pszemraj commited on
Commit
de4b817
1 Parent(s): 407841f

adding model files

Browse files
Files changed (6) hide show
  1. .gitignore +1 -0
  2. README.md +62 -0
  3. config.json +41 -0
  4. pytorch_model.bin +3 -0
  5. trainer_state.json +2887 -0
  6. training_args.bin +3 -0
.gitignore ADDED
@@ -0,0 +1 @@
 
 
1
+ checkpoint-*/
README.md ADDED
@@ -0,0 +1,62 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: mit
3
+ tags:
4
+ - generated_from_trainer
5
+ model-index:
6
+ - name: gpt2-medium-emailgen_DS-multi-clean-100k_Ep-3_Bs-16
7
+ results: []
8
+ ---
9
+
10
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
11
+ should probably proofread and complete it, then remove this comment. -->
12
+
13
+ # gpt2-medium-emailgen_DS-multi-clean-100k_Ep-3_Bs-16
14
+
15
+ This model is a fine-tuned version of [gpt2-medium](https://huggingface.co/gpt2-medium) on the None dataset.
16
+ It achieves the following results on the evaluation set:
17
+ - Loss: 1.5840
18
+
19
+ ## Model description
20
+
21
+ More information needed
22
+
23
+ ## Intended uses & limitations
24
+
25
+ More information needed
26
+
27
+ ## Training and evaluation data
28
+
29
+ More information needed
30
+
31
+ ## Training procedure
32
+
33
+ ### Training hyperparameters
34
+
35
+ The following hyperparameters were used during training:
36
+ - learning_rate: 0.001
37
+ - train_batch_size: 16
38
+ - eval_batch_size: 16
39
+ - seed: 42
40
+ - distributed_type: multi-GPU
41
+ - gradient_accumulation_steps: 8
42
+ - total_train_batch_size: 128
43
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
44
+ - lr_scheduler_type: cosine
45
+ - lr_scheduler_warmup_ratio: 0.02
46
+ - num_epochs: 3
47
+
48
+ ### Training results
49
+
50
+ | Training Loss | Epoch | Step | Validation Loss |
51
+ |:-------------:|:-----:|:----:|:---------------:|
52
+ | 1.8701 | 1.0 | 789 | 1.8378 |
53
+ | 1.5065 | 2.0 | 1578 | 1.6176 |
54
+ | 1.1873 | 3.0 | 2367 | 1.5840 |
55
+
56
+
57
+ ### Framework versions
58
+
59
+ - Transformers 4.22.2
60
+ - Pytorch 1.10.0+cu113
61
+ - Datasets 2.5.1
62
+ - Tokenizers 0.12.1
config.json ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "gpt2-medium",
3
+ "activation_function": "gelu_new",
4
+ "architectures": [
5
+ "GPT2LMHeadModel"
6
+ ],
7
+ "attn_pdrop": 0.1,
8
+ "bos_token_id": 50256,
9
+ "embd_pdrop": 0.1,
10
+ "eos_token_id": 50256,
11
+ "initializer_range": 0.02,
12
+ "layer_norm_epsilon": 1e-05,
13
+ "model_type": "gpt2",
14
+ "n_ctx": 1024,
15
+ "n_embd": 1024,
16
+ "n_head": 16,
17
+ "n_inner": null,
18
+ "n_layer": 24,
19
+ "n_positions": 1024,
20
+ "n_special": 0,
21
+ "predict_special_tokens": true,
22
+ "reorder_and_upcast_attn": false,
23
+ "resid_pdrop": 0.1,
24
+ "scale_attn_by_inverse_layer_idx": false,
25
+ "scale_attn_weights": true,
26
+ "summary_activation": null,
27
+ "summary_first_dropout": 0.1,
28
+ "summary_proj_to_labels": true,
29
+ "summary_type": "cls_index",
30
+ "summary_use_proj": true,
31
+ "task_specific_params": {
32
+ "text-generation": {
33
+ "do_sample": true,
34
+ "max_length": 50
35
+ }
36
+ },
37
+ "torch_dtype": "float32",
38
+ "transformers_version": "4.22.2",
39
+ "use_cache": false,
40
+ "vocab_size": 50257
41
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c96bfc8f0d48bcfdb8c6eaa8bad49bb0ecf584b28fa6feccdbc6c141a3c4945a
3
+ size 1444566873
trainer_state.json ADDED
@@ -0,0 +1,2887 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.9998415967052114,
5
+ "global_step": 2367,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.01,
12
+ "learning_rate": 0.00010416666666666667,
13
+ "loss": 5.6655,
14
+ "step": 5
15
+ },
16
+ {
17
+ "epoch": 0.01,
18
+ "learning_rate": 0.00020833333333333335,
19
+ "loss": 4.5516,
20
+ "step": 10
21
+ },
22
+ {
23
+ "epoch": 0.02,
24
+ "learning_rate": 0.0003125,
25
+ "loss": 3.7337,
26
+ "step": 15
27
+ },
28
+ {
29
+ "epoch": 0.03,
30
+ "learning_rate": 0.0004166666666666667,
31
+ "loss": 3.2268,
32
+ "step": 20
33
+ },
34
+ {
35
+ "epoch": 0.03,
36
+ "learning_rate": 0.0005208333333333334,
37
+ "loss": 2.9287,
38
+ "step": 25
39
+ },
40
+ {
41
+ "epoch": 0.04,
42
+ "learning_rate": 0.000625,
43
+ "loss": 2.8609,
44
+ "step": 30
45
+ },
46
+ {
47
+ "epoch": 0.04,
48
+ "learning_rate": 0.0007291666666666666,
49
+ "loss": 2.7533,
50
+ "step": 35
51
+ },
52
+ {
53
+ "epoch": 0.05,
54
+ "learning_rate": 0.0008333333333333334,
55
+ "loss": 2.7291,
56
+ "step": 40
57
+ },
58
+ {
59
+ "epoch": 0.06,
60
+ "learning_rate": 0.0009375,
61
+ "loss": 2.5916,
62
+ "step": 45
63
+ },
64
+ {
65
+ "epoch": 0.06,
66
+ "learning_rate": 0.0009999981647383803,
67
+ "loss": 2.605,
68
+ "step": 50
69
+ },
70
+ {
71
+ "epoch": 0.07,
72
+ "learning_rate": 0.0009999775181998842,
73
+ "loss": 2.6466,
74
+ "step": 55
75
+ },
76
+ {
77
+ "epoch": 0.08,
78
+ "learning_rate": 0.0009999339319963169,
79
+ "loss": 2.5633,
80
+ "step": 60
81
+ },
82
+ {
83
+ "epoch": 0.08,
84
+ "learning_rate": 0.000999867408127474,
85
+ "loss": 2.628,
86
+ "step": 65
87
+ },
88
+ {
89
+ "epoch": 0.09,
90
+ "learning_rate": 0.0009997779496455639,
91
+ "loss": 2.5335,
92
+ "step": 70
93
+ },
94
+ {
95
+ "epoch": 0.1,
96
+ "learning_rate": 0.0009996655606550656,
97
+ "loss": 2.5589,
98
+ "step": 75
99
+ },
100
+ {
101
+ "epoch": 0.1,
102
+ "learning_rate": 0.000999530246312543,
103
+ "loss": 2.5707,
104
+ "step": 80
105
+ },
106
+ {
107
+ "epoch": 0.11,
108
+ "learning_rate": 0.0009993720128264065,
109
+ "loss": 2.428,
110
+ "step": 85
111
+ },
112
+ {
113
+ "epoch": 0.11,
114
+ "learning_rate": 0.000999190867456629,
115
+ "loss": 2.4371,
116
+ "step": 90
117
+ },
118
+ {
119
+ "epoch": 0.12,
120
+ "learning_rate": 0.0009989868185144122,
121
+ "loss": 2.4955,
122
+ "step": 95
123
+ },
124
+ {
125
+ "epoch": 0.13,
126
+ "learning_rate": 0.0009987598753618061,
127
+ "loss": 2.4295,
128
+ "step": 100
129
+ },
130
+ {
131
+ "epoch": 0.13,
132
+ "learning_rate": 0.0009985100484112787,
133
+ "loss": 2.3623,
134
+ "step": 105
135
+ },
136
+ {
137
+ "epoch": 0.14,
138
+ "learning_rate": 0.0009982373491252382,
139
+ "loss": 2.3422,
140
+ "step": 110
141
+ },
142
+ {
143
+ "epoch": 0.15,
144
+ "learning_rate": 0.0009979417900155081,
145
+ "loss": 2.3134,
146
+ "step": 115
147
+ },
148
+ {
149
+ "epoch": 0.15,
150
+ "learning_rate": 0.0009976233846427521,
151
+ "loss": 2.4091,
152
+ "step": 120
153
+ },
154
+ {
155
+ "epoch": 0.16,
156
+ "learning_rate": 0.000997282147615852,
157
+ "loss": 2.4004,
158
+ "step": 125
159
+ },
160
+ {
161
+ "epoch": 0.16,
162
+ "learning_rate": 0.0009969180945912381,
163
+ "loss": 2.3233,
164
+ "step": 130
165
+ },
166
+ {
167
+ "epoch": 0.17,
168
+ "learning_rate": 0.0009965312422721704,
169
+ "loss": 2.3183,
170
+ "step": 135
171
+ },
172
+ {
173
+ "epoch": 0.18,
174
+ "learning_rate": 0.0009961216084079724,
175
+ "loss": 2.3771,
176
+ "step": 140
177
+ },
178
+ {
179
+ "epoch": 0.18,
180
+ "learning_rate": 0.0009956892117932161,
181
+ "loss": 2.2779,
182
+ "step": 145
183
+ },
184
+ {
185
+ "epoch": 0.19,
186
+ "learning_rate": 0.000995234072266861,
187
+ "loss": 2.303,
188
+ "step": 150
189
+ },
190
+ {
191
+ "epoch": 0.2,
192
+ "learning_rate": 0.0009947562107113423,
193
+ "loss": 2.3452,
194
+ "step": 155
195
+ },
196
+ {
197
+ "epoch": 0.2,
198
+ "learning_rate": 0.000994255649051614,
199
+ "loss": 2.2872,
200
+ "step": 160
201
+ },
202
+ {
203
+ "epoch": 0.21,
204
+ "learning_rate": 0.0009937324102541423,
205
+ "loss": 2.3187,
206
+ "step": 165
207
+ },
208
+ {
209
+ "epoch": 0.22,
210
+ "learning_rate": 0.0009931865183258524,
211
+ "loss": 2.2837,
212
+ "step": 170
213
+ },
214
+ {
215
+ "epoch": 0.22,
216
+ "learning_rate": 0.0009926179983130263,
217
+ "loss": 2.4024,
218
+ "step": 175
219
+ },
220
+ {
221
+ "epoch": 0.23,
222
+ "learning_rate": 0.0009920268763001542,
223
+ "loss": 2.347,
224
+ "step": 180
225
+ },
226
+ {
227
+ "epoch": 0.23,
228
+ "learning_rate": 0.0009914131794087377,
229
+ "loss": 2.3126,
230
+ "step": 185
231
+ },
232
+ {
233
+ "epoch": 0.24,
234
+ "learning_rate": 0.0009907769357960449,
235
+ "loss": 2.2042,
236
+ "step": 190
237
+ },
238
+ {
239
+ "epoch": 0.25,
240
+ "learning_rate": 0.0009901181746538195,
241
+ "loss": 2.1674,
242
+ "step": 195
243
+ },
244
+ {
245
+ "epoch": 0.25,
246
+ "learning_rate": 0.0009894369262069401,
247
+ "loss": 2.1963,
248
+ "step": 200
249
+ },
250
+ {
251
+ "epoch": 0.26,
252
+ "learning_rate": 0.0009887332217120347,
253
+ "loss": 2.2971,
254
+ "step": 205
255
+ },
256
+ {
257
+ "epoch": 0.27,
258
+ "learning_rate": 0.0009880070934560458,
259
+ "loss": 2.1949,
260
+ "step": 210
261
+ },
262
+ {
263
+ "epoch": 0.27,
264
+ "learning_rate": 0.0009872585747547494,
265
+ "loss": 2.1544,
266
+ "step": 215
267
+ },
268
+ {
269
+ "epoch": 0.28,
270
+ "learning_rate": 0.0009864876999512264,
271
+ "loss": 2.1866,
272
+ "step": 220
273
+ },
274
+ {
275
+ "epoch": 0.29,
276
+ "learning_rate": 0.0009856945044142864,
277
+ "loss": 2.198,
278
+ "step": 225
279
+ },
280
+ {
281
+ "epoch": 0.29,
282
+ "learning_rate": 0.000984879024536846,
283
+ "loss": 2.2507,
284
+ "step": 230
285
+ },
286
+ {
287
+ "epoch": 0.3,
288
+ "learning_rate": 0.0009840412977342582,
289
+ "loss": 2.1326,
290
+ "step": 235
291
+ },
292
+ {
293
+ "epoch": 0.3,
294
+ "learning_rate": 0.000983181362442595,
295
+ "loss": 2.204,
296
+ "step": 240
297
+ },
298
+ {
299
+ "epoch": 0.31,
300
+ "learning_rate": 0.0009822992581168867,
301
+ "loss": 2.2197,
302
+ "step": 245
303
+ },
304
+ {
305
+ "epoch": 0.32,
306
+ "learning_rate": 0.0009813950252293075,
307
+ "loss": 2.1562,
308
+ "step": 250
309
+ },
310
+ {
311
+ "epoch": 0.32,
312
+ "learning_rate": 0.0009804687052673228,
313
+ "loss": 2.2193,
314
+ "step": 255
315
+ },
316
+ {
317
+ "epoch": 0.33,
318
+ "learning_rate": 0.0009795203407317828,
319
+ "loss": 2.2668,
320
+ "step": 260
321
+ },
322
+ {
323
+ "epoch": 0.34,
324
+ "learning_rate": 0.0009785499751349732,
325
+ "loss": 2.2434,
326
+ "step": 265
327
+ },
328
+ {
329
+ "epoch": 0.34,
330
+ "learning_rate": 0.0009775576529986198,
331
+ "loss": 2.1776,
332
+ "step": 270
333
+ },
334
+ {
335
+ "epoch": 0.35,
336
+ "learning_rate": 0.0009765434198518448,
337
+ "loss": 2.2592,
338
+ "step": 275
339
+ },
340
+ {
341
+ "epoch": 0.35,
342
+ "learning_rate": 0.0009755073222290775,
343
+ "loss": 2.1072,
344
+ "step": 280
345
+ },
346
+ {
347
+ "epoch": 0.36,
348
+ "learning_rate": 0.0009744494076679204,
349
+ "loss": 2.0784,
350
+ "step": 285
351
+ },
352
+ {
353
+ "epoch": 0.37,
354
+ "learning_rate": 0.0009733697247069677,
355
+ "loss": 2.1776,
356
+ "step": 290
357
+ },
358
+ {
359
+ "epoch": 0.37,
360
+ "learning_rate": 0.0009722683228835774,
361
+ "loss": 2.1752,
362
+ "step": 295
363
+ },
364
+ {
365
+ "epoch": 0.38,
366
+ "learning_rate": 0.0009711452527315997,
367
+ "loss": 2.0958,
368
+ "step": 300
369
+ },
370
+ {
371
+ "epoch": 0.39,
372
+ "learning_rate": 0.0009700005657790578,
373
+ "loss": 2.1055,
374
+ "step": 305
375
+ },
376
+ {
377
+ "epoch": 0.39,
378
+ "learning_rate": 0.0009688343145457837,
379
+ "loss": 2.1135,
380
+ "step": 310
381
+ },
382
+ {
383
+ "epoch": 0.4,
384
+ "learning_rate": 0.0009676465525410088,
385
+ "loss": 2.0275,
386
+ "step": 315
387
+ },
388
+ {
389
+ "epoch": 0.41,
390
+ "learning_rate": 0.0009664373342609086,
391
+ "loss": 2.0031,
392
+ "step": 320
393
+ },
394
+ {
395
+ "epoch": 0.41,
396
+ "learning_rate": 0.0009652067151861023,
397
+ "loss": 2.1489,
398
+ "step": 325
399
+ },
400
+ {
401
+ "epoch": 0.42,
402
+ "learning_rate": 0.0009639547517791077,
403
+ "loss": 2.1305,
404
+ "step": 330
405
+ },
406
+ {
407
+ "epoch": 0.42,
408
+ "learning_rate": 0.0009626815014817496,
409
+ "loss": 2.0894,
410
+ "step": 335
411
+ },
412
+ {
413
+ "epoch": 0.43,
414
+ "learning_rate": 0.0009613870227125259,
415
+ "loss": 2.0822,
416
+ "step": 340
417
+ },
418
+ {
419
+ "epoch": 0.44,
420
+ "learning_rate": 0.0009600713748639258,
421
+ "loss": 2.1499,
422
+ "step": 345
423
+ },
424
+ {
425
+ "epoch": 0.44,
426
+ "learning_rate": 0.0009587346182997054,
427
+ "loss": 2.0848,
428
+ "step": 350
429
+ },
430
+ {
431
+ "epoch": 0.45,
432
+ "learning_rate": 0.0009573768143521182,
433
+ "loss": 2.0754,
434
+ "step": 355
435
+ },
436
+ {
437
+ "epoch": 0.46,
438
+ "learning_rate": 0.000955998025319101,
439
+ "loss": 2.1364,
440
+ "step": 360
441
+ },
442
+ {
443
+ "epoch": 0.46,
444
+ "learning_rate": 0.0009545983144614152,
445
+ "loss": 2.066,
446
+ "step": 365
447
+ },
448
+ {
449
+ "epoch": 0.47,
450
+ "learning_rate": 0.0009531777459997453,
451
+ "loss": 2.1016,
452
+ "step": 370
453
+ },
454
+ {
455
+ "epoch": 0.48,
456
+ "learning_rate": 0.0009517363851117511,
457
+ "loss": 2.0479,
458
+ "step": 375
459
+ },
460
+ {
461
+ "epoch": 0.48,
462
+ "learning_rate": 0.0009502742979290782,
463
+ "loss": 2.0646,
464
+ "step": 380
465
+ },
466
+ {
467
+ "epoch": 0.49,
468
+ "learning_rate": 0.0009487915515343235,
469
+ "loss": 2.1756,
470
+ "step": 385
471
+ },
472
+ {
473
+ "epoch": 0.49,
474
+ "learning_rate": 0.0009472882139579572,
475
+ "loss": 2.0339,
476
+ "step": 390
477
+ },
478
+ {
479
+ "epoch": 0.5,
480
+ "learning_rate": 0.0009457643541752017,
481
+ "loss": 2.1112,
482
+ "step": 395
483
+ },
484
+ {
485
+ "epoch": 0.51,
486
+ "learning_rate": 0.0009442200421028666,
487
+ "loss": 2.1327,
488
+ "step": 400
489
+ },
490
+ {
491
+ "epoch": 0.51,
492
+ "learning_rate": 0.0009426553485961416,
493
+ "loss": 2.0274,
494
+ "step": 405
495
+ },
496
+ {
497
+ "epoch": 0.52,
498
+ "learning_rate": 0.0009410703454453442,
499
+ "loss": 1.9823,
500
+ "step": 410
501
+ },
502
+ {
503
+ "epoch": 0.53,
504
+ "learning_rate": 0.0009394651053726273,
505
+ "loss": 1.9665,
506
+ "step": 415
507
+ },
508
+ {
509
+ "epoch": 0.53,
510
+ "learning_rate": 0.0009378397020286417,
511
+ "loss": 2.0468,
512
+ "step": 420
513
+ },
514
+ {
515
+ "epoch": 0.54,
516
+ "learning_rate": 0.0009361942099891571,
517
+ "loss": 1.9778,
518
+ "step": 425
519
+ },
520
+ {
521
+ "epoch": 0.54,
522
+ "learning_rate": 0.0009345287047516411,
523
+ "loss": 2.0264,
524
+ "step": 430
525
+ },
526
+ {
527
+ "epoch": 0.55,
528
+ "learning_rate": 0.0009328432627317937,
529
+ "loss": 2.1288,
530
+ "step": 435
531
+ },
532
+ {
533
+ "epoch": 0.56,
534
+ "learning_rate": 0.0009311379612600432,
535
+ "loss": 1.9488,
536
+ "step": 440
537
+ },
538
+ {
539
+ "epoch": 0.56,
540
+ "learning_rate": 0.0009294128785779969,
541
+ "loss": 1.9984,
542
+ "step": 445
543
+ },
544
+ {
545
+ "epoch": 0.57,
546
+ "learning_rate": 0.0009276680938348512,
547
+ "loss": 2.0063,
548
+ "step": 450
549
+ },
550
+ {
551
+ "epoch": 0.58,
552
+ "learning_rate": 0.000925903687083761,
553
+ "loss": 2.052,
554
+ "step": 455
555
+ },
556
+ {
557
+ "epoch": 0.58,
558
+ "learning_rate": 0.0009241197392781662,
559
+ "loss": 2.0739,
560
+ "step": 460
561
+ },
562
+ {
563
+ "epoch": 0.59,
564
+ "learning_rate": 0.000922316332268077,
565
+ "loss": 1.9291,
566
+ "step": 465
567
+ },
568
+ {
569
+ "epoch": 0.6,
570
+ "learning_rate": 0.00092049354879632,
571
+ "loss": 2.0575,
572
+ "step": 470
573
+ },
574
+ {
575
+ "epoch": 0.6,
576
+ "learning_rate": 0.0009186514724947399,
577
+ "loss": 1.9736,
578
+ "step": 475
579
+ },
580
+ {
581
+ "epoch": 0.61,
582
+ "learning_rate": 0.0009167901878803637,
583
+ "loss": 2.019,
584
+ "step": 480
585
+ },
586
+ {
587
+ "epoch": 0.61,
588
+ "learning_rate": 0.0009149097803515225,
589
+ "loss": 1.8767,
590
+ "step": 485
591
+ },
592
+ {
593
+ "epoch": 0.62,
594
+ "learning_rate": 0.0009130103361839337,
595
+ "loss": 2.144,
596
+ "step": 490
597
+ },
598
+ {
599
+ "epoch": 0.63,
600
+ "learning_rate": 0.0009110919425267414,
601
+ "loss": 1.9369,
602
+ "step": 495
603
+ },
604
+ {
605
+ "epoch": 0.63,
606
+ "learning_rate": 0.0009091546873985196,
607
+ "loss": 2.0787,
608
+ "step": 500
609
+ },
610
+ {
611
+ "epoch": 0.64,
612
+ "learning_rate": 0.0009071986596832321,
613
+ "loss": 1.8462,
614
+ "step": 505
615
+ },
616
+ {
617
+ "epoch": 0.65,
618
+ "learning_rate": 0.0009052239491261558,
619
+ "loss": 1.9782,
620
+ "step": 510
621
+ },
622
+ {
623
+ "epoch": 0.65,
624
+ "learning_rate": 0.0009032306463297619,
625
+ "loss": 1.912,
626
+ "step": 515
627
+ },
628
+ {
629
+ "epoch": 0.66,
630
+ "learning_rate": 0.0009012188427495597,
631
+ "loss": 2.0624,
632
+ "step": 520
633
+ },
634
+ {
635
+ "epoch": 0.67,
636
+ "learning_rate": 0.0008991886306899002,
637
+ "loss": 1.9532,
638
+ "step": 525
639
+ },
640
+ {
641
+ "epoch": 0.67,
642
+ "learning_rate": 0.0008971401032997409,
643
+ "loss": 2.0872,
644
+ "step": 530
645
+ },
646
+ {
647
+ "epoch": 0.68,
648
+ "learning_rate": 0.0008950733545683725,
649
+ "loss": 2.0329,
650
+ "step": 535
651
+ },
652
+ {
653
+ "epoch": 0.68,
654
+ "learning_rate": 0.0008929884793211061,
655
+ "loss": 1.9024,
656
+ "step": 540
657
+ },
658
+ {
659
+ "epoch": 0.69,
660
+ "learning_rate": 0.0008908855732149224,
661
+ "loss": 1.9682,
662
+ "step": 545
663
+ },
664
+ {
665
+ "epoch": 0.7,
666
+ "learning_rate": 0.0008887647327340832,
667
+ "loss": 1.9982,
668
+ "step": 550
669
+ },
670
+ {
671
+ "epoch": 0.7,
672
+ "learning_rate": 0.0008866260551857045,
673
+ "loss": 1.9342,
674
+ "step": 555
675
+ },
676
+ {
677
+ "epoch": 0.71,
678
+ "learning_rate": 0.0008844696386952916,
679
+ "loss": 1.8963,
680
+ "step": 560
681
+ },
682
+ {
683
+ "epoch": 0.72,
684
+ "learning_rate": 0.0008822955822022372,
685
+ "loss": 1.9586,
686
+ "step": 565
687
+ },
688
+ {
689
+ "epoch": 0.72,
690
+ "learning_rate": 0.0008801039854552822,
691
+ "loss": 1.8984,
692
+ "step": 570
693
+ },
694
+ {
695
+ "epoch": 0.73,
696
+ "learning_rate": 0.0008778949490079383,
697
+ "loss": 1.8707,
698
+ "step": 575
699
+ },
700
+ {
701
+ "epoch": 0.73,
702
+ "learning_rate": 0.0008756685742138751,
703
+ "loss": 1.9667,
704
+ "step": 580
705
+ },
706
+ {
707
+ "epoch": 0.74,
708
+ "learning_rate": 0.0008734249632222702,
709
+ "loss": 1.931,
710
+ "step": 585
711
+ },
712
+ {
713
+ "epoch": 0.75,
714
+ "learning_rate": 0.0008711642189731211,
715
+ "loss": 1.9109,
716
+ "step": 590
717
+ },
718
+ {
719
+ "epoch": 0.75,
720
+ "learning_rate": 0.0008688864451925238,
721
+ "loss": 1.8908,
722
+ "step": 595
723
+ },
724
+ {
725
+ "epoch": 0.76,
726
+ "learning_rate": 0.0008665917463879124,
727
+ "loss": 2.0963,
728
+ "step": 600
729
+ },
730
+ {
731
+ "epoch": 0.77,
732
+ "learning_rate": 0.0008642802278432649,
733
+ "loss": 1.9331,
734
+ "step": 605
735
+ },
736
+ {
737
+ "epoch": 0.77,
738
+ "learning_rate": 0.0008619519956142725,
739
+ "loss": 2.0053,
740
+ "step": 610
741
+ },
742
+ {
743
+ "epoch": 0.78,
744
+ "learning_rate": 0.0008596071565234733,
745
+ "loss": 1.9626,
746
+ "step": 615
747
+ },
748
+ {
749
+ "epoch": 0.79,
750
+ "learning_rate": 0.0008572458181553517,
751
+ "loss": 1.9638,
752
+ "step": 620
753
+ },
754
+ {
755
+ "epoch": 0.79,
756
+ "learning_rate": 0.0008548680888514013,
757
+ "loss": 1.8846,
758
+ "step": 625
759
+ },
760
+ {
761
+ "epoch": 0.8,
762
+ "learning_rate": 0.0008524740777051554,
763
+ "loss": 2.0264,
764
+ "step": 630
765
+ },
766
+ {
767
+ "epoch": 0.8,
768
+ "learning_rate": 0.0008500638945571804,
769
+ "loss": 1.9709,
770
+ "step": 635
771
+ },
772
+ {
773
+ "epoch": 0.81,
774
+ "learning_rate": 0.0008476376499900369,
775
+ "loss": 1.8799,
776
+ "step": 640
777
+ },
778
+ {
779
+ "epoch": 0.82,
780
+ "learning_rate": 0.0008451954553232055,
781
+ "loss": 1.9257,
782
+ "step": 645
783
+ },
784
+ {
785
+ "epoch": 0.82,
786
+ "learning_rate": 0.0008427374226079803,
787
+ "loss": 1.9897,
788
+ "step": 650
789
+ },
790
+ {
791
+ "epoch": 0.83,
792
+ "learning_rate": 0.0008402636646223262,
793
+ "loss": 1.9573,
794
+ "step": 655
795
+ },
796
+ {
797
+ "epoch": 0.84,
798
+ "learning_rate": 0.000837774294865706,
799
+ "loss": 1.9529,
800
+ "step": 660
801
+ },
802
+ {
803
+ "epoch": 0.84,
804
+ "learning_rate": 0.0008352694275538723,
805
+ "loss": 2.0161,
806
+ "step": 665
807
+ },
808
+ {
809
+ "epoch": 0.85,
810
+ "learning_rate": 0.0008327491776136267,
811
+ "loss": 1.942,
812
+ "step": 670
813
+ },
814
+ {
815
+ "epoch": 0.86,
816
+ "learning_rate": 0.0008302136606775479,
817
+ "loss": 1.8917,
818
+ "step": 675
819
+ },
820
+ {
821
+ "epoch": 0.86,
822
+ "learning_rate": 0.000827662993078685,
823
+ "loss": 1.9157,
824
+ "step": 680
825
+ },
826
+ {
827
+ "epoch": 0.87,
828
+ "learning_rate": 0.0008250972918452207,
829
+ "loss": 1.9969,
830
+ "step": 685
831
+ },
832
+ {
833
+ "epoch": 0.87,
834
+ "learning_rate": 0.0008225166746951022,
835
+ "loss": 1.8957,
836
+ "step": 690
837
+ },
838
+ {
839
+ "epoch": 0.88,
840
+ "learning_rate": 0.0008199212600306396,
841
+ "loss": 1.8494,
842
+ "step": 695
843
+ },
844
+ {
845
+ "epoch": 0.89,
846
+ "learning_rate": 0.0008173111669330733,
847
+ "loss": 1.8495,
848
+ "step": 700
849
+ },
850
+ {
851
+ "epoch": 0.89,
852
+ "learning_rate": 0.0008146865151571108,
853
+ "loss": 1.8407,
854
+ "step": 705
855
+ },
856
+ {
857
+ "epoch": 0.9,
858
+ "learning_rate": 0.0008120474251254319,
859
+ "loss": 1.7461,
860
+ "step": 710
861
+ },
862
+ {
863
+ "epoch": 0.91,
864
+ "learning_rate": 0.0008093940179231643,
865
+ "loss": 1.8608,
866
+ "step": 715
867
+ },
868
+ {
869
+ "epoch": 0.91,
870
+ "learning_rate": 0.0008067264152923268,
871
+ "loss": 1.9029,
872
+ "step": 720
873
+ },
874
+ {
875
+ "epoch": 0.92,
876
+ "learning_rate": 0.0008040447396262442,
877
+ "loss": 1.8156,
878
+ "step": 725
879
+ },
880
+ {
881
+ "epoch": 0.93,
882
+ "learning_rate": 0.0008013491139639324,
883
+ "loss": 1.8412,
884
+ "step": 730
885
+ },
886
+ {
887
+ "epoch": 0.93,
888
+ "learning_rate": 0.0007986396619844518,
889
+ "loss": 1.8522,
890
+ "step": 735
891
+ },
892
+ {
893
+ "epoch": 0.94,
894
+ "learning_rate": 0.0007959165080012341,
895
+ "loss": 1.9613,
896
+ "step": 740
897
+ },
898
+ {
899
+ "epoch": 0.94,
900
+ "learning_rate": 0.0007931797769563777,
901
+ "loss": 1.917,
902
+ "step": 745
903
+ },
904
+ {
905
+ "epoch": 0.95,
906
+ "learning_rate": 0.0007904295944149156,
907
+ "loss": 1.8676,
908
+ "step": 750
909
+ },
910
+ {
911
+ "epoch": 0.96,
912
+ "learning_rate": 0.0007876660865590539,
913
+ "loss": 1.839,
914
+ "step": 755
915
+ },
916
+ {
917
+ "epoch": 0.96,
918
+ "learning_rate": 0.0007848893801823833,
919
+ "loss": 1.8689,
920
+ "step": 760
921
+ },
922
+ {
923
+ "epoch": 0.97,
924
+ "learning_rate": 0.0007820996026840606,
925
+ "loss": 1.9667,
926
+ "step": 765
927
+ },
928
+ {
929
+ "epoch": 0.98,
930
+ "learning_rate": 0.000779296882062964,
931
+ "loss": 1.8947,
932
+ "step": 770
933
+ },
934
+ {
935
+ "epoch": 0.98,
936
+ "learning_rate": 0.0007764813469118201,
937
+ "loss": 1.8329,
938
+ "step": 775
939
+ },
940
+ {
941
+ "epoch": 0.99,
942
+ "learning_rate": 0.0007736531264113041,
943
+ "loss": 1.7263,
944
+ "step": 780
945
+ },
946
+ {
947
+ "epoch": 0.99,
948
+ "learning_rate": 0.0007708123503241124,
949
+ "loss": 1.8701,
950
+ "step": 785
951
+ },
952
+ {
953
+ "epoch": 1.0,
954
+ "eval_loss": 1.8378170728683472,
955
+ "eval_runtime": 367.2627,
956
+ "eval_samples_per_second": 34.874,
957
+ "eval_steps_per_second": 2.181,
958
+ "step": 789
959
+ },
960
+ {
961
+ "epoch": 1.0,
962
+ "learning_rate": 0.0007679591489890098,
963
+ "loss": 1.8124,
964
+ "step": 790
965
+ },
966
+ {
967
+ "epoch": 1.01,
968
+ "learning_rate": 0.0007650936533148485,
969
+ "loss": 1.7062,
970
+ "step": 795
971
+ },
972
+ {
973
+ "epoch": 1.01,
974
+ "learning_rate": 0.0007622159947745617,
975
+ "loss": 1.6682,
976
+ "step": 800
977
+ },
978
+ {
979
+ "epoch": 1.02,
980
+ "learning_rate": 0.0007593263053991325,
981
+ "loss": 1.6572,
982
+ "step": 805
983
+ },
984
+ {
985
+ "epoch": 1.03,
986
+ "learning_rate": 0.0007564247177715349,
987
+ "loss": 1.6605,
988
+ "step": 810
989
+ },
990
+ {
991
+ "epoch": 1.03,
992
+ "learning_rate": 0.0007535113650206517,
993
+ "loss": 1.7364,
994
+ "step": 815
995
+ },
996
+ {
997
+ "epoch": 1.04,
998
+ "learning_rate": 0.0007505863808151661,
999
+ "loss": 1.628,
1000
+ "step": 820
1001
+ },
1002
+ {
1003
+ "epoch": 1.05,
1004
+ "learning_rate": 0.0007476498993574277,
1005
+ "loss": 1.5975,
1006
+ "step": 825
1007
+ },
1008
+ {
1009
+ "epoch": 1.05,
1010
+ "learning_rate": 0.0007447020553772973,
1011
+ "loss": 1.6411,
1012
+ "step": 830
1013
+ },
1014
+ {
1015
+ "epoch": 1.06,
1016
+ "learning_rate": 0.0007417429841259631,
1017
+ "loss": 1.6513,
1018
+ "step": 835
1019
+ },
1020
+ {
1021
+ "epoch": 1.06,
1022
+ "learning_rate": 0.0007387728213697366,
1023
+ "loss": 1.6971,
1024
+ "step": 840
1025
+ },
1026
+ {
1027
+ "epoch": 1.07,
1028
+ "learning_rate": 0.0007357917033838228,
1029
+ "loss": 1.6577,
1030
+ "step": 845
1031
+ },
1032
+ {
1033
+ "epoch": 1.08,
1034
+ "learning_rate": 0.0007327997669460682,
1035
+ "loss": 1.6068,
1036
+ "step": 850
1037
+ },
1038
+ {
1039
+ "epoch": 1.08,
1040
+ "learning_rate": 0.0007297971493306848,
1041
+ "loss": 1.5742,
1042
+ "step": 855
1043
+ },
1044
+ {
1045
+ "epoch": 1.09,
1046
+ "learning_rate": 0.0007267839883019515,
1047
+ "loss": 1.6666,
1048
+ "step": 860
1049
+ },
1050
+ {
1051
+ "epoch": 1.1,
1052
+ "learning_rate": 0.0007237604221078942,
1053
+ "loss": 1.7017,
1054
+ "step": 865
1055
+ },
1056
+ {
1057
+ "epoch": 1.1,
1058
+ "learning_rate": 0.000720726589473942,
1059
+ "loss": 1.6573,
1060
+ "step": 870
1061
+ },
1062
+ {
1063
+ "epoch": 1.11,
1064
+ "learning_rate": 0.000717682629596563,
1065
+ "loss": 1.4819,
1066
+ "step": 875
1067
+ },
1068
+ {
1069
+ "epoch": 1.12,
1070
+ "learning_rate": 0.0007146286821368767,
1071
+ "loss": 1.6235,
1072
+ "step": 880
1073
+ },
1074
+ {
1075
+ "epoch": 1.12,
1076
+ "learning_rate": 0.0007115648872142474,
1077
+ "loss": 1.6736,
1078
+ "step": 885
1079
+ },
1080
+ {
1081
+ "epoch": 1.13,
1082
+ "learning_rate": 0.0007084913853998542,
1083
+ "loss": 1.6477,
1084
+ "step": 890
1085
+ },
1086
+ {
1087
+ "epoch": 1.13,
1088
+ "learning_rate": 0.0007054083177102423,
1089
+ "loss": 1.6538,
1090
+ "step": 895
1091
+ },
1092
+ {
1093
+ "epoch": 1.14,
1094
+ "learning_rate": 0.0007023158256008521,
1095
+ "loss": 1.5688,
1096
+ "step": 900
1097
+ },
1098
+ {
1099
+ "epoch": 1.15,
1100
+ "learning_rate": 0.0006992140509595303,
1101
+ "loss": 1.5525,
1102
+ "step": 905
1103
+ },
1104
+ {
1105
+ "epoch": 1.15,
1106
+ "learning_rate": 0.0006961031361000184,
1107
+ "loss": 1.7425,
1108
+ "step": 910
1109
+ },
1110
+ {
1111
+ "epoch": 1.16,
1112
+ "learning_rate": 0.0006929832237554241,
1113
+ "loss": 1.559,
1114
+ "step": 915
1115
+ },
1116
+ {
1117
+ "epoch": 1.17,
1118
+ "learning_rate": 0.0006898544570716722,
1119
+ "loss": 1.5379,
1120
+ "step": 920
1121
+ },
1122
+ {
1123
+ "epoch": 1.17,
1124
+ "learning_rate": 0.0006867169796009369,
1125
+ "loss": 1.586,
1126
+ "step": 925
1127
+ },
1128
+ {
1129
+ "epoch": 1.18,
1130
+ "learning_rate": 0.0006835709352950558,
1131
+ "loss": 1.6424,
1132
+ "step": 930
1133
+ },
1134
+ {
1135
+ "epoch": 1.19,
1136
+ "learning_rate": 0.000680416468498924,
1137
+ "loss": 1.6002,
1138
+ "step": 935
1139
+ },
1140
+ {
1141
+ "epoch": 1.19,
1142
+ "learning_rate": 0.0006772537239438732,
1143
+ "loss": 1.5361,
1144
+ "step": 940
1145
+ },
1146
+ {
1147
+ "epoch": 1.2,
1148
+ "learning_rate": 0.0006740828467410294,
1149
+ "loss": 1.5436,
1150
+ "step": 945
1151
+ },
1152
+ {
1153
+ "epoch": 1.2,
1154
+ "learning_rate": 0.0006709039823746564,
1155
+ "loss": 1.5845,
1156
+ "step": 950
1157
+ },
1158
+ {
1159
+ "epoch": 1.21,
1160
+ "learning_rate": 0.0006677172766954794,
1161
+ "loss": 1.5275,
1162
+ "step": 955
1163
+ },
1164
+ {
1165
+ "epoch": 1.22,
1166
+ "learning_rate": 0.0006645228759139949,
1167
+ "loss": 1.5879,
1168
+ "step": 960
1169
+ },
1170
+ {
1171
+ "epoch": 1.22,
1172
+ "learning_rate": 0.0006613209265937611,
1173
+ "loss": 1.5809,
1174
+ "step": 965
1175
+ },
1176
+ {
1177
+ "epoch": 1.23,
1178
+ "learning_rate": 0.0006581115756446733,
1179
+ "loss": 1.6111,
1180
+ "step": 970
1181
+ },
1182
+ {
1183
+ "epoch": 1.24,
1184
+ "learning_rate": 0.000654894970316224,
1185
+ "loss": 1.5794,
1186
+ "step": 975
1187
+ },
1188
+ {
1189
+ "epoch": 1.24,
1190
+ "learning_rate": 0.0006516712581907471,
1191
+ "loss": 1.5756,
1192
+ "step": 980
1193
+ },
1194
+ {
1195
+ "epoch": 1.25,
1196
+ "learning_rate": 0.000648440587176646,
1197
+ "loss": 1.5913,
1198
+ "step": 985
1199
+ },
1200
+ {
1201
+ "epoch": 1.25,
1202
+ "learning_rate": 0.0006452031055016072,
1203
+ "loss": 1.615,
1204
+ "step": 990
1205
+ },
1206
+ {
1207
+ "epoch": 1.26,
1208
+ "learning_rate": 0.0006419589617058008,
1209
+ "loss": 1.6578,
1210
+ "step": 995
1211
+ },
1212
+ {
1213
+ "epoch": 1.27,
1214
+ "learning_rate": 0.000638708304635063,
1215
+ "loss": 1.6285,
1216
+ "step": 1000
1217
+ },
1218
+ {
1219
+ "epoch": 1.27,
1220
+ "learning_rate": 0.0006354512834340694,
1221
+ "loss": 1.5911,
1222
+ "step": 1005
1223
+ },
1224
+ {
1225
+ "epoch": 1.28,
1226
+ "learning_rate": 0.0006321880475394899,
1227
+ "loss": 1.6383,
1228
+ "step": 1010
1229
+ },
1230
+ {
1231
+ "epoch": 1.29,
1232
+ "learning_rate": 0.0006289187466731334,
1233
+ "loss": 1.5337,
1234
+ "step": 1015
1235
+ },
1236
+ {
1237
+ "epoch": 1.29,
1238
+ "learning_rate": 0.0006256435308350785,
1239
+ "loss": 1.6334,
1240
+ "step": 1020
1241
+ },
1242
+ {
1243
+ "epoch": 1.3,
1244
+ "learning_rate": 0.0006223625502967903,
1245
+ "loss": 1.7418,
1246
+ "step": 1025
1247
+ },
1248
+ {
1249
+ "epoch": 1.31,
1250
+ "learning_rate": 0.000619075955594227,
1251
+ "loss": 1.5987,
1252
+ "step": 1030
1253
+ },
1254
+ {
1255
+ "epoch": 1.31,
1256
+ "learning_rate": 0.0006157838975209322,
1257
+ "loss": 1.6512,
1258
+ "step": 1035
1259
+ },
1260
+ {
1261
+ "epoch": 1.32,
1262
+ "learning_rate": 0.0006124865271211165,
1263
+ "loss": 1.68,
1264
+ "step": 1040
1265
+ },
1266
+ {
1267
+ "epoch": 1.32,
1268
+ "learning_rate": 0.0006091839956827278,
1269
+ "loss": 1.5891,
1270
+ "step": 1045
1271
+ },
1272
+ {
1273
+ "epoch": 1.33,
1274
+ "learning_rate": 0.0006058764547305088,
1275
+ "loss": 1.6131,
1276
+ "step": 1050
1277
+ },
1278
+ {
1279
+ "epoch": 1.34,
1280
+ "learning_rate": 0.0006025640560190467,
1281
+ "loss": 1.6289,
1282
+ "step": 1055
1283
+ },
1284
+ {
1285
+ "epoch": 1.34,
1286
+ "learning_rate": 0.0005992469515258089,
1287
+ "loss": 1.5907,
1288
+ "step": 1060
1289
+ },
1290
+ {
1291
+ "epoch": 1.35,
1292
+ "learning_rate": 0.0005959252934441706,
1293
+ "loss": 1.6156,
1294
+ "step": 1065
1295
+ },
1296
+ {
1297
+ "epoch": 1.36,
1298
+ "learning_rate": 0.0005925992341764323,
1299
+ "loss": 1.5184,
1300
+ "step": 1070
1301
+ },
1302
+ {
1303
+ "epoch": 1.36,
1304
+ "learning_rate": 0.000589268926326827,
1305
+ "loss": 1.5729,
1306
+ "step": 1075
1307
+ },
1308
+ {
1309
+ "epoch": 1.37,
1310
+ "learning_rate": 0.0005859345226945177,
1311
+ "loss": 1.5053,
1312
+ "step": 1080
1313
+ },
1314
+ {
1315
+ "epoch": 1.38,
1316
+ "learning_rate": 0.000582596176266589,
1317
+ "loss": 1.6205,
1318
+ "step": 1085
1319
+ },
1320
+ {
1321
+ "epoch": 1.38,
1322
+ "learning_rate": 0.0005792540402110257,
1323
+ "loss": 1.5363,
1324
+ "step": 1090
1325
+ },
1326
+ {
1327
+ "epoch": 1.39,
1328
+ "learning_rate": 0.0005759082678696857,
1329
+ "loss": 1.6992,
1330
+ "step": 1095
1331
+ },
1332
+ {
1333
+ "epoch": 1.39,
1334
+ "learning_rate": 0.0005725590127512657,
1335
+ "loss": 1.5543,
1336
+ "step": 1100
1337
+ },
1338
+ {
1339
+ "epoch": 1.4,
1340
+ "learning_rate": 0.0005692064285242558,
1341
+ "loss": 1.6408,
1342
+ "step": 1105
1343
+ },
1344
+ {
1345
+ "epoch": 1.41,
1346
+ "learning_rate": 0.0005658506690098916,
1347
+ "loss": 1.6324,
1348
+ "step": 1110
1349
+ },
1350
+ {
1351
+ "epoch": 1.41,
1352
+ "learning_rate": 0.0005624918881750943,
1353
+ "loss": 1.5604,
1354
+ "step": 1115
1355
+ },
1356
+ {
1357
+ "epoch": 1.42,
1358
+ "learning_rate": 0.0005591302401254076,
1359
+ "loss": 1.5888,
1360
+ "step": 1120
1361
+ },
1362
+ {
1363
+ "epoch": 1.43,
1364
+ "learning_rate": 0.000555765879097928,
1365
+ "loss": 1.661,
1366
+ "step": 1125
1367
+ },
1368
+ {
1369
+ "epoch": 1.43,
1370
+ "learning_rate": 0.0005523989594542258,
1371
+ "loss": 1.6155,
1372
+ "step": 1130
1373
+ },
1374
+ {
1375
+ "epoch": 1.44,
1376
+ "learning_rate": 0.0005490296356732649,
1377
+ "loss": 1.5069,
1378
+ "step": 1135
1379
+ },
1380
+ {
1381
+ "epoch": 1.44,
1382
+ "learning_rate": 0.0005456580623443145,
1383
+ "loss": 1.4833,
1384
+ "step": 1140
1385
+ },
1386
+ {
1387
+ "epoch": 1.45,
1388
+ "learning_rate": 0.000542284394159856,
1389
+ "loss": 1.4966,
1390
+ "step": 1145
1391
+ },
1392
+ {
1393
+ "epoch": 1.46,
1394
+ "learning_rate": 0.0005389087859084854,
1395
+ "loss": 1.6364,
1396
+ "step": 1150
1397
+ },
1398
+ {
1399
+ "epoch": 1.46,
1400
+ "learning_rate": 0.0005355313924678123,
1401
+ "loss": 1.6311,
1402
+ "step": 1155
1403
+ },
1404
+ {
1405
+ "epoch": 1.47,
1406
+ "learning_rate": 0.0005321523687973526,
1407
+ "loss": 1.5597,
1408
+ "step": 1160
1409
+ },
1410
+ {
1411
+ "epoch": 1.48,
1412
+ "learning_rate": 0.00052877186993142,
1413
+ "loss": 1.6759,
1414
+ "step": 1165
1415
+ },
1416
+ {
1417
+ "epoch": 1.48,
1418
+ "learning_rate": 0.0005253900509720118,
1419
+ "loss": 1.5734,
1420
+ "step": 1170
1421
+ },
1422
+ {
1423
+ "epoch": 1.49,
1424
+ "learning_rate": 0.0005220070670816932,
1425
+ "loss": 1.6767,
1426
+ "step": 1175
1427
+ },
1428
+ {
1429
+ "epoch": 1.5,
1430
+ "learning_rate": 0.0005186230734764782,
1431
+ "loss": 1.603,
1432
+ "step": 1180
1433
+ },
1434
+ {
1435
+ "epoch": 1.5,
1436
+ "learning_rate": 0.0005152382254187076,
1437
+ "loss": 1.5121,
1438
+ "step": 1185
1439
+ },
1440
+ {
1441
+ "epoch": 1.51,
1442
+ "learning_rate": 0.0005118526782099258,
1443
+ "loss": 1.5978,
1444
+ "step": 1190
1445
+ },
1446
+ {
1447
+ "epoch": 1.51,
1448
+ "learning_rate": 0.0005084665871837553,
1449
+ "loss": 1.4273,
1450
+ "step": 1195
1451
+ },
1452
+ {
1453
+ "epoch": 1.52,
1454
+ "learning_rate": 0.0005050801076987699,
1455
+ "loss": 1.5823,
1456
+ "step": 1200
1457
+ },
1458
+ {
1459
+ "epoch": 1.53,
1460
+ "learning_rate": 0.0005016933951313659,
1461
+ "loss": 1.6582,
1462
+ "step": 1205
1463
+ },
1464
+ {
1465
+ "epoch": 1.53,
1466
+ "learning_rate": 0.0004983066048686342,
1467
+ "loss": 1.5263,
1468
+ "step": 1210
1469
+ },
1470
+ {
1471
+ "epoch": 1.54,
1472
+ "learning_rate": 0.0004949198923012302,
1473
+ "loss": 1.6287,
1474
+ "step": 1215
1475
+ },
1476
+ {
1477
+ "epoch": 1.55,
1478
+ "learning_rate": 0.0004915334128162448,
1479
+ "loss": 1.5244,
1480
+ "step": 1220
1481
+ },
1482
+ {
1483
+ "epoch": 1.55,
1484
+ "learning_rate": 0.00048814732179007443,
1485
+ "loss": 1.5054,
1486
+ "step": 1225
1487
+ },
1488
+ {
1489
+ "epoch": 1.56,
1490
+ "learning_rate": 0.0004847617745812925,
1491
+ "loss": 1.5065,
1492
+ "step": 1230
1493
+ },
1494
+ {
1495
+ "epoch": 1.57,
1496
+ "learning_rate": 0.00048137692652352196,
1497
+ "loss": 1.649,
1498
+ "step": 1235
1499
+ },
1500
+ {
1501
+ "epoch": 1.57,
1502
+ "learning_rate": 0.0004779929329183068,
1503
+ "loss": 1.5417,
1504
+ "step": 1240
1505
+ },
1506
+ {
1507
+ "epoch": 1.58,
1508
+ "learning_rate": 0.0004746099490279883,
1509
+ "loss": 1.4152,
1510
+ "step": 1245
1511
+ },
1512
+ {
1513
+ "epoch": 1.58,
1514
+ "learning_rate": 0.00047122813006858017,
1515
+ "loss": 1.5171,
1516
+ "step": 1250
1517
+ },
1518
+ {
1519
+ "epoch": 1.59,
1520
+ "learning_rate": 0.0004678476312026475,
1521
+ "loss": 1.5013,
1522
+ "step": 1255
1523
+ },
1524
+ {
1525
+ "epoch": 1.6,
1526
+ "learning_rate": 0.00046446860753218797,
1527
+ "loss": 1.5518,
1528
+ "step": 1260
1529
+ },
1530
+ {
1531
+ "epoch": 1.6,
1532
+ "learning_rate": 0.0004610912140915146,
1533
+ "loss": 1.5121,
1534
+ "step": 1265
1535
+ },
1536
+ {
1537
+ "epoch": 1.61,
1538
+ "learning_rate": 0.00045771560584014414,
1539
+ "loss": 1.4661,
1540
+ "step": 1270
1541
+ },
1542
+ {
1543
+ "epoch": 1.62,
1544
+ "learning_rate": 0.0004543419376556855,
1545
+ "loss": 1.5079,
1546
+ "step": 1275
1547
+ },
1548
+ {
1549
+ "epoch": 1.62,
1550
+ "learning_rate": 0.00045097036432673515,
1551
+ "loss": 1.5807,
1552
+ "step": 1280
1553
+ },
1554
+ {
1555
+ "epoch": 1.63,
1556
+ "learning_rate": 0.0004476010405457744,
1557
+ "loss": 1.5803,
1558
+ "step": 1285
1559
+ },
1560
+ {
1561
+ "epoch": 1.63,
1562
+ "learning_rate": 0.00044423412090207216,
1563
+ "loss": 1.5011,
1564
+ "step": 1290
1565
+ },
1566
+ {
1567
+ "epoch": 1.64,
1568
+ "learning_rate": 0.0004408697598745924,
1569
+ "loss": 1.5691,
1570
+ "step": 1295
1571
+ },
1572
+ {
1573
+ "epoch": 1.65,
1574
+ "learning_rate": 0.0004375081118249058,
1575
+ "loss": 1.5332,
1576
+ "step": 1300
1577
+ },
1578
+ {
1579
+ "epoch": 1.65,
1580
+ "learning_rate": 0.0004341493309901085,
1581
+ "loss": 1.5385,
1582
+ "step": 1305
1583
+ },
1584
+ {
1585
+ "epoch": 1.66,
1586
+ "learning_rate": 0.00043079357147574434,
1587
+ "loss": 1.6153,
1588
+ "step": 1310
1589
+ },
1590
+ {
1591
+ "epoch": 1.67,
1592
+ "learning_rate": 0.0004274409872487345,
1593
+ "loss": 1.6695,
1594
+ "step": 1315
1595
+ },
1596
+ {
1597
+ "epoch": 1.67,
1598
+ "learning_rate": 0.0004240917321303143,
1599
+ "loss": 1.5654,
1600
+ "step": 1320
1601
+ },
1602
+ {
1603
+ "epoch": 1.68,
1604
+ "learning_rate": 0.0004207459597889744,
1605
+ "loss": 1.5585,
1606
+ "step": 1325
1607
+ },
1608
+ {
1609
+ "epoch": 1.69,
1610
+ "learning_rate": 0.000417403823733411,
1611
+ "loss": 1.5072,
1612
+ "step": 1330
1613
+ },
1614
+ {
1615
+ "epoch": 1.69,
1616
+ "learning_rate": 0.0004140654773054824,
1617
+ "loss": 1.5723,
1618
+ "step": 1335
1619
+ },
1620
+ {
1621
+ "epoch": 1.7,
1622
+ "learning_rate": 0.00041073107367317315,
1623
+ "loss": 1.6242,
1624
+ "step": 1340
1625
+ },
1626
+ {
1627
+ "epoch": 1.7,
1628
+ "learning_rate": 0.0004074007658235678,
1629
+ "loss": 1.5305,
1630
+ "step": 1345
1631
+ },
1632
+ {
1633
+ "epoch": 1.71,
1634
+ "learning_rate": 0.00040407470655582934,
1635
+ "loss": 1.5084,
1636
+ "step": 1350
1637
+ },
1638
+ {
1639
+ "epoch": 1.72,
1640
+ "learning_rate": 0.00040075304847419114,
1641
+ "loss": 1.5424,
1642
+ "step": 1355
1643
+ },
1644
+ {
1645
+ "epoch": 1.72,
1646
+ "learning_rate": 0.0003974359439809534,
1647
+ "loss": 1.5136,
1648
+ "step": 1360
1649
+ },
1650
+ {
1651
+ "epoch": 1.73,
1652
+ "learning_rate": 0.0003941235452694913,
1653
+ "loss": 1.5349,
1654
+ "step": 1365
1655
+ },
1656
+ {
1657
+ "epoch": 1.74,
1658
+ "learning_rate": 0.0003908160043172725,
1659
+ "loss": 1.4194,
1660
+ "step": 1370
1661
+ },
1662
+ {
1663
+ "epoch": 1.74,
1664
+ "learning_rate": 0.0003875134728788834,
1665
+ "loss": 1.6139,
1666
+ "step": 1375
1667
+ },
1668
+ {
1669
+ "epoch": 1.75,
1670
+ "learning_rate": 0.0003842161024790679,
1671
+ "loss": 1.4876,
1672
+ "step": 1380
1673
+ },
1674
+ {
1675
+ "epoch": 1.76,
1676
+ "learning_rate": 0.00038092404440577315,
1677
+ "loss": 1.5042,
1678
+ "step": 1385
1679
+ },
1680
+ {
1681
+ "epoch": 1.76,
1682
+ "learning_rate": 0.0003776374497032097,
1683
+ "loss": 1.5229,
1684
+ "step": 1390
1685
+ },
1686
+ {
1687
+ "epoch": 1.77,
1688
+ "learning_rate": 0.0003743564691649216,
1689
+ "loss": 1.5647,
1690
+ "step": 1395
1691
+ },
1692
+ {
1693
+ "epoch": 1.77,
1694
+ "learning_rate": 0.0003710812533268666,
1695
+ "loss": 1.5142,
1696
+ "step": 1400
1697
+ },
1698
+ {
1699
+ "epoch": 1.78,
1700
+ "learning_rate": 0.0003678119524605102,
1701
+ "loss": 1.5012,
1702
+ "step": 1405
1703
+ },
1704
+ {
1705
+ "epoch": 1.79,
1706
+ "learning_rate": 0.0003645487165659305,
1707
+ "loss": 1.4997,
1708
+ "step": 1410
1709
+ },
1710
+ {
1711
+ "epoch": 1.79,
1712
+ "learning_rate": 0.000361291695364937,
1713
+ "loss": 1.3816,
1714
+ "step": 1415
1715
+ },
1716
+ {
1717
+ "epoch": 1.8,
1718
+ "learning_rate": 0.00035804103829419943,
1719
+ "loss": 1.4572,
1720
+ "step": 1420
1721
+ },
1722
+ {
1723
+ "epoch": 1.81,
1724
+ "learning_rate": 0.0003547968944983927,
1725
+ "loss": 1.5515,
1726
+ "step": 1425
1727
+ },
1728
+ {
1729
+ "epoch": 1.81,
1730
+ "learning_rate": 0.00035155941282335415,
1731
+ "loss": 1.6266,
1732
+ "step": 1430
1733
+ },
1734
+ {
1735
+ "epoch": 1.82,
1736
+ "learning_rate": 0.0003483287418092528,
1737
+ "loss": 1.4554,
1738
+ "step": 1435
1739
+ },
1740
+ {
1741
+ "epoch": 1.82,
1742
+ "learning_rate": 0.000345105029683776,
1743
+ "loss": 1.4924,
1744
+ "step": 1440
1745
+ },
1746
+ {
1747
+ "epoch": 1.83,
1748
+ "learning_rate": 0.00034188842435532686,
1749
+ "loss": 1.4861,
1750
+ "step": 1445
1751
+ },
1752
+ {
1753
+ "epoch": 1.84,
1754
+ "learning_rate": 0.000338679073406239,
1755
+ "loss": 1.4455,
1756
+ "step": 1450
1757
+ },
1758
+ {
1759
+ "epoch": 1.84,
1760
+ "learning_rate": 0.0003354771240860052,
1761
+ "loss": 1.5319,
1762
+ "step": 1455
1763
+ },
1764
+ {
1765
+ "epoch": 1.85,
1766
+ "learning_rate": 0.0003322827233045206,
1767
+ "loss": 1.5121,
1768
+ "step": 1460
1769
+ },
1770
+ {
1771
+ "epoch": 1.86,
1772
+ "learning_rate": 0.0003290960176253438,
1773
+ "loss": 1.4594,
1774
+ "step": 1465
1775
+ },
1776
+ {
1777
+ "epoch": 1.86,
1778
+ "learning_rate": 0.0003259171532589707,
1779
+ "loss": 1.5755,
1780
+ "step": 1470
1781
+ },
1782
+ {
1783
+ "epoch": 1.87,
1784
+ "learning_rate": 0.00032274627605612685,
1785
+ "loss": 1.6141,
1786
+ "step": 1475
1787
+ },
1788
+ {
1789
+ "epoch": 1.88,
1790
+ "learning_rate": 0.00031958353150107597,
1791
+ "loss": 1.4925,
1792
+ "step": 1480
1793
+ },
1794
+ {
1795
+ "epoch": 1.88,
1796
+ "learning_rate": 0.0003164290647049443,
1797
+ "loss": 1.5114,
1798
+ "step": 1485
1799
+ },
1800
+ {
1801
+ "epoch": 1.89,
1802
+ "learning_rate": 0.0003132830203990631,
1803
+ "loss": 1.4687,
1804
+ "step": 1490
1805
+ },
1806
+ {
1807
+ "epoch": 1.89,
1808
+ "learning_rate": 0.00031014554292832795,
1809
+ "loss": 1.392,
1810
+ "step": 1495
1811
+ },
1812
+ {
1813
+ "epoch": 1.9,
1814
+ "learning_rate": 0.0003070167762445759,
1815
+ "loss": 1.4322,
1816
+ "step": 1500
1817
+ },
1818
+ {
1819
+ "epoch": 1.91,
1820
+ "learning_rate": 0.0003038968638999816,
1821
+ "loss": 1.5359,
1822
+ "step": 1505
1823
+ },
1824
+ {
1825
+ "epoch": 1.91,
1826
+ "learning_rate": 0.00030078594904046967,
1827
+ "loss": 1.5745,
1828
+ "step": 1510
1829
+ },
1830
+ {
1831
+ "epoch": 1.92,
1832
+ "learning_rate": 0.0002976841743991479,
1833
+ "loss": 1.4865,
1834
+ "step": 1515
1835
+ },
1836
+ {
1837
+ "epoch": 1.93,
1838
+ "learning_rate": 0.0002945916822897579,
1839
+ "loss": 1.4958,
1840
+ "step": 1520
1841
+ },
1842
+ {
1843
+ "epoch": 1.93,
1844
+ "learning_rate": 0.0002915086146001458,
1845
+ "loss": 1.4678,
1846
+ "step": 1525
1847
+ },
1848
+ {
1849
+ "epoch": 1.94,
1850
+ "learning_rate": 0.0002884351127857526,
1851
+ "loss": 1.4629,
1852
+ "step": 1530
1853
+ },
1854
+ {
1855
+ "epoch": 1.95,
1856
+ "learning_rate": 0.00028537131786312333,
1857
+ "loss": 1.5381,
1858
+ "step": 1535
1859
+ },
1860
+ {
1861
+ "epoch": 1.95,
1862
+ "learning_rate": 0.00028231737040343717,
1863
+ "loss": 1.5322,
1864
+ "step": 1540
1865
+ },
1866
+ {
1867
+ "epoch": 1.96,
1868
+ "learning_rate": 0.000279273410526058,
1869
+ "loss": 1.4523,
1870
+ "step": 1545
1871
+ },
1872
+ {
1873
+ "epoch": 1.96,
1874
+ "learning_rate": 0.000276239577892106,
1875
+ "loss": 1.552,
1876
+ "step": 1550
1877
+ },
1878
+ {
1879
+ "epoch": 1.97,
1880
+ "learning_rate": 0.00027321601169804873,
1881
+ "loss": 1.4431,
1882
+ "step": 1555
1883
+ },
1884
+ {
1885
+ "epoch": 1.98,
1886
+ "learning_rate": 0.00027020285066931525,
1887
+ "loss": 1.5036,
1888
+ "step": 1560
1889
+ },
1890
+ {
1891
+ "epoch": 1.98,
1892
+ "learning_rate": 0.0002672002330539318,
1893
+ "loss": 1.5313,
1894
+ "step": 1565
1895
+ },
1896
+ {
1897
+ "epoch": 1.99,
1898
+ "learning_rate": 0.0002642082966161772,
1899
+ "loss": 1.5092,
1900
+ "step": 1570
1901
+ },
1902
+ {
1903
+ "epoch": 2.0,
1904
+ "learning_rate": 0.0002612271786302636,
1905
+ "loss": 1.5065,
1906
+ "step": 1575
1907
+ },
1908
+ {
1909
+ "epoch": 2.0,
1910
+ "eval_loss": 1.6175580024719238,
1911
+ "eval_runtime": 366.4829,
1912
+ "eval_samples_per_second": 34.948,
1913
+ "eval_steps_per_second": 2.186,
1914
+ "step": 1578
1915
+ },
1916
+ {
1917
+ "epoch": 2.0,
1918
+ "learning_rate": 0.0002582570158740372,
1919
+ "loss": 1.5297,
1920
+ "step": 1580
1921
+ },
1922
+ {
1923
+ "epoch": 2.01,
1924
+ "learning_rate": 0.00025529794462270285,
1925
+ "loss": 1.3664,
1926
+ "step": 1585
1927
+ },
1928
+ {
1929
+ "epoch": 2.02,
1930
+ "learning_rate": 0.0002523501006425724,
1931
+ "loss": 1.2522,
1932
+ "step": 1590
1933
+ },
1934
+ {
1935
+ "epoch": 2.02,
1936
+ "learning_rate": 0.000249413619184834,
1937
+ "loss": 1.2119,
1938
+ "step": 1595
1939
+ },
1940
+ {
1941
+ "epoch": 2.03,
1942
+ "learning_rate": 0.00024648863497934824,
1943
+ "loss": 1.3337,
1944
+ "step": 1600
1945
+ },
1946
+ {
1947
+ "epoch": 2.03,
1948
+ "learning_rate": 0.00024357528222846515,
1949
+ "loss": 1.2604,
1950
+ "step": 1605
1951
+ },
1952
+ {
1953
+ "epoch": 2.04,
1954
+ "learning_rate": 0.00024067369460086769,
1955
+ "loss": 1.2815,
1956
+ "step": 1610
1957
+ },
1958
+ {
1959
+ "epoch": 2.05,
1960
+ "learning_rate": 0.00023778400522543846,
1961
+ "loss": 1.2383,
1962
+ "step": 1615
1963
+ },
1964
+ {
1965
+ "epoch": 2.05,
1966
+ "learning_rate": 0.00023490634668515154,
1967
+ "loss": 1.2959,
1968
+ "step": 1620
1969
+ },
1970
+ {
1971
+ "epoch": 2.06,
1972
+ "learning_rate": 0.00023204085101099015,
1973
+ "loss": 1.3136,
1974
+ "step": 1625
1975
+ },
1976
+ {
1977
+ "epoch": 2.07,
1978
+ "learning_rate": 0.00022918764967588763,
1979
+ "loss": 1.2596,
1980
+ "step": 1630
1981
+ },
1982
+ {
1983
+ "epoch": 2.07,
1984
+ "learning_rate": 0.00022634687358869595,
1985
+ "loss": 1.272,
1986
+ "step": 1635
1987
+ },
1988
+ {
1989
+ "epoch": 2.08,
1990
+ "learning_rate": 0.00022351865308817991,
1991
+ "loss": 1.218,
1992
+ "step": 1640
1993
+ },
1994
+ {
1995
+ "epoch": 2.08,
1996
+ "learning_rate": 0.00022070311793703608,
1997
+ "loss": 1.3847,
1998
+ "step": 1645
1999
+ },
2000
+ {
2001
+ "epoch": 2.09,
2002
+ "learning_rate": 0.00021790039731593946,
2003
+ "loss": 1.2601,
2004
+ "step": 1650
2005
+ },
2006
+ {
2007
+ "epoch": 2.1,
2008
+ "learning_rate": 0.00021511061981761666,
2009
+ "loss": 1.3791,
2010
+ "step": 1655
2011
+ },
2012
+ {
2013
+ "epoch": 2.1,
2014
+ "learning_rate": 0.0002123339134409461,
2015
+ "loss": 1.2843,
2016
+ "step": 1660
2017
+ },
2018
+ {
2019
+ "epoch": 2.11,
2020
+ "learning_rate": 0.0002095704055850846,
2021
+ "loss": 1.2707,
2022
+ "step": 1665
2023
+ },
2024
+ {
2025
+ "epoch": 2.12,
2026
+ "learning_rate": 0.00020682022304362226,
2027
+ "loss": 1.2781,
2028
+ "step": 1670
2029
+ },
2030
+ {
2031
+ "epoch": 2.12,
2032
+ "learning_rate": 0.00020408349199876587,
2033
+ "loss": 1.3632,
2034
+ "step": 1675
2035
+ },
2036
+ {
2037
+ "epoch": 2.13,
2038
+ "learning_rate": 0.0002013603380155482,
2039
+ "loss": 1.2773,
2040
+ "step": 1680
2041
+ },
2042
+ {
2043
+ "epoch": 2.14,
2044
+ "learning_rate": 0.00019865088603606773,
2045
+ "loss": 1.2757,
2046
+ "step": 1685
2047
+ },
2048
+ {
2049
+ "epoch": 2.14,
2050
+ "learning_rate": 0.00019595526037375588,
2051
+ "loss": 1.299,
2052
+ "step": 1690
2053
+ },
2054
+ {
2055
+ "epoch": 2.15,
2056
+ "learning_rate": 0.00019327358470767325,
2057
+ "loss": 1.2752,
2058
+ "step": 1695
2059
+ },
2060
+ {
2061
+ "epoch": 2.15,
2062
+ "learning_rate": 0.0001906059820768357,
2063
+ "loss": 1.3311,
2064
+ "step": 1700
2065
+ },
2066
+ {
2067
+ "epoch": 2.16,
2068
+ "learning_rate": 0.00018795257487456797,
2069
+ "loss": 1.2696,
2070
+ "step": 1705
2071
+ },
2072
+ {
2073
+ "epoch": 2.17,
2074
+ "learning_rate": 0.00018531348484288925,
2075
+ "loss": 1.2492,
2076
+ "step": 1710
2077
+ },
2078
+ {
2079
+ "epoch": 2.17,
2080
+ "learning_rate": 0.00018268883306692674,
2081
+ "loss": 1.3153,
2082
+ "step": 1715
2083
+ },
2084
+ {
2085
+ "epoch": 2.18,
2086
+ "learning_rate": 0.0001800787399693604,
2087
+ "loss": 1.274,
2088
+ "step": 1720
2089
+ },
2090
+ {
2091
+ "epoch": 2.19,
2092
+ "learning_rate": 0.0001774833253048978,
2093
+ "loss": 1.3302,
2094
+ "step": 1725
2095
+ },
2096
+ {
2097
+ "epoch": 2.19,
2098
+ "learning_rate": 0.0001749027081547792,
2099
+ "loss": 1.1726,
2100
+ "step": 1730
2101
+ },
2102
+ {
2103
+ "epoch": 2.2,
2104
+ "learning_rate": 0.0001723370069213151,
2105
+ "loss": 1.2367,
2106
+ "step": 1735
2107
+ },
2108
+ {
2109
+ "epoch": 2.21,
2110
+ "learning_rate": 0.0001697863393224522,
2111
+ "loss": 1.2647,
2112
+ "step": 1740
2113
+ },
2114
+ {
2115
+ "epoch": 2.21,
2116
+ "learning_rate": 0.0001672508223863732,
2117
+ "loss": 1.2689,
2118
+ "step": 1745
2119
+ },
2120
+ {
2121
+ "epoch": 2.22,
2122
+ "learning_rate": 0.00016473057244612778,
2123
+ "loss": 1.2523,
2124
+ "step": 1750
2125
+ },
2126
+ {
2127
+ "epoch": 2.22,
2128
+ "learning_rate": 0.00016222570513429403,
2129
+ "loss": 1.2597,
2130
+ "step": 1755
2131
+ },
2132
+ {
2133
+ "epoch": 2.23,
2134
+ "learning_rate": 0.00015973633537767396,
2135
+ "loss": 1.2528,
2136
+ "step": 1760
2137
+ },
2138
+ {
2139
+ "epoch": 2.24,
2140
+ "learning_rate": 0.00015726257739201994,
2141
+ "loss": 1.2704,
2142
+ "step": 1765
2143
+ },
2144
+ {
2145
+ "epoch": 2.24,
2146
+ "learning_rate": 0.0001548045446767945,
2147
+ "loss": 1.1525,
2148
+ "step": 1770
2149
+ },
2150
+ {
2151
+ "epoch": 2.25,
2152
+ "learning_rate": 0.0001523623500099633,
2153
+ "loss": 1.2283,
2154
+ "step": 1775
2155
+ },
2156
+ {
2157
+ "epoch": 2.26,
2158
+ "learning_rate": 0.0001499361054428196,
2159
+ "loss": 1.3046,
2160
+ "step": 1780
2161
+ },
2162
+ {
2163
+ "epoch": 2.26,
2164
+ "learning_rate": 0.00014752592229484463,
2165
+ "loss": 1.3047,
2166
+ "step": 1785
2167
+ },
2168
+ {
2169
+ "epoch": 2.27,
2170
+ "learning_rate": 0.00014513191114859869,
2171
+ "loss": 1.2188,
2172
+ "step": 1790
2173
+ },
2174
+ {
2175
+ "epoch": 2.27,
2176
+ "learning_rate": 0.0001427541818446484,
2177
+ "loss": 1.2769,
2178
+ "step": 1795
2179
+ },
2180
+ {
2181
+ "epoch": 2.28,
2182
+ "learning_rate": 0.0001403928434765267,
2183
+ "loss": 1.2688,
2184
+ "step": 1800
2185
+ },
2186
+ {
2187
+ "epoch": 2.29,
2188
+ "learning_rate": 0.00013804800438572745,
2189
+ "loss": 1.2727,
2190
+ "step": 1805
2191
+ },
2192
+ {
2193
+ "epoch": 2.29,
2194
+ "learning_rate": 0.00013571977215673508,
2195
+ "loss": 1.1562,
2196
+ "step": 1810
2197
+ },
2198
+ {
2199
+ "epoch": 2.3,
2200
+ "learning_rate": 0.00013340825361208758,
2201
+ "loss": 1.3305,
2202
+ "step": 1815
2203
+ },
2204
+ {
2205
+ "epoch": 2.31,
2206
+ "learning_rate": 0.00013111355480747622,
2207
+ "loss": 1.2663,
2208
+ "step": 1820
2209
+ },
2210
+ {
2211
+ "epoch": 2.31,
2212
+ "learning_rate": 0.0001288357810268791,
2213
+ "loss": 1.2736,
2214
+ "step": 1825
2215
+ },
2216
+ {
2217
+ "epoch": 2.32,
2218
+ "learning_rate": 0.00012657503677772992,
2219
+ "loss": 1.2877,
2220
+ "step": 1830
2221
+ },
2222
+ {
2223
+ "epoch": 2.33,
2224
+ "learning_rate": 0.00012433142578612493,
2225
+ "loss": 1.3585,
2226
+ "step": 1835
2227
+ },
2228
+ {
2229
+ "epoch": 2.33,
2230
+ "learning_rate": 0.00012210505099206177,
2231
+ "loss": 1.3242,
2232
+ "step": 1840
2233
+ },
2234
+ {
2235
+ "epoch": 2.34,
2236
+ "learning_rate": 0.00011989601454471788,
2237
+ "loss": 1.3037,
2238
+ "step": 1845
2239
+ },
2240
+ {
2241
+ "epoch": 2.34,
2242
+ "learning_rate": 0.00011770441779776281,
2243
+ "loss": 1.2969,
2244
+ "step": 1850
2245
+ },
2246
+ {
2247
+ "epoch": 2.35,
2248
+ "learning_rate": 0.0001155303613047084,
2249
+ "loss": 1.2783,
2250
+ "step": 1855
2251
+ },
2252
+ {
2253
+ "epoch": 2.36,
2254
+ "learning_rate": 0.00011337394481429563,
2255
+ "loss": 1.2409,
2256
+ "step": 1860
2257
+ },
2258
+ {
2259
+ "epoch": 2.36,
2260
+ "learning_rate": 0.00011123526726591682,
2261
+ "loss": 1.2276,
2262
+ "step": 1865
2263
+ },
2264
+ {
2265
+ "epoch": 2.37,
2266
+ "learning_rate": 0.00010911442678507771,
2267
+ "loss": 1.261,
2268
+ "step": 1870
2269
+ },
2270
+ {
2271
+ "epoch": 2.38,
2272
+ "learning_rate": 0.00010701152067889408,
2273
+ "loss": 1.2741,
2274
+ "step": 1875
2275
+ },
2276
+ {
2277
+ "epoch": 2.38,
2278
+ "learning_rate": 0.0001049266454316275,
2279
+ "loss": 1.2337,
2280
+ "step": 1880
2281
+ },
2282
+ {
2283
+ "epoch": 2.39,
2284
+ "learning_rate": 0.00010285989670025914,
2285
+ "loss": 1.2259,
2286
+ "step": 1885
2287
+ },
2288
+ {
2289
+ "epoch": 2.4,
2290
+ "learning_rate": 0.00010081136931009982,
2291
+ "loss": 1.3068,
2292
+ "step": 1890
2293
+ },
2294
+ {
2295
+ "epoch": 2.4,
2296
+ "learning_rate": 9.878115725044045e-05,
2297
+ "loss": 1.2444,
2298
+ "step": 1895
2299
+ },
2300
+ {
2301
+ "epoch": 2.41,
2302
+ "learning_rate": 9.676935367023826e-05,
2303
+ "loss": 1.2301,
2304
+ "step": 1900
2305
+ },
2306
+ {
2307
+ "epoch": 2.41,
2308
+ "learning_rate": 9.477605087384428e-05,
2309
+ "loss": 1.2882,
2310
+ "step": 1905
2311
+ },
2312
+ {
2313
+ "epoch": 2.42,
2314
+ "learning_rate": 9.280134031676795e-05,
2315
+ "loss": 1.2423,
2316
+ "step": 1910
2317
+ },
2318
+ {
2319
+ "epoch": 2.43,
2320
+ "learning_rate": 9.084531260148043e-05,
2321
+ "loss": 1.2471,
2322
+ "step": 1915
2323
+ },
2324
+ {
2325
+ "epoch": 2.43,
2326
+ "learning_rate": 8.890805747325864e-05,
2327
+ "loss": 1.2215,
2328
+ "step": 1920
2329
+ },
2330
+ {
2331
+ "epoch": 2.44,
2332
+ "learning_rate": 8.698966381606633e-05,
2333
+ "loss": 1.359,
2334
+ "step": 1925
2335
+ },
2336
+ {
2337
+ "epoch": 2.45,
2338
+ "learning_rate": 8.509021964847752e-05,
2339
+ "loss": 1.2103,
2340
+ "step": 1930
2341
+ },
2342
+ {
2343
+ "epoch": 2.45,
2344
+ "learning_rate": 8.320981211963647e-05,
2345
+ "loss": 1.291,
2346
+ "step": 1935
2347
+ },
2348
+ {
2349
+ "epoch": 2.46,
2350
+ "learning_rate": 8.134852750526017e-05,
2351
+ "loss": 1.2588,
2352
+ "step": 1940
2353
+ },
2354
+ {
2355
+ "epoch": 2.47,
2356
+ "learning_rate": 7.950645120368011e-05,
2357
+ "loss": 1.3012,
2358
+ "step": 1945
2359
+ },
2360
+ {
2361
+ "epoch": 2.47,
2362
+ "learning_rate": 7.768366773192286e-05,
2363
+ "loss": 1.2732,
2364
+ "step": 1950
2365
+ },
2366
+ {
2367
+ "epoch": 2.48,
2368
+ "learning_rate": 7.58802607218339e-05,
2369
+ "loss": 1.2364,
2370
+ "step": 1955
2371
+ },
2372
+ {
2373
+ "epoch": 2.48,
2374
+ "learning_rate": 7.409631291623903e-05,
2375
+ "loss": 1.3049,
2376
+ "step": 1960
2377
+ },
2378
+ {
2379
+ "epoch": 2.49,
2380
+ "learning_rate": 7.23319061651489e-05,
2381
+ "loss": 1.334,
2382
+ "step": 1965
2383
+ },
2384
+ {
2385
+ "epoch": 2.5,
2386
+ "learning_rate": 7.058712142200324e-05,
2387
+ "loss": 1.1297,
2388
+ "step": 1970
2389
+ },
2390
+ {
2391
+ "epoch": 2.5,
2392
+ "learning_rate": 6.886203873995672e-05,
2393
+ "loss": 1.3076,
2394
+ "step": 1975
2395
+ },
2396
+ {
2397
+ "epoch": 2.51,
2398
+ "learning_rate": 6.715673726820626e-05,
2399
+ "loss": 1.2343,
2400
+ "step": 1980
2401
+ },
2402
+ {
2403
+ "epoch": 2.52,
2404
+ "learning_rate": 6.547129524835904e-05,
2405
+ "loss": 1.2474,
2406
+ "step": 1985
2407
+ },
2408
+ {
2409
+ "epoch": 2.52,
2410
+ "learning_rate": 6.38057900108428e-05,
2411
+ "loss": 1.2312,
2412
+ "step": 1990
2413
+ },
2414
+ {
2415
+ "epoch": 2.53,
2416
+ "learning_rate": 6.216029797135842e-05,
2417
+ "loss": 1.1307,
2418
+ "step": 1995
2419
+ },
2420
+ {
2421
+ "epoch": 2.53,
2422
+ "learning_rate": 6.053489462737283e-05,
2423
+ "loss": 1.2532,
2424
+ "step": 2000
2425
+ },
2426
+ {
2427
+ "epoch": 2.54,
2428
+ "learning_rate": 5.8929654554655985e-05,
2429
+ "loss": 1.2598,
2430
+ "step": 2005
2431
+ },
2432
+ {
2433
+ "epoch": 2.55,
2434
+ "learning_rate": 5.734465140385864e-05,
2435
+ "loss": 1.1858,
2436
+ "step": 2010
2437
+ },
2438
+ {
2439
+ "epoch": 2.55,
2440
+ "learning_rate": 5.5779957897133415e-05,
2441
+ "loss": 1.225,
2442
+ "step": 2015
2443
+ },
2444
+ {
2445
+ "epoch": 2.56,
2446
+ "learning_rate": 5.423564582479845e-05,
2447
+ "loss": 1.2459,
2448
+ "step": 2020
2449
+ },
2450
+ {
2451
+ "epoch": 2.57,
2452
+ "learning_rate": 5.2711786042042854e-05,
2453
+ "loss": 1.1665,
2454
+ "step": 2025
2455
+ },
2456
+ {
2457
+ "epoch": 2.57,
2458
+ "learning_rate": 5.120844846567657e-05,
2459
+ "loss": 1.2943,
2460
+ "step": 2030
2461
+ },
2462
+ {
2463
+ "epoch": 2.58,
2464
+ "learning_rate": 4.972570207092186e-05,
2465
+ "loss": 1.3287,
2466
+ "step": 2035
2467
+ },
2468
+ {
2469
+ "epoch": 2.59,
2470
+ "learning_rate": 4.826361488824898e-05,
2471
+ "loss": 1.2245,
2472
+ "step": 2040
2473
+ },
2474
+ {
2475
+ "epoch": 2.59,
2476
+ "learning_rate": 4.68222540002548e-05,
2477
+ "loss": 1.1462,
2478
+ "step": 2045
2479
+ },
2480
+ {
2481
+ "epoch": 2.6,
2482
+ "learning_rate": 4.5401685538584756e-05,
2483
+ "loss": 1.2929,
2484
+ "step": 2050
2485
+ },
2486
+ {
2487
+ "epoch": 2.6,
2488
+ "learning_rate": 4.400197468089906e-05,
2489
+ "loss": 1.2585,
2490
+ "step": 2055
2491
+ },
2492
+ {
2493
+ "epoch": 2.61,
2494
+ "learning_rate": 4.26231856478817e-05,
2495
+ "loss": 1.1936,
2496
+ "step": 2060
2497
+ },
2498
+ {
2499
+ "epoch": 2.62,
2500
+ "learning_rate": 4.126538170029453e-05,
2501
+ "loss": 1.3289,
2502
+ "step": 2065
2503
+ },
2504
+ {
2505
+ "epoch": 2.62,
2506
+ "learning_rate": 3.99286251360742e-05,
2507
+ "loss": 1.2575,
2508
+ "step": 2070
2509
+ },
2510
+ {
2511
+ "epoch": 2.63,
2512
+ "learning_rate": 3.8612977287474093e-05,
2513
+ "loss": 1.2051,
2514
+ "step": 2075
2515
+ },
2516
+ {
2517
+ "epoch": 2.64,
2518
+ "learning_rate": 3.731849851825042e-05,
2519
+ "loss": 1.265,
2520
+ "step": 2080
2521
+ },
2522
+ {
2523
+ "epoch": 2.64,
2524
+ "learning_rate": 3.604524822089245e-05,
2525
+ "loss": 1.273,
2526
+ "step": 2085
2527
+ },
2528
+ {
2529
+ "epoch": 2.65,
2530
+ "learning_rate": 3.479328481389771e-05,
2531
+ "loss": 1.2193,
2532
+ "step": 2090
2533
+ },
2534
+ {
2535
+ "epoch": 2.66,
2536
+ "learning_rate": 3.35626657390915e-05,
2537
+ "loss": 1.3077,
2538
+ "step": 2095
2539
+ },
2540
+ {
2541
+ "epoch": 2.66,
2542
+ "learning_rate": 3.235344745899116e-05,
2543
+ "loss": 1.177,
2544
+ "step": 2100
2545
+ },
2546
+ {
2547
+ "epoch": 2.67,
2548
+ "learning_rate": 3.1165685454216333e-05,
2549
+ "loss": 1.317,
2550
+ "step": 2105
2551
+ },
2552
+ {
2553
+ "epoch": 2.67,
2554
+ "learning_rate": 2.9999434220942255e-05,
2555
+ "loss": 1.2496,
2556
+ "step": 2110
2557
+ },
2558
+ {
2559
+ "epoch": 2.68,
2560
+ "learning_rate": 2.8854747268400318e-05,
2561
+ "loss": 1.2356,
2562
+ "step": 2115
2563
+ },
2564
+ {
2565
+ "epoch": 2.69,
2566
+ "learning_rate": 2.773167711642266e-05,
2567
+ "loss": 1.2738,
2568
+ "step": 2120
2569
+ },
2570
+ {
2571
+ "epoch": 2.69,
2572
+ "learning_rate": 2.663027529303236e-05,
2573
+ "loss": 1.2312,
2574
+ "step": 2125
2575
+ },
2576
+ {
2577
+ "epoch": 2.7,
2578
+ "learning_rate": 2.5550592332079515e-05,
2579
+ "loss": 1.2528,
2580
+ "step": 2130
2581
+ },
2582
+ {
2583
+ "epoch": 2.71,
2584
+ "learning_rate": 2.449267777092251e-05,
2585
+ "loss": 1.1745,
2586
+ "step": 2135
2587
+ },
2588
+ {
2589
+ "epoch": 2.71,
2590
+ "learning_rate": 2.3456580148155272e-05,
2591
+ "loss": 1.2588,
2592
+ "step": 2140
2593
+ },
2594
+ {
2595
+ "epoch": 2.72,
2596
+ "learning_rate": 2.2442347001380148e-05,
2597
+ "loss": 1.1721,
2598
+ "step": 2145
2599
+ },
2600
+ {
2601
+ "epoch": 2.72,
2602
+ "learning_rate": 2.145002486502684e-05,
2603
+ "loss": 1.2337,
2604
+ "step": 2150
2605
+ },
2606
+ {
2607
+ "epoch": 2.73,
2608
+ "learning_rate": 2.0479659268217378e-05,
2609
+ "loss": 1.2546,
2610
+ "step": 2155
2611
+ },
2612
+ {
2613
+ "epoch": 2.74,
2614
+ "learning_rate": 1.953129473267723e-05,
2615
+ "loss": 1.2842,
2616
+ "step": 2160
2617
+ },
2618
+ {
2619
+ "epoch": 2.74,
2620
+ "learning_rate": 1.8604974770692508e-05,
2621
+ "loss": 1.1871,
2622
+ "step": 2165
2623
+ },
2624
+ {
2625
+ "epoch": 2.75,
2626
+ "learning_rate": 1.770074188311349e-05,
2627
+ "loss": 1.1609,
2628
+ "step": 2170
2629
+ },
2630
+ {
2631
+ "epoch": 2.76,
2632
+ "learning_rate": 1.681863755740487e-05,
2633
+ "loss": 1.1607,
2634
+ "step": 2175
2635
+ },
2636
+ {
2637
+ "epoch": 2.76,
2638
+ "learning_rate": 1.5958702265741977e-05,
2639
+ "loss": 1.2254,
2640
+ "step": 2180
2641
+ },
2642
+ {
2643
+ "epoch": 2.77,
2644
+ "learning_rate": 1.512097546315394e-05,
2645
+ "loss": 1.2322,
2646
+ "step": 2185
2647
+ },
2648
+ {
2649
+ "epoch": 2.78,
2650
+ "learning_rate": 1.430549558571359e-05,
2651
+ "loss": 1.1937,
2652
+ "step": 2190
2653
+ },
2654
+ {
2655
+ "epoch": 2.78,
2656
+ "learning_rate": 1.3512300048773685e-05,
2657
+ "loss": 1.2344,
2658
+ "step": 2195
2659
+ },
2660
+ {
2661
+ "epoch": 2.79,
2662
+ "learning_rate": 1.2741425245250626e-05,
2663
+ "loss": 1.2383,
2664
+ "step": 2200
2665
+ },
2666
+ {
2667
+ "epoch": 2.79,
2668
+ "learning_rate": 1.199290654395424e-05,
2669
+ "loss": 1.2553,
2670
+ "step": 2205
2671
+ },
2672
+ {
2673
+ "epoch": 2.8,
2674
+ "learning_rate": 1.1266778287965297e-05,
2675
+ "loss": 1.381,
2676
+ "step": 2210
2677
+ },
2678
+ {
2679
+ "epoch": 2.81,
2680
+ "learning_rate": 1.056307379305993e-05,
2681
+ "loss": 1.1926,
2682
+ "step": 2215
2683
+ },
2684
+ {
2685
+ "epoch": 2.81,
2686
+ "learning_rate": 9.88182534618054e-06,
2687
+ "loss": 1.1419,
2688
+ "step": 2220
2689
+ },
2690
+ {
2691
+ "epoch": 2.82,
2692
+ "learning_rate": 9.223064203955078e-06,
2693
+ "loss": 1.2016,
2694
+ "step": 2225
2695
+ },
2696
+ {
2697
+ "epoch": 2.83,
2698
+ "learning_rate": 8.586820591262367e-06,
2699
+ "loss": 1.2364,
2700
+ "step": 2230
2701
+ },
2702
+ {
2703
+ "epoch": 2.83,
2704
+ "learning_rate": 7.973123699845819e-06,
2705
+ "loss": 1.2936,
2706
+ "step": 2235
2707
+ },
2708
+ {
2709
+ "epoch": 2.84,
2710
+ "learning_rate": 7.382001686973782e-06,
2711
+ "loss": 1.2675,
2712
+ "step": 2240
2713
+ },
2714
+ {
2715
+ "epoch": 2.85,
2716
+ "learning_rate": 6.8134816741476364e-06,
2717
+ "loss": 1.2742,
2718
+ "step": 2245
2719
+ },
2720
+ {
2721
+ "epoch": 2.85,
2722
+ "learning_rate": 6.267589745857727e-06,
2723
+ "loss": 1.2404,
2724
+ "step": 2250
2725
+ },
2726
+ {
2727
+ "epoch": 2.86,
2728
+ "learning_rate": 5.744350948386101e-06,
2729
+ "loss": 1.311,
2730
+ "step": 2255
2731
+ },
2732
+ {
2733
+ "epoch": 2.86,
2734
+ "learning_rate": 5.243789288657818e-06,
2735
+ "loss": 1.2043,
2736
+ "step": 2260
2737
+ },
2738
+ {
2739
+ "epoch": 2.87,
2740
+ "learning_rate": 4.765927733139108e-06,
2741
+ "loss": 1.2024,
2742
+ "step": 2265
2743
+ },
2744
+ {
2745
+ "epoch": 2.88,
2746
+ "learning_rate": 4.310788206783822e-06,
2747
+ "loss": 1.3395,
2748
+ "step": 2270
2749
+ },
2750
+ {
2751
+ "epoch": 2.88,
2752
+ "learning_rate": 3.87839159202763e-06,
2753
+ "loss": 1.2397,
2754
+ "step": 2275
2755
+ },
2756
+ {
2757
+ "epoch": 2.89,
2758
+ "learning_rate": 3.4687577278295635e-06,
2759
+ "loss": 1.2176,
2760
+ "step": 2280
2761
+ },
2762
+ {
2763
+ "epoch": 2.9,
2764
+ "learning_rate": 3.0819054087619093e-06,
2765
+ "loss": 1.1944,
2766
+ "step": 2285
2767
+ },
2768
+ {
2769
+ "epoch": 2.9,
2770
+ "learning_rate": 2.7178523841481227e-06,
2771
+ "loss": 1.2135,
2772
+ "step": 2290
2773
+ },
2774
+ {
2775
+ "epoch": 2.91,
2776
+ "learning_rate": 2.3766153572480353e-06,
2777
+ "loss": 1.259,
2778
+ "step": 2295
2779
+ },
2780
+ {
2781
+ "epoch": 2.91,
2782
+ "learning_rate": 2.05820998449191e-06,
2783
+ "loss": 1.1943,
2784
+ "step": 2300
2785
+ },
2786
+ {
2787
+ "epoch": 2.92,
2788
+ "learning_rate": 1.7626508747617954e-06,
2789
+ "loss": 1.2961,
2790
+ "step": 2305
2791
+ },
2792
+ {
2793
+ "epoch": 2.93,
2794
+ "learning_rate": 1.4899515887213943e-06,
2795
+ "loss": 1.157,
2796
+ "step": 2310
2797
+ },
2798
+ {
2799
+ "epoch": 2.93,
2800
+ "learning_rate": 1.2401246381938958e-06,
2801
+ "loss": 1.1454,
2802
+ "step": 2315
2803
+ },
2804
+ {
2805
+ "epoch": 2.94,
2806
+ "learning_rate": 1.0131814855877663e-06,
2807
+ "loss": 1.1584,
2808
+ "step": 2320
2809
+ },
2810
+ {
2811
+ "epoch": 2.95,
2812
+ "learning_rate": 8.091325433710606e-07,
2813
+ "loss": 1.1784,
2814
+ "step": 2325
2815
+ },
2816
+ {
2817
+ "epoch": 2.95,
2818
+ "learning_rate": 6.27987173593525e-07,
2819
+ "loss": 1.2977,
2820
+ "step": 2330
2821
+ },
2822
+ {
2823
+ "epoch": 2.96,
2824
+ "learning_rate": 4.6975368745705296e-07,
2825
+ "loss": 1.1467,
2826
+ "step": 2335
2827
+ },
2828
+ {
2829
+ "epoch": 2.97,
2830
+ "learning_rate": 3.3443934493443407e-07,
2831
+ "loss": 1.2271,
2832
+ "step": 2340
2833
+ },
2834
+ {
2835
+ "epoch": 2.97,
2836
+ "learning_rate": 2.2205035443623178e-07,
2837
+ "loss": 1.2055,
2838
+ "step": 2345
2839
+ },
2840
+ {
2841
+ "epoch": 2.98,
2842
+ "learning_rate": 1.325918725258446e-07,
2843
+ "loss": 1.166,
2844
+ "step": 2350
2845
+ },
2846
+ {
2847
+ "epoch": 2.98,
2848
+ "learning_rate": 6.60680036831396e-08,
2849
+ "loss": 1.2394,
2850
+ "step": 2355
2851
+ },
2852
+ {
2853
+ "epoch": 2.99,
2854
+ "learning_rate": 2.248180011588108e-08,
2855
+ "loss": 1.2422,
2856
+ "step": 2360
2857
+ },
2858
+ {
2859
+ "epoch": 3.0,
2860
+ "learning_rate": 1.8352616197314652e-09,
2861
+ "loss": 1.1873,
2862
+ "step": 2365
2863
+ },
2864
+ {
2865
+ "epoch": 3.0,
2866
+ "eval_loss": 1.5839784145355225,
2867
+ "eval_runtime": 366.1985,
2868
+ "eval_samples_per_second": 34.976,
2869
+ "eval_steps_per_second": 2.187,
2870
+ "step": 2367
2871
+ },
2872
+ {
2873
+ "epoch": 3.0,
2874
+ "step": 2367,
2875
+ "total_flos": 5.627889061063557e+17,
2876
+ "train_loss": 1.6624892393546513,
2877
+ "train_runtime": 33680.2242,
2878
+ "train_samples_per_second": 8.997,
2879
+ "train_steps_per_second": 0.07
2880
+ }
2881
+ ],
2882
+ "max_steps": 2367,
2883
+ "num_train_epochs": 3,
2884
+ "total_flos": 5.627889061063557e+17,
2885
+ "trial_name": null,
2886
+ "trial_params": null
2887
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8b1c6f55cc5ce9c3968ab93367e8864fd942687d4c5120a4cba47bd5b9498bb4
3
+ size 3567