Commit
•
332b32a
1
Parent(s):
e8fa6cc
End of training
Browse files- README.md +107 -0
- adapter_config.json +26 -0
- adapter_model.safetensors +3 -0
- generation_config.json +18 -0
- trainer_state.json +1842 -0
- training_args.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,107 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: HuggingFaceM4/idefics2-8b
|
| 4 |
+
tags:
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
model-index:
|
| 7 |
+
- name: idefics2_ft_augmented_dataset
|
| 8 |
+
results: []
|
| 9 |
+
---
|
| 10 |
+
|
| 11 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 12 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 13 |
+
|
| 14 |
+
# idefics2_ft_augmented_dataset
|
| 15 |
+
|
| 16 |
+
This model is a fine-tuned version of [HuggingFaceM4/idefics2-8b](https://huggingface.co/HuggingFaceM4/idefics2-8b) on an unknown dataset.
|
| 17 |
+
It achieves the following results on the evaluation set:
|
| 18 |
+
- Loss: 0.1871
|
| 19 |
+
|
| 20 |
+
## Model description
|
| 21 |
+
|
| 22 |
+
More information needed
|
| 23 |
+
|
| 24 |
+
## Intended uses & limitations
|
| 25 |
+
|
| 26 |
+
More information needed
|
| 27 |
+
|
| 28 |
+
## Training and evaluation data
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Training procedure
|
| 33 |
+
|
| 34 |
+
### Training hyperparameters
|
| 35 |
+
|
| 36 |
+
The following hyperparameters were used during training:
|
| 37 |
+
- learning_rate: 2.5e-05
|
| 38 |
+
- train_batch_size: 4
|
| 39 |
+
- eval_batch_size: 8
|
| 40 |
+
- seed: 42
|
| 41 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 42 |
+
- lr_scheduler_type: linear
|
| 43 |
+
- lr_scheduler_warmup_steps: 25
|
| 44 |
+
- training_steps: 5000
|
| 45 |
+
|
| 46 |
+
### Training results
|
| 47 |
+
|
| 48 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 49 |
+
|:-------------:|:------:|:----:|:---------------:|
|
| 50 |
+
| 0.6338 | 0.0234 | 100 | 0.6474 |
|
| 51 |
+
| 0.5748 | 0.0468 | 200 | 0.5899 |
|
| 52 |
+
| 0.4678 | 0.0702 | 300 | 0.5603 |
|
| 53 |
+
| 0.428 | 0.0936 | 400 | 0.5249 |
|
| 54 |
+
| 0.3798 | 0.1170 | 500 | 0.4984 |
|
| 55 |
+
| 0.3665 | 0.1404 | 600 | 0.4733 |
|
| 56 |
+
| 0.4406 | 0.1637 | 700 | 0.4510 |
|
| 57 |
+
| 0.4723 | 0.1871 | 800 | 0.4245 |
|
| 58 |
+
| 0.4807 | 0.2105 | 900 | 0.4158 |
|
| 59 |
+
| 0.4196 | 0.2339 | 1000 | 0.3971 |
|
| 60 |
+
| 0.3443 | 0.2573 | 1100 | 0.3738 |
|
| 61 |
+
| 0.4133 | 0.2807 | 1200 | 0.3631 |
|
| 62 |
+
| 0.2838 | 0.3041 | 1300 | 0.3334 |
|
| 63 |
+
| 0.4134 | 0.3275 | 1400 | 0.3264 |
|
| 64 |
+
| 0.2838 | 0.3509 | 1500 | 0.3125 |
|
| 65 |
+
| 0.275 | 0.3743 | 1600 | 0.2944 |
|
| 66 |
+
| 0.4141 | 0.3977 | 1700 | 0.2839 |
|
| 67 |
+
| 0.2498 | 0.4211 | 1800 | 0.2749 |
|
| 68 |
+
| 0.2817 | 0.4444 | 1900 | 0.2606 |
|
| 69 |
+
| 0.2899 | 0.4678 | 2000 | 0.2526 |
|
| 70 |
+
| 0.2695 | 0.4912 | 2100 | 0.2521 |
|
| 71 |
+
| 0.2619 | 0.5146 | 2200 | 0.2424 |
|
| 72 |
+
| 0.2238 | 0.5380 | 2300 | 0.2373 |
|
| 73 |
+
| 0.3049 | 0.5614 | 2400 | 0.2301 |
|
| 74 |
+
| 0.1308 | 0.5848 | 2500 | 0.2292 |
|
| 75 |
+
| 0.1936 | 0.6082 | 2600 | 0.2190 |
|
| 76 |
+
| 0.2479 | 0.6316 | 2700 | 0.2191 |
|
| 77 |
+
| 0.1575 | 0.6550 | 2800 | 0.2165 |
|
| 78 |
+
| 0.193 | 0.6784 | 2900 | 0.2107 |
|
| 79 |
+
| 0.2526 | 0.7018 | 3000 | 0.2114 |
|
| 80 |
+
| 0.1574 | 0.7251 | 3100 | 0.2087 |
|
| 81 |
+
| 0.1989 | 0.7485 | 3200 | 0.2051 |
|
| 82 |
+
| 0.1761 | 0.7719 | 3300 | 0.2013 |
|
| 83 |
+
| 0.2223 | 0.7953 | 3400 | 0.1996 |
|
| 84 |
+
| 0.2127 | 0.8187 | 3500 | 0.1966 |
|
| 85 |
+
| 0.2477 | 0.8421 | 3600 | 0.1923 |
|
| 86 |
+
| 0.1931 | 0.8655 | 3700 | 0.1908 |
|
| 87 |
+
| 0.182 | 0.8889 | 3800 | 0.1888 |
|
| 88 |
+
| 0.1693 | 0.9123 | 3900 | 0.1878 |
|
| 89 |
+
| 0.1346 | 0.9357 | 4000 | 0.1853 |
|
| 90 |
+
| 0.1484 | 0.9591 | 4100 | 0.1849 |
|
| 91 |
+
| 0.1217 | 0.9825 | 4200 | 0.1838 |
|
| 92 |
+
| 0.0669 | 1.0058 | 4300 | 0.1844 |
|
| 93 |
+
| 0.1292 | 1.0292 | 4400 | 0.1877 |
|
| 94 |
+
| 0.1106 | 1.0526 | 4500 | 0.1876 |
|
| 95 |
+
| 0.0828 | 1.0760 | 4600 | 0.1875 |
|
| 96 |
+
| 0.0485 | 1.0994 | 4700 | 0.1871 |
|
| 97 |
+
| 0.0624 | 1.1228 | 4800 | 0.1874 |
|
| 98 |
+
| 0.0895 | 1.1462 | 4900 | 0.1871 |
|
| 99 |
+
| 0.1 | 1.1696 | 5000 | 0.1871 |
|
| 100 |
+
|
| 101 |
+
|
| 102 |
+
### Framework versions
|
| 103 |
+
|
| 104 |
+
- Transformers 4.41.2
|
| 105 |
+
- Pytorch 2.3.1+cu121
|
| 106 |
+
- Datasets 2.19.2
|
| 107 |
+
- Tokenizers 0.19.1
|
adapter_config.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "HuggingFaceM4/idefics2-8b",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": false,
|
| 7 |
+
"inference_mode": false,
|
| 8 |
+
"init_lora_weights": "gaussian",
|
| 9 |
+
"layer_replication": null,
|
| 10 |
+
"layers_pattern": null,
|
| 11 |
+
"layers_to_transform": null,
|
| 12 |
+
"loftq_config": {},
|
| 13 |
+
"lora_alpha": 64,
|
| 14 |
+
"lora_dropout": 0.1,
|
| 15 |
+
"megatron_config": null,
|
| 16 |
+
"megatron_core": "megatron.core",
|
| 17 |
+
"modules_to_save": null,
|
| 18 |
+
"peft_type": "LORA",
|
| 19 |
+
"r": 32,
|
| 20 |
+
"rank_pattern": {},
|
| 21 |
+
"revision": null,
|
| 22 |
+
"target_modules": ".*(text_model|modality_projection|perceiver_resampler).*(down_proj|gate_proj|up_proj|k_proj|q_proj|v_proj|o_proj).*$",
|
| 23 |
+
"task_type": null,
|
| 24 |
+
"use_dora": true,
|
| 25 |
+
"use_rslora": false
|
| 26 |
+
}
|
adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2f4e4e2c8426cd5e7abd1a135f4a29ca6ae37bb43e2a799f9c7721c4c6a13b43
|
| 3 |
+
size 189801888
|
generation_config.json
ADDED
|
@@ -0,0 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bad_words_ids": [
|
| 4 |
+
[
|
| 5 |
+
32000
|
| 6 |
+
],
|
| 7 |
+
[
|
| 8 |
+
32001
|
| 9 |
+
]
|
| 10 |
+
],
|
| 11 |
+
"bos_token_id": 1,
|
| 12 |
+
"eos_token_id": [
|
| 13 |
+
2,
|
| 14 |
+
32002
|
| 15 |
+
],
|
| 16 |
+
"pad_token_id": 0,
|
| 17 |
+
"transformers_version": "4.41.2"
|
| 18 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1842 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.1695906432748537,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 5000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.005847953216374269,
|
| 13 |
+
"grad_norm": 19.25,
|
| 14 |
+
"learning_rate": 2.5e-05,
|
| 15 |
+
"loss": 1.3094,
|
| 16 |
+
"step": 25
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.011695906432748537,
|
| 20 |
+
"grad_norm": 20.515625,
|
| 21 |
+
"learning_rate": 2.4874371859296484e-05,
|
| 22 |
+
"loss": 0.6781,
|
| 23 |
+
"step": 50
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.017543859649122806,
|
| 27 |
+
"grad_norm": 9.015625,
|
| 28 |
+
"learning_rate": 2.4748743718592964e-05,
|
| 29 |
+
"loss": 0.6127,
|
| 30 |
+
"step": 75
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.023391812865497075,
|
| 34 |
+
"grad_norm": 9.4765625,
|
| 35 |
+
"learning_rate": 2.462311557788945e-05,
|
| 36 |
+
"loss": 0.6338,
|
| 37 |
+
"step": 100
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.023391812865497075,
|
| 41 |
+
"eval_loss": 0.6474161744117737,
|
| 42 |
+
"eval_runtime": 125.9492,
|
| 43 |
+
"eval_samples_per_second": 3.573,
|
| 44 |
+
"eval_steps_per_second": 0.453,
|
| 45 |
+
"step": 100
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.029239766081871343,
|
| 49 |
+
"grad_norm": 16.71875,
|
| 50 |
+
"learning_rate": 2.449748743718593e-05,
|
| 51 |
+
"loss": 0.506,
|
| 52 |
+
"step": 125
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.03508771929824561,
|
| 56 |
+
"grad_norm": 21.125,
|
| 57 |
+
"learning_rate": 2.4371859296482413e-05,
|
| 58 |
+
"loss": 0.501,
|
| 59 |
+
"step": 150
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.04093567251461988,
|
| 63 |
+
"grad_norm": 17.953125,
|
| 64 |
+
"learning_rate": 2.4246231155778896e-05,
|
| 65 |
+
"loss": 0.6208,
|
| 66 |
+
"step": 175
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.04678362573099415,
|
| 70 |
+
"grad_norm": 7.9609375,
|
| 71 |
+
"learning_rate": 2.4120603015075376e-05,
|
| 72 |
+
"loss": 0.5748,
|
| 73 |
+
"step": 200
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.04678362573099415,
|
| 77 |
+
"eval_loss": 0.5898649096488953,
|
| 78 |
+
"eval_runtime": 129.2682,
|
| 79 |
+
"eval_samples_per_second": 3.481,
|
| 80 |
+
"eval_steps_per_second": 0.441,
|
| 81 |
+
"step": 200
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.05263157894736842,
|
| 85 |
+
"grad_norm": 7.4921875,
|
| 86 |
+
"learning_rate": 2.3994974874371863e-05,
|
| 87 |
+
"loss": 0.5821,
|
| 88 |
+
"step": 225
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 0.05847953216374269,
|
| 92 |
+
"grad_norm": 8.703125,
|
| 93 |
+
"learning_rate": 2.3869346733668342e-05,
|
| 94 |
+
"loss": 0.5764,
|
| 95 |
+
"step": 250
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 0.06432748538011696,
|
| 99 |
+
"grad_norm": 10.3828125,
|
| 100 |
+
"learning_rate": 2.3743718592964825e-05,
|
| 101 |
+
"loss": 0.5921,
|
| 102 |
+
"step": 275
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 0.07017543859649122,
|
| 106 |
+
"grad_norm": 14.90625,
|
| 107 |
+
"learning_rate": 2.361809045226131e-05,
|
| 108 |
+
"loss": 0.4678,
|
| 109 |
+
"step": 300
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.07017543859649122,
|
| 113 |
+
"eval_loss": 0.5602560639381409,
|
| 114 |
+
"eval_runtime": 129.0871,
|
| 115 |
+
"eval_samples_per_second": 3.486,
|
| 116 |
+
"eval_steps_per_second": 0.442,
|
| 117 |
+
"step": 300
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 0.07602339181286549,
|
| 121 |
+
"grad_norm": 16.6875,
|
| 122 |
+
"learning_rate": 2.3492462311557788e-05,
|
| 123 |
+
"loss": 0.5118,
|
| 124 |
+
"step": 325
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 0.08187134502923976,
|
| 128 |
+
"grad_norm": 17.921875,
|
| 129 |
+
"learning_rate": 2.3366834170854275e-05,
|
| 130 |
+
"loss": 0.5199,
|
| 131 |
+
"step": 350
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 0.08771929824561403,
|
| 135 |
+
"grad_norm": 16.5625,
|
| 136 |
+
"learning_rate": 2.3241206030150754e-05,
|
| 137 |
+
"loss": 0.4611,
|
| 138 |
+
"step": 375
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 0.0935672514619883,
|
| 142 |
+
"grad_norm": 10.5,
|
| 143 |
+
"learning_rate": 2.3115577889447238e-05,
|
| 144 |
+
"loss": 0.428,
|
| 145 |
+
"step": 400
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.0935672514619883,
|
| 149 |
+
"eval_loss": 0.524933397769928,
|
| 150 |
+
"eval_runtime": 127.1828,
|
| 151 |
+
"eval_samples_per_second": 3.538,
|
| 152 |
+
"eval_steps_per_second": 0.448,
|
| 153 |
+
"step": 400
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.09941520467836257,
|
| 157 |
+
"grad_norm": 11.6328125,
|
| 158 |
+
"learning_rate": 2.298994974874372e-05,
|
| 159 |
+
"loss": 0.4953,
|
| 160 |
+
"step": 425
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 0.10526315789473684,
|
| 164 |
+
"grad_norm": 5.4296875,
|
| 165 |
+
"learning_rate": 2.28643216080402e-05,
|
| 166 |
+
"loss": 0.4173,
|
| 167 |
+
"step": 450
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 0.1111111111111111,
|
| 171 |
+
"grad_norm": 8.8125,
|
| 172 |
+
"learning_rate": 2.2738693467336687e-05,
|
| 173 |
+
"loss": 0.4225,
|
| 174 |
+
"step": 475
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.11695906432748537,
|
| 178 |
+
"grad_norm": 8.84375,
|
| 179 |
+
"learning_rate": 2.2613065326633167e-05,
|
| 180 |
+
"loss": 0.3798,
|
| 181 |
+
"step": 500
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.11695906432748537,
|
| 185 |
+
"eval_loss": 0.49835124611854553,
|
| 186 |
+
"eval_runtime": 128.3365,
|
| 187 |
+
"eval_samples_per_second": 3.506,
|
| 188 |
+
"eval_steps_per_second": 0.444,
|
| 189 |
+
"step": 500
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.12280701754385964,
|
| 193 |
+
"grad_norm": 11.4609375,
|
| 194 |
+
"learning_rate": 2.248743718592965e-05,
|
| 195 |
+
"loss": 0.4386,
|
| 196 |
+
"step": 525
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 0.1286549707602339,
|
| 200 |
+
"grad_norm": 16.4375,
|
| 201 |
+
"learning_rate": 2.2361809045226133e-05,
|
| 202 |
+
"loss": 0.4537,
|
| 203 |
+
"step": 550
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 0.13450292397660818,
|
| 207 |
+
"grad_norm": 10.140625,
|
| 208 |
+
"learning_rate": 2.2236180904522613e-05,
|
| 209 |
+
"loss": 0.3514,
|
| 210 |
+
"step": 575
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 0.14035087719298245,
|
| 214 |
+
"grad_norm": 3.265625,
|
| 215 |
+
"learning_rate": 2.21105527638191e-05,
|
| 216 |
+
"loss": 0.3665,
|
| 217 |
+
"step": 600
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 0.14035087719298245,
|
| 221 |
+
"eval_loss": 0.47331658005714417,
|
| 222 |
+
"eval_runtime": 128.8126,
|
| 223 |
+
"eval_samples_per_second": 3.493,
|
| 224 |
+
"eval_steps_per_second": 0.443,
|
| 225 |
+
"step": 600
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 0.14619883040935672,
|
| 229 |
+
"grad_norm": 12.3046875,
|
| 230 |
+
"learning_rate": 2.198492462311558e-05,
|
| 231 |
+
"loss": 0.2409,
|
| 232 |
+
"step": 625
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 0.15204678362573099,
|
| 236 |
+
"grad_norm": 9.703125,
|
| 237 |
+
"learning_rate": 2.1859296482412062e-05,
|
| 238 |
+
"loss": 0.4553,
|
| 239 |
+
"step": 650
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 0.15789473684210525,
|
| 243 |
+
"grad_norm": 10.65625,
|
| 244 |
+
"learning_rate": 2.1733668341708545e-05,
|
| 245 |
+
"loss": 0.4242,
|
| 246 |
+
"step": 675
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 0.16374269005847952,
|
| 250 |
+
"grad_norm": 4.96484375,
|
| 251 |
+
"learning_rate": 2.1608040201005025e-05,
|
| 252 |
+
"loss": 0.4406,
|
| 253 |
+
"step": 700
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 0.16374269005847952,
|
| 257 |
+
"eval_loss": 0.45101454854011536,
|
| 258 |
+
"eval_runtime": 129.1251,
|
| 259 |
+
"eval_samples_per_second": 3.485,
|
| 260 |
+
"eval_steps_per_second": 0.441,
|
| 261 |
+
"step": 700
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.1695906432748538,
|
| 265 |
+
"grad_norm": 6.20703125,
|
| 266 |
+
"learning_rate": 2.1482412060301508e-05,
|
| 267 |
+
"loss": 0.408,
|
| 268 |
+
"step": 725
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.17543859649122806,
|
| 272 |
+
"grad_norm": 9.875,
|
| 273 |
+
"learning_rate": 2.135678391959799e-05,
|
| 274 |
+
"loss": 0.5067,
|
| 275 |
+
"step": 750
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.18128654970760233,
|
| 279 |
+
"grad_norm": 7.1875,
|
| 280 |
+
"learning_rate": 2.1231155778894474e-05,
|
| 281 |
+
"loss": 0.4338,
|
| 282 |
+
"step": 775
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.1871345029239766,
|
| 286 |
+
"grad_norm": 12.765625,
|
| 287 |
+
"learning_rate": 2.1105527638190957e-05,
|
| 288 |
+
"loss": 0.4723,
|
| 289 |
+
"step": 800
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.1871345029239766,
|
| 293 |
+
"eval_loss": 0.4244983494281769,
|
| 294 |
+
"eval_runtime": 127.3581,
|
| 295 |
+
"eval_samples_per_second": 3.533,
|
| 296 |
+
"eval_steps_per_second": 0.448,
|
| 297 |
+
"step": 800
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.19298245614035087,
|
| 301 |
+
"grad_norm": 26.03125,
|
| 302 |
+
"learning_rate": 2.0979899497487437e-05,
|
| 303 |
+
"loss": 0.4424,
|
| 304 |
+
"step": 825
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.19883040935672514,
|
| 308 |
+
"grad_norm": 5.234375,
|
| 309 |
+
"learning_rate": 2.085427135678392e-05,
|
| 310 |
+
"loss": 0.3599,
|
| 311 |
+
"step": 850
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.2046783625730994,
|
| 315 |
+
"grad_norm": 4.1875,
|
| 316 |
+
"learning_rate": 2.0728643216080403e-05,
|
| 317 |
+
"loss": 0.3399,
|
| 318 |
+
"step": 875
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.21052631578947367,
|
| 322 |
+
"grad_norm": 9.6640625,
|
| 323 |
+
"learning_rate": 2.0603015075376886e-05,
|
| 324 |
+
"loss": 0.4807,
|
| 325 |
+
"step": 900
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.21052631578947367,
|
| 329 |
+
"eval_loss": 0.41580215096473694,
|
| 330 |
+
"eval_runtime": 128.5483,
|
| 331 |
+
"eval_samples_per_second": 3.501,
|
| 332 |
+
"eval_steps_per_second": 0.443,
|
| 333 |
+
"step": 900
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 0.21637426900584794,
|
| 337 |
+
"grad_norm": 9.03125,
|
| 338 |
+
"learning_rate": 2.047738693467337e-05,
|
| 339 |
+
"loss": 0.3341,
|
| 340 |
+
"step": 925
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 0.2222222222222222,
|
| 344 |
+
"grad_norm": 10.3515625,
|
| 345 |
+
"learning_rate": 2.035175879396985e-05,
|
| 346 |
+
"loss": 0.4407,
|
| 347 |
+
"step": 950
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 0.22807017543859648,
|
| 351 |
+
"grad_norm": 13.046875,
|
| 352 |
+
"learning_rate": 2.0226130653266332e-05,
|
| 353 |
+
"loss": 0.391,
|
| 354 |
+
"step": 975
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.23391812865497075,
|
| 358 |
+
"grad_norm": 13.0859375,
|
| 359 |
+
"learning_rate": 2.0100502512562815e-05,
|
| 360 |
+
"loss": 0.4196,
|
| 361 |
+
"step": 1000
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.23391812865497075,
|
| 365 |
+
"eval_loss": 0.39709773659706116,
|
| 366 |
+
"eval_runtime": 129.6284,
|
| 367 |
+
"eval_samples_per_second": 3.471,
|
| 368 |
+
"eval_steps_per_second": 0.44,
|
| 369 |
+
"step": 1000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.23976608187134502,
|
| 373 |
+
"grad_norm": 2.962890625,
|
| 374 |
+
"learning_rate": 1.9974874371859298e-05,
|
| 375 |
+
"loss": 0.3042,
|
| 376 |
+
"step": 1025
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 0.24561403508771928,
|
| 380 |
+
"grad_norm": 11.9765625,
|
| 381 |
+
"learning_rate": 1.984924623115578e-05,
|
| 382 |
+
"loss": 0.3731,
|
| 383 |
+
"step": 1050
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 0.25146198830409355,
|
| 387 |
+
"grad_norm": 5.42578125,
|
| 388 |
+
"learning_rate": 1.972361809045226e-05,
|
| 389 |
+
"loss": 0.3032,
|
| 390 |
+
"step": 1075
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 0.2573099415204678,
|
| 394 |
+
"grad_norm": 8.9140625,
|
| 395 |
+
"learning_rate": 1.9597989949748744e-05,
|
| 396 |
+
"loss": 0.3443,
|
| 397 |
+
"step": 1100
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 0.2573099415204678,
|
| 401 |
+
"eval_loss": 0.3737768530845642,
|
| 402 |
+
"eval_runtime": 126.3694,
|
| 403 |
+
"eval_samples_per_second": 3.561,
|
| 404 |
+
"eval_steps_per_second": 0.451,
|
| 405 |
+
"step": 1100
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 0.2631578947368421,
|
| 409 |
+
"grad_norm": 14.0625,
|
| 410 |
+
"learning_rate": 1.9472361809045227e-05,
|
| 411 |
+
"loss": 0.2804,
|
| 412 |
+
"step": 1125
|
| 413 |
+
},
|
| 414 |
+
{
|
| 415 |
+
"epoch": 0.26900584795321636,
|
| 416 |
+
"grad_norm": 5.19921875,
|
| 417 |
+
"learning_rate": 1.934673366834171e-05,
|
| 418 |
+
"loss": 0.3257,
|
| 419 |
+
"step": 1150
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 0.27485380116959063,
|
| 423 |
+
"grad_norm": 9.78125,
|
| 424 |
+
"learning_rate": 1.9221105527638193e-05,
|
| 425 |
+
"loss": 0.3625,
|
| 426 |
+
"step": 1175
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 0.2807017543859649,
|
| 430 |
+
"grad_norm": 12.890625,
|
| 431 |
+
"learning_rate": 1.9095477386934673e-05,
|
| 432 |
+
"loss": 0.4133,
|
| 433 |
+
"step": 1200
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 0.2807017543859649,
|
| 437 |
+
"eval_loss": 0.3631095290184021,
|
| 438 |
+
"eval_runtime": 125.638,
|
| 439 |
+
"eval_samples_per_second": 3.582,
|
| 440 |
+
"eval_steps_per_second": 0.454,
|
| 441 |
+
"step": 1200
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 0.28654970760233917,
|
| 445 |
+
"grad_norm": 7.71875,
|
| 446 |
+
"learning_rate": 1.8969849246231156e-05,
|
| 447 |
+
"loss": 0.5163,
|
| 448 |
+
"step": 1225
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 0.29239766081871343,
|
| 452 |
+
"grad_norm": 7.25390625,
|
| 453 |
+
"learning_rate": 1.884422110552764e-05,
|
| 454 |
+
"loss": 0.2991,
|
| 455 |
+
"step": 1250
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 0.2982456140350877,
|
| 459 |
+
"grad_norm": 6.5234375,
|
| 460 |
+
"learning_rate": 1.8718592964824123e-05,
|
| 461 |
+
"loss": 0.431,
|
| 462 |
+
"step": 1275
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 0.30409356725146197,
|
| 466 |
+
"grad_norm": 3.8515625,
|
| 467 |
+
"learning_rate": 1.8592964824120602e-05,
|
| 468 |
+
"loss": 0.2838,
|
| 469 |
+
"step": 1300
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 0.30409356725146197,
|
| 473 |
+
"eval_loss": 0.33338162302970886,
|
| 474 |
+
"eval_runtime": 127.3115,
|
| 475 |
+
"eval_samples_per_second": 3.535,
|
| 476 |
+
"eval_steps_per_second": 0.448,
|
| 477 |
+
"step": 1300
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 0.30994152046783624,
|
| 481 |
+
"grad_norm": 9.828125,
|
| 482 |
+
"learning_rate": 1.8467336683417085e-05,
|
| 483 |
+
"loss": 0.2653,
|
| 484 |
+
"step": 1325
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 0.3157894736842105,
|
| 488 |
+
"grad_norm": 5.34765625,
|
| 489 |
+
"learning_rate": 1.834170854271357e-05,
|
| 490 |
+
"loss": 0.2515,
|
| 491 |
+
"step": 1350
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 0.3216374269005848,
|
| 495 |
+
"grad_norm": 12.15625,
|
| 496 |
+
"learning_rate": 1.821608040201005e-05,
|
| 497 |
+
"loss": 0.2983,
|
| 498 |
+
"step": 1375
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 0.32748538011695905,
|
| 502 |
+
"grad_norm": 9.171875,
|
| 503 |
+
"learning_rate": 1.8090452261306535e-05,
|
| 504 |
+
"loss": 0.4134,
|
| 505 |
+
"step": 1400
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 0.32748538011695905,
|
| 509 |
+
"eval_loss": 0.32642024755477905,
|
| 510 |
+
"eval_runtime": 127.196,
|
| 511 |
+
"eval_samples_per_second": 3.538,
|
| 512 |
+
"eval_steps_per_second": 0.448,
|
| 513 |
+
"step": 1400
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.3333333333333333,
|
| 517 |
+
"grad_norm": 7.16015625,
|
| 518 |
+
"learning_rate": 1.7964824120603014e-05,
|
| 519 |
+
"loss": 0.3015,
|
| 520 |
+
"step": 1425
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.3391812865497076,
|
| 524 |
+
"grad_norm": 15.6640625,
|
| 525 |
+
"learning_rate": 1.7839195979899497e-05,
|
| 526 |
+
"loss": 0.3098,
|
| 527 |
+
"step": 1450
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.34502923976608185,
|
| 531 |
+
"grad_norm": 13.8984375,
|
| 532 |
+
"learning_rate": 1.771356783919598e-05,
|
| 533 |
+
"loss": 0.4022,
|
| 534 |
+
"step": 1475
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.3508771929824561,
|
| 538 |
+
"grad_norm": 9.90625,
|
| 539 |
+
"learning_rate": 1.7587939698492464e-05,
|
| 540 |
+
"loss": 0.2838,
|
| 541 |
+
"step": 1500
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.3508771929824561,
|
| 545 |
+
"eval_loss": 0.3124904930591583,
|
| 546 |
+
"eval_runtime": 128.6196,
|
| 547 |
+
"eval_samples_per_second": 3.499,
|
| 548 |
+
"eval_steps_per_second": 0.443,
|
| 549 |
+
"step": 1500
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.3567251461988304,
|
| 553 |
+
"grad_norm": 18.234375,
|
| 554 |
+
"learning_rate": 1.7462311557788947e-05,
|
| 555 |
+
"loss": 0.3792,
|
| 556 |
+
"step": 1525
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.36257309941520466,
|
| 560 |
+
"grad_norm": 1.482421875,
|
| 561 |
+
"learning_rate": 1.7336683417085427e-05,
|
| 562 |
+
"loss": 0.4708,
|
| 563 |
+
"step": 1550
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.3684210526315789,
|
| 567 |
+
"grad_norm": 10.8046875,
|
| 568 |
+
"learning_rate": 1.721105527638191e-05,
|
| 569 |
+
"loss": 0.3695,
|
| 570 |
+
"step": 1575
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.3742690058479532,
|
| 574 |
+
"grad_norm": 13.6328125,
|
| 575 |
+
"learning_rate": 1.7085427135678393e-05,
|
| 576 |
+
"loss": 0.275,
|
| 577 |
+
"step": 1600
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.3742690058479532,
|
| 581 |
+
"eval_loss": 0.294376403093338,
|
| 582 |
+
"eval_runtime": 125.5965,
|
| 583 |
+
"eval_samples_per_second": 3.583,
|
| 584 |
+
"eval_steps_per_second": 0.454,
|
| 585 |
+
"step": 1600
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 0.38011695906432746,
|
| 589 |
+
"grad_norm": 10.3359375,
|
| 590 |
+
"learning_rate": 1.6959798994974876e-05,
|
| 591 |
+
"loss": 0.3134,
|
| 592 |
+
"step": 1625
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 0.38596491228070173,
|
| 596 |
+
"grad_norm": 4.671875,
|
| 597 |
+
"learning_rate": 1.683417085427136e-05,
|
| 598 |
+
"loss": 0.1961,
|
| 599 |
+
"step": 1650
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 0.391812865497076,
|
| 603 |
+
"grad_norm": 3.734375,
|
| 604 |
+
"learning_rate": 1.670854271356784e-05,
|
| 605 |
+
"loss": 0.2686,
|
| 606 |
+
"step": 1675
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 0.39766081871345027,
|
| 610 |
+
"grad_norm": 11.5859375,
|
| 611 |
+
"learning_rate": 1.6582914572864322e-05,
|
| 612 |
+
"loss": 0.4141,
|
| 613 |
+
"step": 1700
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 0.39766081871345027,
|
| 617 |
+
"eval_loss": 0.2839096784591675,
|
| 618 |
+
"eval_runtime": 126.4222,
|
| 619 |
+
"eval_samples_per_second": 3.56,
|
| 620 |
+
"eval_steps_per_second": 0.451,
|
| 621 |
+
"step": 1700
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 0.40350877192982454,
|
| 625 |
+
"grad_norm": 12.0234375,
|
| 626 |
+
"learning_rate": 1.6457286432160805e-05,
|
| 627 |
+
"loss": 0.2948,
|
| 628 |
+
"step": 1725
|
| 629 |
+
},
|
| 630 |
+
{
|
| 631 |
+
"epoch": 0.4093567251461988,
|
| 632 |
+
"grad_norm": 6.44140625,
|
| 633 |
+
"learning_rate": 1.6331658291457288e-05,
|
| 634 |
+
"loss": 0.3546,
|
| 635 |
+
"step": 1750
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 0.4152046783625731,
|
| 639 |
+
"grad_norm": 7.5234375,
|
| 640 |
+
"learning_rate": 1.620603015075377e-05,
|
| 641 |
+
"loss": 0.3239,
|
| 642 |
+
"step": 1775
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 0.42105263157894735,
|
| 646 |
+
"grad_norm": 4.30078125,
|
| 647 |
+
"learning_rate": 1.608040201005025e-05,
|
| 648 |
+
"loss": 0.2498,
|
| 649 |
+
"step": 1800
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 0.42105263157894735,
|
| 653 |
+
"eval_loss": 0.2749183773994446,
|
| 654 |
+
"eval_runtime": 126.776,
|
| 655 |
+
"eval_samples_per_second": 3.55,
|
| 656 |
+
"eval_steps_per_second": 0.45,
|
| 657 |
+
"step": 1800
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 0.4269005847953216,
|
| 661 |
+
"grad_norm": 8.4375,
|
| 662 |
+
"learning_rate": 1.5954773869346734e-05,
|
| 663 |
+
"loss": 0.3026,
|
| 664 |
+
"step": 1825
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 0.4327485380116959,
|
| 668 |
+
"grad_norm": 8.140625,
|
| 669 |
+
"learning_rate": 1.5829145728643217e-05,
|
| 670 |
+
"loss": 0.3848,
|
| 671 |
+
"step": 1850
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 0.43859649122807015,
|
| 675 |
+
"grad_norm": 10.046875,
|
| 676 |
+
"learning_rate": 1.57035175879397e-05,
|
| 677 |
+
"loss": 0.2574,
|
| 678 |
+
"step": 1875
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 0.4444444444444444,
|
| 682 |
+
"grad_norm": 9.1640625,
|
| 683 |
+
"learning_rate": 1.5577889447236183e-05,
|
| 684 |
+
"loss": 0.2817,
|
| 685 |
+
"step": 1900
|
| 686 |
+
},
|
| 687 |
+
{
|
| 688 |
+
"epoch": 0.4444444444444444,
|
| 689 |
+
"eval_loss": 0.2605937719345093,
|
| 690 |
+
"eval_runtime": 129.5889,
|
| 691 |
+
"eval_samples_per_second": 3.473,
|
| 692 |
+
"eval_steps_per_second": 0.44,
|
| 693 |
+
"step": 1900
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 0.4502923976608187,
|
| 697 |
+
"grad_norm": 11.3828125,
|
| 698 |
+
"learning_rate": 1.5452261306532663e-05,
|
| 699 |
+
"loss": 0.376,
|
| 700 |
+
"step": 1925
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"epoch": 0.45614035087719296,
|
| 704 |
+
"grad_norm": 9.6953125,
|
| 705 |
+
"learning_rate": 1.5326633165829146e-05,
|
| 706 |
+
"loss": 0.2708,
|
| 707 |
+
"step": 1950
|
| 708 |
+
},
|
| 709 |
+
{
|
| 710 |
+
"epoch": 0.4619883040935672,
|
| 711 |
+
"grad_norm": 10.0859375,
|
| 712 |
+
"learning_rate": 1.5201005025125627e-05,
|
| 713 |
+
"loss": 0.2509,
|
| 714 |
+
"step": 1975
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 0.4678362573099415,
|
| 718 |
+
"grad_norm": 11.4921875,
|
| 719 |
+
"learning_rate": 1.507537688442211e-05,
|
| 720 |
+
"loss": 0.2899,
|
| 721 |
+
"step": 2000
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 0.4678362573099415,
|
| 725 |
+
"eval_loss": 0.2526080906391144,
|
| 726 |
+
"eval_runtime": 127.4275,
|
| 727 |
+
"eval_samples_per_second": 3.531,
|
| 728 |
+
"eval_steps_per_second": 0.447,
|
| 729 |
+
"step": 2000
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 0.47368421052631576,
|
| 733 |
+
"grad_norm": 10.1796875,
|
| 734 |
+
"learning_rate": 1.4949748743718595e-05,
|
| 735 |
+
"loss": 0.3491,
|
| 736 |
+
"step": 2025
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 0.47953216374269003,
|
| 740 |
+
"grad_norm": 4.72265625,
|
| 741 |
+
"learning_rate": 1.4824120603015077e-05,
|
| 742 |
+
"loss": 0.1925,
|
| 743 |
+
"step": 2050
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 0.4853801169590643,
|
| 747 |
+
"grad_norm": 6.58203125,
|
| 748 |
+
"learning_rate": 1.4698492462311558e-05,
|
| 749 |
+
"loss": 0.2187,
|
| 750 |
+
"step": 2075
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 0.49122807017543857,
|
| 754 |
+
"grad_norm": 11.4609375,
|
| 755 |
+
"learning_rate": 1.457286432160804e-05,
|
| 756 |
+
"loss": 0.2695,
|
| 757 |
+
"step": 2100
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 0.49122807017543857,
|
| 761 |
+
"eval_loss": 0.2520677149295807,
|
| 762 |
+
"eval_runtime": 129.417,
|
| 763 |
+
"eval_samples_per_second": 3.477,
|
| 764 |
+
"eval_steps_per_second": 0.44,
|
| 765 |
+
"step": 2100
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.49707602339181284,
|
| 769 |
+
"grad_norm": 4.5,
|
| 770 |
+
"learning_rate": 1.4447236180904523e-05,
|
| 771 |
+
"loss": 0.1898,
|
| 772 |
+
"step": 2125
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.5029239766081871,
|
| 776 |
+
"grad_norm": 7.640625,
|
| 777 |
+
"learning_rate": 1.4321608040201007e-05,
|
| 778 |
+
"loss": 0.3,
|
| 779 |
+
"step": 2150
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.5087719298245614,
|
| 783 |
+
"grad_norm": 3.82421875,
|
| 784 |
+
"learning_rate": 1.4195979899497489e-05,
|
| 785 |
+
"loss": 0.2006,
|
| 786 |
+
"step": 2175
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.5146198830409356,
|
| 790 |
+
"grad_norm": 3.509765625,
|
| 791 |
+
"learning_rate": 1.407035175879397e-05,
|
| 792 |
+
"loss": 0.2619,
|
| 793 |
+
"step": 2200
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.5146198830409356,
|
| 797 |
+
"eval_loss": 0.24239127337932587,
|
| 798 |
+
"eval_runtime": 125.4976,
|
| 799 |
+
"eval_samples_per_second": 3.586,
|
| 800 |
+
"eval_steps_per_second": 0.454,
|
| 801 |
+
"step": 2200
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.52046783625731,
|
| 805 |
+
"grad_norm": 7.35546875,
|
| 806 |
+
"learning_rate": 1.3944723618090452e-05,
|
| 807 |
+
"loss": 0.3202,
|
| 808 |
+
"step": 2225
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.5263157894736842,
|
| 812 |
+
"grad_norm": 5.37109375,
|
| 813 |
+
"learning_rate": 1.3819095477386935e-05,
|
| 814 |
+
"loss": 0.143,
|
| 815 |
+
"step": 2250
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.5321637426900585,
|
| 819 |
+
"grad_norm": 13.203125,
|
| 820 |
+
"learning_rate": 1.369346733668342e-05,
|
| 821 |
+
"loss": 0.3726,
|
| 822 |
+
"step": 2275
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.5380116959064327,
|
| 826 |
+
"grad_norm": 7.08984375,
|
| 827 |
+
"learning_rate": 1.3567839195979901e-05,
|
| 828 |
+
"loss": 0.2238,
|
| 829 |
+
"step": 2300
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.5380116959064327,
|
| 833 |
+
"eval_loss": 0.2372661828994751,
|
| 834 |
+
"eval_runtime": 125.8271,
|
| 835 |
+
"eval_samples_per_second": 3.576,
|
| 836 |
+
"eval_steps_per_second": 0.453,
|
| 837 |
+
"step": 2300
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 0.543859649122807,
|
| 841 |
+
"grad_norm": 7.73046875,
|
| 842 |
+
"learning_rate": 1.3442211055276382e-05,
|
| 843 |
+
"loss": 0.2301,
|
| 844 |
+
"step": 2325
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 0.5497076023391813,
|
| 848 |
+
"grad_norm": 1.9326171875,
|
| 849 |
+
"learning_rate": 1.3316582914572864e-05,
|
| 850 |
+
"loss": 0.2891,
|
| 851 |
+
"step": 2350
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 0.5555555555555556,
|
| 855 |
+
"grad_norm": 5.25,
|
| 856 |
+
"learning_rate": 1.3190954773869347e-05,
|
| 857 |
+
"loss": 0.1861,
|
| 858 |
+
"step": 2375
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 0.5614035087719298,
|
| 862 |
+
"grad_norm": 15.7265625,
|
| 863 |
+
"learning_rate": 1.306532663316583e-05,
|
| 864 |
+
"loss": 0.3049,
|
| 865 |
+
"step": 2400
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 0.5614035087719298,
|
| 869 |
+
"eval_loss": 0.23007704317569733,
|
| 870 |
+
"eval_runtime": 125.6345,
|
| 871 |
+
"eval_samples_per_second": 3.582,
|
| 872 |
+
"eval_steps_per_second": 0.454,
|
| 873 |
+
"step": 2400
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 0.5672514619883041,
|
| 877 |
+
"grad_norm": 12.0078125,
|
| 878 |
+
"learning_rate": 1.2939698492462313e-05,
|
| 879 |
+
"loss": 0.2584,
|
| 880 |
+
"step": 2425
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 0.5730994152046783,
|
| 884 |
+
"grad_norm": 11.8125,
|
| 885 |
+
"learning_rate": 1.2814070351758795e-05,
|
| 886 |
+
"loss": 0.29,
|
| 887 |
+
"step": 2450
|
| 888 |
+
},
|
| 889 |
+
{
|
| 890 |
+
"epoch": 0.5789473684210527,
|
| 891 |
+
"grad_norm": 4.7578125,
|
| 892 |
+
"learning_rate": 1.2688442211055276e-05,
|
| 893 |
+
"loss": 0.2648,
|
| 894 |
+
"step": 2475
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 0.5847953216374269,
|
| 898 |
+
"grad_norm": 10.53125,
|
| 899 |
+
"learning_rate": 1.2562814070351759e-05,
|
| 900 |
+
"loss": 0.1308,
|
| 901 |
+
"step": 2500
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 0.5847953216374269,
|
| 905 |
+
"eval_loss": 0.2292058765888214,
|
| 906 |
+
"eval_runtime": 127.785,
|
| 907 |
+
"eval_samples_per_second": 3.522,
|
| 908 |
+
"eval_steps_per_second": 0.446,
|
| 909 |
+
"step": 2500
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 0.5906432748538012,
|
| 913 |
+
"grad_norm": 8.421875,
|
| 914 |
+
"learning_rate": 1.2437185929648242e-05,
|
| 915 |
+
"loss": 0.3249,
|
| 916 |
+
"step": 2525
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 0.5964912280701754,
|
| 920 |
+
"grad_norm": 10.8828125,
|
| 921 |
+
"learning_rate": 1.2311557788944725e-05,
|
| 922 |
+
"loss": 0.2164,
|
| 923 |
+
"step": 2550
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 0.6023391812865497,
|
| 927 |
+
"grad_norm": 7.421875,
|
| 928 |
+
"learning_rate": 1.2185929648241207e-05,
|
| 929 |
+
"loss": 0.2512,
|
| 930 |
+
"step": 2575
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 0.6081871345029239,
|
| 934 |
+
"grad_norm": 9.6171875,
|
| 935 |
+
"learning_rate": 1.2060301507537688e-05,
|
| 936 |
+
"loss": 0.1936,
|
| 937 |
+
"step": 2600
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 0.6081871345029239,
|
| 941 |
+
"eval_loss": 0.21902820467948914,
|
| 942 |
+
"eval_runtime": 125.3191,
|
| 943 |
+
"eval_samples_per_second": 3.591,
|
| 944 |
+
"eval_steps_per_second": 0.455,
|
| 945 |
+
"step": 2600
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 0.6140350877192983,
|
| 949 |
+
"grad_norm": 1.091796875,
|
| 950 |
+
"learning_rate": 1.1934673366834171e-05,
|
| 951 |
+
"loss": 0.189,
|
| 952 |
+
"step": 2625
|
| 953 |
+
},
|
| 954 |
+
{
|
| 955 |
+
"epoch": 0.6198830409356725,
|
| 956 |
+
"grad_norm": 7.83203125,
|
| 957 |
+
"learning_rate": 1.1809045226130654e-05,
|
| 958 |
+
"loss": 0.2179,
|
| 959 |
+
"step": 2650
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 0.6257309941520468,
|
| 963 |
+
"grad_norm": 8.2109375,
|
| 964 |
+
"learning_rate": 1.1683417085427137e-05,
|
| 965 |
+
"loss": 0.224,
|
| 966 |
+
"step": 2675
|
| 967 |
+
},
|
| 968 |
+
{
|
| 969 |
+
"epoch": 0.631578947368421,
|
| 970 |
+
"grad_norm": 5.1640625,
|
| 971 |
+
"learning_rate": 1.1557788944723619e-05,
|
| 972 |
+
"loss": 0.2479,
|
| 973 |
+
"step": 2700
|
| 974 |
+
},
|
| 975 |
+
{
|
| 976 |
+
"epoch": 0.631578947368421,
|
| 977 |
+
"eval_loss": 0.21907079219818115,
|
| 978 |
+
"eval_runtime": 126.3038,
|
| 979 |
+
"eval_samples_per_second": 3.563,
|
| 980 |
+
"eval_steps_per_second": 0.451,
|
| 981 |
+
"step": 2700
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 0.6374269005847953,
|
| 985 |
+
"grad_norm": 4.47265625,
|
| 986 |
+
"learning_rate": 1.14321608040201e-05,
|
| 987 |
+
"loss": 0.219,
|
| 988 |
+
"step": 2725
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 0.6432748538011696,
|
| 992 |
+
"grad_norm": 3.30078125,
|
| 993 |
+
"learning_rate": 1.1306532663316583e-05,
|
| 994 |
+
"loss": 0.2741,
|
| 995 |
+
"step": 2750
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 0.6491228070175439,
|
| 999 |
+
"grad_norm": 4.21484375,
|
| 1000 |
+
"learning_rate": 1.1180904522613066e-05,
|
| 1001 |
+
"loss": 0.1153,
|
| 1002 |
+
"step": 2775
|
| 1003 |
+
},
|
| 1004 |
+
{
|
| 1005 |
+
"epoch": 0.6549707602339181,
|
| 1006 |
+
"grad_norm": 5.125,
|
| 1007 |
+
"learning_rate": 1.105527638190955e-05,
|
| 1008 |
+
"loss": 0.1575,
|
| 1009 |
+
"step": 2800
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 0.6549707602339181,
|
| 1013 |
+
"eval_loss": 0.2165260761976242,
|
| 1014 |
+
"eval_runtime": 125.8096,
|
| 1015 |
+
"eval_samples_per_second": 3.577,
|
| 1016 |
+
"eval_steps_per_second": 0.453,
|
| 1017 |
+
"step": 2800
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.6608187134502924,
|
| 1021 |
+
"grad_norm": 0.83837890625,
|
| 1022 |
+
"learning_rate": 1.0929648241206031e-05,
|
| 1023 |
+
"loss": 0.2238,
|
| 1024 |
+
"step": 2825
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.6666666666666666,
|
| 1028 |
+
"grad_norm": 6.61328125,
|
| 1029 |
+
"learning_rate": 1.0804020100502512e-05,
|
| 1030 |
+
"loss": 0.3554,
|
| 1031 |
+
"step": 2850
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.672514619883041,
|
| 1035 |
+
"grad_norm": 12.875,
|
| 1036 |
+
"learning_rate": 1.0678391959798995e-05,
|
| 1037 |
+
"loss": 0.096,
|
| 1038 |
+
"step": 2875
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.6783625730994152,
|
| 1042 |
+
"grad_norm": 3.28125,
|
| 1043 |
+
"learning_rate": 1.0552763819095479e-05,
|
| 1044 |
+
"loss": 0.193,
|
| 1045 |
+
"step": 2900
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.6783625730994152,
|
| 1049 |
+
"eval_loss": 0.21065188944339752,
|
| 1050 |
+
"eval_runtime": 125.6087,
|
| 1051 |
+
"eval_samples_per_second": 3.583,
|
| 1052 |
+
"eval_steps_per_second": 0.454,
|
| 1053 |
+
"step": 2900
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.6842105263157895,
|
| 1057 |
+
"grad_norm": 8.7265625,
|
| 1058 |
+
"learning_rate": 1.042713567839196e-05,
|
| 1059 |
+
"loss": 0.1936,
|
| 1060 |
+
"step": 2925
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.6900584795321637,
|
| 1064 |
+
"grad_norm": 4.0859375,
|
| 1065 |
+
"learning_rate": 1.0301507537688443e-05,
|
| 1066 |
+
"loss": 0.2103,
|
| 1067 |
+
"step": 2950
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.695906432748538,
|
| 1071 |
+
"grad_norm": 15.5703125,
|
| 1072 |
+
"learning_rate": 1.0175879396984924e-05,
|
| 1073 |
+
"loss": 0.2701,
|
| 1074 |
+
"step": 2975
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.7017543859649122,
|
| 1078 |
+
"grad_norm": 3.73046875,
|
| 1079 |
+
"learning_rate": 1.0050251256281408e-05,
|
| 1080 |
+
"loss": 0.2526,
|
| 1081 |
+
"step": 3000
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.7017543859649122,
|
| 1085 |
+
"eval_loss": 0.21144379675388336,
|
| 1086 |
+
"eval_runtime": 128.1153,
|
| 1087 |
+
"eval_samples_per_second": 3.512,
|
| 1088 |
+
"eval_steps_per_second": 0.445,
|
| 1089 |
+
"step": 3000
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 0.7076023391812866,
|
| 1093 |
+
"grad_norm": 1.537109375,
|
| 1094 |
+
"learning_rate": 9.92462311557789e-06,
|
| 1095 |
+
"loss": 0.2087,
|
| 1096 |
+
"step": 3025
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 0.7134502923976608,
|
| 1100 |
+
"grad_norm": 0.55908203125,
|
| 1101 |
+
"learning_rate": 9.798994974874372e-06,
|
| 1102 |
+
"loss": 0.1532,
|
| 1103 |
+
"step": 3050
|
| 1104 |
+
},
|
| 1105 |
+
{
|
| 1106 |
+
"epoch": 0.7192982456140351,
|
| 1107 |
+
"grad_norm": 9.7578125,
|
| 1108 |
+
"learning_rate": 9.673366834170855e-06,
|
| 1109 |
+
"loss": 0.1985,
|
| 1110 |
+
"step": 3075
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"epoch": 0.7251461988304093,
|
| 1114 |
+
"grad_norm": 9.390625,
|
| 1115 |
+
"learning_rate": 9.547738693467337e-06,
|
| 1116 |
+
"loss": 0.1574,
|
| 1117 |
+
"step": 3100
|
| 1118 |
+
},
|
| 1119 |
+
{
|
| 1120 |
+
"epoch": 0.7251461988304093,
|
| 1121 |
+
"eval_loss": 0.20868618786334991,
|
| 1122 |
+
"eval_runtime": 129.4237,
|
| 1123 |
+
"eval_samples_per_second": 3.477,
|
| 1124 |
+
"eval_steps_per_second": 0.44,
|
| 1125 |
+
"step": 3100
|
| 1126 |
+
},
|
| 1127 |
+
{
|
| 1128 |
+
"epoch": 0.7309941520467836,
|
| 1129 |
+
"grad_norm": 4.6171875,
|
| 1130 |
+
"learning_rate": 9.42211055276382e-06,
|
| 1131 |
+
"loss": 0.2921,
|
| 1132 |
+
"step": 3125
|
| 1133 |
+
},
|
| 1134 |
+
{
|
| 1135 |
+
"epoch": 0.7368421052631579,
|
| 1136 |
+
"grad_norm": 10.859375,
|
| 1137 |
+
"learning_rate": 9.296482412060301e-06,
|
| 1138 |
+
"loss": 0.1932,
|
| 1139 |
+
"step": 3150
|
| 1140 |
+
},
|
| 1141 |
+
{
|
| 1142 |
+
"epoch": 0.7426900584795322,
|
| 1143 |
+
"grad_norm": 5.97265625,
|
| 1144 |
+
"learning_rate": 9.170854271356784e-06,
|
| 1145 |
+
"loss": 0.2999,
|
| 1146 |
+
"step": 3175
|
| 1147 |
+
},
|
| 1148 |
+
{
|
| 1149 |
+
"epoch": 0.7485380116959064,
|
| 1150 |
+
"grad_norm": 4.85546875,
|
| 1151 |
+
"learning_rate": 9.045226130653267e-06,
|
| 1152 |
+
"loss": 0.1989,
|
| 1153 |
+
"step": 3200
|
| 1154 |
+
},
|
| 1155 |
+
{
|
| 1156 |
+
"epoch": 0.7485380116959064,
|
| 1157 |
+
"eval_loss": 0.20511174201965332,
|
| 1158 |
+
"eval_runtime": 127.5209,
|
| 1159 |
+
"eval_samples_per_second": 3.529,
|
| 1160 |
+
"eval_steps_per_second": 0.447,
|
| 1161 |
+
"step": 3200
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 0.7543859649122807,
|
| 1165 |
+
"grad_norm": 8.7265625,
|
| 1166 |
+
"learning_rate": 8.919597989949749e-06,
|
| 1167 |
+
"loss": 0.1795,
|
| 1168 |
+
"step": 3225
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 0.7602339181286549,
|
| 1172 |
+
"grad_norm": 3.490234375,
|
| 1173 |
+
"learning_rate": 8.793969849246232e-06,
|
| 1174 |
+
"loss": 0.2444,
|
| 1175 |
+
"step": 3250
|
| 1176 |
+
},
|
| 1177 |
+
{
|
| 1178 |
+
"epoch": 0.7660818713450293,
|
| 1179 |
+
"grad_norm": 6.671875,
|
| 1180 |
+
"learning_rate": 8.668341708542713e-06,
|
| 1181 |
+
"loss": 0.2775,
|
| 1182 |
+
"step": 3275
|
| 1183 |
+
},
|
| 1184 |
+
{
|
| 1185 |
+
"epoch": 0.7719298245614035,
|
| 1186 |
+
"grad_norm": 5.1796875,
|
| 1187 |
+
"learning_rate": 8.542713567839196e-06,
|
| 1188 |
+
"loss": 0.1761,
|
| 1189 |
+
"step": 3300
|
| 1190 |
+
},
|
| 1191 |
+
{
|
| 1192 |
+
"epoch": 0.7719298245614035,
|
| 1193 |
+
"eval_loss": 0.20133711397647858,
|
| 1194 |
+
"eval_runtime": 129.2927,
|
| 1195 |
+
"eval_samples_per_second": 3.48,
|
| 1196 |
+
"eval_steps_per_second": 0.441,
|
| 1197 |
+
"step": 3300
|
| 1198 |
+
},
|
| 1199 |
+
{
|
| 1200 |
+
"epoch": 0.7777777777777778,
|
| 1201 |
+
"grad_norm": 0.83935546875,
|
| 1202 |
+
"learning_rate": 8.41708542713568e-06,
|
| 1203 |
+
"loss": 0.216,
|
| 1204 |
+
"step": 3325
|
| 1205 |
+
},
|
| 1206 |
+
{
|
| 1207 |
+
"epoch": 0.783625730994152,
|
| 1208 |
+
"grad_norm": 2.580078125,
|
| 1209 |
+
"learning_rate": 8.291457286432161e-06,
|
| 1210 |
+
"loss": 0.0873,
|
| 1211 |
+
"step": 3350
|
| 1212 |
+
},
|
| 1213 |
+
{
|
| 1214 |
+
"epoch": 0.7894736842105263,
|
| 1215 |
+
"grad_norm": 5.87109375,
|
| 1216 |
+
"learning_rate": 8.165829145728644e-06,
|
| 1217 |
+
"loss": 0.1145,
|
| 1218 |
+
"step": 3375
|
| 1219 |
+
},
|
| 1220 |
+
{
|
| 1221 |
+
"epoch": 0.7953216374269005,
|
| 1222 |
+
"grad_norm": 7.125,
|
| 1223 |
+
"learning_rate": 8.040201005025125e-06,
|
| 1224 |
+
"loss": 0.2223,
|
| 1225 |
+
"step": 3400
|
| 1226 |
+
},
|
| 1227 |
+
{
|
| 1228 |
+
"epoch": 0.7953216374269005,
|
| 1229 |
+
"eval_loss": 0.19959186017513275,
|
| 1230 |
+
"eval_runtime": 125.4382,
|
| 1231 |
+
"eval_samples_per_second": 3.587,
|
| 1232 |
+
"eval_steps_per_second": 0.454,
|
| 1233 |
+
"step": 3400
|
| 1234 |
+
},
|
| 1235 |
+
{
|
| 1236 |
+
"epoch": 0.8011695906432749,
|
| 1237 |
+
"grad_norm": 3.94921875,
|
| 1238 |
+
"learning_rate": 7.914572864321608e-06,
|
| 1239 |
+
"loss": 0.1845,
|
| 1240 |
+
"step": 3425
|
| 1241 |
+
},
|
| 1242 |
+
{
|
| 1243 |
+
"epoch": 0.8070175438596491,
|
| 1244 |
+
"grad_norm": 5.99609375,
|
| 1245 |
+
"learning_rate": 7.788944723618092e-06,
|
| 1246 |
+
"loss": 0.104,
|
| 1247 |
+
"step": 3450
|
| 1248 |
+
},
|
| 1249 |
+
{
|
| 1250 |
+
"epoch": 0.8128654970760234,
|
| 1251 |
+
"grad_norm": 7.953125,
|
| 1252 |
+
"learning_rate": 7.663316582914573e-06,
|
| 1253 |
+
"loss": 0.1119,
|
| 1254 |
+
"step": 3475
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 0.8187134502923976,
|
| 1258 |
+
"grad_norm": 7.5859375,
|
| 1259 |
+
"learning_rate": 7.537688442211055e-06,
|
| 1260 |
+
"loss": 0.2127,
|
| 1261 |
+
"step": 3500
|
| 1262 |
+
},
|
| 1263 |
+
{
|
| 1264 |
+
"epoch": 0.8187134502923976,
|
| 1265 |
+
"eval_loss": 0.1966124027967453,
|
| 1266 |
+
"eval_runtime": 126.749,
|
| 1267 |
+
"eval_samples_per_second": 3.55,
|
| 1268 |
+
"eval_steps_per_second": 0.45,
|
| 1269 |
+
"step": 3500
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.8245614035087719,
|
| 1273 |
+
"grad_norm": 25.390625,
|
| 1274 |
+
"learning_rate": 7.412060301507538e-06,
|
| 1275 |
+
"loss": 0.1597,
|
| 1276 |
+
"step": 3525
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.8304093567251462,
|
| 1280 |
+
"grad_norm": 7.55078125,
|
| 1281 |
+
"learning_rate": 7.28643216080402e-06,
|
| 1282 |
+
"loss": 0.0942,
|
| 1283 |
+
"step": 3550
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.8362573099415205,
|
| 1287 |
+
"grad_norm": 6.5390625,
|
| 1288 |
+
"learning_rate": 7.160804020100504e-06,
|
| 1289 |
+
"loss": 0.236,
|
| 1290 |
+
"step": 3575
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.8421052631578947,
|
| 1294 |
+
"grad_norm": 9.390625,
|
| 1295 |
+
"learning_rate": 7.035175879396985e-06,
|
| 1296 |
+
"loss": 0.2477,
|
| 1297 |
+
"step": 3600
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.8421052631578947,
|
| 1301 |
+
"eval_loss": 0.1922728419303894,
|
| 1302 |
+
"eval_runtime": 125.5954,
|
| 1303 |
+
"eval_samples_per_second": 3.583,
|
| 1304 |
+
"eval_steps_per_second": 0.454,
|
| 1305 |
+
"step": 3600
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.847953216374269,
|
| 1309 |
+
"grad_norm": 0.7783203125,
|
| 1310 |
+
"learning_rate": 6.909547738693467e-06,
|
| 1311 |
+
"loss": 0.3097,
|
| 1312 |
+
"step": 3625
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.8538011695906432,
|
| 1316 |
+
"grad_norm": 5.91796875,
|
| 1317 |
+
"learning_rate": 6.7839195979899505e-06,
|
| 1318 |
+
"loss": 0.2097,
|
| 1319 |
+
"step": 3650
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.8596491228070176,
|
| 1323 |
+
"grad_norm": 2.55078125,
|
| 1324 |
+
"learning_rate": 6.658291457286432e-06,
|
| 1325 |
+
"loss": 0.1837,
|
| 1326 |
+
"step": 3675
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.8654970760233918,
|
| 1330 |
+
"grad_norm": 5.1015625,
|
| 1331 |
+
"learning_rate": 6.532663316582915e-06,
|
| 1332 |
+
"loss": 0.1931,
|
| 1333 |
+
"step": 3700
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.8654970760233918,
|
| 1337 |
+
"eval_loss": 0.1908126324415207,
|
| 1338 |
+
"eval_runtime": 125.264,
|
| 1339 |
+
"eval_samples_per_second": 3.592,
|
| 1340 |
+
"eval_steps_per_second": 0.455,
|
| 1341 |
+
"step": 3700
|
| 1342 |
+
},
|
| 1343 |
+
{
|
| 1344 |
+
"epoch": 0.8713450292397661,
|
| 1345 |
+
"grad_norm": 1.0830078125,
|
| 1346 |
+
"learning_rate": 6.407035175879397e-06,
|
| 1347 |
+
"loss": 0.1688,
|
| 1348 |
+
"step": 3725
|
| 1349 |
+
},
|
| 1350 |
+
{
|
| 1351 |
+
"epoch": 0.8771929824561403,
|
| 1352 |
+
"grad_norm": 2.021484375,
|
| 1353 |
+
"learning_rate": 6.2814070351758795e-06,
|
| 1354 |
+
"loss": 0.1635,
|
| 1355 |
+
"step": 3750
|
| 1356 |
+
},
|
| 1357 |
+
{
|
| 1358 |
+
"epoch": 0.8830409356725146,
|
| 1359 |
+
"grad_norm": 24.90625,
|
| 1360 |
+
"learning_rate": 6.155778894472363e-06,
|
| 1361 |
+
"loss": 0.2588,
|
| 1362 |
+
"step": 3775
|
| 1363 |
+
},
|
| 1364 |
+
{
|
| 1365 |
+
"epoch": 0.8888888888888888,
|
| 1366 |
+
"grad_norm": 4.9453125,
|
| 1367 |
+
"learning_rate": 6.030150753768844e-06,
|
| 1368 |
+
"loss": 0.182,
|
| 1369 |
+
"step": 3800
|
| 1370 |
+
},
|
| 1371 |
+
{
|
| 1372 |
+
"epoch": 0.8888888888888888,
|
| 1373 |
+
"eval_loss": 0.18878790736198425,
|
| 1374 |
+
"eval_runtime": 126.7176,
|
| 1375 |
+
"eval_samples_per_second": 3.551,
|
| 1376 |
+
"eval_steps_per_second": 0.45,
|
| 1377 |
+
"step": 3800
|
| 1378 |
+
},
|
| 1379 |
+
{
|
| 1380 |
+
"epoch": 0.8947368421052632,
|
| 1381 |
+
"grad_norm": 7.25,
|
| 1382 |
+
"learning_rate": 5.904522613065327e-06,
|
| 1383 |
+
"loss": 0.2793,
|
| 1384 |
+
"step": 3825
|
| 1385 |
+
},
|
| 1386 |
+
{
|
| 1387 |
+
"epoch": 0.9005847953216374,
|
| 1388 |
+
"grad_norm": 9.59375,
|
| 1389 |
+
"learning_rate": 5.778894472361809e-06,
|
| 1390 |
+
"loss": 0.1608,
|
| 1391 |
+
"step": 3850
|
| 1392 |
+
},
|
| 1393 |
+
{
|
| 1394 |
+
"epoch": 0.9064327485380117,
|
| 1395 |
+
"grad_norm": 2.943359375,
|
| 1396 |
+
"learning_rate": 5.653266331658292e-06,
|
| 1397 |
+
"loss": 0.2136,
|
| 1398 |
+
"step": 3875
|
| 1399 |
+
},
|
| 1400 |
+
{
|
| 1401 |
+
"epoch": 0.9122807017543859,
|
| 1402 |
+
"grad_norm": 0.48681640625,
|
| 1403 |
+
"learning_rate": 5.527638190954775e-06,
|
| 1404 |
+
"loss": 0.1693,
|
| 1405 |
+
"step": 3900
|
| 1406 |
+
},
|
| 1407 |
+
{
|
| 1408 |
+
"epoch": 0.9122807017543859,
|
| 1409 |
+
"eval_loss": 0.18779730796813965,
|
| 1410 |
+
"eval_runtime": 126.6799,
|
| 1411 |
+
"eval_samples_per_second": 3.552,
|
| 1412 |
+
"eval_steps_per_second": 0.45,
|
| 1413 |
+
"step": 3900
|
| 1414 |
+
},
|
| 1415 |
+
{
|
| 1416 |
+
"epoch": 0.9181286549707602,
|
| 1417 |
+
"grad_norm": 3.2109375,
|
| 1418 |
+
"learning_rate": 5.402010050251256e-06,
|
| 1419 |
+
"loss": 0.0918,
|
| 1420 |
+
"step": 3925
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"epoch": 0.9239766081871345,
|
| 1424 |
+
"grad_norm": 1.541015625,
|
| 1425 |
+
"learning_rate": 5.276381909547739e-06,
|
| 1426 |
+
"loss": 0.2076,
|
| 1427 |
+
"step": 3950
|
| 1428 |
+
},
|
| 1429 |
+
{
|
| 1430 |
+
"epoch": 0.9298245614035088,
|
| 1431 |
+
"grad_norm": 4.7421875,
|
| 1432 |
+
"learning_rate": 5.1507537688442215e-06,
|
| 1433 |
+
"loss": 0.2429,
|
| 1434 |
+
"step": 3975
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 0.935672514619883,
|
| 1438 |
+
"grad_norm": 4.375,
|
| 1439 |
+
"learning_rate": 5.025125628140704e-06,
|
| 1440 |
+
"loss": 0.1346,
|
| 1441 |
+
"step": 4000
|
| 1442 |
+
},
|
| 1443 |
+
{
|
| 1444 |
+
"epoch": 0.935672514619883,
|
| 1445 |
+
"eval_loss": 0.18527910113334656,
|
| 1446 |
+
"eval_runtime": 129.3266,
|
| 1447 |
+
"eval_samples_per_second": 3.48,
|
| 1448 |
+
"eval_steps_per_second": 0.441,
|
| 1449 |
+
"step": 4000
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 0.9415204678362573,
|
| 1453 |
+
"grad_norm": 4.875,
|
| 1454 |
+
"learning_rate": 4.899497487437186e-06,
|
| 1455 |
+
"loss": 0.2457,
|
| 1456 |
+
"step": 4025
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 0.9473684210526315,
|
| 1460 |
+
"grad_norm": 3.013671875,
|
| 1461 |
+
"learning_rate": 4.773869346733668e-06,
|
| 1462 |
+
"loss": 0.187,
|
| 1463 |
+
"step": 4050
|
| 1464 |
+
},
|
| 1465 |
+
{
|
| 1466 |
+
"epoch": 0.9532163742690059,
|
| 1467 |
+
"grad_norm": 4.3125,
|
| 1468 |
+
"learning_rate": 4.6482412060301506e-06,
|
| 1469 |
+
"loss": 0.1546,
|
| 1470 |
+
"step": 4075
|
| 1471 |
+
},
|
| 1472 |
+
{
|
| 1473 |
+
"epoch": 0.9590643274853801,
|
| 1474 |
+
"grad_norm": 2.5,
|
| 1475 |
+
"learning_rate": 4.522613065326634e-06,
|
| 1476 |
+
"loss": 0.1484,
|
| 1477 |
+
"step": 4100
|
| 1478 |
+
},
|
| 1479 |
+
{
|
| 1480 |
+
"epoch": 0.9590643274853801,
|
| 1481 |
+
"eval_loss": 0.18491099774837494,
|
| 1482 |
+
"eval_runtime": 126.5822,
|
| 1483 |
+
"eval_samples_per_second": 3.555,
|
| 1484 |
+
"eval_steps_per_second": 0.45,
|
| 1485 |
+
"step": 4100
|
| 1486 |
+
},
|
| 1487 |
+
{
|
| 1488 |
+
"epoch": 0.9649122807017544,
|
| 1489 |
+
"grad_norm": 5.03515625,
|
| 1490 |
+
"learning_rate": 4.396984924623116e-06,
|
| 1491 |
+
"loss": 0.1739,
|
| 1492 |
+
"step": 4125
|
| 1493 |
+
},
|
| 1494 |
+
{
|
| 1495 |
+
"epoch": 0.9707602339181286,
|
| 1496 |
+
"grad_norm": 7.1015625,
|
| 1497 |
+
"learning_rate": 4.271356783919598e-06,
|
| 1498 |
+
"loss": 0.3268,
|
| 1499 |
+
"step": 4150
|
| 1500 |
+
},
|
| 1501 |
+
{
|
| 1502 |
+
"epoch": 0.9766081871345029,
|
| 1503 |
+
"grad_norm": 0.397216796875,
|
| 1504 |
+
"learning_rate": 4.1457286432160804e-06,
|
| 1505 |
+
"loss": 0.2178,
|
| 1506 |
+
"step": 4175
|
| 1507 |
+
},
|
| 1508 |
+
{
|
| 1509 |
+
"epoch": 0.9824561403508771,
|
| 1510 |
+
"grad_norm": 5.00390625,
|
| 1511 |
+
"learning_rate": 4.020100502512563e-06,
|
| 1512 |
+
"loss": 0.1217,
|
| 1513 |
+
"step": 4200
|
| 1514 |
+
},
|
| 1515 |
+
{
|
| 1516 |
+
"epoch": 0.9824561403508771,
|
| 1517 |
+
"eval_loss": 0.1838068664073944,
|
| 1518 |
+
"eval_runtime": 127.461,
|
| 1519 |
+
"eval_samples_per_second": 3.53,
|
| 1520 |
+
"eval_steps_per_second": 0.447,
|
| 1521 |
+
"step": 4200
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.9883040935672515,
|
| 1525 |
+
"grad_norm": 0.1112060546875,
|
| 1526 |
+
"learning_rate": 3.894472361809046e-06,
|
| 1527 |
+
"loss": 0.0894,
|
| 1528 |
+
"step": 4225
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.9941520467836257,
|
| 1532 |
+
"grad_norm": 1.748046875,
|
| 1533 |
+
"learning_rate": 3.7688442211055276e-06,
|
| 1534 |
+
"loss": 0.1733,
|
| 1535 |
+
"step": 4250
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 1.0,
|
| 1539 |
+
"grad_norm": 6.53125,
|
| 1540 |
+
"learning_rate": 3.64321608040201e-06,
|
| 1541 |
+
"loss": 0.1281,
|
| 1542 |
+
"step": 4275
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 1.0058479532163742,
|
| 1546 |
+
"grad_norm": 0.9111328125,
|
| 1547 |
+
"learning_rate": 3.5175879396984926e-06,
|
| 1548 |
+
"loss": 0.0669,
|
| 1549 |
+
"step": 4300
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 1.0058479532163742,
|
| 1553 |
+
"eval_loss": 0.18437370657920837,
|
| 1554 |
+
"eval_runtime": 128.6228,
|
| 1555 |
+
"eval_samples_per_second": 3.499,
|
| 1556 |
+
"eval_steps_per_second": 0.443,
|
| 1557 |
+
"step": 4300
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 1.0116959064327484,
|
| 1561 |
+
"grad_norm": 0.486328125,
|
| 1562 |
+
"learning_rate": 3.3919597989949752e-06,
|
| 1563 |
+
"loss": 0.0748,
|
| 1564 |
+
"step": 4325
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 1.0175438596491229,
|
| 1568 |
+
"grad_norm": 7.703125,
|
| 1569 |
+
"learning_rate": 3.2663316582914575e-06,
|
| 1570 |
+
"loss": 0.071,
|
| 1571 |
+
"step": 4350
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 1.023391812865497,
|
| 1575 |
+
"grad_norm": 8.9765625,
|
| 1576 |
+
"learning_rate": 3.1407035175879398e-06,
|
| 1577 |
+
"loss": 0.1588,
|
| 1578 |
+
"step": 4375
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 1.0292397660818713,
|
| 1582 |
+
"grad_norm": 0.38427734375,
|
| 1583 |
+
"learning_rate": 3.015075376884422e-06,
|
| 1584 |
+
"loss": 0.1292,
|
| 1585 |
+
"step": 4400
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 1.0292397660818713,
|
| 1589 |
+
"eval_loss": 0.18767422437667847,
|
| 1590 |
+
"eval_runtime": 126.9252,
|
| 1591 |
+
"eval_samples_per_second": 3.545,
|
| 1592 |
+
"eval_steps_per_second": 0.449,
|
| 1593 |
+
"step": 4400
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 1.0350877192982457,
|
| 1597 |
+
"grad_norm": 6.4375,
|
| 1598 |
+
"learning_rate": 2.8894472361809047e-06,
|
| 1599 |
+
"loss": 0.1419,
|
| 1600 |
+
"step": 4425
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 1.04093567251462,
|
| 1604 |
+
"grad_norm": 13.3515625,
|
| 1605 |
+
"learning_rate": 2.7638190954773874e-06,
|
| 1606 |
+
"loss": 0.084,
|
| 1607 |
+
"step": 4450
|
| 1608 |
+
},
|
| 1609 |
+
{
|
| 1610 |
+
"epoch": 1.0467836257309941,
|
| 1611 |
+
"grad_norm": 9.203125,
|
| 1612 |
+
"learning_rate": 2.6381909547738696e-06,
|
| 1613 |
+
"loss": 0.072,
|
| 1614 |
+
"step": 4475
|
| 1615 |
+
},
|
| 1616 |
+
{
|
| 1617 |
+
"epoch": 1.0526315789473684,
|
| 1618 |
+
"grad_norm": 2.439453125,
|
| 1619 |
+
"learning_rate": 2.512562814070352e-06,
|
| 1620 |
+
"loss": 0.1106,
|
| 1621 |
+
"step": 4500
|
| 1622 |
+
},
|
| 1623 |
+
{
|
| 1624 |
+
"epoch": 1.0526315789473684,
|
| 1625 |
+
"eval_loss": 0.18756501376628876,
|
| 1626 |
+
"eval_runtime": 128.4204,
|
| 1627 |
+
"eval_samples_per_second": 3.504,
|
| 1628 |
+
"eval_steps_per_second": 0.444,
|
| 1629 |
+
"step": 4500
|
| 1630 |
+
},
|
| 1631 |
+
{
|
| 1632 |
+
"epoch": 1.0584795321637426,
|
| 1633 |
+
"grad_norm": 4.671875,
|
| 1634 |
+
"learning_rate": 2.386934673366834e-06,
|
| 1635 |
+
"loss": 0.1226,
|
| 1636 |
+
"step": 4525
|
| 1637 |
+
},
|
| 1638 |
+
{
|
| 1639 |
+
"epoch": 1.064327485380117,
|
| 1640 |
+
"grad_norm": 2.62890625,
|
| 1641 |
+
"learning_rate": 2.261306532663317e-06,
|
| 1642 |
+
"loss": 0.123,
|
| 1643 |
+
"step": 4550
|
| 1644 |
+
},
|
| 1645 |
+
{
|
| 1646 |
+
"epoch": 1.0701754385964912,
|
| 1647 |
+
"grad_norm": 16.640625,
|
| 1648 |
+
"learning_rate": 2.135678391959799e-06,
|
| 1649 |
+
"loss": 0.0593,
|
| 1650 |
+
"step": 4575
|
| 1651 |
+
},
|
| 1652 |
+
{
|
| 1653 |
+
"epoch": 1.0760233918128654,
|
| 1654 |
+
"grad_norm": 6.8828125,
|
| 1655 |
+
"learning_rate": 2.0100502512562813e-06,
|
| 1656 |
+
"loss": 0.0828,
|
| 1657 |
+
"step": 4600
|
| 1658 |
+
},
|
| 1659 |
+
{
|
| 1660 |
+
"epoch": 1.0760233918128654,
|
| 1661 |
+
"eval_loss": 0.1875353455543518,
|
| 1662 |
+
"eval_runtime": 126.5483,
|
| 1663 |
+
"eval_samples_per_second": 3.556,
|
| 1664 |
+
"eval_steps_per_second": 0.45,
|
| 1665 |
+
"step": 4600
|
| 1666 |
+
},
|
| 1667 |
+
{
|
| 1668 |
+
"epoch": 1.0818713450292399,
|
| 1669 |
+
"grad_norm": 4.55078125,
|
| 1670 |
+
"learning_rate": 1.8844221105527638e-06,
|
| 1671 |
+
"loss": 0.1481,
|
| 1672 |
+
"step": 4625
|
| 1673 |
+
},
|
| 1674 |
+
{
|
| 1675 |
+
"epoch": 1.087719298245614,
|
| 1676 |
+
"grad_norm": 3.6796875,
|
| 1677 |
+
"learning_rate": 1.7587939698492463e-06,
|
| 1678 |
+
"loss": 0.1275,
|
| 1679 |
+
"step": 4650
|
| 1680 |
+
},
|
| 1681 |
+
{
|
| 1682 |
+
"epoch": 1.0935672514619883,
|
| 1683 |
+
"grad_norm": 6.02734375,
|
| 1684 |
+
"learning_rate": 1.6331658291457288e-06,
|
| 1685 |
+
"loss": 0.1274,
|
| 1686 |
+
"step": 4675
|
| 1687 |
+
},
|
| 1688 |
+
{
|
| 1689 |
+
"epoch": 1.0994152046783625,
|
| 1690 |
+
"grad_norm": 1.4111328125,
|
| 1691 |
+
"learning_rate": 1.507537688442211e-06,
|
| 1692 |
+
"loss": 0.0485,
|
| 1693 |
+
"step": 4700
|
| 1694 |
+
},
|
| 1695 |
+
{
|
| 1696 |
+
"epoch": 1.0994152046783625,
|
| 1697 |
+
"eval_loss": 0.18709704279899597,
|
| 1698 |
+
"eval_runtime": 129.1721,
|
| 1699 |
+
"eval_samples_per_second": 3.484,
|
| 1700 |
+
"eval_steps_per_second": 0.441,
|
| 1701 |
+
"step": 4700
|
| 1702 |
+
},
|
| 1703 |
+
{
|
| 1704 |
+
"epoch": 1.1052631578947367,
|
| 1705 |
+
"grad_norm": 0.74853515625,
|
| 1706 |
+
"learning_rate": 1.3819095477386937e-06,
|
| 1707 |
+
"loss": 0.0483,
|
| 1708 |
+
"step": 4725
|
| 1709 |
+
},
|
| 1710 |
+
{
|
| 1711 |
+
"epoch": 1.1111111111111112,
|
| 1712 |
+
"grad_norm": 1.044921875,
|
| 1713 |
+
"learning_rate": 1.256281407035176e-06,
|
| 1714 |
+
"loss": 0.0799,
|
| 1715 |
+
"step": 4750
|
| 1716 |
+
},
|
| 1717 |
+
{
|
| 1718 |
+
"epoch": 1.1169590643274854,
|
| 1719 |
+
"grad_norm": 0.171142578125,
|
| 1720 |
+
"learning_rate": 1.1306532663316584e-06,
|
| 1721 |
+
"loss": 0.1273,
|
| 1722 |
+
"step": 4775
|
| 1723 |
+
},
|
| 1724 |
+
{
|
| 1725 |
+
"epoch": 1.1228070175438596,
|
| 1726 |
+
"grad_norm": 0.346923828125,
|
| 1727 |
+
"learning_rate": 1.0050251256281407e-06,
|
| 1728 |
+
"loss": 0.0624,
|
| 1729 |
+
"step": 4800
|
| 1730 |
+
},
|
| 1731 |
+
{
|
| 1732 |
+
"epoch": 1.1228070175438596,
|
| 1733 |
+
"eval_loss": 0.1874168962240219,
|
| 1734 |
+
"eval_runtime": 128.2151,
|
| 1735 |
+
"eval_samples_per_second": 3.51,
|
| 1736 |
+
"eval_steps_per_second": 0.445,
|
| 1737 |
+
"step": 4800
|
| 1738 |
+
},
|
| 1739 |
+
{
|
| 1740 |
+
"epoch": 1.128654970760234,
|
| 1741 |
+
"grad_norm": 0.1552734375,
|
| 1742 |
+
"learning_rate": 8.793969849246231e-07,
|
| 1743 |
+
"loss": 0.1017,
|
| 1744 |
+
"step": 4825
|
| 1745 |
+
},
|
| 1746 |
+
{
|
| 1747 |
+
"epoch": 1.1345029239766082,
|
| 1748 |
+
"grad_norm": 1.619140625,
|
| 1749 |
+
"learning_rate": 7.537688442211055e-07,
|
| 1750 |
+
"loss": 0.1556,
|
| 1751 |
+
"step": 4850
|
| 1752 |
+
},
|
| 1753 |
+
{
|
| 1754 |
+
"epoch": 1.1403508771929824,
|
| 1755 |
+
"grad_norm": 1.958984375,
|
| 1756 |
+
"learning_rate": 6.28140703517588e-07,
|
| 1757 |
+
"loss": 0.1113,
|
| 1758 |
+
"step": 4875
|
| 1759 |
+
},
|
| 1760 |
+
{
|
| 1761 |
+
"epoch": 1.1461988304093567,
|
| 1762 |
+
"grad_norm": 5.375,
|
| 1763 |
+
"learning_rate": 5.025125628140703e-07,
|
| 1764 |
+
"loss": 0.0895,
|
| 1765 |
+
"step": 4900
|
| 1766 |
+
},
|
| 1767 |
+
{
|
| 1768 |
+
"epoch": 1.1461988304093567,
|
| 1769 |
+
"eval_loss": 0.1870710253715515,
|
| 1770 |
+
"eval_runtime": 126.862,
|
| 1771 |
+
"eval_samples_per_second": 3.547,
|
| 1772 |
+
"eval_steps_per_second": 0.449,
|
| 1773 |
+
"step": 4900
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 1.1520467836257309,
|
| 1777 |
+
"grad_norm": 5.984375,
|
| 1778 |
+
"learning_rate": 3.7688442211055275e-07,
|
| 1779 |
+
"loss": 0.1076,
|
| 1780 |
+
"step": 4925
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 1.1578947368421053,
|
| 1784 |
+
"grad_norm": 1.9619140625,
|
| 1785 |
+
"learning_rate": 2.5125628140703517e-07,
|
| 1786 |
+
"loss": 0.1024,
|
| 1787 |
+
"step": 4950
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 1.1637426900584795,
|
| 1791 |
+
"grad_norm": 6.17578125,
|
| 1792 |
+
"learning_rate": 1.2562814070351758e-07,
|
| 1793 |
+
"loss": 0.0698,
|
| 1794 |
+
"step": 4975
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 1.1695906432748537,
|
| 1798 |
+
"grad_norm": 7.06640625,
|
| 1799 |
+
"learning_rate": 0.0,
|
| 1800 |
+
"loss": 0.1,
|
| 1801 |
+
"step": 5000
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 1.1695906432748537,
|
| 1805 |
+
"eval_loss": 0.1871223896741867,
|
| 1806 |
+
"eval_runtime": 128.9773,
|
| 1807 |
+
"eval_samples_per_second": 3.489,
|
| 1808 |
+
"eval_steps_per_second": 0.442,
|
| 1809 |
+
"step": 5000
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 1.1695906432748537,
|
| 1813 |
+
"step": 5000,
|
| 1814 |
+
"total_flos": 2.536648947447456e+17,
|
| 1815 |
+
"train_loss": 0.27304353165626527,
|
| 1816 |
+
"train_runtime": 21409.3382,
|
| 1817 |
+
"train_samples_per_second": 0.934,
|
| 1818 |
+
"train_steps_per_second": 0.234
|
| 1819 |
+
}
|
| 1820 |
+
],
|
| 1821 |
+
"logging_steps": 25,
|
| 1822 |
+
"max_steps": 5000,
|
| 1823 |
+
"num_input_tokens_seen": 0,
|
| 1824 |
+
"num_train_epochs": 2,
|
| 1825 |
+
"save_steps": 100,
|
| 1826 |
+
"stateful_callbacks": {
|
| 1827 |
+
"TrainerControl": {
|
| 1828 |
+
"args": {
|
| 1829 |
+
"should_epoch_stop": false,
|
| 1830 |
+
"should_evaluate": false,
|
| 1831 |
+
"should_log": false,
|
| 1832 |
+
"should_save": true,
|
| 1833 |
+
"should_training_stop": true
|
| 1834 |
+
},
|
| 1835 |
+
"attributes": {}
|
| 1836 |
+
}
|
| 1837 |
+
},
|
| 1838 |
+
"total_flos": 2.536648947447456e+17,
|
| 1839 |
+
"train_batch_size": 4,
|
| 1840 |
+
"trial_name": null,
|
| 1841 |
+
"trial_params": null
|
| 1842 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a3660282e9bbcdcfad19c2bf547dad5d55220b8a1e969aab83a17e938d21e214
|
| 3 |
+
size 5176
|