  2.02it/s]
 77%|███████▋  | 8288/10707 [1:12:37<19:58,  2.02it/s]
 77%|███████▋  | 8289/10707 [1:12:38<19:58,  2.02it/s]
 77%|███████▋  | 8290/10707 [1:12:38<19:58,  2.02it/s]
 77%|███████▋  | 8291/10707 [1:12:39<19:58,  2.02it/s]
 77%|███████▋  | 8292/10707 [1:12:39<19:57,  2.02it/s]
 77%|███████▋  | 8293/10707 [1:12:40<19:56,  2.02it/s]
 77%|███████▋  | 8294/10707 [1:12:40<19:54,  2.02it/s]
 77%|███████▋  | 8295/10707 [1:12:41<19:54,  2.02it/s]
 77%|███████▋  | 8296/10707 [1:12:41<19:53,  2.02it/s]
 77%|███████▋  | 8297/10707 [1:12:42<19:53,  2.02it/s]
 78%|███████▊  | 8298/10707 [1:12:42<19:51,  2.02it/s]
 78%|███████▊  | 8299/10707 [1:12:43<19:51,  2.02it/s]
 78%|███████▊  | 8300/10707 [1:12:43<19:50,  2.02it/s]{'loss': 3.6145, 'grad_norm': 0.20146432518959045, 'learning_rate': 0.00014621614893945757, 'epoch': 0.78}
-
                                                      

 78%|███████▊  | 8300/10707 [1:12:43<19:50,  2.02it/s]
 78%|███████▊  | 8301/10707 [1:12:44<19:53,  2.02it/s]
 78%|███████▊  | 8302/10707 [1:12:44<19:51,  2.02it/s]
 78%|███████▊  | 8303/10707 [1:12:45<19:51,  2.02it/s]
 78%|███████▊  | 8304/10707 [1:12:45<19:49,  2.02it/s]
 78%|███████▊  | 8305/10707 [1:12:46<19:50,  2.02it/s]
 78%|███████▊  | 8306/10707 [1:12:46<19:49,  2.02it/s]
 78%|███████▊  | 8307/10707 [1:12:47<19:48,  2.02it/s]
 78%|███████▊  | 8308/10707 [1:12:47<19:49,  2.02it/s]
 78%|███████▊  | 8309/10707 [1:12:48<19:48,  2.02it/s]
 78%|███████▊  | 8310/10707 [1:12:48<19:48,  2.02it/s]
 78%|███████▊  | 8311/10707 [1:12:49<19:45,  2.02it/s]
 78%|███████▊  | 8312/10707 [1:12:49<19:45,  2.02it/s]
 78%|███████▊  | 8313/10707 [1:12:50<19:44,  2.02it/s]
 78%|███████▊  | 8314/10707 [1:12:50<19:44,  2.02it/s]
 78%|███████▊  | 8315/10707 [1:12:51<19:44,  2.02it/s]
 78%|███████▊  | 8316/10707 [1:12:51<19:44,  2.02it/s]
 78%|███████▊  | 8317/10707 [1:12:51<19:42,  2.02it/s]
 78%|███████▊  | 8318/10707 [1:12:52<19:42,  2.02it/s]
 78%|███████▊  | 8319/10707 [1:12:52<19:41,  2.02it/s]
 78%|███████▊  | 8320/10707 [1:12:53<19:42,  2.02it/s]
 78%|███████▊  | 8321/10707 [1:12:53<19:40,  2.02it/s]
 78%|███████▊  | 8322/10707 [1:12:54<19:41,  2.02it/s]
 78%|███████▊  | 8323/10707 [1:12:54<19:40,  2.02it/s]
 78%|███████▊  | 8324/10707 [1:12:55<19:40,  2.02it/s]
 78%|███████▊  | 8325/10707 [1:12:55<19:38,  2.02it/s]{'loss': 3.6076, 'grad_norm': 0.19519568979740143, 'learning_rate': 0.00014334811636115304, 'epoch': 0.78}
-
                                                      

 78%|███████▊  | 8325/10707 [1:12:55<19:38,  2.02it/s]
 78%|███████▊  | 8326/10707 [1:12:56<19:40,  2.02it/s]
 78%|███████▊  | 8327/10707 [1:12:56<19:37,  2.02it/s]
 78%|███████▊  | 8328/10707 [1:12:57<19:37,  2.02it/s]
 78%|███████▊  | 8329/10707 [1:12:57<19:35,  2.02it/s]
 78%|███████▊  | 8330/10707 [1:12:58<19:36,  2.02it/s]
 78%|███████▊  | 8331/10707 [1:12:58<19:35,  2.02it/s]
 78%|███████▊  | 8332/10707 [1:12:59<19:35,  2.02it/s]
 78%|███████▊  | 8333/10707 [1:12:59<19:35,  2.02it/s]
 78%|███████▊  | 8334/10707 [1:13:00<19:34,  2.02it/s]
 78%|███████▊  | 8335/10707 [1:13:00<19:34,  2.02it/s]
 78%|███████▊  | 8336/10707 [1:13:01<19:33,  2.02it/s]
 78%|███████▊  | 8337/10707 [1:13:01<19:36,  2.01it/s]
 78%|███████▊  | 8338/10707 [1:13:02<19:35,  2.02it/s]
 78%|███████▊  | 8339/10707 [1:13:02<19:35,  2.01it/s]
 78%|███████▊  | 8340/10707 [1:13:03<19:34,  2.02it/s]
 78%|███████▊  | 8341/10707 [1:13:03<19:35,  2.01it/s]
 78%|███████▊  | 8342/10707 [1:13:04<19:34,  2.01it/s]
 78%|███████▊  | 8343/10707 [1:13:04<19:32,  2.02it/s]
 78%|███████▊  | 8344/10707 [1:13:05<19:31,  2.02it/s]
 78%|███████▊  | 8345/10707 [1:13:05<19:30,  2.02it/s]
 78%|███████▊  | 8346/10707 [1:13:06<19:31,  2.02it/s]
 78%|███████▊  | 8347/10707 [1:13:06<19:29,  2.02it/s]
 78%|███████▊  | 8348/10707 [1:13:07<19:28,  2.02it/s]
 78%|███████▊  | 8349/10707 [1:13:07<19:26,  2.02it/s]
 78%|███████▊  | 8350/10707 [1:13:08<19:26,  2.02it/s]{'loss': 3.6072, 'grad_norm': 0.19890965521335602, 'learning_rate': 0.0001405037772325733, 'epoch': 0.78}
-
                                                      

 78%|███████▊  | 8350/10707 [1:13:08<19:26,  2.02it/s]
 78%|███████▊  | 8351/10707 [1:13:08<19:26,  2.02it/s]
 78%|███████▊  | 8352/10707 [1:13:09<19:26,  2.02it/s]
 78%|███████▊  | 8353/10707 [1:13:09<19:26,  2.02it/s]
 78%|███████▊  | 8354/10707 [1:13:10<19:25,  2.02it/s]
 78%|███████▊  | 8355/10707 [1:13:10<19:24,  2.02it/s]
 78%|███████▊  | 8356/10707 [1:13:11<19:23,  2.02it/s]
 78%|███████▊  | 8357/10707 [1:13:11<19:22,  2.02it/s]
 78%|███████▊  | 8358/10707 [1:13:12<19:22,  2.02it/s]
 78%|███████▊  | 8359/10707 [1:13:12<19:21,  2.02it/s]
 78%|███████▊  | 8360/10707 [1:13:13<19:20,  2.02it/s]
 78%|███████▊  | 8361/10707 [1:13:13<19:21,  2.02it/s]
 78%|███████▊  | 8362/10707 [1:13:14<19:19,  2.02it/s]
 78%|███████▊  | 8363/10707 [1:13:14<19:19,  2.02it/s]
 78%|███████▊  | 8364/10707 [1:13:15<19:18,  2.02it/s]
 78%|███████▊  | 8365/10707 [1:13:15<19:18,  2.02it/s]
 78%|███████▊  | 8366/10707 [1:13:16<19:18,  2.02it/s]
 78%|███████▊  | 8367/10707 [1:13:16<19:18,  2.02it/s]
 78%|███████▊  | 8368/10707 [1:13:17<19:18,  2.02it/s]
 78%|███████▊  | 8369/10707 [1:13:17<19:18,  2.02it/s]
 78%|███████▊  | 8370/10707 [1:13:18<19:16,  2.02it/s]
 78%|███████▊  | 8371/10707 [1:13:18<19:17,  2.02it/s]
 78%|███████▊  | 8372/10707 [1:13:19<19:14,  2.02it/s]
 78%|███████▊  | 8373/10707 [1:13:19<19:14,  2.02it/s]
 78%|███████▊  | 8374/10707 [1:13:20<19:13,  2.02it/s]
 78%|███████▊  | 8375/10707 [1:13:20<19:14,  2.02it/s]{'loss': 3.6026, 'grad_norm': 0.19742442667484283, 'learning_rate': 0.00013768332051166088, 'epoch': 0.78}
-
                                                      

 78%|███████▊  | 8375/10707 [1:13:20<19:14,  2.02it/s]
 78%|███████▊  | 8376/10707 [1:13:21<19:15,  2.02it/s]
 78%|███████▊  | 8377/10707 [1:13:21<19:14,  2.02it/s]
 78%|███████▊  | 8378/10707 [1:13:22<19:13,  2.02it/s]
 78%|███████▊  | 8379/10707 [1:13:22<19:13,  2.02it/s]
 78%|███████▊  | 8380/10707 [1:13:23<19:12,  2.02it/s]
 78%|███████▊  | 8381/10707 [1:13:23<19:12,  2.02it/s]
 78%|███████▊  | 8382/10707 [1:13:24<19:10,  2.02it/s]
 78%|███████▊  | 8383/10707 [1:13:24<19:10,  2.02it/s]
 78%|███████▊  | 8384/10707 [1:13:25<19:09,  2.02it/s]
 78%|███████▊  | 8385/10707 [1:13:25<19:08,  2.02it/s]
 78%|███████▊  | 8386/10707 [1:13:26<19:08,  2.02it/s]
 78%|███████▊  | 8387/10707 [1:13:26<19:09,  2.02it/s]
 78%|███████▊  | 8388/10707 [1:13:27<19:08,  2.02it/s]
 78%|███████▊  | 8389/10707 [1:13:27<19:07,  2.02it/s]
 78%|███████▊  | 8390/10707 [1:13:28<19:07,  2.02it/s]
 78%|███████▊  | 8391/10707 [1:13:28<19:06,  2.02it/s]
 78%|███████▊  | 8392/10707 [1:13:29<19:05,  2.02it/s]
 78%|███████▊  | 8393/10707 [1:13:29<19:05,  2.02it/s]
 78%|███████▊  | 8394/10707 [1:13:30<19:04,  2.02it/s]
 78%|███████▊  | 8395/10707 [1:13:30<19:03,  2.02it/s]
 78%|███████▊  | 8396/10707 [1:13:31<19:03,  2.02it/s]
 78%|███████▊  | 8397/10707 [1:13:31<19:02,  2.02it/s]
 78%|███████▊  | 8398/10707 [1:13:32<19:01,  2.02it/s]
 78%|███████▊  | 8399/10707 [1:13:32<19:01,  2.02it/s]
 78%|███████▊  | 8400/10707 [1:13:33<19:00,  2.02it/s]
                                                      {'loss': 3.6065, 'grad_norm': 0.20854869484901428, 'learning_rate': 0.00013488693356977872, 'epoch': 0.78}
-
 78%|███████▊  | 8400/10707 [1:13:33<19:00,  2.02it/s]
 78%|███████▊  | 8401/10707 [1:13:33<19:03,  2.02it/s]
 78%|███████▊  | 8402/10707 [1:13:34<19:04,  2.01it/s]
 78%|███████▊  | 8403/10707 [1:13:34<19:02,  2.02it/s]
 78%|███████▊  | 8404/10707 [1:13:35<19:01,  2.02it/s]
 79%|███████▊  | 8405/10707 [1:13:35<18:59,  2.02it/s]
 79%|███████▊  | 8406/10707 [1:13:36<18:59,  2.02it/s]
 79%|███████▊  | 8407/10707 [1:13:36<18:59,  2.02it/s]
 79%|███████▊  | 8408/10707 [1:13:37<18:59,  2.02it/s]
 79%|███████▊  | 8409/10707 [1:13:37<18:57,  2.02it/s]
 79%|███████▊  | 8410/10707 [1:13:38<18:56,  2.02it/s]
 79%|███████▊  | 8411/10707 [1:13:38<18:55,  2.02it/s]
 79%|███████▊  | 8412/10707 [1:13:39<18:56,  2.02it/s]
 79%|███████▊  | 8413/10707 [1:13:39<18:54,  2.02it/s]
 79%|███████▊  | 8414/10707 [1:13:40<18:54,  2.02it/s]
 79%|███████▊  | 8415/10707 [1:13:40<18:53,  2.02it/s]
 79%|███████▊  | 8416/10707 [1:13:41<18:53,  2.02it/s]
 79%|███████▊  | 8417/10707 [1:13:41<18:51,  2.02it/s]
 79%|███████▊  | 8418/10707 [1:13:42<18:51,  2.02it/s]
 79%|███████▊  | 8419/10707 [1:13:42<18:50,  2.02it/s]
 79%|███████▊  | 8420/10707 [1:13:42<18:52,  2.02it/s]
 79%|███████▊  | 8421/10707 [1:13:43<18:50,  2.02it/s]
 79%|███████▊  | 8422/10707 [1:13:43<18:50,  2.02it/s]
 79%|███████▊  | 8423/10707 [1:13:44<18:48,  2.02it/s]
 79%|███████▊  | 8424/10707 [1:13:44<18:49,  2.02it/s]
 79%|███████▊  | 8425/10707 [1:13:45<18:48,  2.02it/s]{'loss': 3.6008, 'grad_norm': 0.1994030624628067, 'learning_rate': 0.0001321148021792624, 'epoch': 0.79}
-
                                                      

 79%|███████▊  | 8425/10707 [1:13:45<18:48,  2.02it/s]
 79%|███████▊  | 8426/10707 [1:13:45<18:48,  2.02it/s]
 79%|███████▊  | 8427/10707 [1:13:46<18:47,  2.02it/s]
 79%|███████▊  | 8428/10707 [1:13:46<18:47,  2.02it/s]
 79%|███████▊  | 8429/10707 [1:13:47<18:47,  2.02it/s]
 79%|███████▊  | 8430/10707 [1:13:47<18:47,  2.02it/s]
 79%|███████▊  | 8431/10707 [1:13:48<18:46,  2.02it/s]
 79%|███████▉  | 8432/10707 [1:13:48<18:45,  2.02it/s]
 79%|███████▉  | 8433/10707 [1:13:49<18:44,  2.02it/s]
 79%|███████▉  | 8434/10707 [1:13:49<18:44,  2.02it/s]
 79%|███████▉  | 8435/10707 [1:13:50<18:43,  2.02it/s]
 79%|███████▉  | 8436/10707 [1:13:50<18:43,  2.02it/s]
 79%|███████▉  | 8437/10707 [1:13:51<18:43,  2.02it/s]
 79%|███████▉  | 8438/10707 [1:13:51<18:43,  2.02it/s]
 79%|███████▉  | 8439/10707 [1:13:52<18:42,  2.02it/s]
 79%|███████▉  | 8440/10707 [1:13:52<18:42,  2.02it/s]
 79%|███████▉  | 8441/10707 [1:13:53<18:41,  2.02it/s]
 79%|███████▉  | 8442/10707 [1:13:53<18:41,  2.02it/s]
 79%|███████▉  | 8443/10707 [1:13:54<18:40,  2.02it/s]
 79%|███████▉  | 8444/10707 [1:13:54<18:39,  2.02it/s]
 79%|███████▉  | 8445/10707 [1:13:55<18:39,  2.02it/s]
 79%|███████▉  | 8446/10707 [1:13:55<18:38,  2.02it/s]
 79%|███████▉  | 8447/10707 [1:13:56<18:38,  2.02it/s]
 79%|███████▉  | 8448/10707 [1:13:56<18:37,  2.02it/s]
 79%|███████▉  | 8449/10707 [1:13:57<18:36,  2.02it/s]
 79%|███████▉  | 8450/10707 [1:13:57<18:36,  2.02it/s]
                                                      
{'loss': 3.6038, 'grad_norm': 0.20027849078178406, 'learning_rate': 0.00012936711050107874, 'epoch': 0.79}
-
 79%|███████▉  | 8450/10707 [1:13:57<18:36,  2.02it/s]
 79%|███████▉  | 8451/10707 [1:13:58<18:37,  2.02it/s]
 79%|███████▉  | 8452/10707 [1:13:58<18:35,  2.02it/s]
 79%|███████▉  | 8453/10707 [1:13:59<18:34,  2.02it/s]
 79%|███████▉  | 8454/10707 [1:13:59<18:34,  2.02it/s]
 79%|███████▉  | 8455/10707 [1:14:00<18:34,  2.02it/s]
 79%|███████▉  | 8456/10707 [1:14:00<18:34,  2.02it/s]
 79%|███████▉  | 8457/10707 [1:14:01<18:33,  2.02it/s]
 79%|███████▉  | 8458/10707 [1:14:01<18:33,  2.02it/s]
 79%|███████▉  | 8459/10707 [1:14:02<18:32,  2.02it/s]
 79%|███████▉  | 8460/10707 [1:14:02<18:31,  2.02it/s]
 79%|███████▉  | 8461/10707 [1:14:03<18:32,  2.02it/s]
 79%|███████▉  | 8462/10707 [1:14:03<18:33,  2.02it/s]
 79%|███████▉  | 8463/10707 [1:14:04<18:33,  2.01it/s]
 79%|███████▉  | 8464/10707 [1:14:04<18:31,  2.02it/s]
 79%|███████▉  | 8465/10707 [1:14:05<18:30,  2.02it/s]
 79%|███████▉  | 8466/10707 [1:14:05<18:28,  2.02it/s]
 79%|███████▉  | 8467/10707 [1:14:06<18:28,  2.02it/s]
 79%|███████▉  | 8468/10707 [1:14:06<18:27,  2.02it/s]
 79%|███████▉  | 8469/10707 [1:14:07<18:26,  2.02it/s]
 79%|███████▉  | 8470/10707 [1:14:07<18:27,  2.02it/s]
 79%|███████▉  | 8471/10707 [1:14:08<18:27,  2.02it/s]
 79%|███████▉  | 8472/10707 [1:14:08<18:26,  2.02it/s]
 79%|███████▉  | 8473/10707 [1:14:09<18:25,  2.02it/s]
 79%|███████▉  | 8474/10707 [1:14:09<18:24,  2.02it/s]
 79%|█��█████▉  | 8475/10707 [1:14:10<18:24,  2.02it/s]{'loss': 3.5979, 'grad_norm': 0.20495913922786713, 'learning_rate': 0.00012664404107259165, 'epoch': 0.79}
                                                      
-
 79%|███████▉  | 8475/10707 [1:14:10<18:24,  2.02it/s]
 79%|███████▉  | 8476/10707 [1:14:10<18:24,  2.02it/s]
 79%|███████▉  | 8477/10707 [1:14:11<18:23,  2.02it/s]
 79%|███████▉  | 8478/10707 [1:14:11<18:24,  2.02it/s]
 79%|███████▉  | 8479/10707 [1:14:12<18:22,  2.02it/s]
 79%|███████▉  | 8480/10707 [1:14:12<18:22,  2.02it/s]
 79%|███████▉  | 8481/10707 [1:14:13<18:21,  2.02it/s]
 79%|███████▉  | 8482/10707 [1:14:13<18:21,  2.02it/s]
 79%|███████▉  | 8483/10707 [1:14:14<18:20,  2.02it/s]
 79%|███████▉  | 8484/10707 [1:14:14<18:20,  2.02it/s]
 79%|███████▉  | 8485/10707 [1:14:15<18:18,  2.02it/s]
 79%|███████▉  | 8486/10707 [1:14:15<18:19,  2.02it/s]
 79%|███████▉  | 8487/10707 [1:14:16<18:17,  2.02it/s]
 79%|███████▉  | 8488/10707 [1:14:16<18:17,  2.02it/s]
 79%|███████▉  | 8489/10707 [1:14:17<18:17,  2.02it/s]
 79%|███████▉  | 8490/10707 [1:14:17<18:17,  2.02it/s]
 79%|███████▉  | 8491/10707 [1:14:18<18:16,  2.02it/s]
 79%|███████▉  | 8492/10707 [1:14:18<18:16,  2.02it/s]
 79%|███████▉  | 8493/10707 [1:14:19<18:15,  2.02it/s]
 79%|███████▉  | 8494/10707 [1:14:19<18:15,  2.02it/s]
 79%|███████▉  | 8495/10707 [1:14:20<18:14,  2.02it/s]
 79%|███████▉  | 8496/10707 [1:14:20<18:13,  2.02it/s]
 79%|███████▉  | 8497/10707 [1:14:21<18:14,  2.02it/s]
 79%|███████▉  | 8498/10707 [1:14:21<18:12,  2.02it/s]
 79%|███████▉  | 8499/10707 [1:14:22<18:12,  2.02it/s]
 79%|███████▉  | 8500/10707 [1:14:22<18:12,  2.02it/s]{'loss': 3.6088, 'grad_norm': 0.20116840302944183, 'learning_rate': 0.00012394577479543528, 'epoch': 0.79}
-
                                                      

 79%|███████▉  | 8500/10707 [1:14:22<18:12,  2.02it/s]
 79%|███████▉  | 8501/10707 [1:14:23<18:12,  2.02it/s]
 79%|███████▉  | 8502/10707 [1:14:23<18:11,  2.02it/s]
 79%|███████▉  | 8503/10707 [1:14:24<18:11,  2.02it/s]
 79%|███████▉  | 8504/10707 [1:14:24<18:11,  2.02it/s]
 79%|███████▉  | 8505/10707 [1:14:25<18:12,  2.02it/s]
 79%|███████▉  | 8506/10707 [1:14:25<18:10,  2.02it/s]
 79%|███████▉  | 8507/10707 [1:14:26<18:10,  2.02it/s]
 79%|███████▉  | 8508/10707 [1:14:26<18:09,  2.02it/s]
 79%|███████▉  | 8509/10707 [1:14:27<18:08,  2.02it/s]
 79%|███████▉  | 8510/10707 [1:14:27<18:08,  2.02it/s]
 79%|███████▉  | 8511/10707 [1:14:28<18:08,  2.02it/s]
 79%|███████▉  | 8512/10707 [1:14:28<18:06,  2.02it/s]
 80%|███████▉  | 8513/10707 [1:14:29<18:05,  2.02it/s]
 80%|███████▉  | 8514/10707 [1:14:29<18:05,  2.02it/s]
 80%|███████▉  | 8515/10707 [1:14:30<18:05,  2.02it/s]
 80%|███████▉  | 8516/10707 [1:14:30<18:05,  2.02it/s]
 80%|███████▉  | 8517/10707 [1:14:31<18:03,  2.02it/s]
 80%|███████▉  | 8518/10707 [1:14:31<18:03,  2.02it/s]
 80%|███████▉  | 8519/10707 [1:14:31<18:02,  2.02it/s]
 80%|███████▉  | 8520/10707 [1:14:32<18:02,  2.02it/s]
 80%|███████▉  | 8521/10707 [1:14:32<18:01,  2.02it/s]
 80%|███████▉  | 8522/10707 [1:14:33<18:03,  2.02it/s]
 80%|███████▉  | 8523/10707 [1:14:33<18:03,  2.02it/s]
 80%|███████▉  | 8524/10707 [1:14:34<18:02,  2.02it/s]
 80%|███████▉  | 8525/10707 [1:14:34<18:01,  2.02it/s]
                                                      
{'loss': 3.6057, 'grad_norm': 0.197871133685112, 'learning_rate': 0.00012127249092349646, 'epoch': 0.8}
-
 80%|███████▉  | 8525/10707 [1:14:34<18:01,  2.02it/s]
 80%|███████▉  | 8526/10707 [1:14:35<18:02,  2.02it/s]
 80%|███████▉  | 8527/10707 [1:14:35<18:00,  2.02it/s]
 80%|███████▉  | 8528/10707 [1:14:36<18:00,  2.02it/s]
 80%|███████▉  | 8529/10707 [1:14:36<17:59,  2.02it/s]
 80%|███████▉  | 8530/10707 [1:14:37<17:59,  2.02it/s]
 80%|███████▉  | 8531/10707 [1:14:37<17:59,  2.02it/s]
 80%|███████▉  | 8532/10707 [1:14:38<17:58,  2.02it/s]
 80%|███████▉  | 8533/10707 [1:14:38<17:58,  2.02it/s]
 80%|███████▉  | 8534/10707 [1:14:39<17:56,  2.02it/s]
 80%|███████▉  | 8535/10707 [1:14:39<17:56,  2.02it/s]
 80%|███████▉  | 8536/10707 [1:14:40<17:54,  2.02it/s]
 80%|███████▉  | 8537/10707 [1:14:40<17:53,  2.02it/s]
 80%|███████▉  | 8538/10707 [1:14:41<17:54,  2.02it/s]
 80%|███████▉  | 8539/10707 [1:14:41<17:53,  2.02it/s]
 80%|███████▉  | 8540/10707 [1:14:42<17:52,  2.02it/s]
 80%|███████▉  | 8541/10707 [1:14:42<17:52,  2.02it/s]
 80%|███████▉  | 8542/10707 [1:14:43<17:51,  2.02it/s]
 80%|███████▉  | 8543/10707 [1:14:43<17:51,  2.02it/s]
 80%|███████▉  | 8544/10707 [1:14:44<17:49,  2.02it/s]
 80%|███████▉  | 8545/10707 [1:14:44<17:49,  2.02it/s]
 80%|███████▉  | 8546/10707 [1:14:45<17:48,  2.02it/s]
 80%|███████▉  | 8547/10707 [1:14:45<17:48,  2.02it/s]
 80%|███████▉  | 8548/10707 [1:14:46<17:48,  2.02it/s]
 80%|███████▉  | 8549/10707 [1:14:46<17:48,  2.02it/s]
 80%|███████▉  | 8550/10707 [1:14:47<17:46,  2.02it/s]{'loss': 3.5988, 'grad_norm': 0.19787617027759552, 'learning_rate': 0.00011862436705100627, 'epoch': 0.8}
-
                                                      

 80%|███████▉  | 8550/10707 [1:14:47<17:46,  2.02it/s]
 80%|███████▉  | 8551/10707 [1:14:47<17:49,  2.02it/s]
 80%|███████▉  | 8552/10707 [1:14:48<17:47,  2.02it/s]
 80%|███████▉  | 8553/10707 [1:14:48<17:47,  2.02it/s]
 80%|███████▉  | 8554/10707 [1:14:49<17:46,  2.02it/s]
 80%|███████▉  | 8555/10707 [1:14:49<17:45,  2.02it/s]
 80%|███████▉  | 8556/10707 [1:14:50<17:44,  2.02it/s]
 80%|███████▉  | 8557/10707 [1:14:50<17:44,  2.02it/s]
 80%|███████▉  | 8558/10707 [1:14:51<17:44,  2.02it/s]
 80%|███████▉  | 8559/10707 [1:14:51<17:43,  2.02it/s]
 80%|███████▉  | 8560/10707 [1:14:52<17:42,  2.02it/s]
 80%|███████▉  | 8561/10707 [1:14:52<17:42,  2.02it/s]
 80%|███████▉  | 8562/10707 [1:14:53<17:42,  2.02it/s]
 80%|███████▉  | 8563/10707 [1:14:53<17:41,  2.02it/s]
 80%|███████▉  | 8564/10707 [1:14:54<17:41,  2.02it/s]
 80%|███████▉  | 8565/10707 [1:14:54<17:40,  2.02it/s]
 80%|████████  | 8566/10707 [1:14:55<17:39,  2.02it/s]
 80%|████████  | 8567/10707 [1:14:55<17:39,  2.02it/s]
 80%|████████  | 8568/10707 [1:14:56<17:39,  2.02it/s]
 80%|████████  | 8569/10707 [1:14:56<17:39,  2.02it/s]
 80%|████████  | 8570/10707 [1:14:57<17:38,  2.02it/s]
 80%|████████  | 8571/10707 [1:14:57<17:37,  2.02it/s]
 80%|████████  | 8572/10707 [1:14:58<17:38,  2.02it/s]
 80%|████████  | 8573/10707 [1:14:58<17:36,  2.02it/s]
 80%|████████  | 8574/10707 [1:14:59<17:36,  2.02it/s]
 80%|████████  | 8575/10707 [1:14:59<17:34,  2.02it/s]{'loss': 3.5973, 'grad_norm': 0.20584620535373688, 'learning_rate': 0.00011600157910074177, 'epoch': 0.8}
                                                      
-
 80%|████████  | 8575/10707 [1:14:59<17:34,  2.02it/s]
 80%|████████  | 8576/10707 [1:15:00<17:36,  2.02it/s]
 80%|████████  | 8577/10707 [1:15:00<17:35,  2.02it/s]
 80%|████████  | 8578/10707 [1:15:01<17:34,  2.02it/s]
 80%|████████  | 8579/10707 [1:15:01<17:33,  2.02it/s]
 80%|████████  | 8580/10707 [1:15:02<17:32,  2.02it/s]
 80%|████████  | 8581/10707 [1:15:02<17:31,  2.02it/s]
 80%|████████  | 8582/10707 [1:15:03<17:32,  2.02it/s]
 80%|████████  | 8583/10707 [1:15:03<17:34,  2.01it/s]
 80%|████████  | 8584/10707 [1:15:04<17:33,  2.01it/s]
 80%|████████  | 8585/10707 [1:15:04<17:33,  2.01it/s]
 80%|████████  | 8586/10707 [1:15:05<17:31,  2.02it/s]
 80%|████████  | 8587/10707 [1:15:05<17:30,  2.02it/s]
 80%|████████  | 8588/10707 [1:15:06<17:28,  2.02it/s]
 80%|████████  | 8589/10707 [1:15:06<17:28,  2.02it/s]
 80%|████████  | 8590/10707 [1:15:07<17:27,  2.02it/s]
 80%|████████  | 8591/10707 [1:15:07<17:28,  2.02it/s]
 80%|████████  | 8592/10707 [1:15:08<17:26,  2.02it/s]
 80%|████████  | 8593/10707 [1:15:08<17:26,  2.02it/s]
 80%|████████  | 8594/10707 [1:15:09<17:26,  2.02it/s]
 80%|████████  | 8595/10707 [1:15:09<17:26,  2.02it/s]
 80%|████████  | 8596/10707 [1:15:10<17:24,  2.02it/s]
 80%|████████  | 8597/10707 [1:15:10<17:24,  2.02it/s]
 80%|████████  | 8598/10707 [1:15:11<17:23,  2.02it/s]
 80%|████████  | 8599/10707 [1:15:11<17:23,  2.02it/s]
 80%|████████  | 8600/10707 [1:15:12<17:22,  2.02it/s]{'loss': 3.606, 'grad_norm': 0.20237839221954346, 'learning_rate': 0.00011340430131233915, 'epoch': 0.8}
-
                                                      

 80%|████████  | 8600/10707 [1:15:12<17:22,  2.02it/s]
 80%|████████  | 8601/10707 [1:15:12<17:25,  2.01it/s]
 80%|████████  | 8602/10707 [1:15:13<17:22,  2.02it/s]
 80%|████████  | 8603/10707 [1:15:13<17:22,  2.02it/s]
 80%|████████  | 8604/10707 [1:15:14<17:21,  2.02it/s]
 80%|████████  | 8605/10707 [1:15:14<17:20,  2.02it/s]
 80%|████████  | 8606/10707 [1:15:15<17:23,  2.01it/s]
 80%|████████  | 8607/10707 [1:15:15<17:21,  2.02it/s]
 80%|████████  | 8608/10707 [1:15:16<17:21,  2.02it/s]
 80%|████████  | 8609/10707 [1:15:16<17:20,  2.02it/s]
 80%|████████  | 8610/10707 [1:15:17<17:19,  2.02it/s]
 80%|████████  | 8611/10707 [1:15:17<17:19,  2.02it/s]
 80%|████████  | 8612/10707 [1:15:18<17:18,  2.02it/s]
 80%|████████  | 8613/10707 [1:15:18<17:17,  2.02it/s]
 80%|████████  | 8614/10707 [1:15:19<17:16,  2.02it/s]
 80%|████████  | 8615/10707 [1:15:19<17:16,  2.02it/s]
 80%|████████  | 8616/10707 [1:15:20<17:14,  2.02it/s]
 80%|████████  | 8617/10707 [1:15:20<17:14,  2.02it/s]
 80%|████████  | 8618/10707 [1:15:21<17:13,  2.02it/s]
 80%|████████  | 8619/10707 [1:15:21<17:13,  2.02it/s]
 81%|████████  | 8620/10707 [1:15:22<17:12,  2.02it/s]
 81%|████████  | 8621/10707 [1:15:22<17:13,  2.02it/s]
 81%|████████  | 8622/10707 [1:15:23<17:12,  2.02it/s]
 81%|████████  | 8623/10707 [1:15:23<17:13,  2.02it/s]
 81%|████████  | 8624/10707 [1:15:24<17:12,  2.02it/s]
 81%|████████  | 8625/10707 [1:15:24<17:10,  2.02it/s]{'loss': 3.5991, 'grad_norm': 0.1973121464252472, 'learning_rate': 0.00011083270623071829, 'epoch': 0.81}
-
                                                      

 81%|████████  | 8625/10707 [1:15:24<17:10,  2.02it/s]
 81%|████████  | 8626/10707 [1:15:24<17:10,  2.02it/s]
 81%|████████  | 8627/10707 [1:15:25<17:10,  2.02it/s]
 81%|████████  | 8628/10707 [1:15:25<17:09,  2.02it/s]
 81%|████████  | 8629/10707 [1:15:26<17:09,  2.02it/s]
 81%|████████  | 8630/10707 [1:15:26<17:08,  2.02it/s]
 81%|████████  | 8631/10707 [1:15:27<17:08,  2.02it/s]
 81%|████████  | 8632/10707 [1:15:27<17:08,  2.02it/s]
 81%|████████  | 8633/10707 [1:15:28<17:07,  2.02it/s]
 81%|████████  | 8634/10707 [1:15:28<17:07,  2.02it/s]
 81%|████████  | 8635/10707 [1:15:29<17:06,  2.02it/s]
 81%|████████  | 8636/10707 [1:15:29<17:06,  2.02it/s]
 81%|████████  | 8637/10707 [1:15:30<17:06,  2.02it/s]
 81%|████████  | 8638/10707 [1:15:30<17:05,  2.02it/s]
 81%|████████  | 8639/10707 [1:15:31<17:04,  2.02it/s]
 81%|████████  | 8640/10707 [1:15:31<17:02,  2.02it/s]
 81%|████████  | 8641/10707 [1:15:32<17:02,  2.02it/s]
 81%|████████  | 8642/10707 [1:15:32<17:01,  2.02it/s]
 81%|████████  | 8643/10707 [1:15:33<17:02,  2.02it/s]
 81%|████████  | 8644/10707 [1:15:33<17:04,  2.01it/s]
 81%|████████  | 8645/10707 [1:15:34<17:03,  2.02it/s]
 81%|████████  | 8646/10707 [1:15:34<17:02,  2.02it/s]
 81%|████████  | 8647/10707 [1:15:35<17:01,  2.02it/s]
 81%|████████  | 8648/10707 [1:15:35<16:59,  2.02it/s]
 81%|████████  | 8649/10707 [1:15:36<16:59,  2.02it/s]
 81%|████████  | 8650/10707 [1:15:36<16:57,  2.02it/s]{'loss': 3.5951, 'grad_norm': 0.19609494507312775, 'learning_rate': 0.00010828696469462023, 'epoch': 0.81}
-
                                                      

 81%|████████  | 8650/10707 [1:15:36<16:57,  2.02it/s]
 81%|████████  | 8651/10707 [1:15:37<16:59,  2.02it/s]
 81%|████████  | 8652/10707 [1:15:37<16:57,  2.02it/s]
 81%|████████  | 8653/10707 [1:15:38<16:57,  2.02it/s]
 81%|████████  | 8654/10707 [1:15:38<16:56,  2.02it/s]
 81%|████████  | 8655/10707 [1:15:39<16:55,  2.02it/s]
 81%|████████  | 8656/10707 [1:15:39<16:54,  2.02it/s]
 81%|████████  | 8657/10707 [1:15:40<16:54,  2.02it/s]
 81%|████████  | 8658/10707 [1:15:40<16:53,  2.02it/s]
 81%|████████  | 8659/10707 [1:15:41<16:53,  2.02it/s]
 81%|████████  | 8660/10707 [1:15:41<16:51,  2.02it/s]
 81%|████████  | 8661/10707 [1:15:42<16:52,  2.02it/s]
 81%|████████  | 8662/10707 [1:15:42<16:50,  2.02it/s]
 81%|████████  | 8663/10707 [1:15:43<16:51,  2.02it/s]
 81%|████████  | 8664/10707 [1:15:43<16:49,  2.02it/s]
 81%|████████  | 8665/10707 [1:15:44<16:50,  2.02it/s]
 81%|████████  | 8666/10707 [1:15:44<16:49,  2.02it/s]
 81%|████████  | 8667/10707 [1:15:45<16:49,  2.02it/s]
 81%|████████  | 8668/10707 [1:15:45<16:47,  2.02it/s]
 81%|████████  | 8669/10707 [1:15:46<16:48,  2.02it/s]
 81%|████████  | 8670/10707 [1:15:46<16:48,  2.02it/s]
 81%|████████  | 8671/10707 [1:15:47<16:47,  2.02it/s]
 81%|████████  | 8672/10707 [1:15:47<16:47,  2.02it/s]
 81%|████████  | 8673/10707 [1:15:48<16:47,  2.02it/s]
 81%|████████  | 8674/10707 [1:15:48<16:48,  2.02it/s]
 81%|████████  | 8675/10707 [1:15:49<16:47,  2.02it/s]{'loss': 3.5965, 'grad_norm': 0.19613724946975708, 'learning_rate': 0.00010576724582525793, 'epoch': 0.81}
-
                                                      

 81%|████████  | 8675/10707 [1:15:49<16:47,  2.02it/s]
 81%|████████  | 8676/10707 [1:15:49<16:49,  2.01it/s]
 81%|████████  | 8677/10707 [1:15:50<16:48,  2.01it/s]
 81%|████████  | 8678/10707 [1:15:50<16:47,  2.01it/s]
 81%|████████  | 8679/10707 [1:15:51<16:45,  2.02it/s]
 81%|████████  | 8680/10707 [1:15:51<16:45,  2.02it/s]
 81%|████████  | 8681/10707 [1:15:52<16:44,  2.02it/s]
 81%|████████  | 8682/10707 [1:15:52<16:43,  2.02it/s]
 81%|████████  | 8683/10707 [1:15:53<16:42,  2.02it/s]
 81%|████████  | 8684/10707 [1:15:53<16:41,  2.02it/s]
 81%|████████  | 8685/10707 [1:15:54<16:41,  2.02it/s]
 81%|████████  | 8686/10707 [1:15:54<16:41,  2.02it/s]
 81%|████████  | 8687/10707 [1:15:55<16:40,  2.02it/s]
 81%|████████  | 8688/10707 [1:15:55<16:40,  2.02it/s]
 81%|████████  | 8689/10707 [1:15:56<16:40,  2.02it/s]
 81%|████████  | 8690/10707 [1:15:56<16:39,  2.02it/s]
 81%|████████  | 8691/10707 [1:15:57<16:39,  2.02it/s]
 81%|████████  | 8692/10707 [1:15:57<16:38,  2.02it/s]
 81%|████████  | 8693/10707 [1:15:58<16:36,  2.02it/s]
 81%|████████  | 8694/10707 [1:15:58<16:36,  2.02it/s]
 81%|████████  | 8695/10707 [1:15:59<16:35,  2.02it/s]
 81%|████████  | 8696/10707 [1:15:59<16:35,  2.02it/s]
 81%|████████  | 8697/10707 [1:16:00<16:36,  2.02it/s]
 81%|████████  | 8698/10707 [1:16:00<16:34,  2.02it/s]
 81%|████████  | 8699/10707 [1:16:01<16:34,  2.02it/s]
 81%|████████▏ | 8700/10707 [1:16:01<16:32,  2.02it/s]{'loss': 3.601, 'grad_norm': 0.19928880035877228, 'learning_rate': 0.00010327371701508031, 'epoch': 0.81}
-
                                                      

 81%|████████▏ | 8700/10707 [1:16:01<16:32,  2.02it/s]
 81%|████████▏ | 8701/10707 [1:16:02<16:34,  2.02it/s]
 81%|████████▏ | 8702/10707 [1:16:02<16:33,  2.02it/s]
 81%|████████▏ | 8703/10707 [1:16:03<16:32,  2.02it/s]
 81%|████████▏ | 8704/10707 [1:16:03<16:31,  2.02it/s]
 81%|████████▏ | 8705/10707 [1:16:04<16:33,  2.01it/s]
 81%|████████▏ | 8706/10707 [1:16:04<16:32,  2.02it/s]
 81%|████████▏ | 8707/10707 [1:16:05<16:31,  2.02it/s]
 81%|████████▏ | 8708/10707 [1:16:05<16:30,  2.02it/s]
 81%|████████▏ | 8709/10707 [1:16:06<16:28,  2.02it/s]
 81%|████████▏ | 8710/10707 [1:16:06<16:27,  2.02it/s]
 81%|████████▏ | 8711/10707 [1:16:07<16:27,  2.02it/s]
 81%|████████▏ | 8712/10707 [1:16:07<16:27,  2.02it/s]
 81%|████████▏ | 8713/10707 [1:16:08<16:26,  2.02it/s]
 81%|████████▏ | 8714/10707 [1:16:08<16:26,  2.02it/s]
 81%|████████▏ | 8715/10707 [1:16:09<16:26,  2.02it/s]
 81%|████████▏ | 8716/10707 [1:16:09<16:25,  2.02it/s]
 81%|████████▏ | 8717/10707 [1:16:10<16:24,  2.02it/s]
 81%|████████▏ | 8718/10707 [1:16:10<16:24,  2.02it/s]
 81%|████████▏ | 8719/10707 [1:16:11<16:23,  2.02it/s]
 81%|████████▏ | 8720/10707 [1:16:11<16:23,  2.02it/s]
 81%|████████▏ | 8721/10707 [1:16:12<16:22,  2.02it/s]
 81%|████████▏ | 8722/10707 [1:16:12<16:22,  2.02it/s]
 81%|████████▏ | 8723/10707 [1:16:13<16:20,  2.02it/s]
 81%|████████▏ | 8724/10707 [1:16:13<16:20,  2.02it/s]
 81%|████████▏ | 8725/10707 [1:16:14<16:19,  2.02it/s]{'loss': 3.5977, 'grad_norm': 0.19399872422218323, 'learning_rate': 0.0001008065439166529, 'epoch': 0.81}
-
                                                      

 81%|████████▏ | 8725/10707 [1:16:14<16:19,  2.02it/s]
 81%|████████▏ | 8726/10707 [1:16:14<16:21,  2.02it/s]
 82%|████████▏ | 8727/10707 [1:16:15<16:19,  2.02it/s]
 82%|████████▏ | 8728/10707 [1:16:15<16:19,  2.02it/s]
 82%|████████▏ | 8729/10707 [1:16:16<16:18,  2.02it/s]
 82%|████████▏ | 8730/10707 [1:16:16<16:17,  2.02it/s]
 82%|████████▏ | 8731/10707 [1:16:16<16:17,  2.02it/s]
 82%|████████▏ | 8732/10707 [1:16:17<16:17,  2.02it/s]
 82%|████████▏ | 8733/10707 [1:16:17<16:17,  2.02it/s]
 82%|████████▏ | 8734/10707 [1:16:18<16:16,  2.02it/s]
 82%|████████▏ | 8735/10707 [1:16:18<16:15,  2.02it/s]
 82%|████████▏ | 8736/10707 [1:16:19<16:15,  2.02it/s]
 82%|████████▏ | 8737/10707 [1:16:19<16:15,  2.02it/s]
 82%|████████▏ | 8738/10707 [1:16:20<16:14,  2.02it/s]
 82%|████████▏ | 8739/10707 [1:16:20<16:14,  2.02it/s]
 82%|████████▏ | 8740/10707 [1:16:21<16:15,  2.02it/s]
 82%|████████▏ | 8741/10707 [1:16:21<16:14,  2.02it/s]
 82%|████████▏ | 8742/10707 [1:16:22<16:13,  2.02it/s]
 82%|████████▏ | 8743/10707 [1:16:22<16:12,  2.02it/s]
 82%|████████▏ | 8744/10707 [1:16:23<16:11,  2.02it/s]
 82%|████████▏ | 8745/10707 [1:16:23<16:10,  2.02it/s]
 82%|████████▏ | 8746/10707 [1:16:24<16:10,  2.02it/s]
 82%|████████▏ | 8747/10707 [1:16:24<16:09,  2.02it/s]
 82%|████████▏ | 8748/10707 [1:16:25<16:09,  2.02it/s]
 82%|████████▏ | 8749/10707 [1:16:25<16:08,  2.02it/s]
 82%|████████▏ | 8750/10707 [1:16:26<16:08,  2.02it/s]{'loss': 3.5919, 'grad_norm': 0.199885293841362, 'learning_rate': 9.836589043165278e-05, 'epoch': 0.82}
-
                                                      

 82%|████████▏ | 8750/10707 [1:16:26<16:08,  2.02it/s]
 82%|████████▏ | 8751/10707 [1:16:26<16:09,  2.02it/s]
 82%|████████▏ | 8752/10707 [1:16:27<16:08,  2.02it/s]
 82%|████████▏ | 8753/10707 [1:16:27<16:07,  2.02it/s]
 82%|████████▏ | 8754/10707 [1:16:28<16:06,  2.02it/s]
 82%|████████▏ | 8755/10707 [1:16:28<16:06,  2.02it/s]
 82%|████████▏ | 8756/10707 [1:16:29<16:05,  2.02it/s]
 82%|████████▏ | 8757/10707 [1:16:29<16:05,  2.02it/s]
 82%|████████▏ | 8758/10707 [1:16:30<16:04,  2.02it/s]
 82%|████████▏ | 8759/10707 [1:16:30<16:03,  2.02it/s]
 82%|████████▏ | 8760/10707 [1:16:31<16:04,  2.02it/s]
 82%|████████▏ | 8761/10707 [1:16:31<16:04,  2.02it/s]
 82%|████████▏ | 8762/10707 [1:16:32<16:02,  2.02it/s]
 82%|████████▏ | 8763/10707 [1:16:32<16:03,  2.02it/s]
 82%|████████▏ | 8764/10707 [1:16:33<16:02,  2.02it/s]
 82%|████████▏ | 8765/10707 [1:16:33<16:04,  2.01it/s]
 82%|████████▏ | 8766/10707 [1:16:34<16:02,  2.02it/s]
 82%|████████▏ | 8767/10707 [1:16:34<16:02,  2.02it/s]
 82%|████████▏ | 8768/10707 [1:16:35<16:01,  2.02it/s]
 82%|████████▏ | 8769/10707 [1:16:35<15:59,  2.02it/s]
 82%|████████▏ | 8770/10707 [1:16:36<15:58,  2.02it/s]
 82%|████████▏ | 8771/10707 [1:16:36<15:59,  2.02it/s]
 82%|████████▏ | 8772/10707 [1:16:37<15:59,  2.02it/s]
 82%|████████▏ | 8773/10707 [1:16:37<15:58,  2.02it/s]
 82%|████████▏ | 8774/10707 [1:16:38<15:59,  2.02it/s]
 82%|████████▏ | 8775/10707 [1:16:38<15:58,  2.02it/s]{'loss': 3.5957, 'grad_norm': 0.19489921629428864, 'learning_rate': 9.595191869997966e-05, 'epoch': 0.82}
                                                      
-
 82%|████████▏ | 8775/10707 [1:16:38<15:58,  2.02it/s]
 82%|████████▏ | 8776/10707 [1:16:39<15:57,  2.02it/s]
 82%|████████▏ | 8777/10707 [1:16:39<15:58,  2.01it/s]
 82%|████████▏ | 8778/10707 [1:16:40<15:56,  2.02it/s]
 82%|████████▏ | 8779/10707 [1:16:40<15:56,  2.01it/s]
 82%|████████▏ | 8780/10707 [1:16:41<15:56,  2.02it/s]
 82%|████████▏ | 8781/10707 [1:16:41<15:54,  2.02it/s]
 82%|████████▏ | 8782/10707 [1:16:42<15:53,  2.02it/s]
 82%|████████▏ | 8783/10707 [1:16:42<15:53,  2.02it/s]
 82%|████████▏ | 8784/10707 [1:16:43<15:52,  2.02it/s]
 82%|████████▏ | 8785/10707 [1:16:43<15:52,  2.02it/s]
 82%|████████▏ | 8786/10707 [1:16:44<15:52,  2.02it/s]
 82%|████████▏ | 8787/10707 [1:16:44<15:50,  2.02it/s]
 82%|████████▏ | 8788/10707 [1:16:45<15:51,  2.02it/s]
 82%|████████▏ | 8789/10707 [1:16:45<15:50,  2.02it/s]
 82%|███��████▏ | 8790/10707 [1:16:46<15:50,  2.02it/s]
 82%|████████▏ | 8791/10707 [1:16:46<15:48,  2.02it/s]
 82%|████████▏ | 8792/10707 [1:16:47<15:48,  2.02it/s]
 82%|████████▏ | 8793/10707 [1:16:47<15:47,  2.02it/s]
 82%|████████▏ | 8794/10707 [1:16:48<15:48,  2.02it/s]
 82%|████████▏ | 8795/10707 [1:16:48<15:46,  2.02it/s]
 82%|████████▏ | 8796/10707 [1:16:49<15:46,  2.02it/s]
 82%|████████▏ | 8797/10707 [1:16:49<15:46,  2.02it/s]
 82%|████████▏ | 8798/10707 [1:16:50<15:46,  2.02it/s]
 82%|████████▏ | 8799/10707 [1:16:50<15:45,  2.02it/s]
 82%|████████▏ | 8800/10707 [1:16:51<15:44,  2.02it/s]{'loss': 3.5983, 'grad_norm': 0.18889516592025757, 'learning_rate': 9.356478908898435e-05, 'epoch': 0.82}
                                                      
-
 82%|████████▏ | 8800/10707 [1:16:51<15:44,  2.02it/s]
 82%|████████▏ | 8801/10707 [1:16:51<15:45,  2.02it/s]
 82%|████████▏ | 8802/10707 [1:16:52<15:44,  2.02it/s]
 82%|████████▏ | 8803/10707 [1:16:52<15:43,  2.02it/s]
 82%|████████▏ | 8804/10707 [1:16:53<15:42,  2.02it/s]
 82%|████████▏ | 8805/10707 [1:16:53<15:42,  2.02it/s]
 82%|████████▏ | 8806/10707 [1:16:54<15:40,  2.02it/s]
 82%|████████▏ | 8807/10707 [1:16:54<15:40,  2.02it/s]
 82%|████████▏ | 8808/10707 [1:16:55<15:39,  2.02it/s]
 82%|████████▏ | 8809/10707 [1:16:55<15:39,  2.02it/s]
 82%|████████▏ | 8810/10707 [1:16:56<15:38,  2.02it/s]
 82%|████████▏ | 8811/10707 [1:16:56<15:38,  2.02it/s]
 82%|████████▏ | 8812/10707 [1:16:57<15:36,  2.02it/s]
 82%|████████▏ | 8813/10707 [1:16:57<15:36,  2.02it/s]
 82%|████████▏ | 8814/10707 [1:16:58<15:35,  2.02it/s]
 82%|████████▏ | 8815/10707 [1:16:58<15:36,  2.02it/s]
 82%|████████▏ | 8816/10707 [1:16:59<15:35,  2.02it/s]
 82%|████████▏ | 8817/10707 [1:16:59<15:35,  2.02it/s]
 82%|████████▏ | 8818/10707 [1:17:00<15:34,  2.02it/s]
 82%|████████▏ | 8819/10707 [1:17:00<15:34,  2.02it/s]
 82%|████████▏ | 8820/10707 [1:17:01<15:34,  2.02it/s]
 82%|████████▏ | 8821/10707 [1:17:01<15:33,  2.02it/s]
 82%|████████▏ | 8822/10707 [1:17:02<15:32,  2.02it/s]
 82%|████████▏ | 8823/10707 [1:17:02<15:33,  2.02it/s]
 82%|████████▏ | 8824/10707 [1:17:03<15:31,  2.02it/s]
 82%|████████▏ | 8825/10707 [1:17:03<15:33,  2.02it/s]
                                                      {'loss': 3.5896, 'grad_norm': 0.1980924904346466, 'learning_rate': 9.120466018281593e-05, 'epoch': 0.82}
-
 82%|████████▏ | 8825/10707 [1:17:03<15:33,  2.02it/s]
 82%|████████▏ | 8826/10707 [1:17:04<15:35,  2.01it/s]
 82%|████████▏ | 8827/10707 [1:17:04<15:33,  2.01it/s]
 82%|████████▏ | 8828/10707 [1:17:05<15:32,  2.02it/s]
 82%|████████▏ | 8829/10707 [1:17:05<15:31,  2.02it/s]
 82%|████████▏ | 8830/10707 [1:17:06<15:30,  2.02it/s]
 82%|████████▏ | 8831/10707 [1:17:06<15:29,  2.02it/s]
 82%|████████▏ | 8832/10707 [1:17:07<15:28,  2.02it/s]
 82%|████████▏ | 8833/10707 [1:17:07<15:27,  2.02it/s]
 83%|████████▎ | 8834/10707 [1:17:08<15:28,  2.02it/s]
 83%|████████▎ | 8835/10707 [1:17:08<15:27,  2.02it/s]
 83%|████████▎ | 8836/10707 [1:17:09<15:26,  2.02it/s]
 83%|████████▎ | 8837/10707 [1:17:09<15:24,  2.02it/s]
 83%|████████▎ | 8838/10707 [1:17:09<15:24,  2.02it/s]
 83%|████████▎ | 8839/10707 [1:17:10<15:23,  2.02it/s]
 83%|████████▎ | 8840/10707 [1:17:10<15:24,  2.02it/s]
 83%|████████▎ | 8841/10707 [1:17:11<15:23,  2.02it/s]
 83%|████████▎ | 8842/10707 [1:17:11<15:24,  2.02it/s]
 83%|████████▎ | 8843/10707 [1:17:12<15:22,  2.02it/s]
 83%|████████▎ | 8844/10707 [1:17:12<15:22,  2.02it/s]
 83%|████████▎ | 8845/10707 [1:17:13<15:20,  2.02it/s]
 83%|████████▎ | 8846/10707 [1:17:13<15:20,  2.02it/s]
 83%|████████▎ | 8847/10707 [1:17:14<15:19,  2.02it/s]
 83%|████████▎ | 8848/10707 [1:17:14<15:19,  2.02it/s]
 83%|████████▎ | 8849/10707 [1:17:15<15:18,  2.02it/s]
 83%|████████▎ | 8850/10707 [1:17:15<15:18,  2.02it/s]
                                                      
{'loss': 3.5937, 'grad_norm': 0.19781309366226196, 'learning_rate': 8.887168877188607e-05, 'epoch': 0.83}
-
 83%|████████▎ | 8850/10707 [1:17:15<15:18,  2.02it/s]
 83%|████████▎ | 8851/10707 [1:17:16<15:19,  2.02it/s]
 83%|████████▎ | 8852/10707 [1:17:16<15:19,  2.02it/s]
 83%|████████▎ | 8853/10707 [1:17:17<15:18,  2.02it/s]
 83%|████████▎ | 8854/10707 [1:17:17<15:17,  2.02it/s]
 83%|████████▎ | 8855/10707 [1:17:18<15:16,  2.02it/s]
 83%|████████▎ | 8856/10707 [1:17:18<15:16,  2.02it/s]
 83%|████████▎ | 8857/10707 [1:17:19<15:16,  2.02it/s]
 83%|████████▎ | 8858/10707 [1:17:19<15:16,  2.02it/s]
 83%|████████▎ | 8859/10707 [1:17:20<15:15,  2.02it/s]
 83%|████████▎ | 8860/10707 [1:17:20<15:14,  2.02it/s]
 83%|████████▎ | 8861/10707 [1:17:21<15:14,  2.02it/s]
 83%|████████▎ | 8862/10707 [1:17:21<15:13,  2.02it/s]
 83%|████████▎ | 8863/10707 [1:17:22<15:13,  2.02it/s]
 83%|████████▎ | 8864/10707 [1:17:22<15:12,  2.02it/s]
 83%|████████▎ | 8865/10707 [1:17:23<15:11,  2.02it/s]
 83%|████████▎ | 8866/10707 [1:17:23<15:11,  2.02it/s]
 83%|████████▎ | 8867/10707 [1:17:24<15:11,  2.02it/s]
 83%|████████▎ | 8868/10707 [1:17:24<15:10,  2.02it/s]
 83%|████████▎ | 8869/10707 [1:17:25<15:10,  2.02it/s]
 83%|████████▎ | 8870/10707 [1:17:25<15:09,  2.02it/s]
 83%|████████▎ | 8871/10707 [1:17:26<15:08,  2.02it/s]
 83%|████████▎ | 8872/10707 [1:17:26<15:07,  2.02it/s]
 83%|████████▎ | 8873/10707 [1:17:27<15:06,  2.02it/s]
 83%|████████▎ | 8874/10707 [1:17:27<15:06,  2.02it/s]
 83%|████████▎ | 8875/10707 [1:17:28<15:05,  2.02it/s]{'loss': 3.5964, 'grad_norm': 0.20029385387897491, 'learning_rate': 8.656602984245237e-05, 'epoch': 0.83}
                                                      
-
 83%|████████▎ | 8875/10707 [1:17:28<15:05,  2.02it/s]
 83%|████████▎ | 8876/10707 [1:17:28<15:07,  2.02it/s]
 83%|████████▎ | 8877/10707 [1:17:29<15:06,  2.02it/s]
 83%|████████▎ | 8878/10707 [1:17:29<15:06,  2.02it/s]
 83%|████████▎ | 8879/10707 [1:17:30<15:05,  2.02it/s]
 83%|████████▎ | 8880/10707 [1:17:30<15:03,  2.02it/s]
 83%|████████▎ | 8881/10707 [1:17:31<15:04,  2.02it/s]
 83%|████████▎ | 8882/10707 [1:17:31<15:03,  2.02it/s]
 83%|████████▎ | 8883/10707 [1:17:32<15:02,  2.02it/s]
 83%|████████▎ | 8884/10707 [1:17:32<15:01,  2.02it/s]
 83%|████████▎ | 8885/10707 [1:17:33<15:01,  2.02it/s]
 83%|████████▎ | 8886/10707 [1:17:33<15:02,  2.02it/s]
 83%|████████▎ | 8887/10707 [1:17:34<15:02,  2.02it/s]
 83%|████████▎ | 8888/10707 [1:17:34<15:02,  2.02it/s]
 83%|████████▎ | 8889/10707 [1:17:35<15:01,  2.02it/s]
 83%|████████▎ | 8890/10707 [1:17:35<15:01,  2.02it/s]
 83%|████████▎ | 8891/10707 [1:17:36<14:59,  2.02it/s]
 83%|████████▎ | 8892/10707 [1:17:36<14:58,  2.02it/s]
 83%|████████▎ | 8893/10707 [1:17:37<14:57,  2.02it/s]
 83%|████████▎ | 8894/10707 [1:17:37<14:57,  2.02it/s]
 83%|████████▎ | 8895/10707 [1:17:38<14:56,  2.02it/s]
 83%|████████▎ | 8896/10707 [1:17:38<14:55,  2.02it/s]
 83%|████████▎ | 8897/10707 [1:17:39<14:55,  2.02it/s]
 83%|████████▎ | 8898/10707 [1:17:39<14:54,  2.02it/s]
 83%|████████▎ | 8899/10707 [1:17:40<14:54,  2.02it/s]
 83%|████████▎ | 8900/10707 [1:17:40<14:55,  2.02it/s]{'loss': 3.5959, 'grad_norm': 0.19270426034927368, 'learning_rate': 8.428783656632332e-05, 'epoch': 0.83}
-
                                                      

 83%|████████▎ | 8900/10707 [1:17:40<14:55,  2.02it/s]
 83%|████████▎ | 8901/10707 [1:17:41<14:54,  2.02it/s]
 83%|████████▎ | 8902/10707 [1:17:41<14:54,  2.02it/s]
 83%|████████▎ | 8903/10707 [1:17:42<14:52,  2.02it/s]
 83%|████████▎ | 8904/10707 [1:17:42<14:52,  2.02it/s]
 83%|████████▎ | 8905/10707 [1:17:43<14:51,  2.02it/s]
 83%|████████▎ | 8906/10707 [1:17:43<14:51,  2.02it/s]
 83%|████████▎ | 8907/10707 [1:17:44<14:50,  2.02it/s]
 83%|████████▎ | 8908/10707 [1:17:44<14:51,  2.02it/s]
 83%|████████▎ | 8909/10707 [1:17:45<14:49,  2.02it/s]
 83%|████████▎ | 8910/10707 [1:17:45<14:49,  2.02it/s]
 83%|████████▎ | 8911/10707 [1:17:46<14:49,  2.02it/s]
 83%|████████▎ | 8912/10707 [1:17:46<14:48,  2.02it/s]
 83%|████████▎ | 8913/10707 [1:17:47<14:48,  2.02it/s]
 83%|██████��█▎ | 8914/10707 [1:17:47<14:48,  2.02it/s]
 83%|████████▎ | 8915/10707 [1:17:48<14:47,  2.02it/s]
 83%|████████▎ | 8916/10707 [1:17:48<14:47,  2.02it/s]
 83%|████████▎ | 8917/10707 [1:17:49<14:46,  2.02it/s]
 83%|████████▎ | 8918/10707 [1:17:49<14:45,  2.02it/s]
 83%|████████▎ | 8919/10707 [1:17:50<14:45,  2.02it/s]
 83%|████████▎ | 8920/10707 [1:17:50<14:44,  2.02it/s]
 83%|████████▎ | 8921/10707 [1:17:51<14:44,  2.02it/s]
 83%|████████▎ | 8922/10707 [1:17:51<14:43,  2.02it/s]
 83%|████████▎ | 8923/10707 [1:17:52<14:42,  2.02it/s]
 83%|████████▎ | 8924/10707 [1:17:52<14:42,  2.02it/s]
 83%|████████▎ | 8925/10707 [1:17:53<14:43,  2.02it/s]{'loss': 3.6001, 'grad_norm': 0.1941545456647873, 'learning_rate': 8.203726029068148e-05, 'epoch': 0.83}
-
                                                      

 83%|████████▎ | 8925/10707 [1:17:53<14:43,  2.02it/s]
 83%|████████▎ | 8926/10707 [1:17:53<14:44,  2.01it/s]
 83%|████████▎ | 8927/10707 [1:17:54<14:42,  2.02it/s]
 83%|████████▎ | 8928/10707 [1:17:54<14:42,  2.02it/s]
 83%|████████▎ | 8929/10707 [1:17:55<14:40,  2.02it/s]
 83%|████████▎ | 8930/10707 [1:17:55<14:40,  2.02it/s]
 83%|████████▎ | 8931/10707 [1:17:56<14:39,  2.02it/s]
 83%|████████▎ | 8932/10707 [1:17:56<14:38,  2.02it/s]
 83%|████████▎ | 8933/10707 [1:17:57<14:38,  2.02it/s]
 83%|████████▎ | 8934/10707 [1:17:57<14:37,  2.02it/s]
 83%|████████▎ | 8935/10707 [1:17:58<14:36,  2.02it/s]
 83%|████████▎ | 8936/10707 [1:17:58<14:36,  2.02it/s]
 83%|████████▎ | 8937/10707 [1:17:59<16:56,  1.74it/s]
 83%|████████▎ | 8938/10707 [1:17:59<16:13,  1.82it/s]
 83%|████████▎ | 8939/10707 [1:18:00<15:44,  1.87it/s]
 83%|████████▎ | 8940/10707 [1:18:00<15:22,  1.91it/s]
 84%|████████▎ | 8941/10707 [1:18:01<15:07,  1.95it/s]
 84%|████████▎ | 8942/10707 [1:18:01<14:57,  1.97it/s]
 84%|████████▎ | 8943/10707 [1:18:02<14:49,  1.98it/s]
 84%|████████▎ | 8944/10707 [1:18:02<14:45,  1.99it/s]
 84%|████████▎ | 8945/10707 [1:18:03<14:41,  2.00it/s]
 84%|████████▎ | 8946/10707 [1:18:03<14:40,  2.00it/s]
 84%|████████▎ | 8947/10707 [1:18:04<14:36,  2.01it/s]
 84%|████████▎ | 8948/10707 [1:18:04<14:35,  2.01it/s]
 84%|████████▎ | 8949/10707 [1:18:05<14:32,  2.02it/s]
 84%|████████▎ | 8950/10707 [1:18:05<14:32,  2.01it/s]
                                                      
{'loss': 3.5957, 'grad_norm': 0.195368230342865, 'learning_rate': 7.981445052803037e-05, 'epoch': 0.84}
-
 84%|████████▎ | 8950/10707 [1:18:05<14:32,  2.01it/s]
 84%|████████▎ | 8951/10707 [1:18:06<14:30,  2.02it/s]
 84%|████████▎ | 8952/10707 [1:18:06<14:30,  2.02it/s]
 84%|████████▎ | 8953/10707 [1:18:07<14:28,  2.02it/s]
 84%|████████▎ | 8954/10707 [1:18:07<14:28,  2.02it/s]
 84%|████████▎ | 8955/10707 [1:18:08<14:27,  2.02it/s]
 84%|████████▎ | 8956/10707 [1:18:08<14:27,  2.02it/s]
 84%|████████▎ | 8957/10707 [1:18:09<14:27,  2.02it/s]
 84%|████████▎ | 8958/10707 [1:18:09<14:26,  2.02it/s]
 84%|████████▎ | 8959/10707 [1:18:10<14:26,  2.02it/s]
 84%|████████▎ | 8960/10707 [1:18:10<14:25,  2.02it/s]
 84%|████████▎ | 8961/10707 [1:18:11<14:25,  2.02it/s]
 84%|████████▎ | 8962/10707 [1:18:11<14:24,  2.02it/s]
 84%|████████▎ | 8963/10707 [1:18:12<14:24,  2.02it/s]
 84%|████████▎ | 8964/10707 [1:18:12<14:23,  2.02it/s]
 84%|████████▎ | 8965/10707 [1:18:13<14:23,  2.02it/s]
 84%|████████▎ | 8966/10707 [1:18:13<14:23,  2.02it/s]
 84%|████████▎ | 8967/10707 [1:18:14<14:22,  2.02it/s]
 84%|████████▍ | 8968/10707 [1:18:14<14:23,  2.01it/s]
 84%|████████▍ | 8969/10707 [1:18:15<14:22,  2.02it/s]
 84%|████████▍ | 8970/10707 [1:18:15<14:20,  2.02it/s]
 84%|████████▍ | 8971/10707 [1:18:16<14:20,  2.02it/s]
 84%|████████▍ | 8972/10707 [1:18:16<14:18,  2.02it/s]
 84%|████████▍ | 8973/10707 [1:18:17<14:18,  2.02it/s]
 84%|████████▍ | 8974/10707 [1:18:17<16:33,  1.74it/s]
 84%|████████▍ | 8975/10707 [1:18:18<15:52,  1.82it/s]{'loss': 3.594, 'grad_norm': 0.19900812208652496, 'learning_rate': 7.761955494626027e-05, 'epoch': 0.84}
-
                                                      

 84%|████████▍ | 8975/10707 [1:18:18<15:52,  1.82it/s]
 84%|████████▍ | 8976/10707 [1:18:18<15:24,  1.87it/s]
 84%|████████▍ | 8977/10707 [1:18:19<15:03,  1.91it/s]
 84%|████████▍ | 8978/10707 [1:18:19<14:48,  1.94it/s]
 84%|████████▍ | 8979/10707 [1:18:20<14:39,  1.97it/s]
 84%|████████▍ | 8980/10707 [1:18:20<14:31,  1.98it/s]
 84%|████████▍ | 8981/10707 [1:18:21<14:25,  1.99it/s]
 84%|████████▍ | 8982/10707 [1:18:21<14:22,  2.00it/s]
 84%|████████▍ | 8983/10707 [1:18:22<14:19,  2.00it/s]
 84%|████████▍ | 8984/10707 [1:18:22<14:17,  2.01it/s]
 84%|████████▍ | 8985/10707 [1:18:23<14:16,  2.01it/s]
 84%|████████▍ | 8986/10707 [1:18:23<14:15,  2.01it/s]
 84%|████████▍ | 8987/10707 [1:18:24<14:13,  2.01it/s]
 84%|████████▍ | 8988/10707 [1:18:24<14:12,  2.02it/s]
 84%|████████▍ | 8989/10707 [1:18:25<14:11,  2.02it/s]
 84%|████████▍ | 8990/10707 [1:18:25<14:11,  2.02it/s]
 84%|████████▍ | 8991/10707 [1:18:26<14:09,  2.02it/s]
 84%|████████▍ | 8992/10707 [1:18:26<14:09,  2.02it/s]
 84%|████████▍ | 8993/10707 [1:18:27<14:08,  2.02it/s]
 84%|████████▍ | 8994/10707 [1:18:27<14:07,  2.02it/s]
 84%|████████▍ | 8995/10707 [1:18:28<14:07,  2.02it/s]
 84%|████████▍ | 8996/10707 [1:18:28<14:07,  2.02it/s]
 84%|████████▍ | 8997/10707 [1:18:29<14:06,  2.02it/s]
 84%|████████▍ | 8998/10707 [1:18:29<14:06,  2.02it/s]
 84%|████████▍ | 8999/10707 [1:18:30<14:05,  2.02it/s]
 84%|████████▍ | 9000/10707 [1:18:30<14:04,  2.02it/s]{'loss': 3.5931, 'grad_norm': 0.19717416167259216, 'learning_rate': 7.545271935883957e-05, 'epoch': 0.84}
-
                                                      

 84%|████████▍ | 9000/10707 [1:18:30<14:04,  2.02it/s]
 84%|████████▍ | 9001/10707 [1:18:31<14:04,  2.02it/s]
 84%|████████▍ | 9002/10707 [1:18:31<14:04,  2.02it/s]
 84%|████████▍ | 9003/10707 [1:18:32<14:02,  2.02it/s]
 84%|████████▍ | 9004/10707 [1:18:32<14:03,  2.02it/s]
 84%|████████▍ | 9005/10707 [1:18:33<14:03,  2.02it/s]
 84%|████████▍ | 9006/10707 [1:18:33<14:04,  2.01it/s]
 84%|████████▍ | 9007/10707 [1:18:34<14:03,  2.02it/s]
 84%|████████▍ | 9008/10707 [1:18:34<14:02,  2.02it/s]
 84%|████████▍ | 9009/10707 [1:18:35<14:01,  2.02it/s]
 84%|████████▍ | 9010/10707 [1:18:35<14:00,  2.02it/s]
 84%|████████▍ | 9011/10707 [1:18:36<13:59,  2.02it/s]
 84%|████████▍ | 9012/10707 [1:18:36<13:59,  2.02it/s]
 84%|████████▍ | 9013/10707 [1:18:37<13:58,  2.02it/s]
 84%|████████▍ | 9014/10707 [1:18:37<13:57,  2.02it/s]
 84%|████████▍ | 9015/10707 [1:18:38<13:57,  2.02it/s]
 84%|████████▍ | 9016/10707 [1:18:38<13:56,  2.02it/s]
 84%|████████▍ | 9017/10707 [1:18:39<13:57,  2.02it/s]
 84%|████████▍ | 9018/10707 [1:18:39<13:56,  2.02it/s]
 84%|████████▍ | 9019/10707 [1:18:40<13:56,  2.02it/s]
 84%|████████▍ | 9020/10707 [1:18:40<13:55,  2.02it/s]
 84%|████████▍ | 9021/10707 [1:18:41<13:55,  2.02it/s]
 84%|████████▍ | 9022/10707 [1:18:41<13:53,  2.02it/s]
 84%|████████▍ | 9023/10707 [1:18:42<13:54,  2.02it/s]
 84%|████████▍ | 9024/10707 [1:18:42<13:53,  2.02it/s]
 84%|████████▍ | 9025/10707 [1:18:43<13:53,  2.02it/s]
                                                      
{'loss': 3.5934, 'grad_norm': 0.1970512866973877, 'learning_rate': 7.331408771512733e-05, 'epoch': 0.84}
-
 84%|████████▍ | 9025/10707 [1:18:43<13:53,  2.02it/s]
 84%|████████▍ | 9026/10707 [1:18:43<13:54,  2.01it/s]
 84%|████████▍ | 9027/10707 [1:18:44<13:52,  2.02it/s]
 84%|████████▍ | 9028/10707 [1:18:44<13:53,  2.01it/s]
 84%|████████▍ | 9029/10707 [1:18:45<13:51,  2.02it/s]
 84%|████████▍ | 9030/10707 [1:18:45<13:50,  2.02it/s]
 84%|████████▍ | 9031/10707 [1:18:46<13:50,  2.02it/s]
 84%|████████▍ | 9032/10707 [1:18:46<13:50,  2.02it/s]
 84%|████████▍ | 9033/10707 [1:18:47<13:49,  2.02it/s]
 84%|████████▍ | 9034/10707 [1:18:47<13:48,  2.02it/s]
 84%|████████▍ | 9035/10707 [1:18:48<13:47,  2.02it/s]
 84%|████████▍ | 9036/10707 [1:18:48<13:47,  2.02it/s]
 84%|████████▍ | 9037/10707 [1:18:49<13:46,  2.02it/s]
 84%|████████▍ | 9038/10707 [1:18:49<13:46,  2.02it/s]
 84%|████████▍ | 9039/10707 [1:18:50<13:45,  2.02it/s]
 84%|████████▍ | 9040/10707 [1:18:50<13:45,  2.02it/s]
 84%|████████▍ | 9041/10707 [1:18:51<13:44,  2.02it/s]
 84%|████████▍ | 9042/10707 [1:18:51<13:44,  2.02it/s]
 84%|████████▍ | 9043/10707 [1:18:52<13:44,  2.02it/s]
 84%|████████▍ | 9044/10707 [1:18:52<13:43,  2.02it/s]
 84%|████████▍ | 9045/10707 [1:18:53<13:44,  2.02it/s]
 84%|████████▍ | 9046/10707 [1:18:53<13:43,  2.02it/s]
 84%|████████▍ | 9047/10707 [1:18:54<13:43,  2.02it/s]
 85%|████████▍ | 9048/10707 [1:18:54<13:42,  2.02it/s]
 85%|████████▍ | 9049/10707 [1:18:55<13:42,  2.02it/s]
 85%|████████▍ | 9050/10707 [1:18:55<13:42,  2.02it/s]{'loss': 3.595, 'grad_norm': 0.2010490596294403, 'learning_rate': 7.120380209081034e-05, 'epoch': 0.85}
-
                                                      

 85%|████████▍ | 9050/10707 [1:18:55<13:42,  2.02it/s]
 85%|████████▍ | 9051/10707 [1:18:56<13:41,  2.02it/s]
 85%|████████▍ | 9052/10707 [1:18:56<13:40,  2.02it/s]
 85%|████████▍ | 9053/10707 [1:18:57<13:39,  2.02it/s]
 85%|████████▍ | 9054/10707 [1:18:57<13:38,  2.02it/s]
 85%|████████▍ | 9055/10707 [1:18:57<13:37,  2.02it/s]
 85%|████████▍ | 9056/10707 [1:18:58<13:37,  2.02it/s]
 85%|████████▍ | 9057/10707 [1:18:58<13:36,  2.02it/s]
 85%|████████▍ | 9058/10707 [1:18:59<13:37,  2.02it/s]
 85%|████████▍ | 9059/10707 [1:18:59<13:35,  2.02it/s]
 85%|████████▍ | 9060/10707 [1:19:00<13:34,  2.02it/s]
 85%|████████▍ | 9061/10707 [1:19:00<13:34,  2.02it/s]
 85%|████████▍ | 9062/10707 [1:19:01<13:33,  2.02it/s]
 85%|████████▍ | 9063/10707 [1:19:01<13:33,  2.02it/s]
 85%|████████▍ | 9064/10707 [1:19:02<13:33,  2.02it/s]
 85%|████████▍ | 9065/10707 [1:19:02<13:32,  2.02it/s]
 85%|████████▍ | 9066/10707 [1:19:03<13:32,  2.02it/s]
 85%|████████▍ | 9067/10707 [1:19:03<13:33,  2.01it/s]
 85%|████████▍ | 9068/10707 [1:19:04<13:31,  2.02it/s]
 85%|████████▍ | 9069/10707 [1:19:04<13:31,  2.02it/s]
 85%|████████▍ | 9070/10707 [1:19:05<13:30,  2.02it/s]
 85%|████████▍ | 9071/10707 [1:19:05<13:30,  2.02it/s]
 85%|████████▍ | 9072/10707 [1:19:06<13:28,  2.02it/s]
 85%|████████▍ | 9073/10707 [1:19:06<13:28,  2.02it/s]
 85%|████████▍ | 9074/10707 [1:19:07<13:28,  2.02it/s]
 85%|████████▍ | 9075/10707 [1:19:07<13:27,  2.02it/s]{'loss': 3.5932, 'grad_norm': 0.1938103437423706, 'learning_rate': 6.912200267846474e-05, 'epoch': 0.85}
-
                                                      

 85%|████████▍ | 9075/10707 [1:19:07<13:27,  2.02it/s]
 85%|████████▍ | 9076/10707 [1:19:08<13:28,  2.02it/s]
 85%|████████▍ | 9077/10707 [1:19:08<13:27,  2.02it/s]
 85%|████████▍ | 9078/10707 [1:19:09<13:26,  2.02it/s]
 85%|████████▍ | 9079/10707 [1:19:09<13:26,  2.02it/s]
 85%|████████▍ | 9080/10707 [1:19:10<13:25,  2.02it/s]
 85%|████████▍ | 9081/10707 [1:19:10<13:24,  2.02it/s]
 85%|████████▍ | 9082/10707 [1:19:11<13:24,  2.02it/s]
 85%|████████▍ | 9083/10707 [1:19:11<13:24,  2.02it/s]
 85%|████████▍ | 9084/10707 [1:19:12<13:23,  2.02it/s]
 85%|████████▍ | 9085/10707 [1:19:12<13:23,  2.02it/s]
 85%|████████▍ | 9086/10707 [1:19:13<13:22,  2.02it/s]
 85%|████████▍ | 9087/10707 [1:19:13<13:22,  2.02it/s]
 85%|████████▍ | 9088/10707 [1:19:14<13:20,  2.02it/s]
 85%|████████▍ | 9089/10707 [1:19:14<13:20,  2.02it/s]
 85%|████████▍ | 9090/10707 [1:19:15<13:19,  2.02it/s]
 85%|████████▍ | 9091/10707 [1:19:15<13:20,  2.02it/s]
 85%|████████▍ | 9092/10707 [1:19:16<13:19,  2.02it/s]
 85%|████████▍ | 9093/10707 [1:19:16<13:19,  2.02it/s]
 85%|████████▍ | 9094/10707 [1:19:17<13:17,  2.02it/s]
 85%|████████▍ | 9095/10707 [1:19:17<13:17,  2.02it/s]
 85%|████████▍ | 9096/10707 [1:19:18<13:16,  2.02it/s]
 85%|████████▍ | 9097/10707 [1:19:18<13:16,  2.02it/s]
 85%|████████▍ | 9098/10707 [1:19:19<13:15,  2.02it/s]
 85%|████████▍ | 9099/10707 [1:19:19<13:15,  2.02it/s]
 85%|████████▍ | 9100/10707 [1:19:20<13:14,  2.02it/s]{'loss': 3.5869, 'grad_norm': 0.19811514019966125, 'learning_rate': 6.706882777824257e-05, 'epoch': 0.85}
-
                                                      

 85%|████████▍ | 9100/10707 [1:19:20<13:14,  2.02it/s]
 85%|████████▌ | 9101/10707 [1:19:20<13:16,  2.02it/s]
 85%|████████▌ | 9102/10707 [1:19:21<13:14,  2.02it/s]
 85%|████████▌ | 9103/10707 [1:19:21<13:14,  2.02it/s]
 85%|████████▌ | 9104/10707 [1:19:22<13:13,  2.02it/s]
 85%|████████▌ | 9105/10707 [1:19:22<13:14,  2.02it/s]
 85%|████████▌ | 9106/10707 [1:19:23<13:14,  2.02it/s]
 85%|████████▌ | 9107/10707 [1:19:23<13:13,  2.02it/s]
 85%|████████▌ | 9108/10707 [1:19:24<13:11,  2.02it/s]
 85%|████████▌ | 9109/10707 [1:19:24<13:11,  2.02it/s]
 85%|████████▌ | 9110/10707 [1:19:25<13:09,  2.02it/s]
 85%|████████▌ | 9111/10707 [1:19:25<13:10,  2.02it/s]
 85%|████████▌ | 9112/10707 [1:19:26<13:09,  2.02it/s]
 85%|████████▌ | 9113/10707 [1:19:26<13:08,  2.02it/s]
 85%|████████▌ | 9114/10707 [1:19:27<13:08,  2.02it/s]
 85%|████████▌ | 9115/10707 [1:19:27<13:07,  2.02it/s]
 85%|████████▌ | 9116/10707 [1:19:28<13:07,  2.02it/s]
 85%|████████▌ | 9117/10707 [1:19:28<13:06,  2.02it/s]
 85%|████████▌ | 9118/10707 [1:19:29<13:06,  2.02it/s]
 85%|████████▌ | 9119/10707 [1:19:29<13:05,  2.02it/s]
 85%|████████▌ | 9120/10707 [1:19:30<13:05,  2.02it/s]
 85%|████████▌ | 9121/10707 [1:19:30<13:05,  2.02it/s]
 85%|████████▌ | 9122/10707 [1:19:31<13:04,  2.02it/s]
 85%|████████▌ | 9123/10707 [1:19:31<13:04,  2.02it/s]
 85%|████████▌ | 9124/10707 [1:19:32<13:03,  2.02it/s]
 85%|████████▌ | 9125/10707 [1:19:32<13:03,  2.02it/s]{'loss': 3.5822, 'grad_norm': 0.19582489132881165, 'learning_rate': 6.504441378868392e-05, 'epoch': 0.85}
-
                                                      

 85%|████████▌ | 9125/10707 [1:19:32<13:03,  2.02it/s]
 85%|████████▌ | 9126/10707 [1:19:33<13:04,  2.02it/s]
 85%|████████▌ | 9127/10707 [1:19:33<13:04,  2.01it/s]
 85%|████████▌ | 9128/10707 [1:19:34<13:04,  2.01it/s]
 85%|████████▌ | 9129/10707 [1:19:34<13:03,  2.01it/s]
 85%|████████▌ | 9130/10707 [1:19:35<13:02,  2.02it/s]
 85%|████████▌ | 9131/10707 [1:19:35<13:01,  2.02it/s]
 85%|████████▌ | 9132/10707 [1:19:36<13:00,  2.02it/s]
 85%|████████▌ | 9133/10707 [1:19:36<12:59,  2.02it/s]
 85%|████████▌ | 9134/10707 [1:19:37<12:58,  2.02it/s]
 85%|████████▌ | 9135/10707 [1:19:37<12:58,  2.02it/s]
 85%|████████▌ | 9136/10707 [1:19:38<12:58,  2.02it/s]
 85%|████████▌ | 9137/10707 [1:19:38<12:58,  2.02it/s]
 85%|████████▌ | 9138/10707 [1:19:39<12:57,  2.02it/s]
 85%|████████▌ | 9139/10707 [1:19:39<12:57,  2.02it/s]
 85%|████████▌ | 9140/10707 [1:19:40<12:56,  2.02it/s]
 85%|████████▌ | 9141/10707 [1:19:40<12:55,  2.02it/s]
 85%|████████▌ | 9142/10707 [1:19:41<12:55,  2.02it/s]
 85%|████████▌ | 9143/10707 [1:19:41<12:55,  2.02it/s]
 85%|████████▌ | 9144/10707 [1:19:42<12:55,  2.02it/s]
 85%|████████▌ | 9145/10707 [1:19:42<12:54,  2.02it/s]
 85%|████████▌ | 9146/10707 [1:19:43<12:53,  2.02it/s]
 85%|████████▌ | 9147/10707 [1:19:43<12:52,  2.02it/s]
 85%|████████▌ | 9148/10707 [1:19:44<12:51,  2.02it/s]
 85%|████████▌ | 9149/10707 [1:19:44<12:51,  2.02it/s]
 85%|████████▌ | 9150/10707 [1:19:45<12:50,  2.02it/s]{'loss': 3.5909, 'grad_norm': 0.19437777996063232, 'learning_rate': 6.304889519765577e-05, 'epoch': 0.85}
-
                                                      

 85%|████████▌ | 9150/10707 [1:19:45<12:50,  2.02it/s]
 85%|████████▌ | 9151/10707 [1:19:45<12:50,  2.02it/s]
 85%|████████▌ | 9152/10707 [1:19:46<12:49,  2.02it/s]
 85%|████████▌ | 9153/10707 [1:19:46<12:49,  2.02it/s]
 85%|████████▌ | 9154/10707 [1:19:47<12:48,  2.02it/s]
 86%|████████▌ | 9155/10707 [1:19:47<12:48,  2.02it/s]
 86%|████████▌ | 9156/10707 [1:19:48<12:47,  2.02it/s]
 86%|████████▌ | 9157/10707 [1:19:48<12:46,  2.02it/s]
 86%|████████▌ | 9158/10707 [1:19:49<12:47,  2.02it/s]
 86%|████████▌ | 9159/10707 [1:19:49<12:46,  2.02it/s]
 86%|████████▌ | 9160/10707 [1:19:49<12:45,  2.02it/s]
 86%|████████▌ | 9161/10707 [1:19:50<12:44,  2.02it/s]
 86%|████████▌ | 9162/10707 [1:19:50<12:44,  2.02it/s]
 86%|████████▌ | 9163/10707 [1:19:51<12:44,  2.02it/s]
 86%|████████▌ | 9164/10707 [1:19:51<12:43,  2.02it/s]
 86%|████████▌ | 9165/10707 [1:19:52<12:42,  2.02it/s]
 86%|████████▌ | 9166/10707 [1:19:52<12:42,  2.02it/s]
 86%|████████▌ | 9167/10707 [1:19:53<12:41,  2.02it/s]
 86%|████████▌ | 9168/10707 [1:19:53<12:41,  2.02it/s]
 86%|████████▌ | 9169/10707 [1:19:54<12:41,  2.02it/s]
 86%|████████▌ | 9170/10707 [1:19:54<12:40,  2.02it/s]
 86%|████████▌ | 9171/10707 [1:19:55<12:40,  2.02it/s]
 86%|████████▌ | 9172/10707 [1:19:55<12:39,  2.02it/s]
 86%|████████▌ | 9173/10707 [1:19:56<12:39,  2.02it/s]
 86%|████████▌ | 9174/10707 [1:19:56<12:38,  2.02it/s]
 86%|████████▌ | 9175/10707 [1:19:57<12:38,  2.02it/s]{'loss': 3.5913, 'grad_norm': 0.19037391245365143, 'learning_rate': 6.108240457341751e-05, 'epoch': 0.86}
-
                                                      

 86%|████████▌ | 9175/10707 [1:19:57<12:38,  2.02it/s]
 86%|████████▌ | 9176/10707 [1:19:57<12:38,  2.02it/s]
 86%|████████▌ | 9177/10707 [1:19:58<12:38,  2.02it/s]
 86%|████████▌ | 9178/10707 [1:19:58<12:38,  2.02it/s]
 86%|████████▌ | 9179/10707 [1:19:59<12:37,  2.02it/s]
 86%|████████▌ | 9180/10707 [1:19:59<12:35,  2.02it/s]
 86%|████████▌ | 9181/10707 [1:20:00<12:35,  2.02it/s]
 86%|████████▌ | 9182/10707 [1:20:00<12:34,  2.02it/s]
 86%|████████▌ | 9183/10707 [1:20:01<12:33,  2.02it/s]
 86%|████████▌ | 9184/10707 [1:20:01<12:33,  2.02it/s]
 86%|████████▌ | 9185/10707 [1:20:02<12:33,  2.02it/s]
 86%|████████▌ | 9186/10707 [1:20:02<12:32,  2.02it/s]
 86%|████████▌ | 9187/10707 [1:20:03<12:32,  2.02it/s]
 86%|████████▌ | 9188/10707 [1:20:03<12:33,  2.02it/s]
 86%|████████▌ | 9189/10707 [1:20:04<12:32,  2.02it/s]
 86%|████████▌ | 9190/10707 [1:20:04<12:32,  2.02it/s]
 86%|████████▌ | 9191/10707 [1:20:05<12:31,  2.02it/s]
 86%|████████▌ | 9192/10707 [1:20:05<12:31,  2.02it/s]
 86%|████████▌ | 9193/10707 [1:20:06<12:30,  2.02it/s]
 86%|████████▌ | 9194/10707 [1:20:06<12:30,  2.02it/s]
 86%|████████▌ | 9195/10707 [1:20:07<12:28,  2.02it/s]
 86%|████████▌ | 9196/10707 [1:20:07<12:28,  2.02it/s]
 86%|████████▌ | 9197/10707 [1:20:08<12:27,  2.02it/s]
 86%|████████▌ | 9198/10707 [1:20:08<12:27,  2.02it/s]
 86%|████████▌ | 9199/10707 [1:20:09<12:25,  2.02it/s]
 86%|████████▌ | 9200/10707 [1:20:09<12:25,  2.02it/s]{'loss': 3.5946, 'grad_norm': 0.19603893160820007, 'learning_rate': 5.9145072555813974e-05, 'epoch': 0.86}
-
                                                      

 86%|████████▌ | 9200/10707 [1:20:09<12:25,  2.02it/s]
 86%|████████▌ | 9201/10707 [1:20:10<12:25,  2.02it/s]
 86%|████████▌ | 9202/10707 [1:20:10<12:25,  2.02it/s]
 86%|████████▌ | 9203/10707 [1:20:11<12:24,  2.02it/s]
 86%|████████▌ | 9204/10707 [1:20:11<12:24,  2.02it/s]
 86%|████████▌ | 9205/10707 [1:20:12<12:23,  2.02it/s]
 86%|████████▌ | 9206/10707 [1:20:12<12:23,  2.02it/s]
 86%|████████▌ | 9207/10707 [1:20:13<12:22,  2.02it/s]
 86%|████████▌ | 9208/10707 [1:20:13<12:22,  2.02it/s]
 86%|████████▌ | 9209/10707 [1:20:14<12:22,  2.02it/s]
 86%|████████▌ | 9210/10707 [1:20:14<12:21,  2.02it/s]
 86%|████████▌ | 9211/10707 [1:20:15<12:21,  2.02it/s]
 86%|████████▌ | 9212/10707 [1:20:15<12:19,  2.02it/s]
 86%|████████▌ | 9213/10707 [1:20:16<12:20,  2.02it/s]
 86%|████████▌ | 9214/10707 [1:20:16<12:19,  2.02it/s]
 86%|████████▌ | 9215/10707 [1:20:17<12:19,  2.02it/s]
 86%|████████▌ | 9216/10707 [1:20:17<12:17,  2.02it/s]
 86%|████████▌ | 9217/10707 [1:20:18<12:17,  2.02it/s]
 86%|████████▌ | 9218/10707 [1:20:18<12:17,  2.02it/s]
 86%|████████▌ | 9219/10707 [1:20:19<12:17,  2.02it/s]
 86%|████████▌ | 9220/10707 [1:20:19<12:15,  2.02it/s]
 86%|████████▌ | 9221/10707 [1:20:20<12:15,  2.02it/s]
 86%|████████▌ | 9222/10707 [1:20:20<12:14,  2.02it/s]
 86%|████████▌ | 9223/10707 [1:20:21<12:14,  2.02it/s]
 86%|████████▌ | 9224/10707 [1:20:21<12:14,  2.02it/s]
 86%|████████▌ | 9225/10707 [1:20:22<12:14,  2.02it/s]{'loss': 3.586, 'grad_norm': 0.19919906556606293, 'learning_rate': 5.7237027847596714e-05, 'epoch': 0.86}
-
                                                      

 86%|████████▌ | 9225/10707 [1:20:22<12:14,  2.02it/s]
 86%|████████▌ | 9226/10707 [1:20:22<12:14,  2.02it/s]
 86%|████████▌ | 9227/10707 [1:20:23<12:13,  2.02it/s]
 86%|████████▌ | 9228/10707 [1:20:23<12:12,  2.02it/s]
 86%|████████▌ | 9229/10707 [1:20:24<12:11,  2.02it/s]
 86%|████████▌ | 9230/10707 [1:20:24<12:11,  2.02it/s]
 86%|████████▌ | 9231/10707 [1:20:25<12:10,  2.02it/s]
 86%|████████▌ | 9232/10707 [1:20:25<12:09,  2.02it/s]
 86%|████████▌ | 9233/10707 [1:20:26<12:09,  2.02it/s]
 86%|████████▌ | 9234/10707 [1:20:26<12:09,  2.02it/s]
 86%|████████▋ | 9235/10707 [1:20:27<12:08,  2.02it/s]
 86%|████████▋ | 9236/10707 [1:20:27<12:07,  2.02it/s]
 86%|████████▋ | 9237/10707 [1:20:28<12:08,  2.02it/s]
 86%|████████▋ | 9238/10707 [1:20:28<12:06,  2.02it/s]
 86%|████████▋ | 9239/10707 [1:20:29<12:06,  2.02it/s]
 86%|████████▋ | 9240/10707 [1:20:29<12:05,  2.02it/s]
 86%|████████▋ | 9241/10707 [1:20:30<12:05,  2.02it/s]
 86%|████████▋ | 9242/10707 [1:20:30<12:04,  2.02it/s]
 86%|████████▋ | 9243/10707 [1:20:31<12:05,  2.02it/s]
 86%|████████▋ | 9244/10707 [1:20:31<12:03,  2.02it/s]
 86%|████████▋ | 9245/10707 [1:20:32<12:03,  2.02it/s]
 86%|████████▋ | 9246/10707 [1:20:32<12:02,  2.02it/s]
 86%|████████▋ | 9247/10707 [1:20:33<12:02,  2.02it/s]
 86%|████████▋ | 9248/10707 [1:20:33<12:02,  2.02it/s]
 86%|████████▋ | 9249/10707 [1:20:34<12:03,  2.02it/s]
 86%|████████▋ | 9250/10707 [1:20:34<12:02,  2.02it/s]{'loss': 3.5886, 'grad_norm': 0.20326022803783417, 'learning_rate': 5.535839720587371e-05, 'epoch': 0.86}
-
                                                      

 86%|████████▋ | 9250/10707 [1:20:34<12:02,  2.02it/s]
 86%|████████▋ | 9251/10707 [1:20:35<12:01,  2.02it/s]
 86%|████████▋ | 9252/10707 [1:20:35<12:01,  2.02it/s]
 86%|████████▋ | 9253/10707 [1:20:36<12:00,  2.02it/s]
 86%|████████▋ | 9254/10707 [1:20:36<12:00,  2.02it/s]
 86%|████████▋ | 9255/10707 [1:20:37<11:59,  2.02it/s]
 86%|████████▋ | 9256/10707 [1:20:37<11:58,  2.02it/s]
 86%|████████▋ | 9257/10707 [1:20:38<11:57,  2.02it/s]
 86%|████████▋ | 9258/10707 [1:20:38<11:57,  2.02it/s]
 86%|████████▋ | 9259/10707 [1:20:39<11:56,  2.02it/s]
 86%|████████▋ | 9260/10707 [1:20:39<11:56,  2.02it/s]
 86%|████████▋ | 9261/10707 [1:20:40<11:55,  2.02it/s]
 87%|████████▋ | 9262/10707 [1:20:40<11:55,  2.02it/s]
 87%|████████▋ | 9263/10707 [1:20:40<11:54,  2.02it/s]
 87%|████████▋ | 9264/10707 [1:20:41<11:54,  2.02it/s]
 87%|████████▋ | 9265/10707 [1:20:41<11:53,  2.02it/s]
 87%|████████▋ | 9266/10707 [1:20:42<11:53,  2.02it/s]
 87%|████████▋ | 9267/10707 [1:20:42<11:52,  2.02it/s]
 87%|████████▋ | 9268/10707 [1:20:43<11:52,  2.02it/s]
 87%|████████▋ | 9269/10707 [1:20:43<11:51,  2.02it/s]
 87%|████████▋ | 9270/10707 [1:20:44<11:50,  2.02it/s]
 87%|████████▋ | 9271/10707 [1:20:44<11:50,  2.02it/s]
 87%|████████▋ | 9272/10707 [1:20:45<11:50,  2.02it/s]
 87%|████████▋ | 9273/10707 [1:20:45<11:49,  2.02it/s]
 87%|████████▋ | 9274/10707 [1:20:46<11:49,  2.02it/s]
 87%|████████▋ | 9275/10707 [1:20:46<11:48,  2.02it/s]{'loss': 3.5886, 'grad_norm': 0.19901801645755768, 'learning_rate': 5.3509305433688926e-05, 'epoch': 0.87}
-
                                                      

 87%|████████▋ | 9275/10707 [1:20:46<11:48,  2.02it/s]
 87%|████████▋ | 9276/10707 [1:20:47<11:49,  2.02it/s]
 87%|████████▋ | 9277/10707 [1:20:47<11:48,  2.02it/s]
 87%|████████▋ | 9278/10707 [1:20:48<11:47,  2.02it/s]
 87%|████████▋ | 9279/10707 [1:20:48<11:46,  2.02it/s]
 87%|████████▋ | 9280/10707 [1:20:49<11:46,  2.02it/s]
 87%|████████▋ | 9281/10707 [1:20:49<11:45,  2.02it/s]
 87%|████████▋ | 9282/10707 [1:20:50<11:45,  2.02it/s]
 87%|████████▋ | 9283/10707 [1:20:50<11:44,  2.02it/s]
 87%|████████▋ | 9284/10707 [1:20:51<11:43,  2.02it/s]
 87%|████████▋ | 9285/10707 [1:20:51<11:43,  2.02it/s]
 87%|████████▋ | 9286/10707 [1:20:52<11:43,  2.02it/s]
 87%|████████▋ | 9287/10707 [1:20:52<11:43,  2.02it/s]
 87%|████████▋ | 9288/10707 [1:20:53<11:42,  2.02it/s]
 87%|████████▋ | 9289/10707 [1:20:53<11:42,  2.02it/s]
 87%|████████▋ | 9290/10707 [1:20:54<11:41,  2.02it/s]
 87%|████████▋ | 9291/10707 [1:20:54<11:41,  2.02it/s]
 87%|████████▋ | 9292/10707 [1:20:55<11:40,  2.02it/s]
 87%|████████▋ | 9293/10707 [1:20:55<11:40,  2.02it/s]
 87%|████████▋ | 9294/10707 [1:20:56<11:39,  2.02it/s]
 87%|████████▋ | 9295/10707 [1:20:56<11:40,  2.02it/s]
 87%|████████▋ | 9296/10707 [1:20:57<11:39,  2.02it/s]
 87%|████████▋ | 9297/10707 [1:20:57<11:38,  2.02it/s]
 87%|████████▋ | 9298/10707 [1:20:58<11:37,  2.02it/s]
 87%|████████▋ | 9299/10707 [1:20:58<11:37,  2.02it/s]
 87%|████████▋ | 9300/10707 [1:20:59<11:35,  2.02it/s]{'loss': 3.5866, 'grad_norm': 0.19509033858776093, 'learning_rate': 5.1689875371730756e-05, 'epoch': 0.87}
-
                                                      

 87%|████████▋ | 9300/10707 [1:20:59<11:35,  2.02it/s]
 87%|████████▋ | 9301/10707 [1:20:59<11:36,  2.02it/s]
 87%|████████▋ | 9302/10707 [1:21:00<11:35,  2.02it/s]
 87%|████████▋ | 9303/10707 [1:21:00<11:35,  2.02it/s]
 87%|████████▋ | 9304/10707 [1:21:01<11:34,  2.02it/s]
 87%|████████▋ | 9305/10707 [1:21:01<11:34,  2.02it/s]
 87%|████████▋ | 9306/10707 [1:21:02<11:33,  2.02it/s]
 87%|████████▋ | 9307/10707 [1:21:02<11:33,  2.02it/s]
 87%|████████▋ | 9308/10707 [1:21:03<11:32,  2.02it/s]
 87%|████████▋ | 9309/10707 [1:21:03<11:32,  2.02it/s]
 87%|████████▋ | 9310/10707 [1:21:04<11:33,  2.01it/s]
 87%|████████▋ | 9311/10707 [1:21:04<11:31,  2.02it/s]
 87%|████████▋ | 9312/10707 [1:21:05<11:31,  2.02it/s]
 87%|████████▋ | 9313/10707 [1:21:05<11:30,  2.02it/s]
 87%|████████▋ | 9314/10707 [1:21:06<11:30,  2.02it/s]
 87%|████████▋ | 9315/10707 [1:21:06<11:30,  2.02it/s]
 87%|████████▋ | 9316/10707 [1:21:07<11:29,  2.02it/s]
 87%|████████▋ | 9317/10707 [1:21:07<11:29,  2.02it/s]
 87%|████████▋ | 9318/10707 [1:21:08<11:28,  2.02it/s]
 87%|████████▋ | 9319/10707 [1:21:08<11:27,  2.02it/s]
 87%|████████▋ | 9320/10707 [1:21:09<11:27,  2.02it/s]
 87%|████████▋ | 9321/10707 [1:21:09<11:26,  2.02it/s]
 87%|████████▋ | 9322/10707 [1:21:10<11:26,  2.02it/s]
 87%|████████▋ | 9323/10707 [1:21:10<11:25,  2.02it/s]
 87%|████████▋ | 9324/10707 [1:21:11<11:24,  2.02it/s]
 87%|████████▋ | 9325/10707 [1:21:11<11:25,  2.02it/s]{'loss': 3.5869, 'grad_norm': 0.192819744348526, 'learning_rate': 4.990022789017185e-05, 'epoch': 0.87}
-
                                                      

 87%|████████▋ | 9325/10707 [1:21:11<11:25,  2.02it/s]
 87%|████████▋ | 9326/10707 [1:21:12<11:25,  2.02it/s]
 87%|████████▋ | 9327/10707 [1:21:12<11:24,  2.02it/s]
 87%|████████▋ | 9328/10707 [1:21:13<11:24,  2.02it/s]
 87%|████████▋ | 9329/10707 [1:21:13<11:22,  2.02it/s]
 87%|████████▋ | 9330/10707 [1:21:14<11:22,  2.02it/s]
 87%|████████▋ | 9331/10707 [1:21:14<11:21,  2.02it/s]
 87%|████████▋ | 9332/10707 [1:21:15<11:21,  2.02it/s]
 87%|████████▋ | 9333/10707 [1:21:15<11:20,  2.02it/s]
 87%|████████▋ | 9334/10707 [1:21:16<11:19,  2.02it/s]
 87%|████████▋ | 9335/10707 [1:21:16<11:19,  2.02it/s]
 87%|████████▋ | 9336/10707 [1:21:17<11:19,  2.02it/s]
 87%|████████▋ | 9337/10707 [1:21:17<11:18,  2.02it/s]
 87%|████████▋ | 9338/10707 [1:21:18<11:18,  2.02it/s]
 87%|████████▋ | 9339/10707 [1:21:18<11:17,  2.02it/s]
 87%|████████▋ | 9340/10707 [1:21:19<11:17,  2.02it/s]
 87%|████████▋ | 9341/10707 [1:21:19<11:17,  2.02it/s]
 87%|████████▋ | 9342/10707 [1:21:20<11:16,  2.02it/s]
 87%|████████▋ | 9343/10707 [1:21:20<11:15,  2.02it/s]
 87%|████████▋ | 9344/10707 [1:21:21<11:14,  2.02it/s]
 87%|████████▋ | 9345/10707 [1:21:21<11:14,  2.02it/s]
 87%|████████▋ | 9346/10707 [1:21:22<11:14,  2.02it/s]
 87%|████████▋ | 9347/10707 [1:21:22<11:12,  2.02it/s]
 87%|████████▋ | 9348/10707 [1:21:23<11:12,  2.02it/s]
 87%|████████▋ | 9349/10707 [1:21:23<11:12,  2.02it/s]
 87%|████████▋ | 9350/10707 [1:21:24<11:12,  2.02it/s]{'loss': 3.5913, 'grad_norm': 0.19467785954475403, 'learning_rate': 4.814048188063869e-05, 'epoch': 0.87}
-
                                                      

 87%|████████▋ | 9350/10707 [1:21:24<11:12,  2.02it/s]
 87%|████████▋ | 9351/10707 [1:21:24<11:12,  2.02it/s]
 87%|████████▋ | 9352/10707 [1:21:25<11:11,  2.02it/s]
 87%|████████▋ | 9353/10707 [1:21:25<11:11,  2.02it/s]
 87%|████████▋ | 9354/10707 [1:21:26<11:10,  2.02it/s]
 87%|████████▋ | 9355/10707 [1:21:26<11:09,  2.02it/s]
 87%|████████▋ | 9356/10707 [1:21:27<11:08,  2.02it/s]
 87%|████████▋ | 9357/10707 [1:21:27<11:08,  2.02it/s]
 87%|████████▋ | 9358/10707 [1:21:28<11:07,  2.02it/s]
 87%|████████▋ | 9359/10707 [1:21:28<11:06,  2.02it/s]
 87%|████████▋ | 9360/10707 [1:21:29<11:07,  2.02it/s]
 87%|████████▋ | 9361/10707 [1:21:29<11:06,  2.02it/s]
 87%|████████▋ | 9362/10707 [1:21:30<11:06,  2.02it/s]
 87%|████████▋ | 9363/10707 [1:21:30<11:05,  2.02it/s]
 87%|████████▋ | 9364/10707 [1:21:31<11:04,  2.02it/s]
 87%|████████▋ | 9365/10707 [1:21:31<11:04,  2.02it/s]
 87%|████████▋ | 9366/10707 [1:21:32<11:04,  2.02it/s]
 87%|████████▋ | 9367/10707 [1:21:32<11:03,  2.02it/s]
 87%|████████▋ | 9368/10707 [1:21:32<11:03,  2.02it/s]
 88%|████████▊ | 9369/10707 [1:21:33<11:03,  2.02it/s]
 88%|████████▊ | 9370/10707 [1:21:33<11:03,  2.01it/s]
 88%|████████▊ | 9371/10707 [1:21:34<11:02,  2.02it/s]
 88%|████████▊ | 9372/10707 [1:21:34<11:01,  2.02it/s]
 88%|████████▊ | 9373/10707 [1:21:35<11:01,  2.02it/s]
 88%|████████▊ | 9374/10707 [1:21:35<11:00,  2.02it/s]
 88%|████████▊ | 9375/10707 [1:21:36<10:59,  2.02it/s]{'loss': 3.5847, 'grad_norm': 0.1919584423303604, 'learning_rate': 4.64107542483142e-05, 'epoch': 0.88}
                                                      
-
 88%|████████▊ | 9375/10707 [1:21:36<10:59,  2.02it/s]
 88%|████████▊ | 9376/10707 [1:21:36<11:00,  2.01it/s]
 88%|████████▊ | 9377/10707 [1:21:37<10:59,  2.02it/s]
 88%|████████▊ | 9378/10707 [1:21:37<10:59,  2.02it/s]
 88%|████████▊ | 9379/10707 [1:21:38<10:58,  2.02it/s]
 88%|████████▊ | 9380/10707 [1:21:38<10:57,  2.02it/s]
 88%|████████▊ | 9381/10707 [1:21:39<10:57,  2.02it/s]
 88%|████████▊ | 9382/10707 [1:21:39<10:56,  2.02it/s]
 88%|████████▊ | 9383/10707 [1:21:40<10:55,  2.02it/s]
 88%|████████▊ | 9384/10707 [1:21:40<10:54,  2.02it/s]
 88%|████████▊ | 9385/10707 [1:21:41<10:53,  2.02it/s]
 88%|████████▊ | 9386/10707 [1:21:41<10:53,  2.02it/s]
 88%|████████▊ | 9387/10707 [1:21:42<10:53,  2.02it/s]
 88%|████████▊ | 9388/10707 [1:21:42<10:52,  2.02it/s]
 88%|████████▊ | 9389/10707 [1:21:43<10:52,  2.02it/s]
 88%|████████▊ | 9390/10707 [1:21:43<10:51,  2.02it/s]
 88%|████████▊ | 9391/10707 [1:21:44<10:51,  2.02it/s]
 88%|████████▊ | 9392/10707 [1:21:44<10:50,  2.02it/s]
 88%|████████▊ | 9393/10707 [1:21:45<10:50,  2.02it/s]
 88%|████████▊ | 9394/10707 [1:21:45<10:49,  2.02it/s]
 88%|████████▊ | 9395/10707 [1:21:46<10:49,  2.02it/s]
 88%|████████▊ | 9396/10707 [1:21:46<10:48,  2.02it/s]
 88%|████████▊ | 9397/10707 [1:21:47<10:48,  2.02it/s]
 88%|████████▊ | 9398/10707 [1:21:47<10:47,  2.02it/s]
 88%|████████▊ | 9399/10707 [1:21:48<10:47,  2.02it/s]
 88%|████████▊ | 9400/10707 [1:21:48<10:46,  2.02it/s]{'loss': 3.5828, 'grad_norm': 0.19641809165477753, 'learning_rate': 4.471115990417057e-05, 'epoch': 0.88}
-
                                                      

 88%|████████▊ | 9400/10707 [1:21:48<10:46,  2.02it/s]
 88%|████████▊ | 9401/10707 [1:21:49<10:47,  2.02it/s]
 88%|████████▊ | 9402/10707 [1:21:49<10:46,  2.02it/s]
 88%|████████▊ | 9403/10707 [1:21:50<10:45,  2.02it/s]
 88%|████████▊ | 9404/10707 [1:21:50<10:45,  2.02it/s]
 88%|████████▊ | 9405/10707 [1:21:51<10:44,  2.02it/s]
 88%|████████▊ | 9406/10707 [1:21:51<10:44,  2.02it/s]
 88%|████████▊ | 9407/10707 [1:21:52<10:43,  2.02it/s]
 88%|████████▊ | 9408/10707 [1:21:52<10:43,  2.02it/s]
 88%|████████▊ | 9409/10707 [1:21:53<10:42,  2.02it/s]
 88%|████████▊ | 9410/10707 [1:21:53<10:41,  2.02it/s]
 88%|████████▊ | 9411/10707 [1:21:54<10:41,  2.02it/s]
 88%|████████▊ | 9412/10707 [1:21:54<10:40,  2.02it/s]
 88%|████████▊ | 9413/10707 [1:21:55<10:40,  2.02it/s]
 88%|████████▊ | 9414/10707 [1:21:55<10:40,  2.02it/s]
 88%|████████▊ | 9415/10707 [1:21:56<10:39,  2.02it/s]
 88%|████████▊ | 9416/10707 [1:21:56<10:38,  2.02it/s]
 88%|████████▊ | 9417/10707 [1:21:57<10:38,  2.02it/s]
 88%|████████▊ | 9418/10707 [1:21:57<10:37,  2.02it/s]
 88%|████████▊ | 9419/10707 [1:21:58<10:36,  2.02it/s]
 88%|████████▊ | 9420/10707 [1:21:58<10:36,  2.02it/s]
 88%|████████▊ | 9421/10707 [1:21:59<10:36,  2.02it/s]
 88%|████████▊ | 9422/10707 [1:21:59<10:35,  2.02it/s]
 88%|████████▊ | 9423/10707 [1:22:00<10:35,  2.02it/s]
 88%|████████▊ | 9424/10707 [1:22:00<10:34,  2.02it/s]
 88%|████████▊ | 9425/10707 [1:22:01<10:34,  2.02it/s]{'loss': 3.5882, 'grad_norm': 0.1931866854429245, 'learning_rate': 4.304181175733557e-05, 'epoch': 0.88}
-
                                                      

 88%|████████▊ | 9425/10707 [1:22:01<10:34,  2.02it/s]
 88%|████████▊ | 9426/10707 [1:22:01<10:35,  2.02it/s]
 88%|████████▊ | 9427/10707 [1:22:02<10:34,  2.02it/s]
 88%|████████▊ | 9428/10707 [1:22:02<10:33,  2.02it/s]
 88%|████████▊ | 9429/10707 [1:22:03<10:32,  2.02it/s]
 88%|████████▊ | 9430/10707 [1:22:03<10:33,  2.02it/s]
 88%|████████▊ | 9431/10707 [1:22:04<10:32,  2.02it/s]
 88%|████████▊ | 9432/10707 [1:22:04<10:31,  2.02it/s]
 88%|████████▊ | 9433/10707 [1:22:05<10:30,  2.02it/s]
 88%|████████▊ | 9434/10707 [1:22:05<10:31,  2.02it/s]
 88%|████████▊ | 9435/10707 [1:22:06<10:29,  2.02it/s]
 88%|████████▊ | 9436/10707 [1:22:06<10:29,  2.02it/s]
 88%|████████▊ | 9437/10707 [1:22:07<10:28,  2.02it/s]
 88%|████████▊ | 9438/10707 [1:22:07<10:28,  2.02it/s]
 88%|████████▊ | 9439/10707 [1:22:08<10:27,  2.02it/s]
 88%|████████▊ | 9440/10707 [1:22:08<10:27,  2.02it/s]
 88%|████████▊ | 9441/10707 [1:22:09<10:26,  2.02it/s]
 88%|████████▊ | 9442/10707 [1:22:09<10:27,  2.02it/s]
 88%|████████▊ | 9443/10707 [1:22:10<10:26,  2.02it/s]
 88%|████████▊ | 9444/10707 [1:22:10<10:25,  2.02it/s]
 88%|████████▊ | 9445/10707 [1:22:11<10:25,  2.02it/s]
 88%|████████▊ | 9446/10707 [1:22:11<10:24,  2.02it/s]
 88%|████████▊ | 9447/10707 [1:22:12<10:23,  2.02it/s]
 88%|████████▊ | 9448/10707 [1:22:12<10:23,  2.02it/s]
 88%|████████▊ | 9449/10707 [1:22:13<10:23,  2.02it/s]
 88%|████████▊ | 9450/10707 [1:22:13<10:22,  2.02it/s]{'loss': 3.5859, 'grad_norm': 0.1912146359682083, 'learning_rate': 4.1402820707591826e-05, 'epoch': 0.88}
-
                                                      

 88%|████████▊ | 9450/10707 [1:22:13<10:22,  2.02it/s]
 88%|████████▊ | 9451/10707 [1:22:14<10:23,  2.01it/s]
 88%|████████▊ | 9452/10707 [1:22:14<10:22,  2.01it/s]
 88%|████████▊ | 9453/10707 [1:22:15<10:22,  2.01it/s]
 88%|████████▊ | 9454/10707 [1:22:15<10:21,  2.02it/s]
 88%|████████▊ | 9455/10707 [1:22:16<10:20,  2.02it/s]
 88%|████████▊ | 9456/10707 [1:22:16<10:20,  2.02it/s]
 88%|████████▊ | 9457/10707 [1:22:17<10:19,  2.02it/s]
 88%|████████▊ | 9458/10707 [1:22:17<10:18,  2.02it/s]
 88%|████████▊ | 9459/10707 [1:22:18<10:17,  2.02it/s]
 88%|████████▊ | 9460/10707 [1:22:18<10:17,  2.02it/s]
 88%|████████▊ | 9461/10707 [1:22:19<10:16,  2.02it/s]
 88%|████████▊ | 9462/10707 [1:22:19<10:16,  2.02it/s]
 88%|████████▊ | 9463/10707 [1:22:20<10:15,  2.02it/s]
 88%|████████▊ | 9464/10707 [1:22:20<10:15,  2.02it/s]
 88%|████████▊ | 9465/10707 [1:22:21<10:14,  2.02it/s]
 88%|████████▊ | 9466/10707 [1:22:21<10:14,  2.02it/s]
 88%|████████▊ | 9467/10707 [1:22:22<10:13,  2.02it/s]
 88%|████████▊ | 9468/10707 [1:22:22<10:13,  2.02it/s]
 88%|████████▊ | 9469/10707 [1:22:23<10:13,  2.02it/s]
 88%|████████▊ | 9470/10707 [1:22:23<10:13,  2.02it/s]
 88%|████████▊ | 9471/10707 [1:22:24<10:12,  2.02it/s]
 88%|████████▊ | 9472/10707 [1:22:24<10:11,  2.02it/s]
 88%|████████▊ | 9473/10707 [1:22:25<10:11,  2.02it/s]
 88%|████████▊ | 9474/10707 [1:22:25<10:11,  2.02it/s]
 88%|████████▊ | 9475/10707 [1:22:25<10:10,  2.02it/s]{'loss': 3.5864, 'grad_norm': 0.19553107023239136, 'learning_rate': 3.9794295638009684e-05, 'epoch': 0.88}
-
                                                      

 88%|████████▊ | 9475/10707 [1:22:25<10:10,  2.02it/s]
 89%|████████▊ | 9476/10707 [1:22:26<10:11,  2.01it/s]
 89%|████████▊ | 9477/10707 [1:22:26<10:10,  2.01it/s]
 89%|████████▊ | 9478/10707 [1:22:27<10:09,  2.02it/s]
 89%|████████▊ | 9479/10707 [1:22:27<10:09,  2.02it/s]
 89%|████████▊ | 9480/10707 [1:22:28<10:08,  2.02it/s]
 89%|████████▊ | 9481/10707 [1:22:28<10:07,  2.02it/s]
 89%|████████▊ | 9482/10707 [1:22:29<10:06,  2.02it/s]
 89%|████████▊ | 9483/10707 [1:22:29<10:05,  2.02it/s]
 89%|████████▊ | 9484/10707 [1:22:30<10:05,  2.02it/s]
 89%|████████▊ | 9485/10707 [1:22:30<10:04,  2.02it/s]
 89%|████████▊ | 9486/10707 [1:22:31<10:03,  2.02it/s]
 89%|████████▊ | 9487/10707 [1:22:31<10:03,  2.02it/s]
 89%|████████▊ | 9488/10707 [1:22:32<10:02,  2.02it/s]
 89%|████████▊ | 9489/10707 [1:22:32<10:02,  2.02it/s]
 89%|████████▊ | 9490/10707 [1:22:33<10:02,  2.02it/s]
 89%|████████▊ | 9491/10707 [1:22:33<10:02,  2.02it/s]
 89%|████████▊ | 9492/10707 [1:22:34<10:02,  2.02it/s]
 89%|████████▊ | 9493/10707 [1:22:34<10:02,  2.02it/s]
 89%|████████▊ | 9494/10707 [1:22:35<10:01,  2.02it/s]
 89%|████████▊ | 9495/10707 [1:22:35<10:01,  2.02it/s]
 89%|████████▊ | 9496/10707 [1:22:36<10:00,  2.02it/s]
 89%|████████▊ | 9497/10707 [1:22:36<09:59,  2.02it/s]
 89%|████████▊ | 9498/10707 [1:22:37<09:59,  2.02it/s]
 89%|████████▊ | 9499/10707 [1:22:37<09:58,  2.02it/s]
 89%|████████▊ | 9500/10707 [1:22:38<09:57,  2.02it/s]{'loss': 3.5864, 'grad_norm': 0.19720493257045746, 'learning_rate': 3.8216343407713075e-05, 'epoch': 0.89}
-
                                                      

 89%|████████▊ | 9500/10707 [1:22:38<09:57,  2.02it/s]
 89%|████████▊ | 9501/10707 [1:22:38<09:57,  2.02it/s]
 89%|████████▊ | 9502/10707 [1:22:39<09:57,  2.02it/s]
 89%|████████▉ | 9503/10707 [1:22:39<09:56,  2.02it/s]
 89%|████████▉ | 9504/10707 [1:22:40<09:56,  2.02it/s]
 89%|████████▉ | 9505/10707 [1:22:40<09:55,  2.02it/s]
 89%|████████▉ | 9506/10707 [1:22:41<09:54,  2.02it/s]
 89%|████████▉ | 9507/10707 [1:22:41<09:53,  2.02it/s]
 89%|████████▉ | 9508/10707 [1:22:42<09:53,  2.02it/s]
 89%|████████▉ | 9509/10707 [1:22:42<09:52,  2.02it/s]
 89%|████████▉ | 9510/10707 [1:22:43<09:52,  2.02it/s]
 89%|████████▉ | 9511/10707 [1:22:43<09:52,  2.02it/s]
 89%|████████▉ | 9512/10707 [1:22:44<09:52,  2.02it/s]
 89%|████████▉ | 9513/10707 [1:22:44<09:51,  2.02it/s]
 89%|████████▉ | 9514/10707 [1:22:45<09:50,  2.02it/s]
 89%|████████▉ | 9515/10707 [1:22:45<09:50,  2.02it/s]
 89%|████████▉ | 9516/10707 [1:22:46<09:49,  2.02it/s]
 89%|████████▉ | 9517/10707 [1:22:46<09:49,  2.02it/s]
 89%|████████▉ | 9518/10707 [1:22:47<09:48,  2.02it/s]
 89%|████████▉ | 9519/10707 [1:22:47<09:48,  2.02it/s]
 89%|████████▉ | 9520/10707 [1:22:48<09:47,  2.02it/s]
 89%|████████▉ | 9521/10707 [1:22:48<09:47,  2.02it/s]
 89%|████████▉ | 9522/10707 [1:22:49<09:46,  2.02it/s]
 89%|████████▉ | 9523/10707 [1:22:49<09:45,  2.02it/s]
 89%|████████▉ | 9524/10707 [1:22:50<09:45,  2.02it/s]
 89%|████████▉ | 9525/10707 [1:22:50<09:44,  2.02it/s]{'loss': 3.5803, 'grad_norm': 0.19633334875106812, 'learning_rate': 3.6669068844780826e-05, 'epoch': 0.89}
-
                                                      

 89%|████████▉ | 9525/10707 [1:22:50<09:44,  2.02it/s]
 89%|████████▉ | 9526/10707 [1:22:51<09:45,  2.02it/s]
 89%|████████▉ | 9527/10707 [1:22:51<09:45,  2.01it/s]
 89%|████████▉ | 9528/10707 [1:22:52<09:44,  2.02it/s]
 89%|████████▉ | 9529/10707 [1:22:52<09:44,  2.02it/s]
 89%|████████▉ | 9530/10707 [1:22:53<09:43,  2.02it/s]
 89%|████████▉ | 9531/10707 [1:22:53<09:42,  2.02it/s]
 89%|████████▉ | 9532/10707 [1:22:54<09:41,  2.02it/s]
 89%|████████▉ | 9533/10707 [1:22:54<09:41,  2.02it/s]
 89%|████████▉ | 9534/10707 [1:22:55<09:40,  2.02it/s]
 89%|████████▉ | 9535/10707 [1:22:55<09:40,  2.02it/s]
 89%|████████▉ | 9536/10707 [1:22:56<09:40,  2.02it/s]
 89%|████████▉ | 9537/10707 [1:22:56<09:39,  2.02it/s]
 89%|████████▉ | 9538/10707 [1:22:57<09:38,  2.02it/s]
 89%|████████▉ | 9539/10707 [1:22:57<09:37,  2.02it/s]
 89%|████████▉ | 9540/10707 [1:22:58<09:37,  2.02it/s]
 89%|████████▉ | 9541/10707 [1:22:58<09:36,  2.02it/s]
 89%|████████▉ | 9542/10707 [1:22:59<09:36,  2.02it/s]
 89%|████████▉ | 9543/10707 [1:22:59<09:35,  2.02it/s]
 89%|████████▉ | 9544/10707 [1:23:00<09:35,  2.02it/s]
 89%|████████▉ | 9545/10707 [1:23:00<09:34,  2.02it/s]
 89%|████████▉ | 9546/10707 [1:23:01<09:34,  2.02it/s]
 89%|████████▉ | 9547/10707 [1:23:01<09:33,  2.02it/s]
 89%|████████▉ | 9548/10707 [1:23:02<09:33,  2.02it/s]
 89%|████████▉ | 9549/10707 [1:23:02<09:33,  2.02it/s]
 89%|████████▉ | 9550/10707 [1:23:03<09:33,  2.02it/s]{'loss': 3.5907, 'grad_norm': 0.19249987602233887, 'learning_rate': 3.515257473928302e-05, 'epoch': 0.89}
                                                      
-
 89%|████████▉ | 9550/10707 [1:23:03<09:33,  2.02it/s]
 89%|████████▉ | 9551/10707 [1:23:03<09:33,  2.01it/s]
 89%|████████▉ | 9552/10707 [1:23:04<09:33,  2.01it/s]
 89%|████████▉ | 9553/10707 [1:23:04<09:32,  2.02it/s]
 89%|████████▉ | 9554/10707 [1:23:05<09:31,  2.02it/s]
 89%|████████▉ | 9555/10707 [1:23:05<09:31,  2.02it/s]
 89%|████████▉ | 9556/10707 [1:23:06<09:29,  2.02it/s]
 89%|████████▉ | 9557/10707 [1:23:06<09:29,  2.02it/s]
 89%|████████▉ | 9558/10707 [1:23:07<09:28,  2.02it/s]
 89%|████████▉ | 9559/10707 [1:23:07<09:28,  2.02it/s]
 89%|████████▉ | 9560/10707 [1:23:08<09:27,  2.02it/s]
 89%|████████▉ | 9561/10707 [1:23:08<09:27,  2.02it/s]
 89%|████████▉ | 9562/10707 [1:23:09<09:26,  2.02it/s]
 89%|████████▉ | 9563/10707 [1:23:09<09:25,  2.02it/s]
 89%|████████▉ | 9564/10707 [1:23:10<09:25,  2.02it/s]
 89%|████████▉ | 9565/10707 [1:23:10<09:25,  2.02it/s]
 89%|████████▉ | 9566/10707 [1:23:11<09:25,  2.02it/s]
 89%|████████▉ | 9567/10707 [1:23:11<09:24,  2.02it/s]
 89%|████████▉ | 9568/10707 [1:23:12<09:23,  2.02it/s]
 89%|████████▉ | 9569/10707 [1:23:12<09:23,  2.02it/s]
 89%|████████▉ | 9570/10707 [1:23:13<09:22,  2.02it/s]
 89%|████████▉ | 9571/10707 [1:23:13<09:21,  2.02it/s]
 89%|████████▉ | 9572/10707 [1:23:14<09:21,  2.02it/s]
 89%|████████▉ | 9573/10707 [1:23:14<09:20,  2.02it/s]
 89%|████████▉ | 9574/10707 [1:23:15<09:20,  2.02it/s]
 89%|████████▉ | 9575/10707 [1:23:15<09:19,  2.02it/s]{'loss': 3.5891, 'grad_norm': 0.19398942589759827, 'learning_rate': 3.366696183645196e-05, 'epoch': 0.89}
                                                      
-
 89%|████████▉ | 9575/10707 [1:23:15<09:19,  2.02it/s]
 89%|████████▉ | 9576/10707 [1:23:16<09:20,  2.02it/s]
 89%|████████▉ | 9577/10707 [1:23:16<09:19,  2.02it/s]
 89%|████████▉ | 9578/10707 [1:23:16<09:19,  2.02it/s]
 89%|████████▉ | 9579/10707 [1:23:17<09:19,  2.02it/s]
 89%|████████▉ | 9580/10707 [1:23:17<09:18,  2.02it/s]
 89%|████████▉ | 9581/10707 [1:23:18<09:18,  2.02it/s]
 89%|████████▉ | 9582/10707 [1:23:18<09:17,  2.02it/s]
 90%|████████▉ | 9583/10707 [1:23:19<09:16,  2.02it/s]
 90%|████████▉ | 9584/10707 [1:23:19<09:15,  2.02it/s]
 90%|████████▉ | 9585/10707 [1:23:20<09:15,  2.02it/s]
 90%|████████▉ | 9586/10707 [1:23:20<09:14,  2.02it/s]
 90%|████████▉ | 9587/10707 [1:23:21<09:14,  2.02it/s]
 90%|████████▉ | 9588/10707 [1:23:21<09:13,  2.02it/s]
 90%|████████▉ | 9589/10707 [1:23:22<09:13,  2.02it/s]
 90%|████████▉ | 9590/10707 [1:23:22<09:12,  2.02it/s]
 90%|████████▉ | 9591/10707 [1:23:23<09:12,  2.02it/s]
 90%|████████▉ | 9592/10707 [1:23:23<09:11,  2.02it/s]
 90%|████████▉ | 9593/10707 [1:23:24<09:11,  2.02it/s]
 90%|████████▉ | 9594/10707 [1:23:24<09:10,  2.02it/s]
 90%|████████▉ | 9595/10707 [1:23:25<09:10,  2.02it/s]
 90%|████████▉ | 9596/10707 [1:23:25<09:09,  2.02it/s]
 90%|████████▉ | 9597/10707 [1:23:26<09:09,  2.02it/s]
 90%|████████▉ | 9598/10707 [1:23:26<09:08,  2.02it/s]
 90%|████████▉ | 9599/10707 [1:23:27<09:08,  2.02it/s]
 90%|████████▉ | 9600/10707 [1:23:27<09:07,  2.02it/s]
                                                      
{'loss': 3.5815, 'grad_norm': 0.1906517744064331, 'learning_rate': 3.221232882998931e-05, 'epoch': 0.9}
-
 90%|████████▉ | 9600/10707 [1:23:27<09:07,  2.02it/s]
 90%|████████▉ | 9601/10707 [1:23:28<09:08,  2.02it/s]
 90%|████████▉ | 9602/10707 [1:23:28<09:07,  2.02it/s]
 90%|████████▉ | 9603/10707 [1:23:29<09:07,  2.02it/s]
 90%|████████▉ | 9604/10707 [1:23:29<09:06,  2.02it/s]
 90%|████████▉ | 9605/10707 [1:23:30<09:05,  2.02it/s]
 90%|████████▉ | 9606/10707 [1:23:30<09:05,  2.02it/s]
 90%|████████▉ | 9607/10707 [1:23:31<09:04,  2.02it/s]
 90%|████████▉ | 9608/10707 [1:23:31<09:04,  2.02it/s]
 90%|████████▉ | 9609/10707 [1:23:32<09:03,  2.02it/s]
 90%|████████▉ | 9610/10707 [1:23:32<09:02,  2.02it/s]
 90%|████████▉ | 9611/10707 [1:23:33<09:02,  2.02it/s]
 90%|████████▉ | 9612/10707 [1:23:33<09:03,  2.02it/s]
 90%|████████▉ | 9613/10707 [1:23:34<09:03,  2.01it/s]
 90%|████████▉ | 9614/10707 [1:23:34<09:01,  2.02it/s]
 90%|████████▉ | 9615/10707 [1:23:35<09:01,  2.02it/s]
 90%|████████▉ | 9616/10707 [1:23:35<09:00,  2.02it/s]
 90%|████████▉ | 9617/10707 [1:23:36<09:00,  2.02it/s]
 90%|████████▉ | 9618/10707 [1:23:36<08:59,  2.02it/s]
 90%|████████▉ | 9619/10707 [1:23:37<08:58,  2.02it/s]
 90%|████████▉ | 9620/10707 [1:23:37<08:57,  2.02it/s]
 90%|████████▉ | 9621/10707 [1:23:38<08:57,  2.02it/s]
 90%|████████▉ | 9622/10707 [1:23:38<08:56,  2.02it/s]
 90%|████████▉ | 9623/10707 [1:23:39<08:56,  2.02it/s]
 90%|████████▉ | 9624/10707 [1:23:39<08:55,  2.02it/s]
 90%|████████▉ | 9625/10707 [1:23:40<08:55,  2.02it/s]{'loss': 3.5844, 'grad_norm': 0.19804587960243225, 'learning_rate': 3.07887723555097e-05, 'epoch': 0.9}
-
                                                      

 90%|████████▉ | 9625/10707 [1:23:40<08:55,  2.02it/s]
 90%|████████▉ | 9626/10707 [1:23:40<08:55,  2.02it/s]
 90%|████████▉ | 9627/10707 [1:23:41<08:54,  2.02it/s]
 90%|████████▉ | 9628/10707 [1:23:41<08:53,  2.02it/s]
 90%|████████▉ | 9629/10707 [1:23:42<08:53,  2.02it/s]
 90%|████████▉ | 9630/10707 [1:23:42<08:52,  2.02it/s]
 90%|████████▉ | 9631/10707 [1:23:43<08:52,  2.02it/s]
 90%|████████▉ | 9632/10707 [1:23:43<08:51,  2.02it/s]
 90%|████████▉ | 9633/10707 [1:23:44<08:51,  2.02it/s]
 90%|████████▉ | 9634/10707 [1:23:44<08:50,  2.02it/s]
 90%|████████▉ | 9635/10707 [1:23:45<08:50,  2.02it/s]
 90%|████████▉ | 9636/10707 [1:23:45<08:49,  2.02it/s]
 90%|█████████ | 9637/10707 [1:23:46<08:49,  2.02it/s]
 90%|█████████ | 9638/10707 [1:23:46<08:48,  2.02it/s]
 90%|█████████ | 9639/10707 [1:23:47<08:48,  2.02it/s]
 90%|█████████ | 9640/10707 [1:23:47<08:47,  2.02it/s]
 90%|█████████ | 9641/10707 [1:23:48<08:47,  2.02it/s]
 90%|█████████ | 9642/10707 [1:23:48<08:46,  2.02it/s]
 90%|█████████ | 9643/10707 [1:23:49<08:46,  2.02it/s]
 90%|█████████ | 9644/10707 [1:23:49<08:46,  2.02it/s]
 90%|█████████ | 9645/10707 [1:23:50<08:46,  2.02it/s]
 90%|█████████ | 9646/10707 [1:23:50<08:45,  2.02it/s]
 90%|█████████ | 9647/10707 [1:23:51<08:45,  2.02it/s]
 90%|█████████ | 9648/10707 [1:23:51<08:44,  2.02it/s]
 90%|█████████ | 9649/10707 [1:23:52<08:44,  2.02it/s]
 90%|█████████ | 9650/10707 [1:23:52<08:44,  2.02it/s]
                                                      {'loss': 3.5789, 'grad_norm': 0.19353294372558594, 'learning_rate': 2.9396386984121005e-05, 'epoch': 0.9}
-

 90%|█████████ | 9650/10707 [1:23:52<08:44,  2.02it/s]
 90%|█████████ | 9651/10707 [1:23:53<08:44,  2.01it/s]
 90%|█████████ | 9652/10707 [1:23:53<10:06,  1.74it/s]
 90%|█████████ | 9653/10707 [1:23:54<09:41,  1.81it/s]
 90%|█████████ | 9654/10707 [1:23:54<09:23,  1.87it/s]
 90%|█████████ | 9655/10707 [1:23:55<09:09,  1.91it/s]
 90%|█████████ | 9656/10707 [1:23:55<09:00,  1.94it/s]
 90%|█████████ | 9657/10707 [1:23:56<08:53,  1.97it/s]
 90%|█████████ | 9658/10707 [1:23:56<08:49,  1.98it/s]
 90%|█████████ | 9659/10707 [1:23:57<08:45,  1.99it/s]
 90%|█████████ | 9660/10707 [1:23:57<08:42,  2.00it/s]
 90%|█████████ | 9661/10707 [1:23:58<08:41,  2.01it/s]
 90%|█████████ | 9662/10707 [1:23:58<08:39,  2.01it/s]
 90%|█████████ | 9663/10707 [1:23:59<08:38,  2.01it/s]
 90%|█████████ | 9664/10707 [1:23:59<08:37,  2.02it/s]
 90%|█████████ | 9665/10707 [1:24:00<08:36,  2.02it/s]
 90%|█████████ | 9666/10707 [1:24:00<08:35,  2.02it/s]
 90%|█████████ | 9667/10707 [1:24:01<08:35,  2.02it/s]
 90%|█████████ | 9668/10707 [1:24:01<08:34,  2.02it/s]
 90%|█████████ | 9669/10707 [1:24:02<08:33,  2.02it/s]
 90%|█████████ | 9670/10707 [1:24:02<08:33,  2.02it/s]
 90%|█████████ | 9671/10707 [1:24:03<08:32,  2.02it/s]
 90%|█████████ | 9672/10707 [1:24:03<08:32,  2.02it/s]
 90%|█████████ | 9673/10707 [1:24:04<08:32,  2.02it/s]
 90%|█████████ | 9674/10707 [1:24:04<08:31,  2.02it/s]
 90%|█████████ | 9675/10707 [1:24:05<08:31,  2.02it/s]{'loss': 3.579, 'grad_norm': 0.1903531402349472, 'learning_rate': 2.8035265216141802e-05, 'epoch': 0.9}
-
                                                      

 90%|█████████ | 9675/10707 [1:24:05<08:31,  2.02it/s]
 90%|█████████ | 9676/10707 [1:24:05<08:31,  2.02it/s]
 90%|█████████ | 9677/10707 [1:24:06<08:30,  2.02it/s]
 90%|█████████ | 9678/10707 [1:24:06<08:30,  2.02it/s]
 90%|█████████ | 9679/10707 [1:24:07<08:29,  2.02it/s]
 90%|█████████ | 9680/10707 [1:24:07<08:28,  2.02it/s]
 90%|█████████ | 9681/10707 [1:24:08<08:27,  2.02it/s]
 90%|█████████ | 9682/10707 [1:24:08<08:27,  2.02it/s]
 90%|█████████ | 9683/10707 [1:24:09<08:26,  2.02it/s]
 90%|█████████ | 9684/10707 [1:24:09<08:26,  2.02it/s]
 90%|█████████ | 9685/10707 [1:24:10<08:25,  2.02it/s]
 90%|█████████ | 9686/10707 [1:24:10<08:24,  2.02it/s]
 90%|█████████ | 9687/10707 [1:24:11<08:24,  2.02it/s]
 90%|█████████ | 9688/10707 [1:24:11<08:24,  2.02it/s]
 90%|█████████ | 9689/10707 [1:24:12<08:23,  2.02it/s]
 91%|█████████ | 9690/10707 [1:24:12<08:23,  2.02it/s]
 91%|█████████ | 9691/10707 [1:24:13<08:22,  2.02it/s]
 91%|█████████ | 9692/10707 [1:24:13<08:22,  2.02it/s]
 91%|█████████ | 9693/10707 [1:24:14<08:21,  2.02it/s]
 91%|█████████ | 9694/10707 [1:24:14<09:40,  1.74it/s]
 91%|█████████ | 9695/10707 [1:24:15<09:16,  1.82it/s]
 91%|█████████ | 9696/10707 [1:24:15<08:59,  1.87it/s]
 91%|█████████ | 9697/10707 [1:24:16<08:46,  1.92it/s]
 91%|█████████ | 9698/10707 [1:24:16<08:38,  1.95it/s]
 91%|█████████ | 9699/10707 [1:24:17<08:31,  1.97it/s]
 91%|█████████ | 9700/10707 [1:24:17<08:27,  1.98it/s]{'loss': 3.5848, 'grad_norm': 0.19306747615337372, 'learning_rate': 2.6705497474955763e-05, 'epoch': 0.91}
-
                                                      

 91%|█████████ | 9700/10707 [1:24:17<08:27,  1.98it/s]
 91%|█████████ | 9701/10707 [1:24:18<08:24,  1.99it/s]
 91%|█████████ | 9702/10707 [1:24:18<08:22,  2.00it/s]
 91%|█████████ | 9703/10707 [1:24:19<08:20,  2.01it/s]
 91%|█████████ | 9704/10707 [1:24:19<08:18,  2.01it/s]
 91%|█████████ | 9705/10707 [1:24:20<08:17,  2.01it/s]
 91%|█████████ | 9706/10707 [1:24:20<08:16,  2.01it/s]
 91%|█████████ | 9707/10707 [1:24:21<08:15,  2.02it/s]
 91%|█████████ | 9708/10707 [1:24:21<08:15,  2.02it/s]
 91%|█████████ | 9709/10707 [1:24:22<08:14,  2.02it/s]
 91%|█████████ | 9710/10707 [1:24:22<08:14,  2.02it/s]
 91%|█████████ | 9711/10707 [1:24:23<08:13,  2.02it/s]
 91%|█████████ | 9712/10707 [1:24:23<08:12,  2.02it/s]
 91%|█████████ | 9713/10707 [1:24:24<08:12,  2.02it/s]
 91%|█████████ | 9714/10707 [1:24:24<08:12,  2.02it/s]
 91%|█████████ | 9715/10707 [1:24:25<08:11,  2.02it/s]
 91%|█████████ | 9716/10707 [1:24:25<08:10,  2.02it/s]
 91%|█████████ | 9717/10707 [1:24:26<08:10,  2.02it/s]
 91%|█████████ | 9718/10707 [1:24:26<08:09,  2.02it/s]
 91%|█████████ | 9719/10707 [1:24:27<08:09,  2.02it/s]
 91%|█████████ | 9720/10707 [1:24:27<08:08,  2.02it/s]
 91%|█████████ | 9721/10707 [1:24:28<08:07,  2.02it/s]
 91%|█████████ | 9722/10707 [1:24:28<08:07,  2.02it/s]
 91%|█████████ | 9723/10707 [1:24:29<08:07,  2.02it/s]
 91%|█████████ | 9724/10707 [1:24:29<08:06,  2.02it/s]
 91%|█████████ | 9725/10707 [1:24:30<08:06,  2.02it/s]{'loss': 3.5814, 'grad_norm': 0.19483162462711334, 'learning_rate': 2.5407172101005182e-05, 'epoch': 0.91}
-
                                                      

 91%|█████████ | 9725/10707 [1:24:30<08:06,  2.02it/s]
 91%|█████████ | 9726/10707 [1:24:30<08:06,  2.02it/s]
 91%|█████████ | 9727/10707 [1:24:31<08:05,  2.02it/s]
 91%|█████████ | 9728/10707 [1:24:31<08:04,  2.02it/s]
 91%|█████████ | 9729/10707 [1:24:32<08:04,  2.02it/s]
 91%|█████████ | 9730/10707 [1:24:32<08:03,  2.02it/s]
 91%|█████████ | 9731/10707 [1:24:33<08:03,  2.02it/s]
 91%|█████████ | 9732/10707 [1:24:33<08:03,  2.02it/s]
 91%|█████████ | 9733/10707 [1:24:34<08:03,  2.02it/s]
 91%|█████████ | 9734/10707 [1:24:34<08:02,  2.01it/s]
 91%|█████████ | 9735/10707 [1:24:35<08:02,  2.02it/s]
 91%|█████████ | 9736/10707 [1:24:35<08:01,  2.02it/s]
 91%|█████████ | 9737/10707 [1:24:36<08:00,  2.02it/s]
 91%|█████████ | 9738/10707 [1:24:36<07:59,  2.02it/s]
 91%|█████████ | 9739/10707 [1:24:37<07:59,  2.02it/s]
 91%|█████████ | 9740/10707 [1:24:37<07:58,  2.02it/s]
 91%|█████████ | 9741/10707 [1:24:38<07:58,  2.02it/s]
 91%|█████████ | 9742/10707 [1:24:38<07:57,  2.02it/s]
 91%|█████████ | 9743/10707 [1:24:39<07:56,  2.02it/s]
 91%|█████████ | 9744/10707 [1:24:39<07:56,  2.02it/s]
 91%|█████████ | 9745/10707 [1:24:40<07:55,  2.02it/s]
 91%|█████████ | 9746/10707 [1:24:40<07:55,  2.02it/s]
 91%|█████████ | 9747/10707 [1:24:41<07:55,  2.02it/s]
 91%|█████████ | 9748/10707 [1:24:41<07:54,  2.02it/s]
 91%|█████████ | 9749/10707 [1:24:42<07:53,  2.02it/s]
 91%|█████████ | 9750/10707 [1:24:42<07:53,  2.02it/s]
                                                      {'loss': 3.5798, 'grad_norm': 0.19022154808044434, 'learning_rate': 2.4140375345921895e-05, 'epoch': 0.91}
-

 91%|█████████ | 9750/10707 [1:24:42<07:53,  2.02it/s]
 91%|█████████ | 9751/10707 [1:24:43<07:54,  2.01it/s]
 91%|█████████ | 9752/10707 [1:24:43<07:53,  2.02it/s]
 91%|█████████ | 9753/10707 [1:24:44<07:53,  2.02it/s]
 91%|█████████ | 9754/10707 [1:24:44<07:52,  2.02it/s]
 91%|█████████ | 9755/10707 [1:24:45<07:51,  2.02it/s]
 91%|█████████ | 9756/10707 [1:24:45<07:50,  2.02it/s]
 91%|█████████ | 9757/10707 [1:24:46<07:50,  2.02it/s]
 91%|█████████ | 9758/10707 [1:24:46<07:50,  2.02it/s]
 91%|█████████ | 9759/10707 [1:24:47<07:49,  2.02it/s]
 91%|█████████ | 9760/10707 [1:24:47<07:48,  2.02it/s]
 91%|█████████ | 9761/10707 [1:24:48<07:48,  2.02it/s]
 91%|█████████ | 9762/10707 [1:24:48<07:48,  2.02it/s]
 91%|█████████ | 9763/10707 [1:24:49<07:47,  2.02it/s]
 91%|█████████ | 9764/10707 [1:24:49<07:46,  2.02it/s]
 91%|█████████ | 9765/10707 [1:24:50<07:46,  2.02it/s]
 91%|█████████ | 9766/10707 [1:24:50<07:45,  2.02it/s]
 91%|█████████ | 9767/10707 [1:24:51<07:45,  2.02it/s]
 91%|█████████ | 9768/10707 [1:24:51<07:44,  2.02it/s]
 91%|█████████ | 9769/10707 [1:24:52<07:43,  2.02it/s]
 91%|█████████ | 9770/10707 [1:24:52<07:42,  2.02it/s]
 91%|█████████▏| 9771/10707 [1:24:53<07:42,  2.02it/s]
 91%|█████████▏| 9772/10707 [1:24:53<07:42,  2.02it/s]
 91%|█████████▏| 9773/10707 [1:24:54<07:43,  2.02it/s]
 91%|█████████▏| 9774/10707 [1:24:54<07:42,  2.02it/s]
 91%|█████████▏| 9775/10707 [1:24:55<07:42,  2.02it/s]{'loss': 3.5809, 'grad_norm': 0.19282498955726624, 'learning_rate': 2.290519136679753e-05, 'epoch': 0.91}
-
                                                      

 91%|█████████▏| 9775/10707 [1:24:55<07:42,  2.02it/s]
 91%|█████████▏| 9776/10707 [1:24:55<07:42,  2.01it/s]
 91%|█████████▏| 9777/10707 [1:24:56<07:41,  2.01it/s]
 91%|█████████▏| 9778/10707 [1:24:56<07:41,  2.01it/s]
 91%|█████████▏| 9779/10707 [1:24:57<07:40,  2.01it/s]
 91%|█████████▏| 9780/10707 [1:24:57<07:40,  2.01it/s]
 91%|█████████▏| 9781/10707 [1:24:58<07:39,  2.02it/s]
 91%|█████████▏| 9782/10707 [1:24:58<07:38,  2.02it/s]
 91%|█████████▏| 9783/10707 [1:24:59<07:37,  2.02it/s]
 91%|█████████▏| 9784/10707 [1:24:59<07:37,  2.02it/s]
 91%|█████████▏| 9785/10707 [1:25:00<07:36,  2.02it/s]
 91%|█████████▏| 9786/10707 [1:25:00<07:36,  2.02it/s]
 91%|█████████▏| 9787/10707 [1:25:01<07:35,  2.02it/s]
 91%|█████████▏| 9788/10707 [1:25:01<07:35,  2.02it/s]
 91%|█████████▏| 9789/10707 [1:25:02<07:34,  2.02it/s]
 91%|█████████▏| 9790/10707 [1:25:02<07:33,  2.02it/s]
 91%|█████████▏| 9791/10707 [1:25:02<07:33,  2.02it/s]
 91%|█████████▏| 9792/10707 [1:25:03<07:33,  2.02it/s]
 91%|█████████▏| 9793/10707 [1:25:03<07:34,  2.01it/s]
 91%|█████████▏| 9794/10707 [1:25:04<07:33,  2.01it/s]
 91%|█████████▏| 9795/10707 [1:25:04<07:32,  2.02it/s]
 91%|█████████▏| 9796/10707 [1:25:05<07:31,  2.02it/s]
 92%|█████████▏| 9797/10707 [1:25:05<07:31,  2.02it/s]
 92%|█████████▏| 9798/10707 [1:25:06<07:30,  2.02it/s]
 92%|█████████▏| 9799/10707 [1:25:06<07:29,  2.02it/s]
 92%|█████████▏| 9800/10707 [1:25:07<07:28,  2.02it/s]{'loss': 3.585, 'grad_norm': 0.1930411010980606, 'learning_rate': 2.170170222059248e-05, 'epoch': 0.92}
-
                                                      

 92%|█████████▏| 9800/10707 [1:25:07<07:28,  2.02it/s]
 92%|█████████▏| 9801/10707 [1:25:07<07:29,  2.02it/s]
 92%|█████████▏| 9802/10707 [1:25:08<07:27,  2.02it/s]
 92%|█████████▏| 9803/10707 [1:25:08<07:27,  2.02it/s]
 92%|█████████▏| 9804/10707 [1:25:09<07:26,  2.02it/s]
 92%|█████████▏| 9805/10707 [1:25:09<07:26,  2.02it/s]
 92%|█████████▏| 9806/10707 [1:25:10<07:25,  2.02it/s]
 92%|█████████▏| 9807/10707 [1:25:10<07:25,  2.02it/s]
 92%|█████████▏| 9808/10707 [1:25:11<07:24,  2.02it/s]
 92%|█████████▏| 9809/10707 [1:25:11<07:24,  2.02it/s]
 92%|█████████▏| 9810/10707 [1:25:12<07:23,  2.02it/s]
 92%|█████████▏| 9811/10707 [1:25:12<07:23,  2.02it/s]
 92%|█████████▏| 9812/10707 [1:25:13<07:22,  2.02it/s]
 92%|█████████▏| 9813/10707 [1:25:13<07:22,  2.02it/s]
 92%|█████████▏| 9814/10707 [1:25:14<07:21,  2.02it/s]
 92%|█████████▏| 9815/10707 [1:25:14<07:21,  2.02it/s]
 92%|█████████▏| 9816/10707 [1:25:15<07:20,  2.02it/s]
 92%|█████████▏| 9817/10707 [1:25:15<07:20,  2.02it/s]
 92%|█████████▏| 9818/10707 [1:25:16<07:19,  2.02it/s]
 92%|█████████▏| 9819/10707 [1:25:16<07:19,  2.02it/s]
 92%|█████████▏| 9820/10707 [1:25:17<07:18,  2.02it/s]
 92%|█████████▏| 9821/10707 [1:25:17<07:18,  2.02it/s]
 92%|█████████▏| 9822/10707 [1:25:18<07:17,  2.02it/s]
 92%|█████████▏| 9823/10707 [1:25:18<07:17,  2.02it/s]
 92%|█████████▏| 9824/10707 [1:25:19<07:16,  2.02it/s]
 92%|█████████▏| 9825/10707 [1:25:19<07:16,  2.02it/s]{'loss': 3.5757, 'grad_norm': 0.18975821137428284, 'learning_rate': 2.0529987858684762e-05, 'epoch': 0.92}
                                                      
-
 92%|█████████▏| 9825/10707 [1:25:19<07:16,  2.02it/s]
 92%|█████████▏| 9826/10707 [1:25:20<07:17,  2.01it/s]
 92%|█████████▏| 9827/10707 [1:25:20<07:16,  2.01it/s]
 92%|█████████▏| 9828/10707 [1:25:21<07:16,  2.01it/s]
 92%|█████████▏| 9829/10707 [1:25:21<07:15,  2.01it/s]
 92%|█████████▏| 9830/10707 [1:25:22<07:15,  2.02it/s]
 92%|█████████▏| 9831/10707 [1:25:22<07:15,  2.01it/s]
 92%|█████████▏| 9832/10707 [1:25:23<07:13,  2.02it/s]
 92%|█████████▏| 9833/10707 [1:25:23<07:13,  2.02it/s]
 92%|█████████▏| 9834/10707 [1:25:24<07:12,  2.02it/s]
 92%|█████████▏| 9835/10707 [1:25:24<07:12,  2.02it/s]
 92%|█████████▏| 9836/10707 [1:25:25<07:12,  2.02it/s]
 92%|█████████▏| 9837/10707 [1:25:25<07:11,  2.02it/s]
 92%|█████████▏| 9838/10707 [1:25:26<07:10,  2.02it/s]
 92%|█████████▏| 9839/10707 [1:25:26<07:10,  2.02it/s]
 92%|█████████▏| 9840/10707 [1:25:27<07:09,  2.02it/s]
 92%|█████████▏| 9841/10707 [1:25:27<07:08,  2.02it/s]
 92%|█████████▏| 9842/10707 [1:25:28<07:07,  2.02it/s]
 92%|█████████▏| 9843/10707 [1:25:28<07:07,  2.02it/s]
 92%|█████████▏| 9844/10707 [1:25:29<07:06,  2.02it/s]
 92%|█████████▏| 9845/10707 [1:25:29<07:06,  2.02it/s]
 92%|█████████▏| 9846/10707 [1:25:30<07:05,  2.02it/s]
 92%|█████████▏| 9847/10707 [1:25:30<07:05,  2.02it/s]
 92%|█████████▏| 9848/10707 [1:25:31<07:04,  2.02it/s]
 92%|█████████▏| 9849/10707 [1:25:31<07:04,  2.02it/s]
 92%|█████████▏| 9850/10707 [1:25:32<07:04,  2.02it/s]{'loss': 3.5731, 'grad_norm': 0.18976882100105286, 'learning_rate': 1.9390126121558715e-05, 'epoch': 0.92}
-
                                                      

 92%|█████████▏| 9850/10707 [1:25:32<07:04,  2.02it/s]
 92%|█████████▏| 9851/10707 [1:25:32<07:04,  2.02it/s]
 92%|█████████▏| 9852/10707 [1:25:33<07:03,  2.02it/s]
 92%|█████████▏| 9853/10707 [1:25:33<07:04,  2.01it/s]
 92%|█████████▏| 9854/10707 [1:25:34<07:03,  2.01it/s]
 92%|█████████▏| 9855/10707 [1:25:34<07:02,  2.02it/s]
 92%|█████████▏| 9856/10707 [1:25:35<07:01,  2.02it/s]
 92%|█████████▏| 9857/10707 [1:25:35<07:01,  2.02it/s]
 92%|█████████▏| 9858/10707 [1:25:36<07:00,  2.02it/s]
 92%|█████████▏| 9859/10707 [1:25:36<06:59,  2.02it/s]
 92%|█████████▏| 9860/10707 [1:25:37<06:59,  2.02it/s]
 92%|█████████▏| 9861/10707 [1:25:37<06:59,  2.02it/s]
 92%|█████████▏| 9862/10707 [1:25:38<06:58,  2.02it/s]
 92%|█████████▏| 9863/10707 [1:25:38<06:58,  2.02it/s]
 92%|█████████▏| 9864/10707 [1:25:39<06:57,  2.02it/s]
 92%|█████████▏| 9865/10707 [1:25:39<06:56,  2.02it/s]
 92%|█████████▏| 9866/10707 [1:25:40<06:55,  2.02it/s]
 92%|█████████▏| 9867/10707 [1:25:40<06:55,  2.02it/s]
 92%|█████████▏| 9868/10707 [1:25:41<06:54,  2.02it/s]
 92%|█████████▏| 9869/10707 [1:25:41<06:54,  2.02it/s]
 92%|█████████▏| 9870/10707 [1:25:42<06:53,  2.02it/s]
 92%|█████████▏| 9871/10707 [1:25:42<06:53,  2.02it/s]
 92%|█████████▏| 9872/10707 [1:25:43<06:52,  2.02it/s]
 92%|█████████▏| 9873/10707 [1:25:43<06:52,  2.02it/s]
 92%|█████████▏| 9874/10707 [1:25:44<06:52,  2.02it/s]
 92%|█████████▏| 9875/10707 [1:25:44<06:51,  2.02it/s]{'loss': 3.57, 'grad_norm': 0.19149377942085266, 'learning_rate': 1.8282192733633573e-05, 'epoch': 0.92}
                                                      
-
 92%|█████████▏| 9875/10707 [1:25:44<06:51,  2.02it/s]
 92%|█████████▏| 9876/10707 [1:25:45<06:51,  2.02it/s]
 92%|█████████▏| 9877/10707 [1:25:45<06:51,  2.02it/s]
 92%|█████████▏| 9878/10707 [1:25:46<06:50,  2.02it/s]
 92%|█████████▏| 9879/10707 [1:25:46<06:50,  2.02it/s]
 92%|█████████▏| 9880/10707 [1:25:47<06:49,  2.02it/s]
 92%|█████████▏| 9881/10707 [1:25:47<06:49,  2.02it/s]
 92%|█████████▏| 9882/10707 [1:25:48<06:48,  2.02it/s]
 92%|█████████▏| 9883/10707 [1:25:48<06:48,  2.02it/s]
 92%|█████████▏| 9884/10707 [1:25:49<06:47,  2.02it/s]
 92%|█████████▏| 9885/10707 [1:25:49<06:47,  2.02it/s]
 92%|█████████▏| 9886/10707 [1:25:50<06:46,  2.02it/s]
 92%|█████████▏| 9887/10707 [1:25:50<06:46,  2.02it/s]
 92%|█████████▏| 9888/10707 [1:25:51<06:46,  2.02it/s]
 92%|█████████▏| 9889/10707 [1:25:51<06:45,  2.02it/s]
 92%|█████████▏| 9890/10707 [1:25:52<06:44,  2.02it/s]
 92%|█████████▏| 9891/10707 [1:25:52<06:44,  2.02it/s]
 92%|█████████▏| 9892/10707 [1:25:53<06:43,  2.02it/s]
 92%|█████████▏| 9893/10707 [1:25:53<06:42,  2.02it/s]
 92%|█████████▏| 9894/10707 [1:25:54<06:42,  2.02it/s]
 92%|█████████▏| 9895/10707 [1:25:54<06:42,  2.02it/s]
 92%|█████████▏| 9896/10707 [1:25:54<06:41,  2.02it/s]
 92%|█████████▏| 9897/10707 [1:25:55<06:41,  2.02it/s]
 92%|█████████▏| 9898/10707 [1:25:55<06:40,  2.02it/s]
 92%|█████████▏| 9899/10707 [1:25:56<06:39,  2.02it/s]
 92%|█████████▏| 9900/10707 [1:25:56<06:39,  2.02it/s]{'loss': 3.5703, 'grad_norm': 0.19538216292858124, 'learning_rate': 1.7206261298232882e-05, 'epoch': 0.92}
-
                                                      

 92%|█████████▏| 9900/10707 [1:25:56<06:39,  2.02it/s]
 92%|█████████▏| 9901/10707 [1:25:57<06:39,  2.02it/s]
 92%|█████████▏| 9902/10707 [1:25:57<06:38,  2.02it/s]
 92%|█████████▏| 9903/10707 [1:25:58<06:38,  2.02it/s]
 93%|█████████▎| 9904/10707 [1:25:58<06:37,  2.02it/s]
 93%|█████████▎| 9905/10707 [1:25:59<06:37,  2.02it/s]
 93%|█████████▎| 9906/10707 [1:25:59<06:36,  2.02it/s]
 93%|█████████▎| 9907/10707 [1:26:00<06:36,  2.02it/s]
 93%|█████████▎| 9908/10707 [1:26:00<06:36,  2.02it/s]
 93%|█████████▎| 9909/10707 [1:26:01<06:35,  2.02it/s]
 93%|█████████▎| 9910/10707 [1:26:01<06:34,  2.02it/s]
 93%|█████████▎| 9911/10707 [1:26:02<06:33,  2.02it/s]
 93%|█████████▎| 9912/10707 [1:26:02<06:33,  2.02it/s]
 93%|█████████▎| 9913/10707 [1:26:03<06:32,  2.02it/s]
 93%|█████████▎| 9914/10707 [1:26:03<06:33,  2.02it/s]
 93%|█████████▎| 9915/10707 [1:26:04<06:33,  2.01it/s]
 93%|█████████▎| 9916/10707 [1:26:04<06:32,  2.02it/s]
 93%|█████████▎| 9917/10707 [1:26:05<06:31,  2.02it/s]
 93%|█████████▎| 9918/10707 [1:26:05<06:30,  2.02it/s]
 93%|█████████▎| 9919/10707 [1:26:06<06:30,  2.02it/s]
 93%|█████████▎| 9920/10707 [1:26:06<06:29,  2.02it/s]
 93%|█████████▎| 9921/10707 [1:26:07<06:29,  2.02it/s]
 93%|█████████▎| 9922/10707 [1:26:07<06:28,  2.02it/s]
 93%|█████████▎| 9923/10707 [1:26:08<06:28,  2.02it/s]
 93%|█████████▎| 9924/10707 [1:26:08<06:27,  2.02it/s]
 93%|█████████▎| 9925/10707 [1:26:09<06:26,  2.02it/s]{'loss': 3.5824, 'grad_norm': 0.19246040284633636, 'learning_rate': 1.616240329269514e-05, 'epoch': 0.93}
                                                      
-
 93%|█████████▎| 9925/10707 [1:26:09<06:26,  2.02it/s]
 93%|█████████▎| 9926/10707 [1:26:09<06:26,  2.02it/s]
 93%|█████████▎| 9927/10707 [1:26:10<06:26,  2.02it/s]
 93%|█████████▎| 9928/10707 [1:26:10<06:25,  2.02it/s]
 93%|█████████▎| 9929/10707 [1:26:11<06:25,  2.02it/s]
 93%|█████████▎| 9930/10707 [1:26:11<06:24,  2.02it/s]
 93%|█████████▎| 9931/10707 [1:26:12<06:23,  2.02it/s]
 93%|█████████▎| 9932/10707 [1:26:12<06:23,  2.02it/s]
 93%|█████████▎| 9933/10707 [1:26:13<06:23,  2.02it/s]
 93%|█████████▎| 9934/10707 [1:26:13<06:22,  2.02it/s]
 93%|█████████▎| 9935/10707 [1:26:14<06:21,  2.02it/s]
 93%|█████████▎| 9936/10707 [1:26:14<06:21,  2.02it/s]
 93%|█████████▎| 9937/10707 [1:26:15<06:20,  2.02it/s]
 93%|█████████▎| 9938/10707 [1:26:15<06:20,  2.02it/s]
 93%|█████████▎| 9939/10707 [1:26:16<06:19,  2.02it/s]
 93%|█████████▎| 9940/10707 [1:26:16<06:19,  2.02it/s]
 93%|█████████▎| 9941/10707 [1:26:17<06:19,  2.02it/s]
 93%|█████████▎| 9942/10707 [1:26:17<06:18,  2.02it/s]
 93%|█████████▎| 9943/10707 [1:26:18<06:18,  2.02it/s]
 93%|█████████▎| 9944/10707 [1:26:18<06:17,  2.02it/s]
 93%|█████████▎| 9945/10707 [1:26:19<06:17,  2.02it/s]
 93%|█████████▎| 9946/10707 [1:26:19<06:16,  2.02it/s]
 93%|█████████▎| 9947/10707 [1:26:20<06:17,  2.01it/s]
 93%|█████████▎| 9948/10707 [1:26:20<06:16,  2.02it/s]
 93%|█████████▎| 9949/10707 [1:26:21<06:15,  2.02it/s]
 93%|█████████▎| 9950/10707 [1:26:21<06:15,  2.02it/s]{'loss': 3.5747, 'grad_norm': 0.19063609838485718, 'learning_rate': 1.5150688063625018e-05, 'epoch': 0.93}
-
                                                      

 93%|█████████▎| 9950/10707 [1:26:21<06:15,  2.02it/s]
 93%|█████████▎| 9951/10707 [1:26:22<06:15,  2.01it/s]
 93%|█████████▎| 9952/10707 [1:26:22<06:14,  2.02it/s]
 93%|█████████▎| 9953/10707 [1:26:23<06:13,  2.02it/s]
 93%|█████████▎| 9954/10707 [1:26:23<06:13,  2.02it/s]
 93%|█████████▎| 9955/10707 [1:26:24<06:12,  2.02it/s]
 93%|█████████▎| 9956/10707 [1:26:24<06:12,  2.02it/s]
 93%|█████████▎| 9957/10707 [1:26:25<06:11,  2.02it/s]
 93%|█████████▎| 9958/10707 [1:26:25<06:10,  2.02it/s]
 93%|█████████▎| 9959/10707 [1:26:26<06:10,  2.02it/s]
 93%|█████████▎| 9960/10707 [1:26:26<06:09,  2.02it/s]
 93%|█████████▎| 9961/10707 [1:26:27<06:09,  2.02it/s]
 93%|█████████▎| 9962/10707 [1:26:27<06:08,  2.02it/s]
 93%|█████████▎| 9963/10707 [1:26:28<06:08,  2.02it/s]
 93%|█████████▎| 9964/10707 [1:26:28<06:07,  2.02it/s]
 93%|█████████▎| 9965/10707 [1:26:29<06:07,  2.02it/s]
 93%|█████████▎| 9966/10707 [1:26:29<06:06,  2.02it/s]
 93%|█████████▎| 9967/10707 [1:26:30<06:06,  2.02it/s]
 93%|█████████▎| 9968/10707 [1:26:30<06:05,  2.02it/s]
 93%|█████████▎| 9969/10707 [1:26:31<06:05,  2.02it/s]
 93%|█████████▎| 9970/10707 [1:26:31<06:04,  2.02it/s]
 93%|█████████▎| 9971/10707 [1:26:32<06:04,  2.02it/s]
 93%|█████████▎| 9972/10707 [1:26:32<06:03,  2.02it/s]
 93%|█████████▎| 9973/10707 [1:26:33<06:03,  2.02it/s]
 93%|█████████▎| 9974/10707 [1:26:33<06:03,  2.02it/s]
 93%|█████████▎| 9975/10707 [1:26:34<06:03,  2.02it/s]
                                                      
{'loss': 3.5792, 'grad_norm': 0.1900818645954132, 'learning_rate': 1.4171182822286354e-05, 'epoch': 0.93}
-
 93%|█████████▎| 9975/10707 [1:26:34<06:03,  2.02it/s]
 93%|█████████▎| 9976/10707 [1:26:34<06:03,  2.01it/s]
 93%|█████████▎| 9977/10707 [1:26:35<06:02,  2.02it/s]
 93%|█████████▎| 9978/10707 [1:26:35<06:01,  2.02it/s]
 93%|█████████▎| 9979/10707 [1:26:36<06:00,  2.02it/s]
 93%|█████████▎| 9980/10707 [1:26:36<06:00,  2.02it/s]
 93%|█████████▎| 9981/10707 [1:26:37<05:59,  2.02it/s]
 93%|█████████▎| 9982/10707 [1:26:37<05:58,  2.02it/s]
 93%|█████████▎| 9983/10707 [1:26:38<05:58,  2.02it/s]
 93%|█████████▎| 9984/10707 [1:26:38<05:57,  2.02it/s]
 93%|█████████▎| 9985/10707 [1:26:39<05:57,  2.02it/s]
 93%|█████████▎| 9986/10707 [1:26:39<05:57,  2.02it/s]
 93%|█████████▎| 9987/10707 [1:26:40<05:56,  2.02it/s]
 93%|█████████▎| 9988/10707 [1:26:40<05:56,  2.02it/s]
 93%|█████████▎| 9989/10707 [1:26:41<05:55,  2.02it/s]
 93%|█████████▎| 9990/10707 [1:26:41<05:55,  2.02it/s]
 93%|█████████▎| 9991/10707 [1:26:42<05:54,  2.02it/s]
 93%|█████████▎| 9992/10707 [1:26:42<05:54,  2.02it/s]
 93%|█████████▎| 9993/10707 [1:26:43<05:53,  2.02it/s]
 93%|█████████▎| 9994/10707 [1:26:43<05:53,  2.02it/s]
 93%|█████████▎| 9995/10707 [1:26:44<05:52,  2.02it/s]
 93%|█████████▎| 9996/10707 [1:26:44<05:52,  2.02it/s]
 93%|█████████▎| 9997/10707 [1:26:45<05:52,  2.02it/s]
 93%|█████████▎| 9998/10707 [1:26:45<05:51,  2.02it/s]
 93%|█████████▎| 9999/10707 [1:26:46<05:51,  2.02it/s]
 93%|█████████▎| 10000/10707 [1:26:46<05:50,  2.02it/s]
                                                       
{'loss': 3.5807, 'grad_norm': 0.190889373421669, 'learning_rate': 1.3223952640137704e-05, 'epoch': 0.93}
-
 93%|█████████▎| 10000/10707 [1:26:46<05:50,  2.02it/s]
 93%|█████████▎| 10001/10707 [1:26:47<05:50,  2.02it/s]
 93%|█████████▎| 10002/10707 [1:26:47<05:49,  2.02it/s]
 93%|█████████▎| 10003/10707 [1:26:47<05:48,  2.02it/s]
 93%|█████████▎| 10004/10707 [1:26:48<05:47,  2.02it/s]
 93%|█████████▎| 10005/10707 [1:26:48<05:47,  2.02it/s]
 93%|█████████▎| 10006/10707 [1:26:49<05:46,  2.02it/s]
 93%|█████████▎| 10007/10707 [1:26:49<05:46,  2.02it/s]
 93%|█████████▎| 10008/10707 [1:26:50<05:45,  2.02it/s]
 93%|█████████▎| 10009/10707 [1:26:50<05:45,  2.02it/s]
 93%|█████████▎| 10010/10707 [1:26:51<05:44,  2.02it/s]
 93%|█████████▎| 10011/10707 [1:26:51<05:44,  2.02it/s]
 94%|█████████▎| 10012/10707 [1:26:52<05:43,  2.02it/s]
 94%|█████████▎| 10013/10707 [1:26:52<05:43,  2.02it/s]
 94%|█████████▎| 10014/10707 [1:26:53<05:43,  2.02it/s]
 94%|█████████▎| 10015/10707 [1:26:53<05:42,  2.02it/s]
 94%|█████████▎| 10016/10707 [1:26:54<05:41,  2.02it/s]
 94%|█████████▎| 10017/10707 [1:26:54<05:41,  2.02it/s]
 94%|█████████▎| 10018/10707 [1:26:55<05:40,  2.02it/s]
 94%|█████████▎| 10019/10707 [1:26:55<05:40,  2.02it/s]
 94%|█████████▎| 10020/10707 [1:26:56<05:40,  2.02it/s]
 94%|█████████▎| 10021/10707 [1:26:56<05:40,  2.02it/s]
 94%|█████████▎| 10022/10707 [1:26:57<05:39,  2.02it/s]
 94%|█████████▎| 10023/10707 [1:26:57<05:39,  2.02it/s]
 94%|█████████▎| 10024/10707 [1:26:58<05:38,  2.02it/s]
 94%|█████████▎| 10025/10707 [1:26:58<05:37,  2.02it/s]{'loss': 3.5805, 'grad_norm': 0.19147241115570068, 'learning_rate': 1.2309060444508746e-05, 'epoch': 0.94}
-
                                                       

 94%|█████████▎| 10025/10707 [1:26:58<05:37,  2.02it/s]
 94%|█████████▎| 10026/10707 [1:26:59<05:37,  2.02it/s]
 94%|█████████▎| 10027/10707 [1:26:59<05:36,  2.02it/s]
 94%|█████████▎| 10028/10707 [1:27:00<05:36,  2.02it/s]
 94%|█████████▎| 10029/10707 [1:27:00<05:35,  2.02it/s]
 94%|█████████▎| 10030/10707 [1:27:01<05:35,  2.02it/s]
 94%|█████████▎| 10031/10707 [1:27:01<05:34,  2.02it/s]
 94%|█████████▎| 10032/10707 [1:27:02<05:34,  2.02it/s]
 94%|█████████▎| 10033/10707 [1:27:02<05:33,  2.02it/s]
 94%|█████████▎| 10034/10707 [1:27:03<05:33,  2.02it/s]
 94%|█████████▎| 10035/10707 [1:27:03<05:33,  2.01it/s]
 94%|█████████▎| 10036/10707 [1:27:04<05:33,  2.01it/s]
 94%|█████████▎| 10037/10707 [1:27:04<05:32,  2.02it/s]
 94%|█████████▍| 10038/10707 [1:27:05<05:31,  2.02it/s]
 94%|█████████▍| 10039/10707 [1:27:05<05:30,  2.02it/s]
 94%|█████████▍| 10040/10707 [1:27:06<05:29,  2.02it/s]
 94%|█████████▍| 10041/10707 [1:27:06<05:29,  2.02it/s]
 94%|█████████▍| 10042/10707 [1:27:07<05:29,  2.02it/s]
 94%|█████████▍| 10043/10707 [1:27:07<05:29,  2.02it/s]
 94%|█████████▍| 10044/10707 [1:27:08<05:28,  2.02it/s]
 94%|█████████▍| 10045/10707 [1:27:08<05:28,  2.02it/s]
 94%|█████████▍| 10046/10707 [1:27:09<05:27,  2.02it/s]
 94%|█████████▍| 10047/10707 [1:27:09<05:26,  2.02it/s]
 94%|█████████▍| 10048/10707 [1:27:10<05:27,  2.01it/s]
 94%|█████████▍| 10049/10707 [1:27:10<05:26,  2.02it/s]
 94%|█████████▍| 10050/10707 [1:27:11<05:25,  2.02it/s]{'loss': 3.5852, 'grad_norm': 0.19100289046764374, 'learning_rate': 1.1426567014420297e-05, 'epoch': 0.94}
-
                                                       

 94%|█████████▍| 10050/10707 [1:27:11<05:25,  2.02it/s]
 94%|█████████▍| 10051/10707 [1:27:11<05:25,  2.01it/s]
 94%|█████████▍| 10052/10707 [1:27:12<05:24,  2.02it/s]
 94%|█████████▍| 10053/10707 [1:27:12<05:24,  2.02it/s]
 94%|█████████▍| 10054/10707 [1:27:13<05:23,  2.02it/s]
 94%|█████████▍| 10055/10707 [1:27:13<05:23,  2.02it/s]
 94%|█████████▍| 10056/10707 [1:27:14<05:22,  2.02it/s]
 94%|█████████▍| 10057/10707 [1:27:14<05:21,  2.02it/s]
 94%|█████████▍| 10058/10707 [1:27:15<05:21,  2.02it/s]
 94%|█████████▍| 10059/10707 [1:27:15<05:20,  2.02it/s]
 94%|█████████▍| 10060/10707 [1:27:16<05:20,  2.02it/s]
 94%|█████████▍| 10061/10707 [1:27:16<05:19,  2.02it/s]
 94%|█████████▍| 10062/10707 [1:27:17<05:19,  2.02it/s]
 94%|█████████▍| 10063/10707 [1:27:17<05:18,  2.02it/s]
 94%|█████████▍| 10064/10707 [1:27:18<05:18,  2.02it/s]
 94%|█████████▍| 10065/10707 [1:27:18<05:17,  2.02it/s]
 94%|█████████▍| 10066/10707 [1:27:19<05:17,  2.02it/s]
 94%|█████████▍| 10067/10707 [1:27:19<05:16,  2.02it/s]
 94%|█████████▍| 10068/10707 [1:27:20<05:16,  2.02it/s]
 94%|█████████▍| 10069/10707 [1:27:20<05:16,  2.02it/s]
 94%|█████████▍| 10070/10707 [1:27:21<05:15,  2.02it/s]
 94%|█████████▍| 10071/10707 [1:27:21<05:15,  2.02it/s]
 94%|█████████▍| 10072/10707 [1:27:22<05:14,  2.02it/s]
 94%|█████████▍| 10073/10707 [1:27:22<05:13,  2.02it/s]
 94%|█████████▍| 10074/10707 [1:27:23<05:13,  2.02it/s]
 94%|█████████▍| 10075/10707 [1:27:23<05:12,  2.02it/s]
                                                       
{'loss': 3.5766, 'grad_norm': 0.1914224475622177, 'learning_rate': 1.0576530976546262e-05, 'epoch': 0.94}
-
 94%|█████████▍| 10075/10707 [1:27:23<05:12,  2.02it/s]
 94%|█████████▍| 10076/10707 [1:27:24<05:12,  2.02it/s]
 94%|█████████▍| 10077/10707 [1:27:24<05:12,  2.02it/s]
 94%|█████████▍| 10078/10707 [1:27:25<05:11,  2.02it/s]
 94%|█████████▍| 10079/10707 [1:27:25<05:11,  2.02it/s]
 94%|█████████▍| 10080/10707 [1:27:26<05:10,  2.02it/s]
 94%|█████████▍| 10081/10707 [1:27:26<05:10,  2.02it/s]
 94%|█████████▍| 10082/10707 [1:27:27<05:09,  2.02it/s]
 94%|█████████▍| 10083/10707 [1:27:27<05:08,  2.02it/s]
 94%|█████████▍| 10084/10707 [1:27:28<05:08,  2.02it/s]
 94%|█████████▍| 10085/10707 [1:27:28<05:07,  2.02it/s]
 94%|█████████▍| 10086/10707 [1:27:29<05:07,  2.02it/s]
 94%|█████████▍| 10087/10707 [1:27:29<05:06,  2.02it/s]
 94%|█████████▍| 10088/10707 [1:27:30<05:06,  2.02it/s]
 94%|█████████▍| 10089/10707 [1:27:30<05:05,  2.02it/s]
 94%|█████████▍| 10090/10707 [1:27:31<05:05,  2.02it/s]
 94%|█████████▍| 10091/10707 [1:27:31<05:04,  2.02it/s]
 94%|█████████▍| 10092/10707 [1:27:32<05:04,  2.02it/s]
 94%|█████████▍| 10093/10707 [1:27:32<05:04,  2.02it/s]
 94%|█████████▍| 10094/10707 [1:27:33<05:03,  2.02it/s]
 94%|█████████▍| 10095/10707 [1:27:33<05:03,  2.02it/s]
 94%|█████████▍| 10096/10707 [1:27:34<05:03,  2.01it/s]
 94%|█████████▍| 10097/10707 [1:27:34<05:02,  2.02it/s]
 94%|█████████▍| 10098/10707 [1:27:35<05:02,  2.02it/s]
 94%|█████████▍| 10099/10707 [1:27:35<05:01,  2.02it/s]
 94%|█████████▍| 10100/10707 [1:27:36<05:00,  2.02it/s]{'loss': 3.5707, 'grad_norm': 0.19331902265548706, 'learning_rate': 9.759008801319246e-06, 'epoch': 0.94}
-
                                                       

 94%|█████████▍| 10100/10707 [1:27:36<05:00,  2.02it/s]
 94%|█████████▍| 10101/10707 [1:27:36<05:00,  2.02it/s]
 94%|█████████▍| 10102/10707 [1:27:37<05:00,  2.02it/s]
 94%|█████████▍| 10103/10707 [1:27:37<04:59,  2.02it/s]
 94%|█████████▍| 10104/10707 [1:27:38<04:59,  2.02it/s]
 94%|█████████▍| 10105/10707 [1:27:38<04:58,  2.02it/s]
 94%|█████████▍| 10106/10707 [1:27:39<04:58,  2.02it/s]
 94%|█████████▍| 10107/10707 [1:27:39<04:57,  2.02it/s]
 94%|█████████▍| 10108/10707 [1:27:40<04:56,  2.02it/s]
 94%|█████████▍| 10109/10707 [1:27:40<04:56,  2.02it/s]
 94%|█████████▍| 10110/10707 [1:27:40<04:55,  2.02it/s]
 94%|█████████▍| 10111/10707 [1:27:41<04:55,  2.02it/s]
 94%|█████████▍| 10112/10707 [1:27:41<04:54,  2.02it/s]
 94%|█████████▍| 10113/10707 [1:27:42<04:53,  2.02it/s]
 94%|█████████▍| 10114/10707 [1:27:42<04:53,  2.02it/s]
 94%|█████████▍| 10115/10707 [1:27:43<04:53,  2.02it/s]
 94%|█████████▍| 10116/10707 [1:27:43<04:52,  2.02it/s]
 94%|█████████▍| 10117/10707 [1:27:44<04:52,  2.02it/s]
 94%|█████████▍| 10118/10707 [1:27:44<04:51,  2.02it/s]
 95%|█████████▍| 10119/10707 [1:27:45<04:51,  2.02it/s]
 95%|█████████▍| 10120/10707 [1:27:45<04:50,  2.02it/s]
 95%|█████████▍| 10121/10707 [1:27:46<04:50,  2.02it/s]
 95%|█████████▍| 10122/10707 [1:27:46<04:49,  2.02it/s]
 95%|█████████▍| 10123/10707 [1:27:47<04:49,  2.02it/s]
 95%|█████████▍| 10124/10707 [1:27:47<04:48,  2.02it/s]
 95%|█████████▍| 10125/10707 [1:27:48<04:48,  2.02it/s]{'loss': 3.5829, 'grad_norm': 0.1911199986934662, 'learning_rate': 8.974054799178844e-06, 'epoch': 0.95}
-
                                                       

 95%|█████████▍| 10125/10707 [1:27:48<04:48,  2.02it/s]
 95%|█████████▍| 10126/10707 [1:27:48<04:48,  2.02it/s]
 95%|█████████▍| 10127/10707 [1:27:49<04:47,  2.02it/s]
 95%|█████████▍| 10128/10707 [1:27:49<04:46,  2.02it/s]
 95%|█████████▍| 10129/10707 [1:27:50<04:46,  2.02it/s]
 95%|█████████▍| 10130/10707 [1:27:50<04:45,  2.02it/s]
 95%|█████████▍| 10131/10707 [1:27:51<04:45,  2.02it/s]
 95%|█████████▍| 10132/10707 [1:27:51<04:44,  2.02it/s]
 95%|█████████▍| 10133/10707 [1:27:52<04:43,  2.02it/s]
 95%|█████████▍| 10134/10707 [1:27:52<04:43,  2.02it/s]
 95%|█████████▍| 10135/10707 [1:27:53<04:43,  2.02it/s]
 95%|█████████▍| 10136/10707 [1:27:53<04:42,  2.02it/s]
 95%|█████████▍| 10137/10707 [1:27:54<04:42,  2.02it/s]
 95%|█████████▍| 10138/10707 [1:27:54<04:41,  2.02it/s]
 95%|█████████▍| 10139/10707 [1:27:55<04:41,  2.02it/s]
 95%|█████████▍| 10140/10707 [1:27:55<04:40,  2.02it/s]
 95%|█████████▍| 10141/10707 [1:27:56<04:40,  2.02it/s]
 95%|█████████▍| 10142/10707 [1:27:56<04:39,  2.02it/s]
 95%|█████████▍| 10143/10707 [1:27:57<04:39,  2.02it/s]
 95%|█████████▍| 10144/10707 [1:27:57<04:38,  2.02it/s]
 95%|█████████▍| 10145/10707 [1:27:58<04:37,  2.02it/s]
 95%|█████████▍| 10146/10707 [1:27:58<04:37,  2.02it/s]
 95%|█████████▍| 10147/10707 [1:27:59<04:36,  2.02it/s]
 95%|█████████▍| 10148/10707 [1:27:59<04:36,  2.02it/s]
 95%|█████████▍| 10149/10707 [1:28:00<04:35,  2.02it/s]
 95%|█████████▍| 10150/10707 [1:28:00<04:35,  2.02it/s]
                                                       {'loss': 3.5795, 'grad_norm': 0.1932750940322876, 'learning_rate': 8.221721116963455e-06, 'epoch': 0.95}
-

 95%|█████████▍| 10150/10707 [1:28:00<04:35,  2.02it/s]
 95%|█████████▍| 10151/10707 [1:28:01<04:35,  2.02it/s]
 95%|█████████▍| 10152/10707 [1:28:01<04:34,  2.02it/s]
 95%|█████████▍| 10153/10707 [1:28:02<04:33,  2.02it/s]
 95%|█████████▍| 10154/10707 [1:28:02<04:33,  2.02it/s]
 95%|█████████▍| 10155/10707 [1:28:03<04:33,  2.02it/s]
 95%|█████████▍| 10156/10707 [1:28:03<04:33,  2.02it/s]
 95%|█████████▍| 10157/10707 [1:28:04<04:33,  2.01it/s]
 95%|█████████▍| 10158/10707 [1:28:04<04:32,  2.02it/s]
 95%|█████████▍| 10159/10707 [1:28:05<04:31,  2.02it/s]
 95%|█████████▍| 10160/10707 [1:28:05<04:30,  2.02it/s]
 95%|█████████▍| 10161/10707 [1:28:06<04:30,  2.02it/s]
 95%|█████████▍| 10162/10707 [1:28:06<04:29,  2.02it/s]
 95%|█████████▍| 10163/10707 [1:28:07<04:29,  2.02it/s]
 95%|█████████▍| 10164/10707 [1:28:07<04:29,  2.02it/s]
 95%|█████████▍| 10165/10707 [1:28:08<04:28,  2.02it/s]
 95%|█████████▍| 10166/10707 [1:28:08<04:27,  2.02it/s]
 95%|█████████▍| 10167/10707 [1:28:09<04:27,  2.02it/s]
 95%|█████████▍| 10168/10707 [1:28:09<04:26,  2.02it/s]
 95%|█████████▍| 10169/10707 [1:28:10<04:26,  2.02it/s]
 95%|█████████▍| 10170/10707 [1:28:10<04:25,  2.02it/s]
 95%|█████████▍| 10171/10707 [1:28:11<04:25,  2.02it/s]
 95%|█████████▌| 10172/10707 [1:28:11<04:24,  2.02it/s]
 95%|█████████▌| 10173/10707 [1:28:12<04:24,  2.02it/s]
 95%|█████████▌| 10174/10707 [1:28:12<04:23,  2.02it/s]
 95%|█████████▌| 10175/10707 [1:28:13<04:23,  2.02it/s]{'loss': 3.5696, 'grad_norm': 0.18948020040988922, 'learning_rate': 7.502057734446344e-06, 'epoch': 0.95}
-
                                                       

 95%|█████████▌| 10175/10707 [1:28:13<04:23,  2.02it/s]
 95%|█████████▌| 10176/10707 [1:28:13<04:23,  2.02it/s]
 95%|█████████▌| 10177/10707 [1:28:14<04:22,  2.02it/s]
 95%|█████████▌| 10178/10707 [1:28:14<04:22,  2.01it/s]
 95%|█████████▌| 10179/10707 [1:28:15<04:21,  2.02it/s]
 95%|█████████▌| 10180/10707 [1:28:15<04:21,  2.02it/s]
 95%|█████████▌| 10181/10707 [1:28:16<04:20,  2.02it/s]
 95%|█████████▌| 10182/10707 [1:28:16<04:20,  2.02it/s]
 95%|█████████▌| 10183/10707 [1:28:17<04:19,  2.02it/s]
 95%|█████████▌| 10184/10707 [1:28:17<04:19,  2.02it/s]
 95%|█████████▌| 10185/10707 [1:28:18<04:18,  2.02it/s]
 95%|█████████▌| 10186/10707 [1:28:18<04:18,  2.02it/s]
 95%|█████████▌| 10187/10707 [1:28:19<04:17,  2.02it/s]
 95%|█████████▌| 10188/10707 [1:28:19<04:17,  2.02it/s]
 95%|█████████▌| 10189/10707 [1:28:20<04:16,  2.02it/s]
 95%|█████████▌| 10190/10707 [1:28:20<04:16,  2.02it/s]
 95%|█████████▌| 10191/10707 [1:28:21<04:15,  2.02it/s]
 95%|█████████▌| 10192/10707 [1:28:21<04:15,  2.02it/s]
 95%|█████████▌| 10193/10707 [1:28:22<04:14,  2.02it/s]
 95%|█████████▌| 10194/10707 [1:28:22<04:14,  2.02it/s]
 95%|█████████▌| 10195/10707 [1:28:23<04:13,  2.02it/s]
 95%|█████████▌| 10196/10707 [1:28:23<04:12,  2.02it/s]
 95%|█████████▌| 10197/10707 [1:28:24<04:12,  2.02it/s]
 95%|█████████▌| 10198/10707 [1:28:24<04:11,  2.02it/s]
 95%|█████████▌| 10199/10707 [1:28:25<04:11,  2.02it/s]
 95%|█████████▌| 10200/10707 [1:28:25<04:11,  2.02it/s]{'loss': 3.574, 'grad_norm': 0.19181020557880402, 'learning_rate': 6.8151124610152385e-06, 'epoch': 0.95}
-
                                                       

 95%|█████████▌| 10200/10707 [1:28:25<04:11,  2.02it/s]
 95%|█████████▌| 10201/10707 [1:28:26<04:11,  2.02it/s]
 95%|█████████▌| 10202/10707 [1:28:26<04:10,  2.02it/s]
 95%|█████████▌| 10203/10707 [1:28:27<04:09,  2.02it/s]
 95%|█████████▌| 10204/10707 [1:28:27<04:09,  2.02it/s]
 95%|█████████▌| 10205/10707 [1:28:28<04:08,  2.02it/s]
 95%|█████████▌| 10206/10707 [1:28:28<04:08,  2.02it/s]
 95%|█████████▌| 10207/10707 [1:28:29<04:07,  2.02it/s]
 95%|█████████▌| 10208/10707 [1:28:29<04:06,  2.02it/s]
 95%|█████████▌| 10209/10707 [1:28:30<04:06,  2.02it/s]
 95%|█████████▌| 10210/10707 [1:28:30<04:05,  2.02it/s]
 95%|█████████▌| 10211/10707 [1:28:31<04:05,  2.02it/s]
 95%|█████████▌| 10212/10707 [1:28:31<04:04,  2.02it/s]
 95%|█████████▌| 10213/10707 [1:28:31<04:04,  2.02it/s]
 95%|█████████▌| 10214/10707 [1:28:32<04:04,  2.02it/s]
 95%|█████████▌| 10215/10707 [1:28:32<04:03,  2.02it/s]
 95%|█████████▌| 10216/10707 [1:28:33<04:03,  2.02it/s]
 95%|█████████▌| 10217/10707 [1:28:33<04:03,  2.01it/s]
 95%|█████████▌| 10218/10707 [1:28:34<04:02,  2.02it/s]
 95%|█████████▌| 10219/10707 [1:28:34<04:02,  2.02it/s]
 95%|█████████▌| 10220/10707 [1:28:35<04:00,  2.02it/s]
 95%|█████████▌| 10221/10707 [1:28:35<04:00,  2.02it/s]
 95%|█████████▌| 10222/10707 [1:28:36<03:59,  2.02it/s]
 95%|█████████▌| 10223/10707 [1:28:36<03:59,  2.02it/s]
 95%|█████████▌| 10224/10707 [1:28:37<03:59,  2.02it/s]
 95%|█████████▌| 10225/10707 [1:28:37<03:58,  2.02it/s]{'loss': 3.5741, 'grad_norm': 0.1904952973127365, 'learning_rate': 6.1609309324962e-06, 'epoch': 0.95}
-
                                                       

 95%|█████████▌| 10225/10707 [1:28:37<03:58,  2.02it/s]
 96%|█████████▌| 10226/10707 [1:28:38<03:58,  2.02it/s]
 96%|█████████▌| 10227/10707 [1:28:38<03:57,  2.02it/s]
 96%|█████████▌| 10228/10707 [1:28:39<03:57,  2.02it/s]
 96%|█████████▌| 10229/10707 [1:28:39<03:56,  2.02it/s]
 96%|█████████▌| 10230/10707 [1:28:40<03:56,  2.02it/s]
 96%|█████████▌| 10231/10707 [1:28:40<03:56,  2.02it/s]
 96%|█████████▌| 10232/10707 [1:28:41<03:55,  2.02it/s]
 96%|█████████▌| 10233/10707 [1:28:41<03:54,  2.02it/s]
 96%|█████████▌| 10234/10707 [1:28:42<03:54,  2.02it/s]
 96%|█████████▌| 10235/10707 [1:28:42<03:53,  2.02it/s]
 96%|█████████▌| 10236/10707 [1:28:43<03:53,  2.02it/s]
 96%|█████████▌| 10237/10707 [1:28:43<03:53,  2.02it/s]
 96%|█████████▌| 10238/10707 [1:28:44<03:52,  2.02it/s]
 96%|█████████▌| 10239/10707 [1:28:44<03:51,  2.02it/s]
 96%|█████████▌| 10240/10707 [1:28:45<03:51,  2.02it/s]
 96%|█████████▌| 10241/10707 [1:28:45<03:50,  2.02it/s]
 96%|█████████▌| 10242/10707 [1:28:46<03:50,  2.02it/s]
 96%|█████████▌| 10243/10707 [1:28:46<03:49,  2.02it/s]
 96%|█████████▌| 10244/10707 [1:28:47<03:49,  2.02it/s]
 96%|█████████▌| 10245/10707 [1:28:47<03:48,  2.02it/s]
 96%|█████████▌| 10246/10707 [1:28:48<03:48,  2.02it/s]
 96%|█████████▌| 10247/10707 [1:28:48<03:47,  2.02it/s]
 96%|█████████▌| 10248/10707 [1:28:49<03:47,  2.02it/s]
 96%|█████████▌| 10249/10707 [1:28:49<03:46,  2.02it/s]
 96%|█████████▌| 10250/10707 [1:28:50<03:45,  2.02it/s]{'loss': 3.5757, 'grad_norm': 0.19374090433120728, 'learning_rate': 5.53955660812161e-06, 'epoch': 0.96}
-
                                                       

 96%|█████████▌| 10250/10707 [1:28:50<03:45,  2.02it/s]
 96%|█████████▌| 10251/10707 [1:28:50<03:45,  2.02it/s]
 96%|█████████▌| 10252/10707 [1:28:51<03:45,  2.02it/s]
 96%|█████████▌| 10253/10707 [1:28:51<03:44,  2.02it/s]
 96%|█████████▌| 10254/10707 [1:28:52<03:44,  2.02it/s]
 96%|█████████▌| 10255/10707 [1:28:52<03:43,  2.02it/s]
 96%|█████████▌| 10256/10707 [1:28:53<03:43,  2.02it/s]
 96%|█████████▌| 10257/10707 [1:28:53<03:42,  2.02it/s]
 96%|█████████▌| 10258/10707 [1:28:54<03:42,  2.02it/s]
 96%|█████████▌| 10259/10707 [1:28:54<03:42,  2.02it/s]
 96%|█████████▌| 10260/10707 [1:28:55<03:41,  2.02it/s]
 96%|█████████▌| 10261/10707 [1:28:55<03:40,  2.02it/s]
 96%|█████████▌| 10262/10707 [1:28:56<03:40,  2.02it/s]
 96%|█████████▌| 10263/10707 [1:28:56<03:39,  2.02it/s]
 96%|█████████▌| 10264/10707 [1:28:57<03:39,  2.02it/s]
 96%|█████████▌| 10265/10707 [1:28:57<03:38,  2.02it/s]
 96%|█████████▌| 10266/10707 [1:28:58<03:38,  2.02it/s]
 96%|█████████▌| 10267/10707 [1:28:58<03:37,  2.02it/s]
 96%|█████████▌| 10268/10707 [1:28:59<03:37,  2.02it/s]
 96%|█████████▌| 10269/10707 [1:28:59<03:36,  2.02it/s]
 96%|█████████▌| 10270/10707 [1:29:00<03:36,  2.02it/s]
 96%|█████████▌| 10271/10707 [1:29:00<03:35,  2.02it/s]
 96%|█████████▌| 10272/10707 [1:29:01<03:35,  2.02it/s]
 96%|█████████▌| 10273/10707 [1:29:01<03:34,  2.02it/s]
 96%|█████████▌| 10274/10707 [1:29:02<03:34,  2.02it/s]
 96%|█████████▌| 10275/10707 [1:29:02<03:33,  2.02it/s]
                                                       
{'loss': 3.5763, 'grad_norm': 0.19099155068397522, 'learning_rate': 4.951030767643638e-06, 'epoch': 0.96}
-
 96%|█████████▌| 10275/10707 [1:29:02<03:33,  2.02it/s]
 96%|█████████▌| 10276/10707 [1:29:03<03:34,  2.01it/s]
 96%|█████████▌| 10277/10707 [1:29:03<03:34,  2.01it/s]
 96%|█████████▌| 10278/10707 [1:29:04<03:33,  2.01it/s]
 96%|█████████▌| 10279/10707 [1:29:04<03:32,  2.01it/s]
 96%|█████████▌| 10280/10707 [1:29:05<03:31,  2.02it/s]
 96%|█████████▌| 10281/10707 [1:29:05<03:31,  2.02it/s]
 96%|█████████▌| 10282/10707 [1:29:06<03:30,  2.02it/s]
 96%|█████████▌| 10283/10707 [1:29:06<03:30,  2.02it/s]
 96%|█████████▌| 10284/10707 [1:29:07<03:29,  2.02it/s]
 96%|█████████▌| 10285/10707 [1:29:07<03:29,  2.02it/s]
 96%|█████████▌| 10286/10707 [1:29:08<03:28,  2.02it/s]
 96%|█████████▌| 10287/10707 [1:29:08<03:27,  2.02it/s]
 96%|█████████▌| 10288/10707 [1:29:09<03:27,  2.02it/s]
 96%|█████████▌| 10289/10707 [1:29:09<03:27,  2.02it/s]
 96%|█████████▌| 10290/10707 [1:29:10<03:26,  2.02it/s]
 96%|█████████▌| 10291/10707 [1:29:10<03:25,  2.02it/s]
 96%|█████████▌| 10292/10707 [1:29:11<03:25,  2.02it/s]
 96%|█████████▌| 10293/10707 [1:29:11<03:24,  2.02it/s]
 96%|█████████▌| 10294/10707 [1:29:12<03:24,  2.02it/s]
 96%|█████████▌| 10295/10707 [1:29:12<03:23,  2.02it/s]
 96%|█████████▌| 10296/10707 [1:29:13<03:23,  2.02it/s]
 96%|█████████▌| 10297/10707 [1:29:13<03:22,  2.02it/s]
 96%|█████████▌| 10298/10707 [1:29:14<03:22,  2.02it/s]
 96%|█████████▌| 10299/10707 [1:29:14<03:22,  2.02it/s]
 96%|█████████▌| 10300/10707 [1:29:15<03:21,  2.02it/s]{'loss': 3.5784, 'grad_norm': 0.18864507973194122, 'learning_rate': 4.395392508591389e-06, 'epoch': 0.96}
-
                                                       

 96%|█████████▌| 10300/10707 [1:29:15<03:21,  2.02it/s]
 96%|█████████▌| 10301/10707 [1:29:15<03:21,  2.02it/s]
 96%|█████████▌| 10302/10707 [1:29:16<03:20,  2.02it/s]
 96%|█████████▌| 10303/10707 [1:29:16<03:20,  2.02it/s]
 96%|█████████▌| 10304/10707 [1:29:17<03:19,  2.02it/s]
 96%|█████████▌| 10305/10707 [1:29:17<03:18,  2.02it/s]
 96%|█████████▋| 10306/10707 [1:29:18<03:18,  2.02it/s]
 96%|█████████▋| 10307/10707 [1:29:18<03:17,  2.02it/s]
 96%|█████████▋| 10308/10707 [1:29:19<03:17,  2.02it/s]
 96%|█████████▋| 10309/10707 [1:29:19<03:16,  2.02it/s]
 96%|█████████▋| 10310/10707 [1:29:20<03:16,  2.02it/s]
 96%|█████████▋| 10311/10707 [1:29:20<03:15,  2.02it/s]
 96%|█████████▋| 10312/10707 [1:29:21<03:15,  2.02it/s]
 96%|█████████▋| 10313/10707 [1:29:21<03:14,  2.02it/s]
 96%|█████████▋| 10314/10707 [1:29:22<03:14,  2.02it/s]
 96%|█████████▋| 10315/10707 [1:29:22<03:14,  2.02it/s]
 96%|█████████▋| 10316/10707 [1:29:23<03:13,  2.02it/s]
 96%|█████████▋| 10317/10707 [1:29:23<03:13,  2.02it/s]
 96%|█████████▋| 10318/10707 [1:29:24<03:12,  2.02it/s]
 96%|█████████▋| 10319/10707 [1:29:24<03:12,  2.02it/s]
 96%|█████████▋| 10320/10707 [1:29:25<03:11,  2.02it/s]
 96%|█████████▋| 10321/10707 [1:29:25<03:11,  2.02it/s]
 96%|█████████▋| 10322/10707 [1:29:25<03:10,  2.02it/s]
 96%|█████████▋| 10323/10707 [1:29:26<03:10,  2.02it/s]
 96%|█████████▋| 10324/10707 [1:29:26<03:09,  2.02it/s]
 96%|█████████▋| 10325/10707 [1:29:27<03:09,  2.02it/s]
                                                       
{'loss': 3.5789, 'grad_norm': 0.19068458676338196, 'learning_rate': 3.872678743673752e-06, 'epoch': 0.96}
-
 96%|█████████▋| 10325/10707 [1:29:27<03:09,  2.02it/s]
 96%|█████████▋| 10326/10707 [1:29:27<03:08,  2.02it/s]
 96%|█████████▋| 10327/10707 [1:29:28<03:08,  2.02it/s]
 96%|█████████▋| 10328/10707 [1:29:28<03:07,  2.02it/s]
 96%|█████████▋| 10329/10707 [1:29:29<03:07,  2.02it/s]
 96%|█████████▋| 10330/10707 [1:29:29<03:06,  2.02it/s]
 96%|█████████▋| 10331/10707 [1:29:30<03:06,  2.02it/s]
 96%|█████████▋| 10332/10707 [1:29:30<03:05,  2.02it/s]
 97%|█████████▋| 10333/10707 [1:29:31<03:05,  2.02it/s]
 97%|█████████▋| 10334/10707 [1:29:31<03:04,  2.02it/s]
 97%|█████████▋| 10335/10707 [1:29:32<03:04,  2.02it/s]
 97%|█████████▋| 10336/10707 [1:29:32<03:03,  2.02it/s]
 97%|█████████▋| 10337/10707 [1:29:33<03:03,  2.02it/s]
 97%|█████████▋| 10338/10707 [1:29:33<03:03,  2.01it/s]
 97%|█████████▋| 10339/10707 [1:29:34<03:02,  2.01it/s]
 97%|█████████▋| 10340/10707 [1:29:34<03:01,  2.02it/s]
 97%|█████████▋| 10341/10707 [1:29:35<03:01,  2.02it/s]
 97%|█████████▋| 10342/10707 [1:29:35<03:00,  2.02it/s]
 97%|█████████▋| 10343/10707 [1:29:36<03:00,  2.02it/s]
 97%|█████████▋| 10344/10707 [1:29:36<02:59,  2.02it/s]
 97%|█████████▋| 10345/10707 [1:29:37<02:59,  2.02it/s]
 97%|█████████▋| 10346/10707 [1:29:37<02:58,  2.02it/s]
 97%|█████████▋| 10347/10707 [1:29:38<02:58,  2.02it/s]
 97%|█████████▋| 10348/10707 [1:29:38<02:57,  2.02it/s]
 97%|█████████▋| 10349/10707 [1:29:39<02:57,  2.02it/s]
 97%|█████████▋| 10350/10707 [1:29:39<02:56,  2.02it/s]{'loss': 3.5772, 'grad_norm': 0.19155988097190857, 'learning_rate': 3.3829241983270866e-06, 'epoch': 0.97}
-
                                                       

 97%|█████████▋| 10350/10707 [1:29:39<02:56,  2.02it/s]
 97%|█████████▋| 10351/10707 [1:29:40<02:56,  2.02it/s]
 97%|█████████▋| 10352/10707 [1:29:40<02:55,  2.02it/s]
 97%|█████████▋| 10353/10707 [1:29:41<02:55,  2.02it/s]
 97%|█████████▋| 10354/10707 [1:29:41<02:54,  2.02it/s]
 97%|█████████▋| 10355/10707 [1:29:42<02:54,  2.02it/s]
 97%|█████████▋| 10356/10707 [1:29:42<02:53,  2.02it/s]
 97%|█████████▋| 10357/10707 [1:29:43<02:53,  2.02it/s]
 97%|█████████▋| 10358/10707 [1:29:43<02:52,  2.02it/s]
 97%|█████████▋| 10359/10707 [1:29:44<02:52,  2.02it/s]
 97%|█████████▋| 10360/10707 [1:29:44<02:51,  2.02it/s]
 97%|█████████▋| 10361/10707 [1:29:45<02:51,  2.02it/s]
 97%|█████████▋| 10362/10707 [1:29:45<02:50,  2.02it/s]
 97%|█████████▋| 10363/10707 [1:29:46<02:50,  2.02it/s]
 97%|█████████▋| 10364/10707 [1:29:46<02:49,  2.02it/s]
 97%|█████████▋| 10365/10707 [1:29:47<02:49,  2.02it/s]
 97%|█████████▋| 10366/10707 [1:29:47<02:48,  2.02it/s]
 97%|█████████▋| 10367/10707 [1:29:48<02:48,  2.02it/s]
 97%|█████████▋| 10368/10707 [1:29:48<02:47,  2.02it/s]
 97%|█████████▋| 10369/10707 [1:29:49<02:47,  2.02it/s]
 97%|█████████▋| 10370/10707 [1:29:49<02:46,  2.02it/s]
 97%|█████████▋| 10371/10707 [1:29:50<02:46,  2.02it/s]
 97%|█████████▋| 10372/10707 [1:29:51<03:12,  1.74it/s]
 97%|█████████▋| 10373/10707 [1:29:51<03:03,  1.82it/s]
 97%|█████████▋| 10374/10707 [1:29:52<02:57,  1.87it/s]
 97%|█████████▋| 10375/10707 [1:29:52<02:53,  1.91it/s]
                                                       
{'loss': 3.5761, 'grad_norm': 0.1905076801776886, 'learning_rate': 2.926161408408512e-06, 'epoch': 0.97}
-
 97%|█████████▋| 10375/10707 [1:29:52<02:53,  1.91it/s]
 97%|█████████▋| 10376/10707 [1:29:53<02:50,  1.94it/s]
 97%|█████████▋| 10377/10707 [1:29:53<02:48,  1.96it/s]
 97%|█████████▋| 10378/10707 [1:29:53<02:46,  1.98it/s]
 97%|█████████▋| 10379/10707 [1:29:54<02:44,  1.99it/s]
 97%|█████████▋| 10380/10707 [1:29:54<02:43,  2.00it/s]
 97%|█████████▋| 10381/10707 [1:29:55<02:42,  2.01it/s]
 97%|█████████▋| 10382/10707 [1:29:55<02:41,  2.01it/s]
 97%|█████████▋| 10383/10707 [1:29:56<02:40,  2.01it/s]
 97%|█████████▋| 10384/10707 [1:29:56<02:40,  2.02it/s]
 97%|█████████▋| 10385/10707 [1:29:57<02:39,  2.02it/s]
 97%|█████████▋| 10386/10707 [1:29:57<02:38,  2.02it/s]
 97%|█████████▋| 10387/10707 [1:29:58<02:38,  2.02it/s]
 97%|█████████▋| 10388/10707 [1:29:58<02:37,  2.02it/s]
 97%|█████████▋| 10389/10707 [1:29:59<02:37,  2.02it/s]
 97%|█████████▋| 10390/10707 [1:29:59<02:36,  2.02it/s]
 97%|█████████▋| 10391/10707 [1:30:00<02:36,  2.02it/s]
 97%|█████████▋| 10392/10707 [1:30:00<02:35,  2.02it/s]
 97%|█████████▋| 10393/10707 [1:30:01<02:35,  2.02it/s]
 97%|█████████▋| 10394/10707 [1:30:01<02:34,  2.02it/s]
 97%|█████████▋| 10395/10707 [1:30:02<02:34,  2.02it/s]
 97%|█████████▋| 10396/10707 [1:30:02<02:33,  2.02it/s]
 97%|█████████▋| 10397/10707 [1:30:03<02:33,  2.02it/s]
 97%|█████████▋| 10398/10707 [1:30:03<02:33,  2.01it/s]
 97%|█████████▋| 10399/10707 [1:30:04<02:32,  2.02it/s]
 97%|█████████▋| 10400/10707 [1:30:04<02:32,  2.02it/s]{'loss': 3.5696, 'grad_norm': 0.19187532365322113, 'learning_rate': 2.502420718034193e-06, 'epoch': 0.97}
-
                                                       

 97%|█████████▋| 10400/10707 [1:30:04<02:32,  2.02it/s]
 97%|█████████▋| 10401/10707 [1:30:05<02:31,  2.02it/s]
 97%|█████████▋| 10402/10707 [1:30:05<02:31,  2.02it/s]
 97%|█████████▋| 10403/10707 [1:30:06<02:30,  2.02it/s]
 97%|█████████▋| 10404/10707 [1:30:06<02:30,  2.02it/s]
 97%|█████████▋| 10405/10707 [1:30:07<02:29,  2.02it/s]
 97%|█████████▋| 10406/10707 [1:30:07<02:29,  2.02it/s]
 97%|█████████▋| 10407/10707 [1:30:08<02:28,  2.02it/s]
 97%|█████████▋| 10408/10707 [1:30:08<02:28,  2.02it/s]
 97%|█████████▋| 10409/10707 [1:30:09<02:27,  2.02it/s]
 97%|█████████▋| 10410/10707 [1:30:09<02:27,  2.02it/s]
 97%|█████████▋| 10411/10707 [1:30:10<02:26,  2.02it/s]
 97%|█████████▋| 10412/10707 [1:30:10<02:26,  2.02it/s]
 97%|█████████▋| 10413/10707 [1:30:11<02:25,  2.02it/s]
 97%|█████████▋| 10414/10707 [1:30:11<02:25,  2.02it/s]
 97%|█████████▋| 10415/10707 [1:30:12<02:47,  1.75it/s]
 97%|█████████▋| 10416/10707 [1:30:13<02:39,  1.82it/s]
 97%|█████████▋| 10417/10707 [1:30:13<02:34,  1.88it/s]
 97%|█████████▋| 10418/10707 [1:30:14<02:30,  1.92it/s]
 97%|█████████▋| 10419/10707 [1:30:14<02:27,  1.95it/s]
 97%|█████████▋| 10420/10707 [1:30:15<02:25,  1.97it/s]
 97%|█████████▋| 10421/10707 [1:30:15<02:24,  1.98it/s]
 97%|█████████▋| 10422/10707 [1:30:16<02:22,  2.00it/s]
 97%|█████████▋| 10423/10707 [1:30:16<02:21,  2.00it/s]
 97%|█████████▋| 10424/10707 [1:30:17<02:21,  2.01it/s]
 97%|█████████▋| 10425/10707 [1:30:17<02:20,  2.01it/s]{'loss': 3.5715, 'grad_norm': 0.18919752538204193, 'learning_rate': 2.111730277563617e-06, 'epoch': 0.97}
-
                                                       

 97%|█████████▋| 10425/10707 [1:30:17<02:20,  2.01it/s]
 97%|█████████▋| 10426/10707 [1:30:18<02:19,  2.01it/s]
 97%|█████████▋| 10427/10707 [1:30:18<02:19,  2.01it/s]
 97%|█████████▋| 10428/10707 [1:30:19<02:18,  2.01it/s]
 97%|█████████▋| 10429/10707 [1:30:19<02:17,  2.02it/s]
 97%|█████████▋| 10430/10707 [1:30:20<02:17,  2.02it/s]
 97%|█████████▋| 10431/10707 [1:30:20<02:16,  2.02it/s]
 97%|█████████▋| 10432/10707 [1:30:20<02:16,  2.02it/s]
 97%|█████████▋| 10433/10707 [1:30:21<02:15,  2.02it/s]
 97%|█████████▋| 10434/10707 [1:30:21<02:15,  2.02it/s]
 97%|█████████▋| 10435/10707 [1:30:22<02:14,  2.02it/s]
 97%|█████████▋| 10436/10707 [1:30:22<02:14,  2.02it/s]
 97%|█████████▋| 10437/10707 [1:30:23<02:13,  2.02it/s]
 97%|█████████▋| 10438/10707 [1:30:23<02:13,  2.02it/s]
 97%|█████████▋| 10439/10707 [1:30:24<02:12,  2.02it/s]
 98%|█████████▊| 10440/10707 [1:30:24<02:12,  2.02it/s]
 98%|█████████▊| 10441/10707 [1:30:25<02:11,  2.02it/s]
 98%|█████████▊| 10442/10707 [1:30:25<02:11,  2.02it/s]
 98%|█████████▊| 10443/10707 [1:30:26<02:10,  2.02it/s]
 98%|█████████▊| 10444/10707 [1:30:26<02:10,  2.02it/s]
 98%|█████████▊| 10445/10707 [1:30:27<02:09,  2.02it/s]
 98%|█████████▊| 10446/10707 [1:30:27<02:09,  2.02it/s]
 98%|█████████▊| 10447/10707 [1:30:28<02:08,  2.02it/s]
 98%|█████████▊| 10448/10707 [1:30:28<02:08,  2.02it/s]
 98%|█████████▊| 10449/10707 [1:30:29<02:07,  2.02it/s]
 98%|█████████▊| 10450/10707 [1:30:29<02:07,  2.02it/s]{'loss': 3.5793, 'grad_norm': 0.19072982668876648, 'learning_rate': 1.7541160417295365e-06, 'epoch': 0.98}
                                                       
-
 98%|█████████▊| 10450/10707 [1:30:29<02:07,  2.02it/s]
 98%|█████████▊| 10451/10707 [1:30:30<02:06,  2.02it/s]
 98%|█████████▊| 10452/10707 [1:30:30<02:06,  2.02it/s]
 98%|█████████▊| 10453/10707 [1:30:31<02:05,  2.02it/s]
 98%|█████████▊| 10454/10707 [1:30:31<02:05,  2.02it/s]
 98%|█████████▊| 10455/10707 [1:30:32<02:04,  2.02it/s]
 98%|█████████▊| 10456/10707 [1:30:32<02:04,  2.02it/s]
 98%|█████████▊| 10457/10707 [1:30:33<02:03,  2.02it/s]
 98%|█████████▊| 10458/10707 [1:30:33<02:03,  2.01it/s]
 98%|█████████▊| 10459/10707 [1:30:34<02:03,  2.01it/s]
 98%|█████████▊| 10460/10707 [1:30:34<02:02,  2.02it/s]
 98%|█████████▊| 10461/10707 [1:30:35<02:01,  2.02it/s]
 98%|█████████▊| 10462/10707 [1:30:35<02:01,  2.02it/s]
 98%|█████████▊| 10463/10707 [1:30:36<02:00,  2.02it/s]
 98%|█████████▊| 10464/10707 [1:30:36<02:00,  2.02it/s]
 98%|█████████▊| 10465/10707 [1:30:37<01:59,  2.02it/s]
 98%|█████████▊| 10466/10707 [1:30:37<01:59,  2.02it/s]
 98%|█████████▊| 10467/10707 [1:30:38<01:58,  2.02it/s]
 98%|█████████▊| 10468/10707 [1:30:38<01:58,  2.02it/s]
 98%|█████████▊| 10469/10707 [1:30:39<01:57,  2.02it/s]
 98%|█████████▊| 10470/10707 [1:30:39<01:57,  2.02it/s]
 98%|█████████▊| 10471/10707 [1:30:40<01:56,  2.02it/s]
 98%|█████████▊| 10472/10707 [1:30:40<01:56,  2.02it/s]
 98%|█████████▊| 10473/10707 [1:30:41<01:55,  2.02it/s]
 98%|█████████▊| 10474/10707 [1:30:41<01:55,  2.02it/s]
 98%|█████████▊| 10475/10707 [1:30:42<01:54,  2.02it/s]{'loss': 3.5785, 'grad_norm': 0.18979234993457794, 'learning_rate': 1.4296017679135686e-06, 'epoch': 0.98}
-
                                                       

 98%|█████████▊| 10475/10707 [1:30:42<01:54,  2.02it/s]
 98%|█████████▊| 10476/10707 [1:30:42<01:54,  2.02it/s]
 98%|█████████▊| 10477/10707 [1:30:43<01:54,  2.02it/s]
 98%|█████████▊| 10478/10707 [1:30:43<01:53,  2.02it/s]
 98%|█████████▊| 10479/10707 [1:30:44<01:53,  2.02it/s]
 98%|█████████▊| 10480/10707 [1:30:44<01:52,  2.02it/s]
 98%|█████████▊| 10481/10707 [1:30:45<01:51,  2.02it/s]
 98%|█████████▊| 10482/10707 [1:30:45<01:51,  2.02it/s]
 98%|█████████▊| 10483/10707 [1:30:46<01:50,  2.02it/s]
 98%|█████████▊| 10484/10707 [1:30:46<01:50,  2.02it/s]
 98%|█████████▊| 10485/10707 [1:30:47<01:49,  2.02it/s]
 98%|█████████▊| 10486/10707 [1:30:47<01:49,  2.02it/s]
 98%|█████████▊| 10487/10707 [1:30:48<01:48,  2.02it/s]
 98%|█████████▊| 10488/10707 [1:30:48<01:48,  2.02it/s]
 98%|█████████▊| 10489/10707 [1:30:49<01:47,  2.02it/s]
 98%|█████████▊| 10490/10707 [1:30:49<01:47,  2.02it/s]
 98%|█████████▊| 10491/10707 [1:30:50<01:46,  2.02it/s]
 98%|█████████▊| 10492/10707 [1:30:50<01:46,  2.02it/s]
 98%|█████████▊| 10493/10707 [1:30:51<01:45,  2.02it/s]
 98%|█████████▊| 10494/10707 [1:30:51<01:45,  2.02it/s]
 98%|█████████▊| 10495/10707 [1:30:52<01:44,  2.02it/s]
 98%|█████████▊| 10496/10707 [1:30:52<01:44,  2.02it/s]
 98%|█████████▊| 10497/10707 [1:30:53<01:44,  2.02it/s]
 98%|█████████▊| 10498/10707 [1:30:53<01:43,  2.02it/s]
 98%|█████████▊| 10499/10707 [1:30:54<01:43,  2.02it/s]
 98%|█████████▊| 10500/10707 [1:30:54<01:42,  2.02it/s]{'loss': 3.5768, 'grad_norm': 0.18976907432079315, 'learning_rate': 1.1382090145680701e-06, 'epoch': 0.98}
-
                                                       

 98%|█████████▊| 10500/10707 [1:30:54<01:42,  2.02it/s]
 98%|█████████▊| 10501/10707 [1:30:55<01:42,  2.02it/s]
 98%|█████████▊| 10502/10707 [1:30:55<01:41,  2.02it/s]
 98%|█████████▊| 10503/10707 [1:30:56<01:41,  2.02it/s]
 98%|█████████▊| 10504/10707 [1:30:56<01:40,  2.02it/s]
 98%|█████████▊| 10505/10707 [1:30:57<01:39,  2.02it/s]
 98%|█████████▊| 10506/10707 [1:30:57<01:39,  2.02it/s]
 98%|█████████▊| 10507/10707 [1:30:58<01:39,  2.02it/s]
 98%|█████████▊| 10508/10707 [1:30:58<01:38,  2.02it/s]
 98%|█████████▊| 10509/10707 [1:30:59<01:37,  2.02it/s]
 98%|█████████▊| 10510/10707 [1:30:59<01:37,  2.02it/s]
 98%|█████████▊| 10511/10707 [1:31:00<01:37,  2.02it/s]
 98%|█████████▊| 10512/10707 [1:31:00<01:36,  2.02it/s]
 98%|█████████▊| 10513/10707 [1:31:01<01:36,  2.02it/s]
 98%|█████████▊| 10514/10707 [1:31:01<01:35,  2.02it/s]
 98%|█████████▊| 10515/10707 [1:31:02<01:35,  2.02it/s]
 98%|█████████▊| 10516/10707 [1:31:02<01:34,  2.02it/s]
 98%|█████████▊| 10517/10707 [1:31:03<01:33,  2.02it/s]
 98%|█████████▊| 10518/10707 [1:31:03<01:33,  2.02it/s]
 98%|█████████▊| 10519/10707 [1:31:04<01:33,  2.01it/s]
 98%|█████████▊| 10520/10707 [1:31:04<01:32,  2.01it/s]
 98%|█████████▊| 10521/10707 [1:31:05<01:32,  2.02it/s]
 98%|█████████▊| 10522/10707 [1:31:05<01:31,  2.02it/s]
 98%|█████████▊| 10523/10707 [1:31:06<01:31,  2.02it/s]
 98%|█████████▊| 10524/10707 [1:31:06<01:30,  2.02it/s]
 98%|█████████▊| 10525/10707 [1:31:07<01:30,  2.02it/s]{'loss': 3.5817, 'grad_norm': 0.18831270933151245, 'learning_rate': 8.799571397838379e-07, 'epoch': 0.98}
-
                                                       

 98%|█████████▊| 10525/10707 [1:31:07<01:30,  2.02it/s]
 98%|█████████▊| 10526/10707 [1:31:07<01:29,  2.02it/s]
 98%|█████████▊| 10527/10707 [1:31:08<01:29,  2.02it/s]
 98%|█████████▊| 10528/10707 [1:31:08<01:28,  2.02it/s]
 98%|█████████▊| 10529/10707 [1:31:09<01:28,  2.02it/s]
 98%|█████████▊| 10530/10707 [1:31:09<01:27,  2.02it/s]
 98%|█████████▊| 10531/10707 [1:31:10<01:27,  2.02it/s]
 98%|█████████▊| 10532/10707 [1:31:10<01:26,  2.02it/s]
 98%|█████████▊| 10533/10707 [1:31:11<01:26,  2.02it/s]
 98%|█████████▊| 10534/10707 [1:31:11<01:25,  2.02it/s]
 98%|█████████▊| 10535/10707 [1:31:12<01:25,  2.02it/s]
 98%|█████████▊| 10536/10707 [1:31:12<01:24,  2.02it/s]
 98%|█████████▊| 10537/10707 [1:31:12<01:24,  2.02it/s]
 98%|█████████▊| 10538/10707 [1:31:13<01:23,  2.02it/s]
 98%|█████████▊| 10539/10707 [1:31:13<01:23,  2.02it/s]
 98%|█████████▊| 10540/10707 [1:31:14<01:22,  2.02it/s]
 98%|█████████▊| 10541/10707 [1:31:14<01:22,  2.02it/s]
 98%|█████████▊| 10542/10707 [1:31:15<01:21,  2.02it/s]
 98%|█████████▊| 10543/10707 [1:31:15<01:21,  2.02it/s]
 98%|█████████▊| 10544/10707 [1:31:16<01:20,  2.02it/s]
 98%|█████████▊| 10545/10707 [1:31:16<01:20,  2.02it/s]
 98%|█████████▊| 10546/10707 [1:31:17<01:19,  2.01it/s]
 99%|█████████▊| 10547/10707 [1:31:17<01:19,  2.02it/s]
 99%|█████████▊| 10548/10707 [1:31:18<01:18,  2.02it/s]
 99%|█████████▊| 10549/10707 [1:31:18<01:18,  2.02it/s]
 99%|█████████▊| 10550/10707 [1:31:19<01:17,  2.02it/s]{'loss': 3.5773, 'grad_norm': 0.19100594520568848, 'learning_rate': 6.548633000041382e-07, 'epoch': 0.99}
-
                                                       

 99%|█████████▊| 10550/10707 [1:31:19<01:17,  2.02it/s]
 99%|█████████▊| 10551/10707 [1:31:19<01:17,  2.01it/s]
 99%|█████████▊| 10552/10707 [1:31:20<01:16,  2.02it/s]
 99%|█████████▊| 10553/10707 [1:31:20<01:16,  2.01it/s]
 99%|█████████▊| 10554/10707 [1:31:21<01:15,  2.02it/s]
 99%|█████████▊| 10555/10707 [1:31:21<01:15,  2.02it/s]
 99%|█████████▊| 10556/10707 [1:31:22<01:14,  2.02it/s]
 99%|█████████▊| 10557/10707 [1:31:22<01:14,  2.02it/s]
 99%|█████████▊| 10558/10707 [1:31:23<01:13,  2.02it/s]
 99%|█████████▊| 10559/10707 [1:31:23<01:13,  2.02it/s]
 99%|█████████▊| 10560/10707 [1:31:24<01:12,  2.02it/s]
 99%|█████████▊| 10561/10707 [1:31:24<01:12,  2.02it/s]
 99%|█████████▊| 10562/10707 [1:31:25<01:11,  2.02it/s]
 99%|█████████▊| 10563/10707 [1:31:25<01:11,  2.02it/s]
 99%|█████████▊| 10564/10707 [1:31:26<01:10,  2.02it/s]
 99%|█████████▊| 10565/10707 [1:31:26<01:10,  2.02it/s]
 99%|█████████▊| 10566/10707 [1:31:27<01:09,  2.02it/s]
 99%|█████████▊| 10567/10707 [1:31:27<01:09,  2.02it/s]
 99%|█████████▊| 10568/10707 [1:31:28<01:08,  2.02it/s]
 99%|█████████▊| 10569/10707 [1:31:28<01:08,  2.02it/s]
 99%|█████████▊| 10570/10707 [1:31:29<01:07,  2.02it/s]
 99%|█████████▊| 10571/10707 [1:31:29<01:07,  2.02it/s]
 99%|█████████▊| 10572/10707 [1:31:30<01:06,  2.02it/s]
 99%|█████████▊| 10573/10707 [1:31:30<01:06,  2.02it/s]
 99%|█████████▉| 10574/10707 [1:31:31<01:05,  2.02it/s]
 99%|█████████▉| 10575/10707 [1:31:31<01:05,  2.02it/s]{'loss': 3.5767, 'grad_norm': 0.19225679337978363, 'learning_rate': 4.6294244888500647e-07, 'epoch': 0.99}
-
                                                       

 99%|█████████▉| 10575/10707 [1:31:31<01:05,  2.02it/s]
 99%|█████████▉| 10576/10707 [1:31:32<01:04,  2.02it/s]
 99%|█████████▉| 10577/10707 [1:31:32<01:04,  2.02it/s]
 99%|█████████▉| 10578/10707 [1:31:33<01:03,  2.02it/s]
 99%|█████████▉| 10579/10707 [1:31:33<01:03,  2.01it/s]
 99%|█████████▉| 10580/10707 [1:31:34<01:03,  2.01it/s]
 99%|█████████▉| 10581/10707 [1:31:34<01:02,  2.01it/s]
 99%|█████████▉| 10582/10707 [1:31:35<01:02,  2.02it/s]
 99%|█████████▉| 10583/10707 [1:31:35<01:01,  2.02it/s]
 99%|█████████▉| 10584/10707 [1:31:36<01:00,  2.02it/s]
 99%|█████████▉| 10585/10707 [1:31:36<01:00,  2.02it/s]
 99%|█████████▉| 10586/10707 [1:31:37<00:59,  2.02it/s]
 99%|█████████▉| 10587/10707 [1:31:37<00:59,  2.02it/s]
 99%|█████████▉| 10588/10707 [1:31:38<00:58,  2.02it/s]
 99%|█████████▉| 10589/10707 [1:31:38<00:58,  2.02it/s]
 99%|█████████▉| 10590/10707 [1:31:39<00:57,  2.02it/s]
 99%|█████████▉| 10591/10707 [1:31:39<00:57,  2.02it/s]
 99%|█████████▉| 10592/10707 [1:31:40<00:56,  2.02it/s]
 99%|█████████▉| 10593/10707 [1:31:40<00:56,  2.02it/s]
 99%|█████████▉| 10594/10707 [1:31:41<00:55,  2.02it/s]
 99%|█████████▉| 10595/10707 [1:31:41<00:55,  2.02it/s]
 99%|█████████▉| 10596/10707 [1:31:42<00:55,  2.02it/s]
 99%|█████████▉| 10597/10707 [1:31:42<00:54,  2.02it/s]
 99%|█████████▉| 10598/10707 [1:31:43<00:53,  2.02it/s]
 99%|█████████▉| 10599/10707 [1:31:43<00:53,  2.02it/s]
 99%|█████████▉| 10600/10707 [1:31:44<00:52,  2.02it/s]
                                                       {'loss': 3.5798, 'grad_norm': 0.18852044641971588, 'learning_rate': 3.0420733630165397e-07, 'epoch': 0.99}
-

 99%|█████████▉| 10600/10707 [1:31:44<00:52,  2.02it/s]
 99%|█████████▉| 10601/10707 [1:31:44<00:52,  2.02it/s]
 99%|█████████▉| 10602/10707 [1:31:45<00:52,  2.02it/s]
 99%|█████████▉| 10603/10707 [1:31:45<00:51,  2.02it/s]
 99%|█████████▉| 10604/10707 [1:31:46<00:51,  2.02it/s]
 99%|█████████▉| 10605/10707 [1:31:46<00:50,  2.02it/s]
 99%|█████████▉| 10606/10707 [1:31:47<00:50,  2.02it/s]
 99%|█████████▉| 10607/10707 [1:31:47<00:49,  2.02it/s]
 99%|█████████▉| 10608/10707 [1:31:48<00:49,  2.02it/s]
 99%|█████████▉| 10609/10707 [1:31:48<00:48,  2.02it/s]
 99%|█████████▉| 10610/10707 [1:31:49<00:48,  2.02it/s]
 99%|█████████▉| 10611/10707 [1:31:49<00:47,  2.02it/s]
 99%|█████████▉| 10612/10707 [1:31:50<00:47,  2.02it/s]
 99%|█████████▉| 10613/10707 [1:31:50<00:46,  2.02it/s]
 99%|█████████▉| 10614/10707 [1:31:51<00:46,  2.02it/s]
 99%|█████████▉| 10615/10707 [1:31:51<00:45,  2.02it/s]
 99%|█████████▉| 10616/10707 [1:31:52<00:45,  2.02it/s]
 99%|█████████▉| 10617/10707 [1:31:52<00:44,  2.02it/s]
 99%|█████████▉| 10618/10707 [1:31:53<00:44,  2.02it/s]
 99%|█████████▉| 10619/10707 [1:31:53<00:43,  2.02it/s]
 99%|█████████▉| 10620/10707 [1:31:54<00:43,  2.02it/s]
 99%|█████████▉| 10621/10707 [1:31:54<00:42,  2.02it/s]
 99%|█████████▉| 10622/10707 [1:31:55<00:42,  2.01it/s]
 99%|█████████▉| 10623/10707 [1:31:55<00:41,  2.02it/s]
 99%|█████████▉| 10624/10707 [1:31:56<00:41,  2.02it/s]
 99%|█████████▉| 10625/10707 [1:31:56<00:40,  2.02it/s]{'loss': 3.5878, 'grad_norm': 0.19044116139411926, 'learning_rate': 1.7866850750175578e-07, 'epoch': 0.99}
-
                                                       

 99%|█████████▉| 10625/10707 [1:31:56<00:40,  2.02it/s]
 99%|█████████▉| 10626/10707 [1:31:57<00:40,  2.02it/s]
 99%|█████████▉| 10627/10707 [1:31:57<00:39,  2.02it/s]
 99%|█████████▉| 10628/10707 [1:31:58<00:39,  2.02it/s]
 99%|█████████▉| 10629/10707 [1:31:58<00:38,  2.02it/s]
 99%|█████████▉| 10630/10707 [1:31:59<00:38,  2.02it/s]
 99%|█████████▉| 10631/10707 [1:31:59<00:37,  2.02it/s]
 99%|█████████▉| 10632/10707 [1:32:00<00:37,  2.02it/s]
 99%|█████████▉| 10633/10707 [1:32:00<00:36,  2.01it/s]
 99%|█████████▉| 10634/10707 [1:32:01<00:36,  2.01it/s]
 99%|█████████▉| 10635/10707 [1:32:01<00:35,  2.01it/s]
 99%|█████████▉| 10636/10707 [1:32:02<00:35,  2.01it/s]
 99%|█████████▉| 10637/10707 [1:32:02<00:34,  2.01it/s]
 99%|█████████▉| 10638/10707 [1:32:03<00:34,  2.02it/s]
 99%|█████████▉| 10639/10707 [1:32:03<00:33,  2.01it/s]
 99%|█████████▉| 10640/10707 [1:32:04<00:33,  2.01it/s]
 99%|█████████▉| 10641/10707 [1:32:04<00:32,  2.01it/s]
 99%|█████████▉| 10642/10707 [1:32:05<00:32,  2.02it/s]
 99%|█████████▉| 10643/10707 [1:32:05<00:31,  2.01it/s]
 99%|█████████▉| 10644/10707 [1:32:06<00:31,  2.02it/s]
 99%|█████████▉| 10645/10707 [1:32:06<00:30,  2.02it/s]
 99%|█████████▉| 10646/10707 [1:32:07<00:30,  2.02it/s]
 99%|█████████▉| 10647/10707 [1:32:07<00:29,  2.02it/s]
 99%|█████████▉| 10648/10707 [1:32:07<00:29,  2.02it/s]
 99%|█████████▉| 10649/10707 [1:32:08<00:28,  2.01it/s]
 99%|█████████▉| 10650/10707 [1:32:08<00:28,  2.01it/s]
                                                       {'loss': 3.5712, 'grad_norm': 0.18948182463645935, 'learning_rate': 8.633430240451156e-08, 'epoch': 0.99}
-

 99%|█████████▉| 10650/10707 [1:32:08<00:28,  2.01it/s]
 99%|█████████▉| 10651/10707 [1:32:09<00:27,  2.01it/s]
 99%|█████████▉| 10652/10707 [1:32:09<00:27,  2.01it/s]
 99%|█████████▉| 10653/10707 [1:32:10<00:26,  2.01it/s]
100%|█████████▉| 10654/10707 [1:32:10<00:26,  2.01it/s]
100%|█████████▉| 10655/10707 [1:32:11<00:25,  2.01it/s]
100%|█████████▉| 10656/10707 [1:32:11<00:25,  2.02it/s]
100%|█████████▉| 10657/10707 [1:32:12<00:24,  2.01it/s]
100%|█████████▉| 10658/10707 [1:32:12<00:24,  2.01it/s]
100%|█████████▉| 10659/10707 [1:32:13<00:23,  2.01it/s]
100%|█████████▉| 10660/10707 [1:32:13<00:23,  2.01it/s]
100%|█████████▉| 10661/10707 [1:32:14<00:22,  2.02it/s]
100%|█████████▉| 10662/10707 [1:32:14<00:22,  2.02it/s]
100%|█████████▉| 10663/10707 [1:32:15<00:21,  2.02it/s]
100%|█████████▉| 10664/10707 [1:32:15<00:21,  2.02it/s]
100%|█████████▉| 10665/10707 [1:32:16<00:20,  2.02it/s]
100%|█████████▉| 10666/10707 [1:32:16<00:20,  2.02it/s]
100%|█████████▉| 10667/10707 [1:32:17<00:19,  2.02it/s]
100%|█████████▉| 10668/10707 [1:32:17<00:19,  2.02it/s]
100%|█████████▉| 10669/10707 [1:32:18<00:18,  2.02it/s]
100%|█████████▉| 10670/10707 [1:32:18<00:18,  2.02it/s]
100%|█████████▉| 10671/10707 [1:32:19<00:17,  2.02it/s]
100%|█████████▉| 10672/10707 [1:32:19<00:17,  2.02it/s]
100%|█████████▉| 10673/10707 [1:32:20<00:16,  2.02it/s]
100%|█████████▉| 10674/10707 [1:32:20<00:16,  2.02it/s]
100%|█████████▉| 10675/10707 [1:32:21<00:15,  2.02it/s]
                                                       
{'loss': 3.5792, 'grad_norm': 0.1904916763305664, 'learning_rate': 2.721085504692189e-08, 'epoch': 1.0}
-
100%|█████████▉| 10675/10707 [1:32:21<00:15,  2.02it/s]
100%|█████████▉| 10676/10707 [1:32:21<00:15,  2.02it/s]
100%|█████████▉| 10677/10707 [1:32:22<00:14,  2.02it/s]
100%|█████████▉| 10678/10707 [1:32:22<00:14,  2.02it/s]
100%|█████████▉| 10679/10707 [1:32:23<00:13,  2.02it/s]
100%|█████████▉| 10680/10707 [1:32:23<00:13,  2.02it/s]
100%|█████████▉| 10681/10707 [1:32:24<00:12,  2.01it/s]
100%|█████████▉| 10682/10707 [1:32:24<00:12,  2.01it/s]
100%|█████████▉| 10683/10707 [1:32:25<00:11,  2.02it/s]
100%|█████████▉| 10684/10707 [1:32:25<00:11,  2.02it/s]
100%|█████████▉| 10685/10707 [1:32:26<00:10,  2.02it/s]
100%|█████████▉| 10686/10707 [1:32:26<00:10,  2.02it/s]
100%|█████████▉| 10687/10707 [1:32:27<00:09,  2.02it/s]
100%|█████████▉| 10688/10707 [1:32:27<00:09,  2.02it/s]
100%|█████████▉| 10689/10707 [1:32:28<00:08,  2.02it/s]
100%|█████████▉| 10690/10707 [1:32:28<00:08,  2.02it/s]
100%|█████████▉| 10691/10707 [1:32:29<00:07,  2.02it/s]
100%|█████████▉| 10692/10707 [1:32:29<00:07,  2.02it/s]
100%|█████████▉| 10693/10707 [1:32:30<00:06,  2.02it/s]
100%|█████████▉| 10694/10707 [1:32:30<00:06,  2.02it/s]
100%|█████████▉| 10695/10707 [1:32:31<00:05,  2.02it/s]
100%|█████████▉| 10696/10707 [1:32:31<00:05,  2.02it/s]
100%|█████████▉| 10697/10707 [1:32:32<00:04,  2.02it/s]
100%|█████████▉| 10698/10707 [1:32:32<00:04,  2.02it/s]
100%|█████████▉| 10699/10707 [1:32:33<00:03,  2.02it/s]
100%|█████████▉| 10700/10707 [1:32:33<00:03,  2.01it/s]{'loss': 3.5731, 'grad_norm': 0.19028829038143158, 'learning_rate': 1.3020931762253163e-09, 'epoch': 1.0}
                                                       
-
100%|█████████▉| 10700/10707 [1:32:33<00:03,  2.01it/s]
100%|█████████▉| 10701/10707 [1:32:34<00:02,  2.01it/s]
100%|█████████▉| 10702/10707 [1:32:34<00:02,  2.01it/s]
100%|█████████▉| 10703/10707 [1:32:35<00:01,  2.02it/s]
100%|█████████▉| 10704/10707 [1:32:35<00:01,  2.01it/s]
100%|█████████▉| 10705/10707 [1:32:36<00:00,  2.01it/s]
100%|█████████▉| 10706/10707 [1:32:36<00:00,  2.02it/s]
100%|██████████| 10707/10707 [1:32:38<00:00,  1.31it/s]
                                                       
{'train_runtime': 5653.3886, 'train_samples_per_second': 1939.363, 'train_steps_per_second': 1.894, 'train_loss': 3.95905240697443, 'epoch': 1.0}
-
100%|██████████| 10707/10707 [1:34:13<00:00,  1.31it/s]
100%|██████████| 10707/10707 [1:34:13<00:00,  1.89it/s]
-Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
+[2024-05-24 13:44:11,920] torch.distributed.run: [WARNING] 
+[2024-05-24 13:44:11,920] torch.distributed.run: [WARNING] *****************************************
+[2024-05-24 13:44:11,920] torch.distributed.run: [WARNING] Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. 
+[2024-05-24 13:44:11,920] torch.distributed.run: [WARNING] *****************************************
+05/24/2024 13:44:15 - INFO - __main__ - Script parameters ScriptArguments(seed=1, dataset_id='/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_initial_init_min_threshold/data/arc_easy', output_dir='/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_initial_init_min_threshold/llms/pythia-70m_arc_easy_1', output_hub_id='pythia-70m_arc_easy', hf_hub_token=True, model_id='EleutherAI/pythia-70m', per_device_train_batch_size=256, num_train_epochs=14.0, learning_rate=0.001, gradient_accumulation_steps=2, from_scratch=True, warmup_ratio=0.1, adam_beta1=0.9, adam_beta2=0.95, adam_epsilon=1e-08, weight_decay=0.01, lr_scheduler_type='cosine', local_rank=0, resume_from_checkpoint=False, deepspeed=None, peft=False)
+05/24/2024 13:44:15 - INFO - __main__ - Script parameters ScriptArguments(seed=1, dataset_id='/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_initial_init_min_threshold/data/arc_easy', output_dir='/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_initial_init_min_threshold/llms/pythia-70m_arc_easy_1', output_hub_id='pythia-70m_arc_easy', hf_hub_token=True, model_id='EleutherAI/pythia-70m', per_device_train_batch_size=256, num_train_epochs=14.0, learning_rate=0.001, gradient_accumulation_steps=2, from_scratch=True, warmup_ratio=0.1, adam_beta1=0.9, adam_beta2=0.95, adam_epsilon=1e-08, weight_decay=0.01, lr_scheduler_type='cosine', local_rank=0, resume_from_checkpoint=False, deepspeed=None, peft=False)
+
  0%|          | 0/11088 [00:00<?, ?it/s][rank0]:[W reducer.cpp:1360] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters in the forward pass. This flag results in an extra traversal of the autograd graph every iteration,  which can adversely affect performance. If your model indeed never has any unused parameters in the forward pass, consider turning this flag off. Note that this warning may be a false positive if your model has flow control causing later iterations to have unused parameters. (function operator())
+[rank1]:[W reducer.cpp:1360] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters in the forward pass. This flag results in an extra traversal of the autograd graph every iteration,  which can adversely affect performance. If your model indeed never has any unused parameters in the forward pass, consider turning this flag off. Note that this warning may be a false positive if your model has flow control causing later iterations to have unused parameters. (function operator())
+
  0%|          | 1/11088 [00:05<18:11:48,  5.91s/it]
  0%|          | 2/11088 [00:07<11:10:40,  3.63s/it]
  0%|          | 3/11088 [00:09<7:59:56,  2.60s/it] 
  0%|          | 4/11088 [00:10<6:17:01,  2.04s/it]
  0%|          | 5/11088 [00:11<5:08:34,  1.67s/it]
  0%|          | 6/11088 [00:12<4:21:17,  1.41s/it]
  0%|          | 7/11088 [00:13<3:48:51,  1.24s/it]
  0%|          | 8/11088 [00:14<3:21:15,  1.09s/it]
  0%|          | 9/11088 [00:14<3:01:54,  1.02it/s]
  0%|          | 10/11088 [00:15<2:47:43,  1.10it/s]
  0%|          | 11/11088 [00:16<2:37:04,  1.18it/s]
  0%|          | 12/11088 [00:16<2:27:00,  1.26it/s]
  0%|          | 13/11088 [00:17<2:18:02,  1.34it/s]
  0%|          | 14/11088 [00:18<2:11:40,  1.40it/s]
  0%|          | 15/11088 [00:18<2:04:46,  1.48it/s]
  0%|          | 16/11088 [00:19<1:59:30,  1.54it/s]
  0%|          | 17/11088 [00:19<1:55:27,  1.60it/s]
  0%|          | 18/11088 [00:20<1:52:06,  1.65it/s]
  0%|          | 19/11088 [00:21<1:49:51,  1.68it/s]
  0%|          | 20/11088 [00:21<1:47:13,  1.72it/s]
  0%|          | 21/11088 [00:22<1:44:37,  1.76it/s]
  0%|          | 22/11088 [00:22<1:43:03,  1.79it/s]
  0%|          | 23/11088 [00:23<1:41:58,  1.81it/s]
  0%|          | 24/11088 [00:23<1:40:25,  1.84it/s]
  0%|          | 25/11088 [00:24<1:39:43,  1.85it/s]
                                                    
{'loss': 10.6741, 'grad_norm': 1.357293725013733, 'learning_rate': 2.2542831379621283e-05, 'epoch': 0.03}
+
  0%|          | 25/11088 [00:24<1:39:43,  1.85it/s]
  0%|          | 26/11088 [00:24<1:39:06,  1.86it/s]
  0%|          | 27/11088 [00:25<1:37:32,  1.89it/s]
  0%|          | 28/11088 [00:25<1:36:50,  1.90it/s]
  0%|          | 29/11088 [00:26<1:36:29,  1.91it/s]
  0%|          | 30/11088 [00:26<1:36:00,  1.92it/s]
  0%|          | 31/11088 [00:27<1:35:50,  1.92it/s]
  0%|          | 32/11088 [00:27<1:35:18,  1.93it/s]
  0%|          | 33/11088 [00:28<1:34:52,  1.94it/s]
  0%|          | 34/11088 [00:28<1:34:35,  1.95it/s]
  0%|          | 35/11088 [00:29<1:34:04,  1.96it/s]
  0%|          | 36/11088 [00:29<1:33:43,  1.97it/s]
  0%|          | 37/11088 [00:30<1:33:44,  1.96it/s]
  0%|          | 38/11088 [00:30<1:33:23,  1.97it/s]
  0%|          | 39/11088 [00:31<1:33:06,  1.98it/s]
  0%|          | 40/11088 [00:31<1:32:50,  1.98it/s]
  0%|          | 41/11088 [00:32<1:32:43,  1.99it/s]
  0%|          | 42/11088 [00:32<1:32:28,  1.99it/s]
  0%|          | 43/11088 [00:33<1:32:19,  1.99it/s]
  0%|          | 44/11088 [00:33<1:32:16,  1.99it/s]
  0%|          | 45/11088 [00:34<1:32:17,  1.99it/s]
  0%|          | 46/11088 [00:34<1:32:11,  2.00it/s]
  0%|          | 47/11088 [00:35<1:32:07,  2.00it/s]
  0%|          | 48/11088 [00:35<1:32:06,  2.00it/s]
  0%|          | 49/11088 [00:36<1:32:09,  2.00it/s]
  0%|          | 50/11088 [00:36<1:32:04,  2.00it/s]
                                                    
{'loss': 9.9577, 'grad_norm': 1.2331771850585938, 'learning_rate': 4.508566275924257e-05, 'epoch': 0.06}
+
  0%|          | 50/11088 [00:37<1:32:04,  2.00it/s]
  0%|          | 51/11088 [00:37<1:31:59,  2.00it/s]
  0%|          | 52/11088 [00:37<1:31:51,  2.00it/s]
  0%|          | 53/11088 [00:38<1:31:46,  2.00it/s]
  0%|          | 54/11088 [00:38<1:31:42,  2.01it/s]
  0%|          | 55/11088 [00:39<1:31:38,  2.01it/s]
  1%|          | 56/11088 [00:39<1:31:27,  2.01it/s]
  1%|          | 57/11088 [00:40<1:31:25,  2.01it/s]
  1%|          | 58/11088 [00:40<1:31:28,  2.01it/s]
  1%|          | 59/11088 [00:41<1:31:30,  2.01it/s]
  1%|          | 60/11088 [00:41<1:31:32,  2.01it/s]
  1%|          | 61/11088 [00:42<1:31:24,  2.01it/s]
  1%|          | 62/11088 [00:42<1:31:25,  2.01it/s]
  1%|          | 63/11088 [00:43<1:31:25,  2.01it/s]
  1%|          | 64/11088 [00:43<1:31:21,  2.01it/s]
  1%|          | 65/11088 [00:44<1:31:23,  2.01it/s]
  1%|          | 66/11088 [00:44<1:31:19,  2.01it/s]
  1%|          | 67/11088 [00:45<1:31:20,  2.01it/s]
  1%|          | 68/11088 [00:45<1:31:21,  2.01it/s]
  1%|          | 69/11088 [00:46<1:31:17,  2.01it/s]
  1%|          | 70/11088 [00:46<1:31:06,  2.02it/s]
  1%|          | 71/11088 [00:47<1:31:11,  2.01it/s]
  1%|          | 72/11088 [00:47<1:31:02,  2.02it/s]
  1%|          | 73/11088 [00:48<1:31:05,  2.02it/s]
  1%|          | 74/11088 [00:48<1:31:06,  2.01it/s]
  1%|          | 75/11088 [00:49<1:31:02,  2.02it/s]
                                                    
{'loss': 9.2363, 'grad_norm': 1.0550384521484375, 'learning_rate': 6.762849413886383e-05, 'epoch': 0.09}
+
  1%|          | 75/11088 [00:49<1:31:02,  2.02it/s]
  1%|          | 76/11088 [00:49<1:31:06,  2.01it/s]
  1%|          | 77/11088 [00:50<1:31:08,  2.01it/s]
  1%|          | 78/11088 [00:50<1:31:05,  2.01it/s]
  1%|          | 79/11088 [00:51<1:30:55,  2.02it/s]
  1%|          | 80/11088 [00:51<1:30:56,  2.02it/s]
  1%|          | 81/11088 [00:52<1:30:54,  2.02it/s]
  1%|          | 82/11088 [00:52<1:30:55,  2.02it/s]
  1%|          | 83/11088 [00:53<1:30:58,  2.02it/s]
  1%|          | 84/11088 [00:53<1:30:54,  2.02it/s]
  1%|          | 85/11088 [00:54<1:30:57,  2.02it/s]
  1%|          | 86/11088 [00:54<1:30:52,  2.02it/s]
  1%|          | 87/11088 [00:55<1:30:51,  2.02it/s]
  1%|          | 88/11088 [00:55<1:30:47,  2.02it/s]
  1%|          | 89/11088 [00:56<1:30:49,  2.02it/s]
  1%|          | 90/11088 [00:56<1:30:45,  2.02it/s]
  1%|          | 91/11088 [00:57<1:30:43,  2.02it/s]
  1%|          | 92/11088 [00:57<1:30:44,  2.02it/s]
  1%|          | 93/11088 [00:58<1:30:42,  2.02it/s]
  1%|          | 94/11088 [00:58<1:30:40,  2.02it/s]
  1%|          | 95/11088 [00:59<1:30:36,  2.02it/s]
  1%|          | 96/11088 [00:59<1:30:42,  2.02it/s]
  1%|          | 97/11088 [01:00<1:30:37,  2.02it/s]
  1%|          | 98/11088 [01:00<1:30:39,  2.02it/s]
  1%|          | 99/11088 [01:01<1:30:41,  2.02it/s]
  1%|          | 100/11088 [01:01<1:30:42,  2.02it/s]
                                                     
{'loss': 8.4477, 'grad_norm': 0.7622587084770203, 'learning_rate': 9.017132551848513e-05, 'epoch': 0.13}
+
  1%|          | 100/11088 [01:01<1:30:42,  2.02it/s]
  1%|          | 101/11088 [01:02<1:30:42,  2.02it/s]
  1%|          | 102/11088 [01:02<1:30:41,  2.02it/s]
  1%|          | 103/11088 [01:03<1:30:38,  2.02it/s]
  1%|          | 104/11088 [01:03<1:30:41,  2.02it/s]
  1%|          | 105/11088 [01:04<1:30:47,  2.02it/s]
  1%|          | 106/11088 [01:04<1:30:40,  2.02it/s]
  1%|          | 107/11088 [01:05<1:30:41,  2.02it/s]
  1%|          | 108/11088 [01:05<1:30:34,  2.02it/s]
  1%|          | 109/11088 [01:06<1:30:37,  2.02it/s]
  1%|          | 110/11088 [01:06<1:30:32,  2.02it/s]
  1%|          | 111/11088 [01:07<1:30:37,  2.02it/s]
  1%|          | 112/11088 [01:07<1:30:37,  2.02it/s]
  1%|          | 113/11088 [01:08<1:30:38,  2.02it/s]
  1%|          | 114/11088 [01:08<1:30:34,  2.02it/s]
  1%|          | 115/11088 [01:09<1:30:32,  2.02it/s]
  1%|          | 116/11088 [01:09<1:30:29,  2.02it/s]
  1%|          | 117/11088 [01:10<1:30:26,  2.02it/s]
  1%|          | 118/11088 [01:10<1:30:26,  2.02it/s]
  1%|          | 119/11088 [01:11<1:30:23,  2.02it/s]
  1%|          | 120/11088 [01:11<1:30:19,  2.02it/s]
  1%|          | 121/11088 [01:12<1:30:26,  2.02it/s]
  1%|          | 122/11088 [01:12<1:30:24,  2.02it/s]
  1%|          | 123/11088 [01:13<1:30:22,  2.02it/s]
  1%|          | 124/11088 [01:13<1:30:24,  2.02it/s]
  1%|          | 125/11088 [01:14<1:30:24,  2.02it/s]
                                                     
{'loss': 7.7892, 'grad_norm': 0.4769901633262634, 'learning_rate': 0.00011271415689810641, 'epoch': 0.16}
+
  1%|          | 125/11088 [01:14<1:30:24,  2.02it/s]
  1%|          | 126/11088 [01:14<1:30:29,  2.02it/s]
  1%|          | 127/11088 [01:15<1:30:23,  2.02it/s]
  1%|          | 128/11088 [01:15<1:30:20,  2.02it/s]
  1%|          | 129/11088 [01:16<1:30:18,  2.02it/s]
  1%|          | 130/11088 [01:16<1:30:16,  2.02it/s]
  1%|          | 131/11088 [01:17<1:30:19,  2.02it/s]
  1%|          | 132/11088 [01:17<1:30:16,  2.02it/s]
  1%|          | 133/11088 [01:18<1:30:17,  2.02it/s]
  1%|          | 134/11088 [01:18<1:30:16,  2.02it/s]
  1%|          | 135/11088 [01:19<1:30:16,  2.02it/s]
  1%|          | 136/11088 [01:19<1:30:17,  2.02it/s]
  1%|          | 137/11088 [01:20<1:30:13,  2.02it/s]
  1%|          | 138/11088 [01:20<1:30:14,  2.02it/s]
  1%|▏         | 139/11088 [01:21<1:30:15,  2.02it/s]
  1%|▏         | 140/11088 [01:21<1:30:16,  2.02it/s]
  1%|▏         | 141/11088 [01:22<1:30:16,  2.02it/s]
  1%|▏         | 142/11088 [01:22<1:30:19,  2.02it/s]
  1%|▏         | 143/11088 [01:23<1:30:15,  2.02it/s]
  1%|▏         | 144/11088 [01:23<1:30:12,  2.02it/s]
  1%|▏         | 145/11088 [01:24<1:30:09,  2.02it/s]
  1%|▏         | 146/11088 [01:24<1:30:08,  2.02it/s]
  1%|▏         | 147/11088 [01:25<1:30:04,  2.02it/s]
  1%|▏         | 148/11088 [01:25<1:30:00,  2.03it/s]
  1%|▏         | 149/11088 [01:26<1:30:04,  2.02it/s]
  1%|▏         | 150/11088 [01:26<1:30:01,  2.02it/s]{'loss': 7.2988, 'grad_norm': 0.42800745368003845, 'learning_rate': 0.00013525698827772766, 'epoch': 0.19}
                                                     
+
  1%|▏         | 150/11088 [01:26<1:30:01,  2.02it/s]
  1%|▏         | 151/11088 [01:27<1:30:07,  2.02it/s]
  1%|▏         | 152/11088 [01:27<1:30:02,  2.02it/s]
  1%|▏         | 153/11088 [01:28<1:30:05,  2.02it/s]
  1%|▏         | 154/11088 [01:28<1:30:04,  2.02it/s]
  1%|▏         | 155/11088 [01:29<1:30:07,  2.02it/s]
  1%|▏         | 156/11088 [01:29<1:29:59,  2.02it/s]
  1%|▏         | 157/11088 [01:30<1:30:05,  2.02it/s]
  1%|▏         | 158/11088 [01:30<1:30:03,  2.02it/s]
  1%|▏         | 159/11088 [01:30<1:30:06,  2.02it/s]
  1%|▏         | 160/11088 [01:31<1:30:00,  2.02it/s]
  1%|▏         | 161/11088 [01:31<1:30:05,  2.02it/s]
  1%|▏         | 162/11088 [01:32<1:29:57,  2.02it/s]
  1%|▏         | 163/11088 [01:32<1:29:58,  2.02it/s]
  1%|▏         | 164/11088 [01:33<1:29:57,  2.02it/s]
  1%|▏         | 165/11088 [01:33<1:29:56,  2.02it/s]
  1%|▏         | 166/11088 [01:34<1:29:56,  2.02it/s]
  2%|▏         | 167/11088 [01:34<1:29:57,  2.02it/s]
  2%|▏         | 168/11088 [01:35<1:29:55,  2.02it/s]
  2%|▏         | 169/11088 [01:35<1:29:50,  2.03it/s]
  2%|▏         | 170/11088 [01:36<1:29:52,  2.02it/s]
  2%|▏         | 171/11088 [01:36<1:29:51,  2.02it/s]
  2%|▏         | 172/11088 [01:37<1:29:53,  2.02it/s]
  2%|▏         | 173/11088 [01:37<1:29:56,  2.02it/s]
  2%|▏         | 174/11088 [01:38<1:29:55,  2.02it/s]
  2%|▏         | 175/11088 [01:38<1:29:55,  2.02it/s]{'loss': 6.8842, 'grad_norm': 0.3714645206928253, 'learning_rate': 0.00015779981965734896, 'epoch': 0.22}
                                                     
+
  2%|▏         | 175/11088 [01:38<1:29:55,  2.02it/s]
  2%|▏         | 176/11088 [01:39<1:30:04,  2.02it/s]
  2%|▏         | 177/11088 [01:39<1:30:01,  2.02it/s]
  2%|▏         | 178/11088 [01:40<1:29:54,  2.02it/s]
  2%|▏         | 179/11088 [01:40<1:29:57,  2.02it/s]
  2%|▏         | 180/11088 [01:41<1:29:50,  2.02it/s]
  2%|▏         | 181/11088 [01:41<1:29:51,  2.02it/s]
  2%|▏         | 182/11088 [01:42<1:29:49,  2.02it/s]
  2%|▏         | 183/11088 [01:42<1:29:47,  2.02it/s]
  2%|▏         | 184/11088 [01:43<1:29:50,  2.02it/s]
  2%|▏         | 185/11088 [01:43<1:29:47,  2.02it/s]
  2%|▏         | 186/11088 [01:44<1:29:50,  2.02it/s]
  2%|▏         | 187/11088 [01:44<1:29:47,  2.02it/s]
  2%|▏         | 188/11088 [01:45<1:29:52,  2.02it/s]
  2%|▏         | 189/11088 [01:45<1:29:53,  2.02it/s]
  2%|▏         | 190/11088 [01:46<1:29:53,  2.02it/s]
  2%|▏         | 191/11088 [01:46<1:29:50,  2.02it/s]
  2%|▏         | 192/11088 [01:47<1:29:49,  2.02it/s]
  2%|▏         | 193/11088 [01:47<1:29:48,  2.02it/s]
  2%|▏         | 194/11088 [01:48<1:29:50,  2.02it/s]
  2%|▏         | 195/11088 [01:48<1:29:51,  2.02it/s]
  2%|▏         | 196/11088 [01:49<1:29:51,  2.02it/s]
  2%|▏         | 197/11088 [01:49<1:29:54,  2.02it/s]
  2%|▏         | 198/11088 [01:50<1:29:49,  2.02it/s]
  2%|▏         | 199/11088 [01:50<1:29:51,  2.02it/s]
  2%|▏         | 200/11088 [01:51<1:29:44,  2.02it/s]
                                                     {'loss': 6.5354, 'grad_norm': 0.5754061341285706, 'learning_rate': 0.00018034265103697027, 'epoch': 0.25}
+

  2%|▏         | 200/11088 [01:51<1:29:44,  2.02it/s]
  2%|▏         | 201/11088 [01:51<1:29:56,  2.02it/s]
  2%|▏         | 202/11088 [01:52<1:29:49,  2.02it/s]
  2%|▏         | 203/11088 [01:52<1:29:53,  2.02it/s]
  2%|▏         | 204/11088 [01:53<1:29:47,  2.02it/s]
  2%|▏         | 205/11088 [01:53<1:29:52,  2.02it/s]
  2%|▏         | 206/11088 [01:54<1:29:42,  2.02it/s]
  2%|▏         | 207/11088 [01:54<1:29:46,  2.02it/s]
  2%|▏         | 208/11088 [01:55<1:29:43,  2.02it/s]
  2%|▏         | 209/11088 [01:55<1:29:45,  2.02it/s]
  2%|▏         | 210/11088 [01:56<1:29:46,  2.02it/s]
  2%|▏         | 211/11088 [01:56<1:29:39,  2.02it/s]
  2%|▏         | 212/11088 [01:57<1:29:33,  2.02it/s]
  2%|▏         | 213/11088 [01:57<1:29:36,  2.02it/s]
  2%|▏         | 214/11088 [01:58<1:29:31,  2.02it/s]
  2%|▏         | 215/11088 [01:58<1:29:33,  2.02it/s]
  2%|▏         | 216/11088 [01:59<1:29:26,  2.03it/s]
  2%|▏         | 217/11088 [01:59<1:29:31,  2.02it/s]
  2%|▏         | 218/11088 [02:00<1:29:27,  2.02it/s]
  2%|▏         | 219/11088 [02:00<1:29:36,  2.02it/s]
  2%|▏         | 220/11088 [02:01<1:29:28,  2.02it/s]
  2%|▏         | 221/11088 [02:01<1:29:33,  2.02it/s]
  2%|▏         | 222/11088 [02:02<1:29:29,  2.02it/s]
  2%|▏         | 223/11088 [02:02<1:29:32,  2.02it/s]
  2%|▏         | 224/11088 [02:03<1:29:24,  2.03it/s]
  2%|▏         | 225/11088 [02:03<1:29:25,  2.02it/s]
                                                     {'loss': 6.2653, 'grad_norm': 0.6229713559150696, 'learning_rate': 0.00020288548241659152, 'epoch': 0.28}
+
  2%|▏         | 225/11088 [02:03<1:29:25,  2.02it/s]
  2%|▏         | 226/11088 [02:04<1:29:31,  2.02it/s]
  2%|▏         | 227/11088 [02:04<1:29:32,  2.02it/s]
  2%|▏         | 228/11088 [02:05<1:29:25,  2.02it/s]
  2%|▏         | 229/11088 [02:05<1:29:30,  2.02it/s]
  2%|▏         | 230/11088 [02:06<1:29:28,  2.02it/s]
  2%|▏         | 231/11088 [02:06<1:29:30,  2.02it/s]
  2%|▏         | 232/11088 [02:07<1:29:26,  2.02it/s]
  2%|▏         | 233/11088 [02:07<1:29:26,  2.02it/s]
  2%|▏         | 234/11088 [02:08<1:29:21,  2.02it/s]
  2%|▏         | 235/11088 [02:08<1:29:20,  2.02it/s]
  2%|▏         | 236/11088 [02:09<1:29:19,  2.02it/s]
  2%|▏         | 237/11088 [02:09<1:29:14,  2.03it/s]
  2%|▏         | 238/11088 [02:10<1:29:18,  2.02it/s]
  2%|▏         | 239/11088 [02:10<1:29:13,  2.03it/s]
  2%|▏         | 240/11088 [02:11<1:29:16,  2.03it/s]
  2%|▏         | 241/11088 [02:11<1:29:13,  2.03it/s]
  2%|▏         | 242/11088 [02:12<1:29:19,  2.02it/s]
  2%|▏         | 243/11088 [02:12<1:29:11,  2.03it/s]
  2%|▏         | 244/11088 [02:13<1:29:18,  2.02it/s]
  2%|▏         | 245/11088 [02:13<1:29:13,  2.03it/s]
  2%|▏         | 246/11088 [02:14<1:29:19,  2.02it/s]
  2%|▏         | 247/11088 [02:14<1:29:13,  2.02it/s]
  2%|▏         | 248/11088 [02:14<1:29:18,  2.02it/s]
  2%|▏         | 249/11088 [02:15<1:29:14,  2.02it/s]
  2%|▏         | 250/11088 [02:15<1:29:14,  2.02it/s]{'loss': 6.0594, 'grad_norm': 0.6278337836265564, 'learning_rate': 0.00022542831379621282, 'epoch': 0.32}
                                                     
+
  2%|▏         | 250/11088 [02:15<1:29:14,  2.02it/s]
  2%|▏         | 251/11088 [02:16<1:29:20,  2.02it/s]
  2%|▏         | 252/11088 [02:16<1:29:18,  2.02it/s]
  2%|▏         | 253/11088 [02:17<1:29:19,  2.02it/s]
  2%|▏         | 254/11088 [02:17<1:29:18,  2.02it/s]
  2%|▏         | 255/11088 [02:18<1:29:18,  2.02it/s]
  2%|▏         | 256/11088 [02:18<1:29:19,  2.02it/s]
  2%|▏         | 257/11088 [02:19<1:29:20,  2.02it/s]
  2%|▏         | 258/11088 [02:19<1:29:19,  2.02it/s]
  2%|▏         | 259/11088 [02:20<1:29:18,  2.02it/s]
  2%|▏         | 260/11088 [02:20<1:29:15,  2.02it/s]
  2%|▏         | 261/11088 [02:21<1:29:17,  2.02it/s]
  2%|▏         | 262/11088 [02:21<1:29:13,  2.02it/s]
  2%|▏         | 263/11088 [02:22<1:29:14,  2.02it/s]
  2%|▏         | 264/11088 [02:22<1:29:15,  2.02it/s]
  2%|▏         | 265/11088 [02:23<1:29:14,  2.02it/s]
  2%|▏         | 266/11088 [02:23<1:29:14,  2.02it/s]
  2%|▏         | 267/11088 [02:24<1:29:12,  2.02it/s]
  2%|▏         | 268/11088 [02:24<1:29:10,  2.02it/s]
  2%|▏         | 269/11088 [02:25<1:29:03,  2.02it/s]
  2%|▏         | 270/11088 [02:25<1:29:06,  2.02it/s]
  2%|▏         | 271/11088 [02:26<1:28:59,  2.03it/s]
  2%|▏         | 272/11088 [02:26<1:29:05,  2.02it/s]
  2%|▏         | 273/11088 [02:27<1:28:59,  2.03it/s]
  2%|▏         | 274/11088 [02:27<1:29:03,  2.02it/s]
  2%|▏         | 275/11088 [02:28<1:29:01,  2.02it/s]
                                                     {'loss': 5.8876, 'grad_norm': 0.6381323337554932, 'learning_rate': 0.00024797114517583407, 'epoch': 0.35}
+
  2%|▏         | 275/11088 [02:28<1:29:01,  2.02it/s]
  2%|▏         | 276/11088 [02:28<1:29:07,  2.02it/s]
  2%|▏         | 277/11088 [02:29<1:29:06,  2.02it/s]
  3%|▎         | 278/11088 [02:29<1:29:04,  2.02it/s]
  3%|▎         | 279/11088 [02:30<1:29:05,  2.02it/s]
  3%|▎         | 280/11088 [02:30<1:29:07,  2.02it/s]
  3%|▎         | 281/11088 [02:31<1:29:02,  2.02it/s]
  3%|▎         | 282/11088 [02:31<1:29:06,  2.02it/s]
  3%|▎         | 283/11088 [02:32<1:29:02,  2.02it/s]
  3%|▎         | 284/11088 [02:32<1:29:03,  2.02it/s]
  3%|▎         | 285/11088 [02:33<1:29:00,  2.02it/s]
  3%|▎         | 286/11088 [02:33<1:29:01,  2.02it/s]
  3%|▎         | 287/11088 [02:34<1:29:00,  2.02it/s]
  3%|▎         | 288/11088 [02:34<1:28:59,  2.02it/s]
  3%|▎         | 289/11088 [02:35<1:28:59,  2.02it/s]
  3%|▎         | 290/11088 [02:35<1:28:59,  2.02it/s]
  3%|▎         | 291/11088 [02:36<1:29:01,  2.02it/s]
  3%|▎         | 292/11088 [02:36<1:29:03,  2.02it/s]
  3%|▎         | 293/11088 [02:37<1:29:01,  2.02it/s]
  3%|▎         | 294/11088 [02:37<1:29:00,  2.02it/s]
  3%|▎         | 295/11088 [02:38<1:28:57,  2.02it/s]
  3%|▎         | 296/11088 [02:38<1:28:58,  2.02it/s]
  3%|▎         | 297/11088 [02:39<1:29:00,  2.02it/s]
  3%|▎         | 298/11088 [02:39<1:28:59,  2.02it/s]
  3%|▎         | 299/11088 [02:40<1:28:59,  2.02it/s]
  3%|▎         | 300/11088 [02:40<1:29:01,  2.02it/s]
                                                     {'loss': 5.7444, 'grad_norm': 1.03485906124115, 'learning_rate': 0.0002705139765554553, 'epoch': 0.38}
+

  3%|▎         | 300/11088 [02:40<1:29:01,  2.02it/s]
  3%|▎         | 301/11088 [02:41<1:29:03,  2.02it/s]
  3%|▎         | 302/11088 [02:41<1:28:59,  2.02it/s]
  3%|▎         | 303/11088 [02:42<1:28:59,  2.02it/s]
  3%|▎         | 304/11088 [02:42<1:28:50,  2.02it/s]
  3%|▎         | 305/11088 [02:43<1:28:52,  2.02it/s]
  3%|▎         | 306/11088 [02:43<1:28:46,  2.02it/s]
  3%|▎         | 307/11088 [02:44<1:28:50,  2.02it/s]
  3%|▎         | 308/11088 [02:44<1:28:46,  2.02it/s]
  3%|▎         | 309/11088 [02:45<1:28:46,  2.02it/s]
  3%|▎         | 310/11088 [02:45<1:28:45,  2.02it/s]
  3%|▎         | 311/11088 [02:46<1:28:46,  2.02it/s]
  3%|▎         | 312/11088 [02:46<1:28:50,  2.02it/s]
  3%|▎         | 313/11088 [02:47<1:28:46,  2.02it/s]
  3%|▎         | 314/11088 [02:47<1:28:41,  2.02it/s]
  3%|▎         | 315/11088 [02:48<1:28:44,  2.02it/s]
  3%|▎         | 316/11088 [02:48<1:28:42,  2.02it/s]
  3%|▎         | 317/11088 [02:49<1:28:42,  2.02it/s]
  3%|▎         | 318/11088 [02:49<1:28:41,  2.02it/s]
  3%|▎         | 319/11088 [02:50<1:28:39,  2.02it/s]
  3%|▎         | 320/11088 [02:50<1:28:39,  2.02it/s]
  3%|▎         | 321/11088 [02:51<1:28:39,  2.02it/s]
  3%|▎         | 322/11088 [02:51<1:28:35,  2.03it/s]
  3%|▎         | 323/11088 [02:52<1:28:38,  2.02it/s]
  3%|▎         | 324/11088 [02:52<1:28:32,  2.03it/s]
  3%|▎         | 325/11088 [02:53<1:28:37,  2.02it/s]
                                                     
{'loss': 5.6197, 'grad_norm': 0.7586981058120728, 'learning_rate': 0.0002930568079350767, 'epoch': 0.41}
+
  3%|▎         | 325/11088 [02:53<1:28:37,  2.02it/s]
  3%|▎         | 326/11088 [02:53<1:28:40,  2.02it/s]
  3%|▎         | 327/11088 [02:54<1:28:41,  2.02it/s]
  3%|▎         | 328/11088 [02:54<1:28:38,  2.02it/s]
  3%|▎         | 329/11088 [02:55<1:28:38,  2.02it/s]
  3%|▎         | 330/11088 [02:55<1:28:40,  2.02it/s]
  3%|▎         | 331/11088 [02:56<1:28:38,  2.02it/s]
  3%|▎         | 332/11088 [02:56<1:28:36,  2.02it/s]
  3%|▎         | 333/11088 [02:57<1:28:35,  2.02it/s]
  3%|▎         | 334/11088 [02:57<1:28:36,  2.02it/s]
  3%|▎         | 335/11088 [02:58<1:28:38,  2.02it/s]
  3%|▎         | 336/11088 [02:58<1:28:28,  2.03it/s]
  3%|▎         | 337/11088 [02:58<1:28:34,  2.02it/s]
  3%|▎         | 338/11088 [02:59<1:28:34,  2.02it/s]
  3%|▎         | 339/11088 [02:59<1:28:36,  2.02it/s]
  3%|▎         | 340/11088 [03:00<1:28:33,  2.02it/s]
  3%|▎         | 341/11088 [03:00<1:28:37,  2.02it/s]
  3%|▎         | 342/11088 [03:01<1:28:33,  2.02it/s]
  3%|▎         | 343/11088 [03:01<1:28:35,  2.02it/s]
  3%|▎         | 344/11088 [03:02<1:28:29,  2.02it/s]
  3%|▎         | 345/11088 [03:02<1:28:35,  2.02it/s]
  3%|▎         | 346/11088 [03:03<1:28:29,  2.02it/s]
  3%|▎         | 347/11088 [03:03<1:28:34,  2.02it/s]
  3%|▎         | 348/11088 [03:04<1:28:27,  2.02it/s]
  3%|▎         | 349/11088 [03:04<1:28:31,  2.02it/s]
  3%|▎         | 350/11088 [03:05<1:28:25,  2.02it/s]
                                                     
{'loss': 5.5272, 'grad_norm': 0.6340382099151611, 'learning_rate': 0.0003155996393146979, 'epoch': 0.44}
+
  3%|▎         | 350/11088 [03:05<1:28:25,  2.02it/s]
  3%|▎         | 351/11088 [03:05<1:28:32,  2.02it/s]
  3%|▎         | 352/11088 [03:06<1:28:25,  2.02it/s]
  3%|▎         | 353/11088 [03:06<1:28:28,  2.02it/s]
  3%|▎         | 354/11088 [03:07<1:28:25,  2.02it/s]
  3%|▎         | 355/11088 [03:07<1:28:26,  2.02it/s]
  3%|▎         | 356/11088 [03:08<1:28:20,  2.02it/s]
  3%|▎         | 357/11088 [03:08<1:28:23,  2.02it/s]
  3%|▎         | 358/11088 [03:09<1:28:20,  2.02it/s]
  3%|▎         | 359/11088 [03:09<1:28:23,  2.02it/s]
  3%|▎         | 360/11088 [03:10<1:28:16,  2.03it/s]
  3%|▎         | 361/11088 [03:10<1:28:23,  2.02it/s]
  3%|▎         | 362/11088 [03:11<1:28:17,  2.02it/s]
  3%|▎         | 363/11088 [03:11<1:28:18,  2.02it/s]
  3%|▎         | 364/11088 [03:12<1:28:20,  2.02it/s]
  3%|▎         | 365/11088 [03:12<1:28:22,  2.02it/s]
  3%|▎         | 366/11088 [03:13<1:28:19,  2.02it/s]
  3%|▎         | 367/11088 [03:13<1:28:21,  2.02it/s]
  3%|▎         | 368/11088 [03:14<1:28:19,  2.02it/s]
  3%|▎         | 369/11088 [03:14<1:28:17,  2.02it/s]
  3%|▎         | 370/11088 [03:15<1:28:23,  2.02it/s]
  3%|▎         | 371/11088 [03:15<1:28:21,  2.02it/s]
  3%|▎         | 372/11088 [03:16<1:28:24,  2.02it/s]
  3%|▎         | 373/11088 [03:16<1:28:29,  2.02it/s]
  3%|▎         | 374/11088 [03:17<1:28:23,  2.02it/s]
  3%|▎         | 375/11088 [03:17<1:28:21,  2.02it/s]
                                                     {'loss': 5.4259, 'grad_norm': 0.9262540936470032, 'learning_rate': 0.0003381424706943192, 'epoch': 0.47}
+
  3%|▎         | 375/11088 [03:17<1:28:21,  2.02it/s]
  3%|▎         | 376/11088 [03:18<1:28:28,  2.02it/s]
  3%|▎         | 377/11088 [03:18<1:28:25,  2.02it/s]
  3%|▎         | 378/11088 [03:19<1:28:18,  2.02it/s]
  3%|▎         | 379/11088 [03:19<1:28:20,  2.02it/s]
  3%|▎         | 380/11088 [03:20<1:28:13,  2.02it/s]
  3%|▎         | 381/11088 [03:20<1:28:18,  2.02it/s]
  3%|▎         | 382/11088 [03:21<1:28:14,  2.02it/s]
  3%|▎         | 383/11088 [03:21<1:28:18,  2.02it/s]
  3%|▎         | 384/11088 [03:22<1:28:14,  2.02it/s]
  3%|▎         | 385/11088 [03:22<1:28:15,  2.02it/s]
  3%|▎         | 386/11088 [03:23<1:28:11,  2.02it/s]
  3%|▎         | 387/11088 [03:23<1:28:11,  2.02it/s]
  3%|▎         | 388/11088 [03:24<1:28:08,  2.02it/s]
  4%|▎         | 389/11088 [03:24<1:28:14,  2.02it/s]
  4%|▎         | 390/11088 [03:25<1:28:07,  2.02it/s]
  4%|▎         | 391/11088 [03:25<1:28:06,  2.02it/s]
  4%|▎         | 392/11088 [03:26<1:28:06,  2.02it/s]
  4%|▎         | 393/11088 [03:26<1:28:11,  2.02it/s]
  4%|▎         | 394/11088 [03:27<1:28:08,  2.02it/s]
  4%|▎         | 395/11088 [03:27<1:28:12,  2.02it/s]
  4%|▎         | 396/11088 [03:28<1:28:11,  2.02it/s]
  4%|▎         | 397/11088 [03:28<1:28:09,  2.02it/s]
  4%|▎         | 398/11088 [03:29<1:28:06,  2.02it/s]
  4%|▎         | 399/11088 [03:29<1:28:08,  2.02it/s]
  4%|▎         | 400/11088 [03:30<1:28:08,  2.02it/s]{'loss': 5.3414, 'grad_norm': 0.6309112310409546, 'learning_rate': 0.00036068530207394053, 'epoch': 0.51}
+
                                                     

  4%|▎         | 400/11088 [03:30<1:28:08,  2.02it/s]
  4%|▎         | 401/11088 [03:30<1:28:09,  2.02it/s]
  4%|▎         | 402/11088 [03:31<1:28:07,  2.02it/s]
  4%|▎         | 403/11088 [03:31<1:28:00,  2.02it/s]
  4%|▎         | 404/11088 [03:32<1:28:06,  2.02it/s]
  4%|▎         | 405/11088 [03:32<1:28:02,  2.02it/s]
  4%|▎         | 406/11088 [03:33<1:28:04,  2.02it/s]
  4%|▎         | 407/11088 [03:33<1:28:01,  2.02it/s]
  4%|▎         | 408/11088 [03:34<1:28:02,  2.02it/s]
  4%|▎         | 409/11088 [03:34<1:28:00,  2.02it/s]
  4%|▎         | 410/11088 [03:35<1:28:00,  2.02it/s]
  4%|▎         | 411/11088 [03:35<1:28:00,  2.02it/s]
  4%|▎         | 412/11088 [03:36<1:28:02,  2.02it/s]
  4%|▎         | 413/11088 [03:36<1:27:57,  2.02it/s]
  4%|▎         | 414/11088 [03:37<1:27:55,  2.02it/s]
  4%|▎         | 415/11088 [03:37<1:27:50,  2.02it/s]
  4%|▍         | 416/11088 [03:38<1:27:52,  2.02it/s]
  4%|▍         | 417/11088 [03:38<1:27:43,  2.03it/s]
  4%|▍         | 418/11088 [03:39<1:27:49,  2.02it/s]
  4%|▍         | 419/11088 [03:39<1:27:45,  2.03it/s]
  4%|▍         | 420/11088 [03:40<1:27:49,  2.02it/s]
  4%|▍         | 421/11088 [03:40<1:27:45,  2.03it/s]
  4%|▍         | 422/11088 [03:41<1:27:38,  2.03it/s]
  4%|▍         | 423/11088 [03:41<1:27:45,  2.03it/s]
  4%|▍         | 424/11088 [03:42<1:27:40,  2.03it/s]
  4%|▍         | 425/11088 [03:42<1:27:47,  2.02it/s]
                                                     
{'loss': 5.2618, 'grad_norm': 0.6182384490966797, 'learning_rate': 0.0003832281334535618, 'epoch': 0.54}
+
  4%|▍         | 425/11088 [03:42<1:27:47,  2.02it/s]
  4%|▍         | 426/11088 [03:43<1:27:45,  2.02it/s]
  4%|▍         | 427/11088 [03:43<1:27:46,  2.02it/s]
  4%|▍         | 428/11088 [03:43<1:27:42,  2.03it/s]
  4%|▍         | 429/11088 [03:44<1:27:47,  2.02it/s]
  4%|▍         | 430/11088 [03:44<1:27:43,  2.02it/s]
  4%|▍         | 431/11088 [03:45<1:27:41,  2.03it/s]
  4%|▍         | 432/11088 [03:45<1:27:44,  2.02it/s]
  4%|▍         | 433/11088 [03:46<1:27:36,  2.03it/s]
  4%|▍         | 434/11088 [03:46<1:27:42,  2.02it/s]
  4%|▍         | 435/11088 [03:47<1:27:34,  2.03it/s]
  4%|▍         | 436/11088 [03:47<1:27:41,  2.02it/s]
  4%|▍         | 437/11088 [03:48<1:27:41,  2.02it/s]
  4%|▍         | 438/11088 [03:48<1:27:45,  2.02it/s]
  4%|▍         | 439/11088 [03:49<1:27:40,  2.02it/s]
  4%|▍         | 440/11088 [03:49<1:27:41,  2.02it/s]
  4%|▍         | 441/11088 [03:50<1:27:40,  2.02it/s]
  4%|▍         | 442/11088 [03:50<1:27:47,  2.02it/s]
  4%|▍         | 443/11088 [03:51<1:27:43,  2.02it/s]
  4%|▍         | 444/11088 [03:51<1:27:44,  2.02it/s]
  4%|▍         | 445/11088 [03:52<1:27:35,  2.03it/s]
  4%|▍         | 446/11088 [03:52<1:27:42,  2.02it/s]
  4%|▍         | 447/11088 [03:53<1:27:35,  2.02it/s]
  4%|▍         | 448/11088 [03:53<1:27:33,  2.03it/s]
  4%|▍         | 449/11088 [03:54<1:27:33,  2.03it/s]
  4%|▍         | 450/11088 [03:54<1:27:30,  2.03it/s]
                                                     
{'loss': 5.1861, 'grad_norm': 0.6628245711326599, 'learning_rate': 0.00040577096483318303, 'epoch': 0.57}
+
  4%|▍         | 450/11088 [03:54<1:27:30,  2.03it/s]
  4%|▍         | 451/11088 [03:55<1:27:41,  2.02it/s]
  4%|▍         | 452/11088 [03:55<1:27:32,  2.02it/s]
  4%|▍         | 453/11088 [03:56<1:27:35,  2.02it/s]
  4%|▍         | 454/11088 [03:56<1:27:28,  2.03it/s]
  4%|▍         | 455/11088 [03:57<1:27:26,  2.03it/s]
  4%|▍         | 456/11088 [03:57<1:27:28,  2.03it/s]
  4%|▍         | 457/11088 [03:58<1:27:21,  2.03it/s]
  4%|▍         | 458/11088 [03:58<1:27:25,  2.03it/s]
  4%|▍         | 459/11088 [03:59<1:27:23,  2.03it/s]
  4%|▍         | 460/11088 [03:59<1:27:29,  2.02it/s]
  4%|▍         | 461/11088 [04:00<1:27:25,  2.03it/s]
  4%|▍         | 462/11088 [04:00<1:27:27,  2.03it/s]
  4%|▍         | 463/11088 [04:01<1:27:33,  2.02it/s]
  4%|▍         | 464/11088 [04:01<1:27:34,  2.02it/s]
  4%|▍         | 465/11088 [04:02<1:27:34,  2.02it/s]
  4%|▍         | 466/11088 [04:02<1:27:35,  2.02it/s]
  4%|▍         | 467/11088 [04:03<1:27:28,  2.02it/s]
  4%|▍         | 468/11088 [04:03<1:27:31,  2.02it/s]
  4%|▍         | 469/11088 [04:04<1:27:25,  2.02it/s]
  4%|▍         | 470/11088 [04:04<1:27:30,  2.02it/s]
  4%|▍         | 471/11088 [04:05<1:27:25,  2.02it/s]
  4%|▍         | 472/11088 [04:05<1:27:28,  2.02it/s]
  4%|▍         | 473/11088 [04:06<1:27:21,  2.03it/s]
  4%|▍         | 474/11088 [04:06<1:27:26,  2.02it/s]
  4%|▍         | 475/11088 [04:07<1:27:20,  2.03it/s]
                                                     
{'loss': 5.1232, 'grad_norm': 0.49333590269088745, 'learning_rate': 0.00042831379621280434, 'epoch': 0.6}
+
  4%|▍         | 475/11088 [04:07<1:27:20,  2.03it/s]
  4%|▍         | 476/11088 [04:07<1:27:29,  2.02it/s]
  4%|▍         | 477/11088 [04:08<1:27:22,  2.02it/s]
  4%|▍         | 478/11088 [04:08<1:27:26,  2.02it/s]
  4%|▍         | 479/11088 [04:09<1:27:21,  2.02it/s]
  4%|▍         | 480/11088 [04:09<1:27:25,  2.02it/s]
  4%|▍         | 481/11088 [04:10<1:27:19,  2.02it/s]
  4%|▍         | 482/11088 [04:10<1:27:20,  2.02it/s]
  4%|▍         | 483/11088 [04:11<1:27:18,  2.02it/s]
  4%|▍         | 484/11088 [04:11<1:27:12,  2.03it/s]
  4%|▍         | 485/11088 [04:12<1:27:14,  2.03it/s]
  4%|▍         | 486/11088 [04:12<1:27:13,  2.03it/s]
  4%|▍         | 487/11088 [04:13<1:27:18,  2.02it/s]
  4%|▍         | 488/11088 [04:13<1:27:13,  2.03it/s]
  4%|▍         | 489/11088 [04:14<1:27:16,  2.02it/s]
  4%|▍         | 490/11088 [04:14<1:27:11,  2.03it/s]
  4%|▍         | 491/11088 [04:15<1:27:12,  2.03it/s]
  4%|▍         | 492/11088 [04:15<1:27:15,  2.02it/s]
  4%|▍         | 493/11088 [04:16<1:27:17,  2.02it/s]
  4%|▍         | 494/11088 [04:16<1:27:12,  2.02it/s]
  4%|▍         | 495/11088 [04:17<1:27:15,  2.02it/s]
  4%|▍         | 496/11088 [04:17<1:27:10,  2.03it/s]
  4%|▍         | 497/11088 [04:18<1:27:13,  2.02it/s]
  4%|▍         | 498/11088 [04:18<1:27:10,  2.02it/s]
  5%|▍         | 499/11088 [04:19<1:27:11,  2.02it/s]
  5%|▍         | 500/11088 [04:19<1:27:10,  2.02it/s]
                                                     
{'loss': 5.062, 'grad_norm': 0.5728343725204468, 'learning_rate': 0.00045085662759242564, 'epoch': 0.63}
+
  5%|▍         | 500/11088 [04:19<1:27:10,  2.02it/s]
  5%|▍         | 501/11088 [04:20<1:27:23,  2.02it/s]
  5%|▍         | 502/11088 [04:20<1:27:18,  2.02it/s]
  5%|▍         | 503/11088 [04:21<1:27:18,  2.02it/s]
  5%|▍         | 504/11088 [04:21<1:27:10,  2.02it/s]
  5%|▍         | 505/11088 [04:22<1:27:11,  2.02it/s]
  5%|▍         | 506/11088 [04:22<1:27:09,  2.02it/s]
  5%|▍         | 507/11088 [04:23<1:27:06,  2.02it/s]
  5%|▍         | 508/11088 [04:23<1:27:11,  2.02it/s]
  5%|▍         | 509/11088 [04:24<1:27:12,  2.02it/s]
  5%|▍         | 510/11088 [04:24<1:27:11,  2.02it/s]
  5%|▍         | 511/11088 [04:24<1:27:10,  2.02it/s]
  5%|▍         | 512/11088 [04:25<1:27:10,  2.02it/s]
  5%|▍         | 513/11088 [04:25<1:27:04,  2.02it/s]
  5%|▍         | 514/11088 [04:26<1:27:06,  2.02it/s]
  5%|▍         | 515/11088 [04:26<1:26:56,  2.03it/s]
  5%|▍         | 516/11088 [04:27<1:27:00,  2.03it/s]
  5%|▍         | 517/11088 [04:27<1:26:56,  2.03it/s]
  5%|▍         | 518/11088 [04:28<1:27:00,  2.02it/s]
  5%|▍         | 519/11088 [04:28<1:27:00,  2.02it/s]
  5%|▍         | 520/11088 [04:29<1:26:56,  2.03it/s]
  5%|▍         | 521/11088 [04:29<1:27:02,  2.02it/s]
  5%|▍         | 522/11088 [04:30<1:26:59,  2.02it/s]
  5%|▍         | 523/11088 [04:30<1:27:00,  2.02it/s]
  5%|▍         | 524/11088 [04:31<1:26:49,  2.03it/s]
  5%|▍         | 525/11088 [04:31<1:26:52,  2.03it/s]{'loss': 5.0021, 'grad_norm': 0.5581721067428589, 'learning_rate': 0.0004733994589720469, 'epoch': 0.66}
                                                     
+
  5%|▍         | 525/11088 [04:31<1:26:52,  2.03it/s]
  5%|▍         | 526/11088 [04:32<1:26:54,  2.03it/s]
  5%|▍         | 527/11088 [04:32<1:26:56,  2.02it/s]
  5%|▍         | 528/11088 [04:33<1:26:52,  2.03it/s]
  5%|▍         | 529/11088 [04:33<1:26:57,  2.02it/s]
  5%|▍         | 530/11088 [04:34<1:26:56,  2.02it/s]
  5%|▍         | 531/11088 [04:34<1:26:58,  2.02it/s]
  5%|▍         | 532/11088 [04:35<1:26:56,  2.02it/s]
  5%|▍         | 533/11088 [04:35<1:26:54,  2.02it/s]
  5%|▍         | 534/11088 [04:36<1:26:55,  2.02it/s]
  5%|▍         | 535/11088 [04:36<1:26:55,  2.02it/s]
  5%|▍         | 536/11088 [04:37<1:26:54,  2.02it/s]
  5%|▍         | 537/11088 [04:37<1:26:52,  2.02it/s]
  5%|▍         | 538/11088 [04:38<1:26:49,  2.03it/s]
  5%|▍         | 539/11088 [04:38<1:26:51,  2.02it/s]
  5%|▍         | 540/11088 [04:39<1:26:46,  2.03it/s]
  5%|▍         | 541/11088 [04:39<1:26:48,  2.02it/s]
  5%|▍         | 542/11088 [04:40<1:26:47,  2.03it/s]
  5%|▍         | 543/11088 [04:40<1:26:49,  2.02it/s]
  5%|▍         | 544/11088 [04:41<1:26:49,  2.02it/s]
  5%|▍         | 545/11088 [04:41<1:26:47,  2.02it/s]
  5%|▍         | 546/11088 [04:42<1:26:48,  2.02it/s]
  5%|▍         | 547/11088 [04:42<1:26:51,  2.02it/s]
  5%|▍         | 548/11088 [04:43<1:26:50,  2.02it/s]
  5%|▍         | 549/11088 [04:43<1:26:43,  2.03it/s]
  5%|▍         | 550/11088 [04:44<1:26:39,  2.03it/s]
                                                     
{'loss': 4.9501, 'grad_norm': 0.5733642578125, 'learning_rate': 0.0004959422903516681, 'epoch': 0.69}
+
  5%|▍         | 550/11088 [04:44<1:26:39,  2.03it/s]
  5%|▍         | 551/11088 [04:44<1:26:41,  2.03it/s]
  5%|▍         | 552/11088 [04:45<1:26:44,  2.02it/s]
  5%|▍         | 553/11088 [04:45<1:26:44,  2.02it/s]
  5%|▍         | 554/11088 [04:46<1:26:42,  2.02it/s]
  5%|▌         | 555/11088 [04:46<1:26:39,  2.03it/s]
  5%|▌         | 556/11088 [04:47<1:26:41,  2.02it/s]
  5%|▌         | 557/11088 [04:47<1:26:41,  2.02it/s]
  5%|▌         | 558/11088 [04:48<1:26:40,  2.02it/s]
  5%|▌         | 559/11088 [04:48<1:26:41,  2.02it/s]
  5%|▌         | 560/11088 [04:49<1:26:41,  2.02it/s]
  5%|▌         | 561/11088 [04:49<1:26:37,  2.03it/s]
  5%|▌         | 562/11088 [04:50<1:26:37,  2.03it/s]
  5%|▌         | 563/11088 [04:50<1:26:34,  2.03it/s]
  5%|▌         | 564/11088 [04:51<1:26:36,  2.03it/s]
  5%|▌         | 565/11088 [04:51<1:26:35,  2.03it/s]
  5%|▌         | 566/11088 [04:52<1:26:38,  2.02it/s]
  5%|▌         | 567/11088 [04:52<1:26:41,  2.02it/s]
  5%|▌         | 568/11088 [04:53<1:26:37,  2.02it/s]
  5%|▌         | 569/11088 [04:53<1:26:38,  2.02it/s]
  5%|▌         | 570/11088 [04:54<1:26:40,  2.02it/s]
  5%|▌         | 571/11088 [04:54<1:26:39,  2.02it/s]
  5%|▌         | 572/11088 [04:55<1:26:38,  2.02it/s]
  5%|▌         | 573/11088 [04:55<1:26:32,  2.03it/s]
  5%|▌         | 574/11088 [04:56<1:26:31,  2.03it/s]
  5%|▌         | 575/11088 [04:56<1:26:29,  2.03it/s]{'loss': 4.9062, 'grad_norm': 0.7264406085014343, 'learning_rate': 0.0005184851217312895, 'epoch': 0.73}
                                                     
+
  5%|▌         | 575/11088 [04:56<1:26:29,  2.03it/s]
  5%|▌         | 576/11088 [04:57<1:26:32,  2.02it/s]
  5%|▌         | 577/11088 [04:57<1:26:31,  2.02it/s]
  5%|▌         | 578/11088 [04:58<1:26:29,  2.03it/s]
  5%|▌         | 579/11088 [04:58<1:26:30,  2.02it/s]
  5%|▌         | 580/11088 [04:59<1:26:23,  2.03it/s]
  5%|▌         | 581/11088 [04:59<1:26:27,  2.03it/s]
  5%|▌         | 582/11088 [05:00<1:26:28,  2.02it/s]
  5%|▌         | 583/11088 [05:00<1:26:21,  2.03it/s]
  5%|▌         | 584/11088 [05:01<1:26:26,  2.03it/s]
  5%|▌         | 585/11088 [05:01<1:26:23,  2.03it/s]
  5%|▌         | 586/11088 [05:02<1:26:26,  2.02it/s]
  5%|▌         | 587/11088 [05:02<1:26:24,  2.03it/s]
  5%|▌         | 588/11088 [05:03<1:26:25,  2.02it/s]
  5%|▌         | 589/11088 [05:03<1:26:27,  2.02it/s]
  5%|▌         | 590/11088 [05:04<1:26:28,  2.02it/s]
  5%|▌         | 591/11088 [05:04<1:26:27,  2.02it/s]
  5%|▌         | 592/11088 [05:05<1:26:21,  2.03it/s]
  5%|▌         | 593/11088 [05:05<1:26:26,  2.02it/s]
  5%|▌         | 594/11088 [05:05<1:26:20,  2.03it/s]
  5%|▌         | 595/11088 [05:06<1:26:23,  2.02it/s]
  5%|▌         | 596/11088 [05:06<1:26:20,  2.03it/s]
  5%|▌         | 597/11088 [05:07<1:26:22,  2.02it/s]
  5%|▌         | 598/11088 [05:07<1:26:20,  2.03it/s]
  5%|▌         | 599/11088 [05:08<1:26:21,  2.02it/s]
  5%|▌         | 600/11088 [05:08<1:26:23,  2.02it/s]
                                                     
{'loss': 4.8559, 'grad_norm': 0.4734659790992737, 'learning_rate': 0.0005410279531109106, 'epoch': 0.76}
+
  5%|▌         | 600/11088 [05:08<1:26:23,  2.02it/s]
  5%|▌         | 601/11088 [05:09<1:26:29,  2.02it/s]
  5%|▌         | 602/11088 [05:09<1:26:28,  2.02it/s]
  5%|▌         | 603/11088 [05:10<1:26:26,  2.02it/s]
  5%|▌         | 604/11088 [05:10<1:26:18,  2.02it/s]
  5%|▌         | 605/11088 [05:11<1:26:20,  2.02it/s]
  5%|▌         | 606/11088 [05:11<1:26:16,  2.02it/s]
  5%|▌         | 607/11088 [05:12<1:26:16,  2.02it/s]
  5%|▌         | 608/11088 [05:12<1:26:13,  2.03it/s]
  5%|▌         | 609/11088 [05:13<1:26:14,  2.03it/s]
  6%|▌         | 610/11088 [05:13<1:26:15,  2.02it/s]
  6%|▌         | 611/11088 [05:14<1:26:10,  2.03it/s]
  6%|▌         | 612/11088 [05:14<1:26:15,  2.02it/s]
  6%|▌         | 613/11088 [05:15<1:26:08,  2.03it/s]
  6%|▌         | 614/11088 [05:15<1:26:14,  2.02it/s]
  6%|▌         | 615/11088 [05:16<1:33:43,  1.86it/s]
  6%|▌         | 616/11088 [05:17<1:38:54,  1.76it/s]
  6%|▌         | 617/11088 [05:17<1:35:01,  1.84it/s]
  6%|▌         | 618/11088 [05:18<1:32:24,  1.89it/s]
  6%|▌         | 619/11088 [05:18<1:30:29,  1.93it/s]
  6%|▌         | 620/11088 [05:19<1:29:11,  1.96it/s]
  6%|▌         | 621/11088 [05:19<1:28:14,  1.98it/s]
  6%|▌         | 622/11088 [05:20<1:27:41,  1.99it/s]
  6%|▌         | 623/11088 [05:20<1:27:13,  2.00it/s]
  6%|▌         | 624/11088 [05:21<1:26:52,  2.01it/s]
  6%|▌         | 625/11088 [05:21<1:26:38,  2.01it/s]
                                                     
{'loss': 4.8127, 'grad_norm': 0.5146079063415527, 'learning_rate': 0.000563570784490532, 'epoch': 0.79}
+
  6%|▌         | 625/11088 [05:21<1:26:38,  2.01it/s]
  6%|▌         | 626/11088 [05:22<1:26:38,  2.01it/s]
  6%|▌         | 627/11088 [05:22<1:26:28,  2.02it/s]
  6%|▌         | 628/11088 [05:23<1:26:23,  2.02it/s]
  6%|▌         | 629/11088 [05:23<1:26:17,  2.02it/s]
  6%|▌         | 630/11088 [05:24<1:26:15,  2.02it/s]
  6%|▌         | 631/11088 [05:24<1:26:13,  2.02it/s]
  6%|▌         | 632/11088 [05:25<1:26:17,  2.02it/s]
  6%|▌         | 633/11088 [05:25<1:26:11,  2.02it/s]
  6%|▌         | 634/11088 [05:26<1:26:09,  2.02it/s]
  6%|▌         | 635/11088 [05:26<1:26:06,  2.02it/s]
  6%|▌         | 636/11088 [05:27<1:26:09,  2.02it/s]
  6%|▌         | 637/11088 [05:27<1:26:08,  2.02it/s]
  6%|▌         | 638/11088 [05:28<1:26:06,  2.02it/s]
  6%|▌         | 639/11088 [05:28<1:26:05,  2.02it/s]
  6%|▌         | 640/11088 [05:29<1:26:07,  2.02it/s]
  6%|▌         | 641/11088 [05:29<1:26:05,  2.02it/s]
  6%|▌         | 642/11088 [05:29<1:26:03,  2.02it/s]
  6%|▌         | 643/11088 [05:30<1:26:05,  2.02it/s]
  6%|▌         | 644/11088 [05:30<1:26:04,  2.02it/s]
  6%|▌         | 645/11088 [05:31<1:26:03,  2.02it/s]
  6%|▌         | 646/11088 [05:31<1:26:02,  2.02it/s]
  6%|▌         | 647/11088 [05:32<1:26:00,  2.02it/s]
  6%|▌         | 648/11088 [05:32<1:26:01,  2.02it/s]
  6%|▌         | 649/11088 [05:33<1:26:04,  2.02it/s]
  6%|▌         | 650/11088 [05:33<1:26:03,  2.02it/s]
                                                     
{'loss': 4.7747, 'grad_norm': 0.5695474743843079, 'learning_rate': 0.0005861136158701534, 'epoch': 0.82}
+
  6%|▌         | 650/11088 [05:33<1:26:03,  2.02it/s]
  6%|▌         | 651/11088 [05:34<1:26:08,  2.02it/s]
  6%|▌         | 652/11088 [05:34<1:26:03,  2.02it/s]
  6%|▌         | 653/11088 [05:35<1:26:03,  2.02it/s]
  6%|▌         | 654/11088 [05:35<1:26:02,  2.02it/s]
  6%|▌         | 655/11088 [05:36<1:26:03,  2.02it/s]
  6%|▌         | 656/11088 [05:36<1:25:59,  2.02it/s]
  6%|▌         | 657/11088 [05:37<1:25:58,  2.02it/s]
  6%|▌         | 658/11088 [05:37<1:25:57,  2.02it/s]
  6%|▌         | 659/11088 [05:38<1:25:57,  2.02it/s]
  6%|▌         | 660/11088 [05:38<1:25:57,  2.02it/s]
  6%|▌         | 661/11088 [05:39<1:25:55,  2.02it/s]
  6%|▌         | 662/11088 [05:39<1:25:57,  2.02it/s]
  6%|▌         | 663/11088 [05:40<1:25:50,  2.02it/s]
  6%|▌         | 664/11088 [05:40<1:25:50,  2.02it/s]
  6%|▌         | 665/11088 [05:41<1:25:42,  2.03it/s]
  6%|▌         | 666/11088 [05:41<1:25:49,  2.02it/s]
  6%|▌         | 667/11088 [05:42<1:25:42,  2.03it/s]
  6%|▌         | 668/11088 [05:42<1:25:43,  2.03it/s]
  6%|▌         | 669/11088 [05:43<1:25:39,  2.03it/s]
  6%|▌         | 670/11088 [05:43<1:25:36,  2.03it/s]
  6%|▌         | 671/11088 [05:44<1:25:38,  2.03it/s]
  6%|▌         | 672/11088 [05:44<1:25:34,  2.03it/s]
  6%|▌         | 673/11088 [05:45<1:25:33,  2.03it/s]
  6%|▌         | 674/11088 [05:45<1:25:35,  2.03it/s]
  6%|▌         | 675/11088 [05:46<1:25:31,  2.03it/s]
                                                     
{'loss': 4.736, 'grad_norm': 0.4409935176372528, 'learning_rate': 0.0006086564472497745, 'epoch': 0.85}
+
  6%|▌         | 675/11088 [05:46<1:25:31,  2.03it/s]
  6%|▌         | 676/11088 [05:46<1:25:39,  2.03it/s]
  6%|▌         | 677/11088 [05:47<1:25:33,  2.03it/s]
  6%|▌         | 678/11088 [05:47<1:25:38,  2.03it/s]
  6%|▌         | 679/11088 [05:48<1:25:37,  2.03it/s]
  6%|▌         | 680/11088 [05:48<1:25:37,  2.03it/s]
  6%|▌         | 681/11088 [05:49<1:25:35,  2.03it/s]
  6%|▌         | 682/11088 [05:49<1:25:39,  2.02it/s]
  6%|▌         | 683/11088 [05:50<1:25:38,  2.03it/s]
  6%|▌         | 684/11088 [05:50<1:25:43,  2.02it/s]
  6%|▌         | 685/11088 [05:51<1:25:42,  2.02it/s]
  6%|▌         | 686/11088 [05:51<1:25:38,  2.02it/s]
  6%|▌         | 687/11088 [05:52<1:25:34,  2.03it/s]
  6%|▌         | 688/11088 [05:52<1:25:31,  2.03it/s]
  6%|▌         | 689/11088 [05:53<1:25:31,  2.03it/s]
  6%|▌         | 690/11088 [05:53<1:25:30,  2.03it/s]
  6%|▌         | 691/11088 [05:54<1:25:30,  2.03it/s]
  6%|▌         | 692/11088 [05:54<1:25:32,  2.03it/s]
  6%|▋         | 693/11088 [05:55<1:25:37,  2.02it/s]
  6%|▋         | 694/11088 [05:55<1:25:36,  2.02it/s]
  6%|▋         | 695/11088 [05:56<1:25:37,  2.02it/s]
  6%|▋         | 696/11088 [05:56<1:25:33,  2.02it/s]
  6%|▋         | 697/11088 [05:57<1:25:32,  2.02it/s]
  6%|▋         | 698/11088 [05:57<1:25:27,  2.03it/s]
  6%|▋         | 699/11088 [05:58<1:25:28,  2.03it/s]
  6%|▋         | 700/11088 [05:58<1:25:28,  2.03it/s]
                                                     {'loss': 4.7106, 'grad_norm': 0.5907058715820312, 'learning_rate': 0.0006311992786293959, 'epoch': 0.88}
+
  6%|▋         | 700/11088 [05:58<1:25:28,  2.03it/s]
  6%|▋         | 701/11088 [05:59<1:25:37,  2.02it/s]
  6%|▋         | 702/11088 [05:59<1:25:35,  2.02it/s]
  6%|▋         | 703/11088 [06:00<1:25:34,  2.02it/s]
  6%|▋         | 704/11088 [06:00<1:25:31,  2.02it/s]
  6%|▋         | 705/11088 [06:01<1:25:32,  2.02it/s]
  6%|▋         | 706/11088 [06:01<1:25:32,  2.02it/s]
  6%|▋         | 707/11088 [06:02<1:25:30,  2.02it/s]
  6%|▋         | 708/11088 [06:02<1:25:31,  2.02it/s]
  6%|▋         | 709/11088 [06:03<1:25:34,  2.02it/s]
  6%|▋         | 710/11088 [06:03<1:25:26,  2.02it/s]
  6%|▋         | 711/11088 [06:04<1:25:32,  2.02it/s]
  6%|▋         | 712/11088 [06:04<1:25:25,  2.02it/s]
  6%|▋         | 713/11088 [06:05<1:25:26,  2.02it/s]
  6%|▋         | 714/11088 [06:05<1:25:23,  2.02it/s]
  6%|▋         | 715/11088 [06:06<1:25:20,  2.03it/s]
  6%|▋         | 716/11088 [06:06<1:25:24,  2.02it/s]
  6%|▋         | 717/11088 [06:07<1:25:18,  2.03it/s]
  6%|▋         | 718/11088 [06:07<1:25:22,  2.02it/s]
  6%|▋         | 719/11088 [06:08<1:25:21,  2.02it/s]
  6%|▋         | 720/11088 [06:08<1:25:20,  2.02it/s]
  7%|▋         | 721/11088 [06:09<1:25:21,  2.02it/s]
  7%|▋         | 722/11088 [06:09<1:25:21,  2.02it/s]
  7%|▋         | 723/11088 [06:10<1:25:24,  2.02it/s]
  7%|▋         | 724/11088 [06:10<1:25:21,  2.02it/s]
  7%|▋         | 725/11088 [06:11<1:25:18,  2.02it/s]
                                                     
{'loss': 4.6677, 'grad_norm': 0.5601474642753601, 'learning_rate': 0.0006537421100090172, 'epoch': 0.92}
+
  7%|▋         | 725/11088 [06:11<1:25:18,  2.02it/s]
  7%|▋         | 726/11088 [06:11<1:25:18,  2.02it/s]
  7%|▋         | 727/11088 [06:11<1:25:17,  2.02it/s]
  7%|▋         | 728/11088 [06:12<1:25:17,  2.02it/s]
  7%|▋         | 729/11088 [06:12<1:25:12,  2.03it/s]
  7%|▋         | 730/11088 [06:13<1:25:16,  2.02it/s]
  7%|▋         | 731/11088 [06:13<1:25:08,  2.03it/s]
  7%|▋         | 732/11088 [06:14<1:25:11,  2.03it/s]
  7%|▋         | 733/11088 [06:14<1:25:11,  2.03it/s]
  7%|▋         | 734/11088 [06:15<1:25:13,  2.02it/s]
  7%|▋         | 735/11088 [06:15<1:25:12,  2.03it/s]
  7%|▋         | 736/11088 [06:16<1:25:12,  2.02it/s]
  7%|▋         | 737/11088 [06:16<1:25:18,  2.02it/s]
  7%|▋         | 738/11088 [06:17<1:25:18,  2.02it/s]
  7%|▋         | 739/11088 [06:17<1:25:19,  2.02it/s]
  7%|▋         | 740/11088 [06:18<1:25:15,  2.02it/s]
  7%|▋         | 741/11088 [06:18<1:25:15,  2.02it/s]
  7%|▋         | 742/11088 [06:19<1:25:10,  2.02it/s]
  7%|▋         | 743/11088 [06:19<1:25:07,  2.03it/s]
  7%|▋         | 744/11088 [06:20<1:25:10,  2.02it/s]
  7%|▋         | 745/11088 [06:20<1:25:08,  2.02it/s]
  7%|▋         | 746/11088 [06:21<1:25:07,  2.02it/s]
  7%|▋         | 747/11088 [06:21<1:25:07,  2.02it/s]
  7%|▋         | 748/11088 [06:22<1:25:09,  2.02it/s]
  7%|▋         | 749/11088 [06:22<1:25:09,  2.02it/s]
  7%|▋         | 750/11088 [06:23<1:25:08,  2.02it/s]
                                                     
{'loss': 4.6427, 'grad_norm': 0.5141180753707886, 'learning_rate': 0.0006762849413886384, 'epoch': 0.95}
+
  7%|▋         | 750/11088 [06:23<1:25:08,  2.02it/s]
  7%|▋         | 751/11088 [06:23<1:25:10,  2.02it/s]
  7%|▋         | 752/11088 [06:24<1:25:13,  2.02it/s]
  7%|▋         | 753/11088 [06:24<1:25:09,  2.02it/s]
  7%|▋         | 754/11088 [06:25<1:25:09,  2.02it/s]
  7%|▋         | 755/11088 [06:25<1:25:10,  2.02it/s]
  7%|▋         | 756/11088 [06:26<1:25:07,  2.02it/s]
  7%|▋         | 757/11088 [06:26<1:25:07,  2.02it/s]
  7%|▋         | 758/11088 [06:27<1:25:06,  2.02it/s]
  7%|▋         | 759/11088 [06:27<1:25:04,  2.02it/s]
  7%|▋         | 760/11088 [06:28<1:25:03,  2.02it/s]
  7%|▋         | 761/11088 [06:28<1:25:04,  2.02it/s]
  7%|▋         | 762/11088 [06:29<1:25:02,  2.02it/s]
  7%|▋         | 763/11088 [06:29<1:25:06,  2.02it/s]
  7%|▋         | 764/11088 [06:30<1:25:03,  2.02it/s]
  7%|▋         | 765/11088 [06:30<1:25:04,  2.02it/s]
  7%|▋         | 766/11088 [06:31<1:25:01,  2.02it/s]
  7%|▋         | 767/11088 [06:31<1:25:01,  2.02it/s]
  7%|▋         | 768/11088 [06:32<1:25:03,  2.02it/s]
  7%|▋         | 769/11088 [06:32<1:25:01,  2.02it/s]
  7%|▋         | 770/11088 [06:33<1:24:59,  2.02it/s]
  7%|▋         | 771/11088 [06:33<1:24:58,  2.02it/s]
  7%|▋         | 772/11088 [06:34<1:24:55,  2.02it/s]
  7%|▋         | 773/11088 [06:34<1:24:59,  2.02it/s]
  7%|▋         | 774/11088 [06:35<1:24:57,  2.02it/s]
  7%|▋         | 775/11088 [06:35<1:24:53,  2.02it/s]
                                                     
{'loss': 4.6105, 'grad_norm': 0.38639992475509644, 'learning_rate': 0.0006988277727682597, 'epoch': 0.98}
+
  7%|▋         | 775/11088 [06:35<1:24:53,  2.02it/s]
  7%|▋         | 776/11088 [06:36<1:25:05,  2.02it/s]
  7%|▋         | 777/11088 [06:36<1:25:02,  2.02it/s]
  7%|▋         | 778/11088 [06:37<1:25:04,  2.02it/s]
  7%|▋         | 779/11088 [06:37<1:25:02,  2.02it/s]
  7%|▋         | 780/11088 [06:38<1:24:58,  2.02it/s]
  7%|▋         | 781/11088 [06:38<1:24:58,  2.02it/s]
  7%|▋         | 782/11088 [06:39<1:24:55,  2.02it/s]
  7%|▋         | 783/11088 [06:39<1:24:54,  2.02it/s]
  7%|▋         | 784/11088 [06:40<1:24:56,  2.02it/s]
  7%|▋         | 785/11088 [06:40<1:24:52,  2.02it/s]
  7%|▋         | 786/11088 [06:41<1:24:45,  2.03it/s]
  7%|▋         | 787/11088 [06:41<1:24:46,  2.03it/s]
  7%|▋         | 788/11088 [06:42<1:24:47,  2.02it/s]
  7%|▋         | 789/11088 [06:42<1:24:44,  2.03it/s]
  7%|▋         | 790/11088 [06:43<1:24:44,  2.03it/s]
  7%|▋         | 791/11088 [06:43<1:24:50,  2.02it/s]
  7%|▋         | 792/11088 [06:44<1:24:18,  2.04it/s]
  7%|▋         | 793/11088 [06:56<11:27:00,  4.00s/it]
  7%|▋         | 794/11088 [06:56<8:26:43,  2.95s/it] 
  7%|▋         | 795/11088 [06:57<6:20:05,  2.22s/it]
  7%|▋         | 796/11088 [06:57<4:51:46,  1.70s/it]
  7%|▋         | 797/11088 [06:58<3:49:43,  1.34s/it]
  7%|▋         | 798/11088 [06:58<3:06:16,  1.09s/it]
  7%|▋         | 799/11088 [06:59<2:35:58,  1.10it/s]
  7%|▋         | 800/11088 [06:59<2:14:34,  1.27it/s]
                                                     
{'loss': 4.5759, 'grad_norm': 0.49646249413490295, 'learning_rate': 0.0007213706041478811, 'epoch': 1.01}
+
  7%|▋         | 800/11088 [06:59<2:14:34,  1.27it/s]
  7%|▋         | 801/11088 [07:00<1:59:48,  1.43it/s]
  7%|▋         | 802/11088 [07:00<1:49:10,  1.57it/s]
  7%|▋         | 803/11088 [07:01<1:41:58,  1.68it/s]
  7%|▋         | 804/11088 [07:01<1:36:45,  1.77it/s]
  7%|▋         | 805/11088 [07:02<1:33:16,  1.84it/s]
  7%|▋         | 806/11088 [07:02<1:30:40,  1.89it/s]
  7%|▋         | 807/11088 [07:03<1:28:48,  1.93it/s]
  7%|▋         | 808/11088 [07:03<1:27:32,  1.96it/s]
  7%|▋         | 809/11088 [07:04<1:26:51,  1.97it/s]
  7%|▋         | 810/11088 [07:04<1:26:09,  1.99it/s]
  7%|▋         | 811/11088 [07:05<1:25:40,  2.00it/s]
  7%|▋         | 812/11088 [07:05<1:25:23,  2.01it/s]
  7%|▋         | 813/11088 [07:06<1:25:16,  2.01it/s]
  7%|▋         | 814/11088 [07:06<1:25:02,  2.01it/s]
  7%|▋         | 815/11088 [07:07<1:25:01,  2.01it/s]
  7%|▋         | 816/11088 [07:07<1:24:50,  2.02it/s]
  7%|▋         | 817/11088 [07:08<1:24:52,  2.02it/s]
  7%|▋         | 818/11088 [07:08<1:24:55,  2.02it/s]
  7%|▋         | 819/11088 [07:09<1:24:49,  2.02it/s]
\ No newline at end of file