Model save
Browse files- README.md +111 -0
- adapter_config.json +25 -0
- adapter_model.safetensors +3 -0
- all_results.json +13 -0
- eval_results.json +8 -0
- special_tokens_map.json +24 -0
- tokenizer.json +0 -0
- tokenizer_config.json +41 -0
- train_results.json +8 -0
- trainer_state.json +1244 -0
- training_args.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,111 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: mistralai/Mistral-7B-v0.1
|
| 4 |
+
tags:
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
model-index:
|
| 7 |
+
- name: zephyr-7b-sft-lora-accum4-lr5e_5
|
| 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 |
+
# zephyr-7b-sft-lora-accum4-lr5e_5
|
| 15 |
+
|
| 16 |
+
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on an unknown dataset.
|
| 17 |
+
It achieves the following results on the evaluation set:
|
| 18 |
+
- Loss: 0.5833
|
| 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: 5e-05
|
| 38 |
+
- train_batch_size: 4
|
| 39 |
+
- eval_batch_size: 8
|
| 40 |
+
- seed: 42
|
| 41 |
+
- distributed_type: multi-GPU
|
| 42 |
+
- num_devices: 2
|
| 43 |
+
- gradient_accumulation_steps: 4
|
| 44 |
+
- total_train_batch_size: 32
|
| 45 |
+
- total_eval_batch_size: 16
|
| 46 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 47 |
+
- lr_scheduler_type: cosine
|
| 48 |
+
- num_epochs: 50.0
|
| 49 |
+
|
| 50 |
+
### Training results
|
| 51 |
+
|
| 52 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 53 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 54 |
+
| 1.8243 | 0.55 | 13 | 1.6684 |
|
| 55 |
+
| 1.5291 | 1.57 | 27 | 1.4003 |
|
| 56 |
+
| 1.2355 | 2.55 | 40 | 1.1808 |
|
| 57 |
+
| 1.1393 | 3.57 | 54 | 1.0949 |
|
| 58 |
+
| 1.0659 | 4.55 | 67 | 1.0457 |
|
| 59 |
+
| 1.0196 | 5.57 | 81 | 1.0065 |
|
| 60 |
+
| 0.9831 | 6.55 | 94 | 0.9686 |
|
| 61 |
+
| 0.9281 | 7.57 | 108 | 0.9255 |
|
| 62 |
+
| 0.8678 | 8.55 | 121 | 0.8814 |
|
| 63 |
+
| 0.8054 | 9.57 | 135 | 0.8275 |
|
| 64 |
+
| 0.7683 | 10.55 | 148 | 0.7861 |
|
| 65 |
+
| 0.6906 | 11.57 | 162 | 0.7272 |
|
| 66 |
+
| 0.6246 | 12.55 | 175 | 0.6795 |
|
| 67 |
+
| 0.5813 | 13.57 | 189 | 0.6364 |
|
| 68 |
+
| 0.5253 | 14.55 | 202 | 0.6078 |
|
| 69 |
+
| 0.5149 | 15.57 | 216 | 0.5811 |
|
| 70 |
+
| 0.4949 | 16.55 | 229 | 0.5605 |
|
| 71 |
+
| 0.4644 | 17.57 | 243 | 0.5462 |
|
| 72 |
+
| 0.458 | 18.55 | 256 | 0.5346 |
|
| 73 |
+
| 0.4294 | 19.57 | 270 | 0.5202 |
|
| 74 |
+
| 0.4143 | 20.55 | 283 | 0.5177 |
|
| 75 |
+
| 0.4161 | 21.57 | 297 | 0.5108 |
|
| 76 |
+
| 0.4128 | 22.55 | 310 | 0.5057 |
|
| 77 |
+
| 0.4055 | 23.57 | 324 | 0.5071 |
|
| 78 |
+
| 0.3937 | 24.55 | 337 | 0.5058 |
|
| 79 |
+
| 0.3967 | 25.57 | 351 | 0.5017 |
|
| 80 |
+
| 0.3754 | 26.55 | 364 | 0.4998 |
|
| 81 |
+
| 0.3742 | 27.57 | 378 | 0.5019 |
|
| 82 |
+
| 0.3756 | 28.55 | 391 | 0.5019 |
|
| 83 |
+
| 0.3652 | 29.57 | 405 | 0.5061 |
|
| 84 |
+
| 0.3597 | 30.55 | 418 | 0.5076 |
|
| 85 |
+
| 0.3609 | 31.57 | 432 | 0.5079 |
|
| 86 |
+
| 0.3581 | 32.55 | 445 | 0.5108 |
|
| 87 |
+
| 0.3426 | 33.57 | 459 | 0.5117 |
|
| 88 |
+
| 0.3481 | 34.55 | 472 | 0.5141 |
|
| 89 |
+
| 0.3435 | 35.57 | 486 | 0.5150 |
|
| 90 |
+
| 0.3317 | 36.55 | 499 | 0.5245 |
|
| 91 |
+
| 0.3387 | 37.57 | 513 | 0.5239 |
|
| 92 |
+
| 0.332 | 38.55 | 526 | 0.5319 |
|
| 93 |
+
| 0.3334 | 39.57 | 540 | 0.5342 |
|
| 94 |
+
| 0.323 | 40.55 | 553 | 0.5388 |
|
| 95 |
+
| 0.3144 | 41.57 | 567 | 0.5423 |
|
| 96 |
+
| 0.3092 | 42.55 | 580 | 0.5465 |
|
| 97 |
+
| 0.3084 | 43.57 | 594 | 0.5481 |
|
| 98 |
+
| 0.3091 | 44.55 | 607 | 0.5605 |
|
| 99 |
+
| 0.3044 | 45.57 | 621 | 0.5606 |
|
| 100 |
+
| 0.303 | 46.55 | 634 | 0.5683 |
|
| 101 |
+
| 0.2896 | 47.57 | 648 | 0.5722 |
|
| 102 |
+
| 0.2854 | 48.55 | 661 | 0.5778 |
|
| 103 |
+
| 0.291 | 49.57 | 675 | 0.5826 |
|
| 104 |
+
|
| 105 |
+
|
| 106 |
+
### Framework versions
|
| 107 |
+
|
| 108 |
+
- Transformers 4.35.0
|
| 109 |
+
- Pytorch 2.1.0
|
| 110 |
+
- Datasets 2.14.6
|
| 111 |
+
- Tokenizers 0.14.1
|
adapter_config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "mistralai/Mistral-7B-v0.1",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": false,
|
| 7 |
+
"inference_mode": true,
|
| 8 |
+
"init_lora_weights": true,
|
| 9 |
+
"layers_pattern": null,
|
| 10 |
+
"layers_to_transform": null,
|
| 11 |
+
"lora_alpha": 16,
|
| 12 |
+
"lora_dropout": 0.1,
|
| 13 |
+
"modules_to_save": null,
|
| 14 |
+
"peft_type": "LORA",
|
| 15 |
+
"r": 64,
|
| 16 |
+
"rank_pattern": {},
|
| 17 |
+
"revision": null,
|
| 18 |
+
"target_modules": [
|
| 19 |
+
"v_proj",
|
| 20 |
+
"q_proj",
|
| 21 |
+
"o_proj",
|
| 22 |
+
"k_proj"
|
| 23 |
+
],
|
| 24 |
+
"task_type": "CAUSAL_LM"
|
| 25 |
+
}
|
adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8784603b903f5c841a3c80e8cf681b621d2c9588345ca26fb67220bdaf9e3b84
|
| 3 |
+
size 109086672
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 49.57,
|
| 3 |
+
"eval_loss": 0.5833138823509216,
|
| 4 |
+
"eval_runtime": 3.9733,
|
| 5 |
+
"eval_samples": 83,
|
| 6 |
+
"eval_samples_per_second": 20.889,
|
| 7 |
+
"eval_steps_per_second": 1.51,
|
| 8 |
+
"train_loss": 0.5519325028525458,
|
| 9 |
+
"train_runtime": 6501.4714,
|
| 10 |
+
"train_samples": 746,
|
| 11 |
+
"train_samples_per_second": 5.737,
|
| 12 |
+
"train_steps_per_second": 0.177
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 49.57,
|
| 3 |
+
"eval_loss": 0.5833138823509216,
|
| 4 |
+
"eval_runtime": 3.9733,
|
| 5 |
+
"eval_samples": 83,
|
| 6 |
+
"eval_samples_per_second": 20.889,
|
| 7 |
+
"eval_steps_per_second": 1.51
|
| 8 |
+
}
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "</s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": "</s>",
|
| 17 |
+
"unk_token": {
|
| 18 |
+
"content": "<unk>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": false,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
}
|
| 24 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,41 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "<unk>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "<s>",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "</s>",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
}
|
| 27 |
+
},
|
| 28 |
+
"additional_special_tokens": [],
|
| 29 |
+
"bos_token": "<s>",
|
| 30 |
+
"chat_template": "{% for message in messages %}\n{% if message['role'] == 'user' %}\n{{ '<|user|>\n' + message['content'] + eos_token }}\n{% elif message['role'] == 'system' %}\n{{ '<|system|>\n' + message['content'] + eos_token }}\n{% elif message['role'] == 'assistant' %}\n{{ '<|assistant|>\n' + message['content'] + eos_token }}\n{% endif %}\n{% if loop.last and add_generation_prompt %}\n{{ '<|assistant|>' }}\n{% endif %}\n{% endfor %}",
|
| 31 |
+
"clean_up_tokenization_spaces": false,
|
| 32 |
+
"eos_token": "</s>",
|
| 33 |
+
"legacy": true,
|
| 34 |
+
"model_max_length": 2048,
|
| 35 |
+
"pad_token": "</s>",
|
| 36 |
+
"sp_model_kwargs": {},
|
| 37 |
+
"spaces_between_special_tokens": false,
|
| 38 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 39 |
+
"unk_token": "<unk>",
|
| 40 |
+
"use_default_system_prompt": false
|
| 41 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 49.57,
|
| 3 |
+
"train_loss": 0.5519325028525458,
|
| 4 |
+
"train_runtime": 6501.4714,
|
| 5 |
+
"train_samples": 746,
|
| 6 |
+
"train_samples_per_second": 5.737,
|
| 7 |
+
"train_steps_per_second": 0.177
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1244 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 49.57446808510638,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 675,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.04,
|
| 13 |
+
"learning_rate": 4.9999906714572185e-05,
|
| 14 |
+
"loss": 2.0607,
|
| 15 |
+
"step": 1
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.21,
|
| 19 |
+
"learning_rate": 4.999766789911305e-05,
|
| 20 |
+
"loss": 2.0068,
|
| 21 |
+
"step": 5
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.43,
|
| 25 |
+
"learning_rate": 4.999067203154777e-05,
|
| 26 |
+
"loss": 1.8243,
|
| 27 |
+
"step": 10
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.55,
|
| 31 |
+
"eval_loss": 1.6684231758117676,
|
| 32 |
+
"eval_runtime": 4.0605,
|
| 33 |
+
"eval_samples_per_second": 20.441,
|
| 34 |
+
"eval_steps_per_second": 1.478,
|
| 35 |
+
"step": 13
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 1.06,
|
| 39 |
+
"learning_rate": 4.9979013702509664e-05,
|
| 40 |
+
"loss": 1.7052,
|
| 41 |
+
"step": 15
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 1.28,
|
| 45 |
+
"learning_rate": 4.99626950870707e-05,
|
| 46 |
+
"loss": 1.6022,
|
| 47 |
+
"step": 20
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 1.49,
|
| 51 |
+
"learning_rate": 4.994171922976348e-05,
|
| 52 |
+
"loss": 1.5291,
|
| 53 |
+
"step": 25
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.57,
|
| 57 |
+
"eval_loss": 1.400346040725708,
|
| 58 |
+
"eval_runtime": 3.9709,
|
| 59 |
+
"eval_samples_per_second": 20.902,
|
| 60 |
+
"eval_steps_per_second": 1.511,
|
| 61 |
+
"step": 27
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 2.13,
|
| 65 |
+
"learning_rate": 4.991609004401324e-05,
|
| 66 |
+
"loss": 1.4168,
|
| 67 |
+
"step": 30
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 2.34,
|
| 71 |
+
"learning_rate": 4.988581231140772e-05,
|
| 72 |
+
"loss": 1.312,
|
| 73 |
+
"step": 35
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 2.55,
|
| 77 |
+
"learning_rate": 4.985089168080509e-05,
|
| 78 |
+
"loss": 1.2355,
|
| 79 |
+
"step": 40
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 2.55,
|
| 83 |
+
"eval_loss": 1.1807862520217896,
|
| 84 |
+
"eval_runtime": 3.9823,
|
| 85 |
+
"eval_samples_per_second": 20.842,
|
| 86 |
+
"eval_steps_per_second": 1.507,
|
| 87 |
+
"step": 40
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 3.19,
|
| 91 |
+
"learning_rate": 4.981133466728004e-05,
|
| 92 |
+
"loss": 1.1754,
|
| 93 |
+
"step": 45
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 3.4,
|
| 97 |
+
"learning_rate": 4.976714865090827e-05,
|
| 98 |
+
"loss": 1.1393,
|
| 99 |
+
"step": 50
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 3.57,
|
| 103 |
+
"eval_loss": 1.0949289798736572,
|
| 104 |
+
"eval_runtime": 3.9683,
|
| 105 |
+
"eval_samples_per_second": 20.916,
|
| 106 |
+
"eval_steps_per_second": 1.512,
|
| 107 |
+
"step": 54
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 4.04,
|
| 111 |
+
"learning_rate": 4.9718341875389625e-05,
|
| 112 |
+
"loss": 1.123,
|
| 113 |
+
"step": 55
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 4.26,
|
| 117 |
+
"learning_rate": 4.966492344651005e-05,
|
| 118 |
+
"loss": 1.0919,
|
| 119 |
+
"step": 60
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 4.47,
|
| 123 |
+
"learning_rate": 4.960690333044279e-05,
|
| 124 |
+
"loss": 1.0659,
|
| 125 |
+
"step": 65
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 4.55,
|
| 129 |
+
"eval_loss": 1.045737624168396,
|
| 130 |
+
"eval_runtime": 3.9807,
|
| 131 |
+
"eval_samples_per_second": 20.85,
|
| 132 |
+
"eval_steps_per_second": 1.507,
|
| 133 |
+
"step": 67
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 5.11,
|
| 137 |
+
"learning_rate": 4.9544292351888966e-05,
|
| 138 |
+
"loss": 1.0601,
|
| 139 |
+
"step": 70
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 5.32,
|
| 143 |
+
"learning_rate": 4.947710219205808e-05,
|
| 144 |
+
"loss": 1.0387,
|
| 145 |
+
"step": 75
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 5.53,
|
| 149 |
+
"learning_rate": 4.9405345386488614e-05,
|
| 150 |
+
"loss": 1.0196,
|
| 151 |
+
"step": 80
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 5.57,
|
| 155 |
+
"eval_loss": 1.0065257549285889,
|
| 156 |
+
"eval_runtime": 3.9679,
|
| 157 |
+
"eval_samples_per_second": 20.918,
|
| 158 |
+
"eval_steps_per_second": 1.512,
|
| 159 |
+
"step": 81
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 6.17,
|
| 163 |
+
"learning_rate": 4.9329035322709386e-05,
|
| 164 |
+
"loss": 0.9975,
|
| 165 |
+
"step": 85
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 6.38,
|
| 169 |
+
"learning_rate": 4.924818623774178e-05,
|
| 170 |
+
"loss": 0.9831,
|
| 171 |
+
"step": 90
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 6.55,
|
| 175 |
+
"eval_loss": 0.9685614109039307,
|
| 176 |
+
"eval_runtime": 3.9721,
|
| 177 |
+
"eval_samples_per_second": 20.896,
|
| 178 |
+
"eval_steps_per_second": 1.511,
|
| 179 |
+
"step": 94
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 7.02,
|
| 183 |
+
"learning_rate": 4.916281321544362e-05,
|
| 184 |
+
"loss": 0.9791,
|
| 185 |
+
"step": 95
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 7.23,
|
| 189 |
+
"learning_rate": 4.907293218369499e-05,
|
| 190 |
+
"loss": 0.9494,
|
| 191 |
+
"step": 100
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 7.45,
|
| 195 |
+
"learning_rate": 4.897855991142658e-05,
|
| 196 |
+
"loss": 0.9281,
|
| 197 |
+
"step": 105
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 7.57,
|
| 201 |
+
"eval_loss": 0.9254654049873352,
|
| 202 |
+
"eval_runtime": 3.9691,
|
| 203 |
+
"eval_samples_per_second": 20.911,
|
| 204 |
+
"eval_steps_per_second": 1.512,
|
| 205 |
+
"step": 108
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 8.09,
|
| 209 |
+
"learning_rate": 4.88797140054912e-05,
|
| 210 |
+
"loss": 0.9186,
|
| 211 |
+
"step": 110
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 8.3,
|
| 215 |
+
"learning_rate": 4.877641290737884e-05,
|
| 216 |
+
"loss": 0.8878,
|
| 217 |
+
"step": 115
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 8.51,
|
| 221 |
+
"learning_rate": 4.8668675889776095e-05,
|
| 222 |
+
"loss": 0.8678,
|
| 223 |
+
"step": 120
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 8.55,
|
| 227 |
+
"eval_loss": 0.8814197182655334,
|
| 228 |
+
"eval_runtime": 3.978,
|
| 229 |
+
"eval_samples_per_second": 20.865,
|
| 230 |
+
"eval_steps_per_second": 1.508,
|
| 231 |
+
"step": 121
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 9.15,
|
| 235 |
+
"learning_rate": 4.855652305297052e-05,
|
| 236 |
+
"loss": 0.8535,
|
| 237 |
+
"step": 125
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 9.36,
|
| 241 |
+
"learning_rate": 4.843997532110051e-05,
|
| 242 |
+
"loss": 0.816,
|
| 243 |
+
"step": 130
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 9.57,
|
| 247 |
+
"learning_rate": 4.831905443825159e-05,
|
| 248 |
+
"loss": 0.8054,
|
| 249 |
+
"step": 135
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 9.57,
|
| 253 |
+
"eval_loss": 0.8274821639060974,
|
| 254 |
+
"eval_runtime": 3.9679,
|
| 255 |
+
"eval_samples_per_second": 20.918,
|
| 256 |
+
"eval_steps_per_second": 1.512,
|
| 257 |
+
"step": 135
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 10.21,
|
| 261 |
+
"learning_rate": 4.819378296439961e-05,
|
| 262 |
+
"loss": 0.7717,
|
| 263 |
+
"step": 140
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 10.43,
|
| 267 |
+
"learning_rate": 4.806418427120179e-05,
|
| 268 |
+
"loss": 0.7683,
|
| 269 |
+
"step": 145
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 10.55,
|
| 273 |
+
"eval_loss": 0.7860919237136841,
|
| 274 |
+
"eval_runtime": 3.9806,
|
| 275 |
+
"eval_samples_per_second": 20.851,
|
| 276 |
+
"eval_steps_per_second": 1.507,
|
| 277 |
+
"step": 148
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 11.06,
|
| 281 |
+
"learning_rate": 4.793028253763633e-05,
|
| 282 |
+
"loss": 0.741,
|
| 283 |
+
"step": 150
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 11.28,
|
| 287 |
+
"learning_rate": 4.779210274549134e-05,
|
| 288 |
+
"loss": 0.7057,
|
| 289 |
+
"step": 155
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 11.49,
|
| 293 |
+
"learning_rate": 4.76496706747041e-05,
|
| 294 |
+
"loss": 0.6906,
|
| 295 |
+
"step": 160
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 11.57,
|
| 299 |
+
"eval_loss": 0.7272327542304993,
|
| 300 |
+
"eval_runtime": 3.9685,
|
| 301 |
+
"eval_samples_per_second": 20.915,
|
| 302 |
+
"eval_steps_per_second": 1.512,
|
| 303 |
+
"step": 162
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 12.13,
|
| 307 |
+
"learning_rate": 4.750301289855128e-05,
|
| 308 |
+
"loss": 0.673,
|
| 309 |
+
"step": 165
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 12.34,
|
| 313 |
+
"learning_rate": 4.735215677869128e-05,
|
| 314 |
+
"loss": 0.6411,
|
| 315 |
+
"step": 170
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 12.55,
|
| 319 |
+
"learning_rate": 4.719713046005938e-05,
|
| 320 |
+
"loss": 0.6246,
|
| 321 |
+
"step": 175
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 12.55,
|
| 325 |
+
"eval_loss": 0.6794944405555725,
|
| 326 |
+
"eval_runtime": 3.9823,
|
| 327 |
+
"eval_samples_per_second": 20.842,
|
| 328 |
+
"eval_steps_per_second": 1.507,
|
| 329 |
+
"step": 175
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 13.19,
|
| 333 |
+
"learning_rate": 4.703796286561679e-05,
|
| 334 |
+
"loss": 0.6086,
|
| 335 |
+
"step": 180
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 13.4,
|
| 339 |
+
"learning_rate": 4.687468369095457e-05,
|
| 340 |
+
"loss": 0.5813,
|
| 341 |
+
"step": 185
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 13.57,
|
| 345 |
+
"eval_loss": 0.6364239454269409,
|
| 346 |
+
"eval_runtime": 3.9657,
|
| 347 |
+
"eval_samples_per_second": 20.93,
|
| 348 |
+
"eval_steps_per_second": 1.513,
|
| 349 |
+
"step": 189
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 14.04,
|
| 353 |
+
"learning_rate": 4.6707323398753346e-05,
|
| 354 |
+
"loss": 0.5586,
|
| 355 |
+
"step": 190
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 14.26,
|
| 359 |
+
"learning_rate": 4.65359132131e-05,
|
| 360 |
+
"loss": 0.5556,
|
| 361 |
+
"step": 195
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 14.47,
|
| 365 |
+
"learning_rate": 4.6360485113662216e-05,
|
| 366 |
+
"loss": 0.5253,
|
| 367 |
+
"step": 200
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 14.55,
|
| 371 |
+
"eval_loss": 0.6078140139579773,
|
| 372 |
+
"eval_runtime": 3.9649,
|
| 373 |
+
"eval_samples_per_second": 20.934,
|
| 374 |
+
"eval_steps_per_second": 1.513,
|
| 375 |
+
"step": 202
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 15.11,
|
| 379 |
+
"learning_rate": 4.618107182972209e-05,
|
| 380 |
+
"loss": 0.5329,
|
| 381 |
+
"step": 205
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 15.32,
|
| 385 |
+
"learning_rate": 4.599770683406991e-05,
|
| 386 |
+
"loss": 0.4948,
|
| 387 |
+
"step": 210
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 15.53,
|
| 391 |
+
"learning_rate": 4.581042433675921e-05,
|
| 392 |
+
"loss": 0.5149,
|
| 393 |
+
"step": 215
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 15.57,
|
| 397 |
+
"eval_loss": 0.5811336636543274,
|
| 398 |
+
"eval_runtime": 3.9681,
|
| 399 |
+
"eval_samples_per_second": 20.917,
|
| 400 |
+
"eval_steps_per_second": 1.512,
|
| 401 |
+
"step": 216
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 16.17,
|
| 405 |
+
"learning_rate": 4.5619259278724214e-05,
|
| 406 |
+
"loss": 0.4857,
|
| 407 |
+
"step": 220
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 16.38,
|
| 411 |
+
"learning_rate": 4.542424732526105e-05,
|
| 412 |
+
"loss": 0.4949,
|
| 413 |
+
"step": 225
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 16.55,
|
| 417 |
+
"eval_loss": 0.5605343580245972,
|
| 418 |
+
"eval_runtime": 3.9813,
|
| 419 |
+
"eval_samples_per_second": 20.847,
|
| 420 |
+
"eval_steps_per_second": 1.507,
|
| 421 |
+
"step": 229
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 17.02,
|
| 425 |
+
"learning_rate": 4.522542485937369e-05,
|
| 426 |
+
"loss": 0.4737,
|
| 427 |
+
"step": 230
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 17.23,
|
| 431 |
+
"learning_rate": 4.5022828974986044e-05,
|
| 432 |
+
"loss": 0.4675,
|
| 433 |
+
"step": 235
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 17.45,
|
| 437 |
+
"learning_rate": 4.4816497470021454e-05,
|
| 438 |
+
"loss": 0.4644,
|
| 439 |
+
"step": 240
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 17.57,
|
| 443 |
+
"eval_loss": 0.5462371110916138,
|
| 444 |
+
"eval_runtime": 3.9643,
|
| 445 |
+
"eval_samples_per_second": 20.937,
|
| 446 |
+
"eval_steps_per_second": 1.514,
|
| 447 |
+
"step": 243
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 18.09,
|
| 451 |
+
"learning_rate": 4.4606468839350785e-05,
|
| 452 |
+
"loss": 0.4565,
|
| 453 |
+
"step": 245
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 18.3,
|
| 457 |
+
"learning_rate": 4.43927822676105e-05,
|
| 458 |
+
"loss": 0.4365,
|
| 459 |
+
"step": 250
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 18.51,
|
| 463 |
+
"learning_rate": 4.417547762189207e-05,
|
| 464 |
+
"loss": 0.458,
|
| 465 |
+
"step": 255
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 18.55,
|
| 469 |
+
"eval_loss": 0.5346133708953857,
|
| 470 |
+
"eval_runtime": 3.9682,
|
| 471 |
+
"eval_samples_per_second": 20.916,
|
| 472 |
+
"eval_steps_per_second": 1.512,
|
| 473 |
+
"step": 256
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 19.15,
|
| 477 |
+
"learning_rate": 4.395459544430407e-05,
|
| 478 |
+
"loss": 0.449,
|
| 479 |
+
"step": 260
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 19.36,
|
| 483 |
+
"learning_rate": 4.373017694440827e-05,
|
| 484 |
+
"loss": 0.4401,
|
| 485 |
+
"step": 265
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 19.57,
|
| 489 |
+
"learning_rate": 4.35022639915313e-05,
|
| 490 |
+
"loss": 0.4294,
|
| 491 |
+
"step": 270
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 19.57,
|
| 495 |
+
"eval_loss": 0.5201942920684814,
|
| 496 |
+
"eval_runtime": 3.9685,
|
| 497 |
+
"eval_samples_per_second": 20.915,
|
| 498 |
+
"eval_steps_per_second": 1.512,
|
| 499 |
+
"step": 270
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 20.21,
|
| 503 |
+
"learning_rate": 4.3270899106953105e-05,
|
| 504 |
+
"loss": 0.4367,
|
| 505 |
+
"step": 275
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 20.43,
|
| 509 |
+
"learning_rate": 4.3036125455973896e-05,
|
| 510 |
+
"loss": 0.4143,
|
| 511 |
+
"step": 280
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 20.55,
|
| 515 |
+
"eval_loss": 0.5177348852157593,
|
| 516 |
+
"eval_runtime": 3.9827,
|
| 517 |
+
"eval_samples_per_second": 20.84,
|
| 518 |
+
"eval_steps_per_second": 1.507,
|
| 519 |
+
"step": 283
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 21.06,
|
| 523 |
+
"learning_rate": 4.279798683986084e-05,
|
| 524 |
+
"loss": 0.4244,
|
| 525 |
+
"step": 285
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 21.28,
|
| 529 |
+
"learning_rate": 4.2556527687676186e-05,
|
| 530 |
+
"loss": 0.4096,
|
| 531 |
+
"step": 290
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 21.49,
|
| 535 |
+
"learning_rate": 4.231179304798815e-05,
|
| 536 |
+
"loss": 0.4161,
|
| 537 |
+
"step": 295
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 21.57,
|
| 541 |
+
"eval_loss": 0.5107729434967041,
|
| 542 |
+
"eval_runtime": 3.9674,
|
| 543 |
+
"eval_samples_per_second": 20.92,
|
| 544 |
+
"eval_steps_per_second": 1.512,
|
| 545 |
+
"step": 297
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 22.13,
|
| 549 |
+
"learning_rate": 4.206382858046636e-05,
|
| 550 |
+
"loss": 0.4075,
|
| 551 |
+
"step": 300
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 22.34,
|
| 555 |
+
"learning_rate": 4.181268054736318e-05,
|
| 556 |
+
"loss": 0.4002,
|
| 557 |
+
"step": 305
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 22.55,
|
| 561 |
+
"learning_rate": 4.1558395804882695e-05,
|
| 562 |
+
"loss": 0.4128,
|
| 563 |
+
"step": 310
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 22.55,
|
| 567 |
+
"eval_loss": 0.5056843757629395,
|
| 568 |
+
"eval_runtime": 3.9783,
|
| 569 |
+
"eval_samples_per_second": 20.863,
|
| 570 |
+
"eval_steps_per_second": 1.508,
|
| 571 |
+
"step": 310
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 23.19,
|
| 575 |
+
"learning_rate": 4.130102179443877e-05,
|
| 576 |
+
"loss": 0.3928,
|
| 577 |
+
"step": 315
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 23.4,
|
| 581 |
+
"learning_rate": 4.1040606533804024e-05,
|
| 582 |
+
"loss": 0.4055,
|
| 583 |
+
"step": 320
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 23.57,
|
| 587 |
+
"eval_loss": 0.5070647597312927,
|
| 588 |
+
"eval_runtime": 3.9713,
|
| 589 |
+
"eval_samples_per_second": 20.9,
|
| 590 |
+
"eval_steps_per_second": 1.511,
|
| 591 |
+
"step": 324
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 24.04,
|
| 595 |
+
"learning_rate": 4.077719860815132e-05,
|
| 596 |
+
"loss": 0.4021,
|
| 597 |
+
"step": 325
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 24.26,
|
| 601 |
+
"learning_rate": 4.051084716098921e-05,
|
| 602 |
+
"loss": 0.3999,
|
| 603 |
+
"step": 330
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 24.47,
|
| 607 |
+
"learning_rate": 4.0241601884993366e-05,
|
| 608 |
+
"loss": 0.3937,
|
| 609 |
+
"step": 335
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 24.55,
|
| 613 |
+
"eval_loss": 0.505768895149231,
|
| 614 |
+
"eval_runtime": 3.9783,
|
| 615 |
+
"eval_samples_per_second": 20.863,
|
| 616 |
+
"eval_steps_per_second": 1.508,
|
| 617 |
+
"step": 337
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 25.11,
|
| 621 |
+
"learning_rate": 3.996951301273557e-05,
|
| 622 |
+
"loss": 0.3927,
|
| 623 |
+
"step": 340
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 25.32,
|
| 627 |
+
"learning_rate": 3.969463130731183e-05,
|
| 628 |
+
"loss": 0.377,
|
| 629 |
+
"step": 345
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 25.53,
|
| 633 |
+
"learning_rate": 3.941700805287168e-05,
|
| 634 |
+
"loss": 0.3967,
|
| 635 |
+
"step": 350
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 25.57,
|
| 639 |
+
"eval_loss": 0.5016723871231079,
|
| 640 |
+
"eval_runtime": 3.9654,
|
| 641 |
+
"eval_samples_per_second": 20.931,
|
| 642 |
+
"eval_steps_per_second": 1.513,
|
| 643 |
+
"step": 351
|
| 644 |
+
},
|
| 645 |
+
{
|
| 646 |
+
"epoch": 26.17,
|
| 647 |
+
"learning_rate": 3.913669504505015e-05,
|
| 648 |
+
"loss": 0.3859,
|
| 649 |
+
"step": 355
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 26.38,
|
| 653 |
+
"learning_rate": 3.885374458130438e-05,
|
| 654 |
+
"loss": 0.3754,
|
| 655 |
+
"step": 360
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 26.55,
|
| 659 |
+
"eval_loss": 0.4997941851615906,
|
| 660 |
+
"eval_runtime": 3.9766,
|
| 661 |
+
"eval_samples_per_second": 20.872,
|
| 662 |
+
"eval_steps_per_second": 1.509,
|
| 663 |
+
"step": 364
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 27.02,
|
| 667 |
+
"learning_rate": 3.856820945115655e-05,
|
| 668 |
+
"loss": 0.3853,
|
| 669 |
+
"step": 365
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 27.23,
|
| 673 |
+
"learning_rate": 3.828014292634509e-05,
|
| 674 |
+
"loss": 0.3852,
|
| 675 |
+
"step": 370
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 27.45,
|
| 679 |
+
"learning_rate": 3.798959875088584e-05,
|
| 680 |
+
"loss": 0.3742,
|
| 681 |
+
"step": 375
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 27.57,
|
| 685 |
+
"eval_loss": 0.5018514394760132,
|
| 686 |
+
"eval_runtime": 3.9719,
|
| 687 |
+
"eval_samples_per_second": 20.897,
|
| 688 |
+
"eval_steps_per_second": 1.511,
|
| 689 |
+
"step": 378
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 28.09,
|
| 693 |
+
"learning_rate": 3.769663113104516e-05,
|
| 694 |
+
"loss": 0.3706,
|
| 695 |
+
"step": 380
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 28.3,
|
| 699 |
+
"learning_rate": 3.74012947252267e-05,
|
| 700 |
+
"loss": 0.3759,
|
| 701 |
+
"step": 385
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 28.51,
|
| 705 |
+
"learning_rate": 3.7103644633774014e-05,
|
| 706 |
+
"loss": 0.3756,
|
| 707 |
+
"step": 390
|
| 708 |
+
},
|
| 709 |
+
{
|
| 710 |
+
"epoch": 28.55,
|
| 711 |
+
"eval_loss": 0.5018544793128967,
|
| 712 |
+
"eval_runtime": 3.9835,
|
| 713 |
+
"eval_samples_per_second": 20.836,
|
| 714 |
+
"eval_steps_per_second": 1.506,
|
| 715 |
+
"step": 391
|
| 716 |
+
},
|
| 717 |
+
{
|
| 718 |
+
"epoch": 29.15,
|
| 719 |
+
"learning_rate": 3.680373638869047e-05,
|
| 720 |
+
"loss": 0.3688,
|
| 721 |
+
"step": 395
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 29.36,
|
| 725 |
+
"learning_rate": 3.6501625943278805e-05,
|
| 726 |
+
"loss": 0.3764,
|
| 727 |
+
"step": 400
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 29.57,
|
| 731 |
+
"learning_rate": 3.619736966170205e-05,
|
| 732 |
+
"loss": 0.3652,
|
| 733 |
+
"step": 405
|
| 734 |
+
},
|
| 735 |
+
{
|
| 736 |
+
"epoch": 29.57,
|
| 737 |
+
"eval_loss": 0.5060749650001526,
|
| 738 |
+
"eval_runtime": 3.9634,
|
| 739 |
+
"eval_samples_per_second": 20.942,
|
| 740 |
+
"eval_steps_per_second": 1.514,
|
| 741 |
+
"step": 405
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 30.21,
|
| 745 |
+
"learning_rate": 3.589102430846773e-05,
|
| 746 |
+
"loss": 0.3688,
|
| 747 |
+
"step": 410
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 30.43,
|
| 751 |
+
"learning_rate": 3.5582647037837445e-05,
|
| 752 |
+
"loss": 0.3597,
|
| 753 |
+
"step": 415
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 30.55,
|
| 757 |
+
"eval_loss": 0.5076042413711548,
|
| 758 |
+
"eval_runtime": 3.9772,
|
| 759 |
+
"eval_samples_per_second": 20.869,
|
| 760 |
+
"eval_steps_per_second": 1.509,
|
| 761 |
+
"step": 418
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 31.06,
|
| 765 |
+
"learning_rate": 3.527229538316371e-05,
|
| 766 |
+
"loss": 0.3635,
|
| 767 |
+
"step": 420
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 31.28,
|
| 771 |
+
"learning_rate": 3.496002724615604e-05,
|
| 772 |
+
"loss": 0.3564,
|
| 773 |
+
"step": 425
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 31.49,
|
| 777 |
+
"learning_rate": 3.464590088607839e-05,
|
| 778 |
+
"loss": 0.3609,
|
| 779 |
+
"step": 430
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 31.57,
|
| 783 |
+
"eval_loss": 0.5078853964805603,
|
| 784 |
+
"eval_runtime": 3.9577,
|
| 785 |
+
"eval_samples_per_second": 20.972,
|
| 786 |
+
"eval_steps_per_second": 1.516,
|
| 787 |
+
"step": 432
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 32.13,
|
| 791 |
+
"learning_rate": 3.4329974908879783e-05,
|
| 792 |
+
"loss": 0.3679,
|
| 793 |
+
"step": 435
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 32.34,
|
| 797 |
+
"learning_rate": 3.401230825626037e-05,
|
| 798 |
+
"loss": 0.3498,
|
| 799 |
+
"step": 440
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 32.55,
|
| 803 |
+
"learning_rate": 3.369296019467473e-05,
|
| 804 |
+
"loss": 0.3581,
|
| 805 |
+
"step": 445
|
| 806 |
+
},
|
| 807 |
+
{
|
| 808 |
+
"epoch": 32.55,
|
| 809 |
+
"eval_loss": 0.5108149647712708,
|
| 810 |
+
"eval_runtime": 3.9725,
|
| 811 |
+
"eval_samples_per_second": 20.894,
|
| 812 |
+
"eval_steps_per_second": 1.51,
|
| 813 |
+
"step": 445
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 33.19,
|
| 817 |
+
"learning_rate": 3.3371990304274656e-05,
|
| 818 |
+
"loss": 0.3623,
|
| 819 |
+
"step": 450
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 33.4,
|
| 823 |
+
"learning_rate": 3.304945846779346e-05,
|
| 824 |
+
"loss": 0.3426,
|
| 825 |
+
"step": 455
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 33.57,
|
| 829 |
+
"eval_loss": 0.511660635471344,
|
| 830 |
+
"eval_runtime": 3.9737,
|
| 831 |
+
"eval_samples_per_second": 20.887,
|
| 832 |
+
"eval_steps_per_second": 1.51,
|
| 833 |
+
"step": 459
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 34.04,
|
| 837 |
+
"learning_rate": 3.272542485937369e-05,
|
| 838 |
+
"loss": 0.3469,
|
| 839 |
+
"step": 460
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 34.26,
|
| 843 |
+
"learning_rate": 3.239994993334059e-05,
|
| 844 |
+
"loss": 0.3513,
|
| 845 |
+
"step": 465
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 34.47,
|
| 849 |
+
"learning_rate": 3.207309441292325e-05,
|
| 850 |
+
"loss": 0.3481,
|
| 851 |
+
"step": 470
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 34.55,
|
| 855 |
+
"eval_loss": 0.5140624046325684,
|
| 856 |
+
"eval_runtime": 3.9754,
|
| 857 |
+
"eval_samples_per_second": 20.878,
|
| 858 |
+
"eval_steps_per_second": 1.509,
|
| 859 |
+
"step": 472
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 35.11,
|
| 863 |
+
"learning_rate": 3.1744919278925605e-05,
|
| 864 |
+
"loss": 0.3408,
|
| 865 |
+
"step": 475
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 35.32,
|
| 869 |
+
"learning_rate": 3.1415485758349346e-05,
|
| 870 |
+
"loss": 0.3469,
|
| 871 |
+
"step": 480
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 35.53,
|
| 875 |
+
"learning_rate": 3.1084855312970896e-05,
|
| 876 |
+
"loss": 0.3435,
|
| 877 |
+
"step": 485
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 35.57,
|
| 881 |
+
"eval_loss": 0.5150405168533325,
|
| 882 |
+
"eval_runtime": 3.9656,
|
| 883 |
+
"eval_samples_per_second": 20.93,
|
| 884 |
+
"eval_steps_per_second": 1.513,
|
| 885 |
+
"step": 486
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 36.17,
|
| 889 |
+
"learning_rate": 3.075308962787466e-05,
|
| 890 |
+
"loss": 0.3419,
|
| 891 |
+
"step": 490
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 36.38,
|
| 895 |
+
"learning_rate": 3.0420250599944523e-05,
|
| 896 |
+
"loss": 0.3317,
|
| 897 |
+
"step": 495
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 36.55,
|
| 901 |
+
"eval_loss": 0.524531900882721,
|
| 902 |
+
"eval_runtime": 3.975,
|
| 903 |
+
"eval_samples_per_second": 20.881,
|
| 904 |
+
"eval_steps_per_second": 1.509,
|
| 905 |
+
"step": 499
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 37.02,
|
| 909 |
+
"learning_rate": 3.008640032631585e-05,
|
| 910 |
+
"loss": 0.342,
|
| 911 |
+
"step": 500
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 37.23,
|
| 915 |
+
"learning_rate": 2.9751601092790184e-05,
|
| 916 |
+
"loss": 0.3352,
|
| 917 |
+
"step": 505
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 37.45,
|
| 921 |
+
"learning_rate": 2.9415915362214692e-05,
|
| 922 |
+
"loss": 0.3387,
|
| 923 |
+
"step": 510
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 37.57,
|
| 927 |
+
"eval_loss": 0.5238548517227173,
|
| 928 |
+
"eval_runtime": 3.9677,
|
| 929 |
+
"eval_samples_per_second": 20.919,
|
| 930 |
+
"eval_steps_per_second": 1.512,
|
| 931 |
+
"step": 513
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 38.09,
|
| 935 |
+
"learning_rate": 2.907940576282856e-05,
|
| 936 |
+
"loss": 0.3257,
|
| 937 |
+
"step": 515
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 38.3,
|
| 941 |
+
"learning_rate": 2.874213507657861e-05,
|
| 942 |
+
"loss": 0.3316,
|
| 943 |
+
"step": 520
|
| 944 |
+
},
|
| 945 |
+
{
|
| 946 |
+
"epoch": 38.51,
|
| 947 |
+
"learning_rate": 2.840416622740617e-05,
|
| 948 |
+
"loss": 0.332,
|
| 949 |
+
"step": 525
|
| 950 |
+
},
|
| 951 |
+
{
|
| 952 |
+
"epoch": 38.55,
|
| 953 |
+
"eval_loss": 0.5318763852119446,
|
| 954 |
+
"eval_runtime": 3.9794,
|
| 955 |
+
"eval_samples_per_second": 20.858,
|
| 956 |
+
"eval_steps_per_second": 1.508,
|
| 957 |
+
"step": 526
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 39.15,
|
| 961 |
+
"learning_rate": 2.8065562269507463e-05,
|
| 962 |
+
"loss": 0.3178,
|
| 963 |
+
"step": 530
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 39.36,
|
| 967 |
+
"learning_rate": 2.7726386375569748e-05,
|
| 968 |
+
"loss": 0.3287,
|
| 969 |
+
"step": 535
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 39.57,
|
| 973 |
+
"learning_rate": 2.7386701824985255e-05,
|
| 974 |
+
"loss": 0.3334,
|
| 975 |
+
"step": 540
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 39.57,
|
| 979 |
+
"eval_loss": 0.5342416167259216,
|
| 980 |
+
"eval_runtime": 3.9681,
|
| 981 |
+
"eval_samples_per_second": 20.917,
|
| 982 |
+
"eval_steps_per_second": 1.512,
|
| 983 |
+
"step": 540
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 40.21,
|
| 987 |
+
"learning_rate": 2.7046571992045334e-05,
|
| 988 |
+
"loss": 0.3303,
|
| 989 |
+
"step": 545
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 40.43,
|
| 993 |
+
"learning_rate": 2.6706060334116777e-05,
|
| 994 |
+
"loss": 0.323,
|
| 995 |
+
"step": 550
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 40.55,
|
| 999 |
+
"eval_loss": 0.538831353187561,
|
| 1000 |
+
"eval_runtime": 3.9807,
|
| 1001 |
+
"eval_samples_per_second": 20.851,
|
| 1002 |
+
"eval_steps_per_second": 1.507,
|
| 1003 |
+
"step": 553
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 41.06,
|
| 1007 |
+
"learning_rate": 2.636523037980275e-05,
|
| 1008 |
+
"loss": 0.3103,
|
| 1009 |
+
"step": 555
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 41.28,
|
| 1013 |
+
"learning_rate": 2.6024145717090358e-05,
|
| 1014 |
+
"loss": 0.3232,
|
| 1015 |
+
"step": 560
|
| 1016 |
+
},
|
| 1017 |
+
{
|
| 1018 |
+
"epoch": 41.49,
|
| 1019 |
+
"learning_rate": 2.5682869981487152e-05,
|
| 1020 |
+
"loss": 0.3144,
|
| 1021 |
+
"step": 565
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"epoch": 41.57,
|
| 1025 |
+
"eval_loss": 0.5423159599304199,
|
| 1026 |
+
"eval_runtime": 3.9774,
|
| 1027 |
+
"eval_samples_per_second": 20.868,
|
| 1028 |
+
"eval_steps_per_second": 1.509,
|
| 1029 |
+
"step": 567
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 42.13,
|
| 1033 |
+
"learning_rate": 2.5341466844148775e-05,
|
| 1034 |
+
"loss": 0.3109,
|
| 1035 |
+
"step": 570
|
| 1036 |
+
},
|
| 1037 |
+
{
|
| 1038 |
+
"epoch": 42.34,
|
| 1039 |
+
"learning_rate": 2.5e-05,
|
| 1040 |
+
"loss": 0.3162,
|
| 1041 |
+
"step": 575
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 42.55,
|
| 1045 |
+
"learning_rate": 2.4658533155851228e-05,
|
| 1046 |
+
"loss": 0.3092,
|
| 1047 |
+
"step": 580
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"epoch": 42.55,
|
| 1051 |
+
"eval_loss": 0.54653400182724,
|
| 1052 |
+
"eval_runtime": 3.9802,
|
| 1053 |
+
"eval_samples_per_second": 20.853,
|
| 1054 |
+
"eval_steps_per_second": 1.507,
|
| 1055 |
+
"step": 580
|
| 1056 |
+
},
|
| 1057 |
+
{
|
| 1058 |
+
"epoch": 43.19,
|
| 1059 |
+
"learning_rate": 2.431713001851286e-05,
|
| 1060 |
+
"loss": 0.3175,
|
| 1061 |
+
"step": 585
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 43.4,
|
| 1065 |
+
"learning_rate": 2.3975854282909644e-05,
|
| 1066 |
+
"loss": 0.3084,
|
| 1067 |
+
"step": 590
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 43.57,
|
| 1071 |
+
"eval_loss": 0.5480648875236511,
|
| 1072 |
+
"eval_runtime": 3.9672,
|
| 1073 |
+
"eval_samples_per_second": 20.921,
|
| 1074 |
+
"eval_steps_per_second": 1.512,
|
| 1075 |
+
"step": 594
|
| 1076 |
+
},
|
| 1077 |
+
{
|
| 1078 |
+
"epoch": 44.04,
|
| 1079 |
+
"learning_rate": 2.3634769620197254e-05,
|
| 1080 |
+
"loss": 0.3039,
|
| 1081 |
+
"step": 595
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 44.26,
|
| 1085 |
+
"learning_rate": 2.329393966588323e-05,
|
| 1086 |
+
"loss": 0.2977,
|
| 1087 |
+
"step": 600
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 44.47,
|
| 1091 |
+
"learning_rate": 2.295342800795468e-05,
|
| 1092 |
+
"loss": 0.3091,
|
| 1093 |
+
"step": 605
|
| 1094 |
+
},
|
| 1095 |
+
{
|
| 1096 |
+
"epoch": 44.55,
|
| 1097 |
+
"eval_loss": 0.5604838728904724,
|
| 1098 |
+
"eval_runtime": 3.9783,
|
| 1099 |
+
"eval_samples_per_second": 20.863,
|
| 1100 |
+
"eval_steps_per_second": 1.508,
|
| 1101 |
+
"step": 607
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 45.11,
|
| 1105 |
+
"learning_rate": 2.261329817501475e-05,
|
| 1106 |
+
"loss": 0.3087,
|
| 1107 |
+
"step": 610
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 45.32,
|
| 1111 |
+
"learning_rate": 2.2273613624430255e-05,
|
| 1112 |
+
"loss": 0.2994,
|
| 1113 |
+
"step": 615
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 45.53,
|
| 1117 |
+
"learning_rate": 2.1934437730492543e-05,
|
| 1118 |
+
"loss": 0.3044,
|
| 1119 |
+
"step": 620
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 45.57,
|
| 1123 |
+
"eval_loss": 0.560636579990387,
|
| 1124 |
+
"eval_runtime": 3.9626,
|
| 1125 |
+
"eval_samples_per_second": 20.946,
|
| 1126 |
+
"eval_steps_per_second": 1.514,
|
| 1127 |
+
"step": 621
|
| 1128 |
+
},
|
| 1129 |
+
{
|
| 1130 |
+
"epoch": 46.17,
|
| 1131 |
+
"learning_rate": 2.159583377259384e-05,
|
| 1132 |
+
"loss": 0.2867,
|
| 1133 |
+
"step": 625
|
| 1134 |
+
},
|
| 1135 |
+
{
|
| 1136 |
+
"epoch": 46.38,
|
| 1137 |
+
"learning_rate": 2.1257864923421404e-05,
|
| 1138 |
+
"loss": 0.303,
|
| 1139 |
+
"step": 630
|
| 1140 |
+
},
|
| 1141 |
+
{
|
| 1142 |
+
"epoch": 46.55,
|
| 1143 |
+
"eval_loss": 0.5683414340019226,
|
| 1144 |
+
"eval_runtime": 3.9884,
|
| 1145 |
+
"eval_samples_per_second": 20.81,
|
| 1146 |
+
"eval_steps_per_second": 1.504,
|
| 1147 |
+
"step": 634
|
| 1148 |
+
},
|
| 1149 |
+
{
|
| 1150 |
+
"epoch": 47.02,
|
| 1151 |
+
"learning_rate": 2.092059423717145e-05,
|
| 1152 |
+
"loss": 0.2971,
|
| 1153 |
+
"step": 635
|
| 1154 |
+
},
|
| 1155 |
+
{
|
| 1156 |
+
"epoch": 47.23,
|
| 1157 |
+
"learning_rate": 2.0584084637785317e-05,
|
| 1158 |
+
"loss": 0.2923,
|
| 1159 |
+
"step": 640
|
| 1160 |
+
},
|
| 1161 |
+
{
|
| 1162 |
+
"epoch": 47.45,
|
| 1163 |
+
"learning_rate": 2.0248398907209826e-05,
|
| 1164 |
+
"loss": 0.2896,
|
| 1165 |
+
"step": 645
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 47.57,
|
| 1169 |
+
"eval_loss": 0.572201669216156,
|
| 1170 |
+
"eval_runtime": 3.9766,
|
| 1171 |
+
"eval_samples_per_second": 20.872,
|
| 1172 |
+
"eval_steps_per_second": 1.509,
|
| 1173 |
+
"step": 648
|
| 1174 |
+
},
|
| 1175 |
+
{
|
| 1176 |
+
"epoch": 48.09,
|
| 1177 |
+
"learning_rate": 1.991359967368416e-05,
|
| 1178 |
+
"loss": 0.2963,
|
| 1179 |
+
"step": 650
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 48.3,
|
| 1183 |
+
"learning_rate": 1.957974940005548e-05,
|
| 1184 |
+
"loss": 0.2849,
|
| 1185 |
+
"step": 655
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 48.51,
|
| 1189 |
+
"learning_rate": 1.9246910372125342e-05,
|
| 1190 |
+
"loss": 0.2854,
|
| 1191 |
+
"step": 660
|
| 1192 |
+
},
|
| 1193 |
+
{
|
| 1194 |
+
"epoch": 48.55,
|
| 1195 |
+
"eval_loss": 0.5778339505195618,
|
| 1196 |
+
"eval_runtime": 3.9935,
|
| 1197 |
+
"eval_samples_per_second": 20.784,
|
| 1198 |
+
"eval_steps_per_second": 1.502,
|
| 1199 |
+
"step": 661
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 49.15,
|
| 1203 |
+
"learning_rate": 1.8915144687029106e-05,
|
| 1204 |
+
"loss": 0.2884,
|
| 1205 |
+
"step": 665
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"epoch": 49.36,
|
| 1209 |
+
"learning_rate": 1.8584514241650666e-05,
|
| 1210 |
+
"loss": 0.2767,
|
| 1211 |
+
"step": 670
|
| 1212 |
+
},
|
| 1213 |
+
{
|
| 1214 |
+
"epoch": 49.57,
|
| 1215 |
+
"learning_rate": 1.825508072107439e-05,
|
| 1216 |
+
"loss": 0.291,
|
| 1217 |
+
"step": 675
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 49.57,
|
| 1221 |
+
"eval_loss": 0.5825899839401245,
|
| 1222 |
+
"eval_runtime": 3.9815,
|
| 1223 |
+
"eval_samples_per_second": 20.846,
|
| 1224 |
+
"eval_steps_per_second": 1.507,
|
| 1225 |
+
"step": 675
|
| 1226 |
+
},
|
| 1227 |
+
{
|
| 1228 |
+
"epoch": 49.57,
|
| 1229 |
+
"step": 675,
|
| 1230 |
+
"total_flos": 6306831714484224.0,
|
| 1231 |
+
"train_loss": 0.5519325028525458,
|
| 1232 |
+
"train_runtime": 6501.4714,
|
| 1233 |
+
"train_samples_per_second": 5.737,
|
| 1234 |
+
"train_steps_per_second": 0.177
|
| 1235 |
+
}
|
| 1236 |
+
],
|
| 1237 |
+
"logging_steps": 5,
|
| 1238 |
+
"max_steps": 1150,
|
| 1239 |
+
"num_train_epochs": 50,
|
| 1240 |
+
"save_steps": 500,
|
| 1241 |
+
"total_flos": 6306831714484224.0,
|
| 1242 |
+
"trial_name": null,
|
| 1243 |
+
"trial_params": null
|
| 1244 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9ed7fec1b3431106775460da2de8fde640cd4b48f3468fa87939bb0abe25c1f6
|
| 3 |
+
size 5688
|