Model save
Browse files- README.md +91 -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 +760 -0
- training_args.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,91 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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_5-30
|
| 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_5-30
|
| 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: 0.5210
|
| 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: 4
|
| 43 |
+
- gradient_accumulation_steps: 2
|
| 44 |
+
- total_train_batch_size: 32
|
| 45 |
+
- total_eval_batch_size: 32
|
| 46 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 47 |
+
- lr_scheduler_type: cosine
|
| 48 |
+
- num_epochs: 30.0
|
| 49 |
+
|
| 50 |
+
### Training results
|
| 51 |
+
|
| 52 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 53 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 54 |
+
| 1.8242 | 0.55 | 13 | 1.6628 |
|
| 55 |
+
| 1.5291 | 1.57 | 27 | 1.3837 |
|
| 56 |
+
| 1.2362 | 2.55 | 40 | 1.1786 |
|
| 57 |
+
| 1.1401 | 3.57 | 54 | 1.0978 |
|
| 58 |
+
| 1.0666 | 4.55 | 67 | 1.0428 |
|
| 59 |
+
| 1.021 | 5.57 | 81 | 1.0063 |
|
| 60 |
+
| 0.9852 | 6.55 | 94 | 0.9629 |
|
| 61 |
+
| 0.9317 | 7.57 | 108 | 0.9246 |
|
| 62 |
+
| 0.8735 | 8.55 | 121 | 0.8885 |
|
| 63 |
+
| 0.8147 | 9.57 | 135 | 0.8335 |
|
| 64 |
+
| 0.763 | 10.55 | 148 | 0.7860 |
|
| 65 |
+
| 0.6926 | 11.57 | 162 | 0.7296 |
|
| 66 |
+
| 0.6332 | 12.55 | 175 | 0.6863 |
|
| 67 |
+
| 0.5898 | 13.57 | 189 | 0.6491 |
|
| 68 |
+
| 0.536 | 14.55 | 202 | 0.6180 |
|
| 69 |
+
| 0.5263 | 15.57 | 216 | 0.5764 |
|
| 70 |
+
| 0.5071 | 16.55 | 229 | 0.5683 |
|
| 71 |
+
| 0.4756 | 17.57 | 243 | 0.5597 |
|
| 72 |
+
| 0.4693 | 18.55 | 256 | 0.5342 |
|
| 73 |
+
| 0.4414 | 19.57 | 270 | 0.5386 |
|
| 74 |
+
| 0.4266 | 20.55 | 283 | 0.5346 |
|
| 75 |
+
| 0.4286 | 21.57 | 297 | 0.5155 |
|
| 76 |
+
| 0.4256 | 22.55 | 310 | 0.5108 |
|
| 77 |
+
| 0.418 | 23.57 | 324 | 0.5230 |
|
| 78 |
+
| 0.407 | 24.55 | 337 | 0.5165 |
|
| 79 |
+
| 0.411 | 25.57 | 351 | 0.5128 |
|
| 80 |
+
| 0.3896 | 26.55 | 364 | 0.5027 |
|
| 81 |
+
| 0.39 | 27.57 | 378 | 0.5063 |
|
| 82 |
+
| 0.3928 | 28.55 | 391 | 0.4946 |
|
| 83 |
+
| 0.3818 | 29.57 | 405 | 0.5063 |
|
| 84 |
+
|
| 85 |
+
|
| 86 |
+
### Framework versions
|
| 87 |
+
|
| 88 |
+
- Transformers 4.35.0
|
| 89 |
+
- Pytorch 2.1.0
|
| 90 |
+
- Datasets 2.14.6
|
| 91 |
+
- 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 |
+
"k_proj",
|
| 22 |
+
"o_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:bce0d9bc95c02f5ae6babcfe7593f0bb2d971991ea3928cb0b23168e5b32fa06
|
| 3 |
+
size 109086672
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 29.57,
|
| 3 |
+
"eval_loss": 0.5209962725639343,
|
| 4 |
+
"eval_runtime": 2.1615,
|
| 5 |
+
"eval_samples": 83,
|
| 6 |
+
"eval_samples_per_second": 38.4,
|
| 7 |
+
"eval_steps_per_second": 1.388,
|
| 8 |
+
"train_loss": 0.7121148815861454,
|
| 9 |
+
"train_runtime": 1989.4524,
|
| 10 |
+
"train_samples": 746,
|
| 11 |
+
"train_samples_per_second": 11.249,
|
| 12 |
+
"train_steps_per_second": 0.347
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 29.57,
|
| 3 |
+
"eval_loss": 0.5209962725639343,
|
| 4 |
+
"eval_runtime": 2.1615,
|
| 5 |
+
"eval_samples": 83,
|
| 6 |
+
"eval_samples_per_second": 38.4,
|
| 7 |
+
"eval_steps_per_second": 1.388
|
| 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": 29.57,
|
| 3 |
+
"train_loss": 0.7121148815861454,
|
| 4 |
+
"train_runtime": 1989.4524,
|
| 5 |
+
"train_samples": 746,
|
| 6 |
+
"train_samples_per_second": 11.249,
|
| 7 |
+
"train_steps_per_second": 0.347
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,760 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 29.574468085106382,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 405,
|
| 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": 4.9999740874098114e-05,
|
| 14 |
+
"loss": 2.0607,
|
| 15 |
+
"step": 1
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.21,
|
| 19 |
+
"learning_rate": 4.999352212103373e-05,
|
| 20 |
+
"loss": 2.0069,
|
| 21 |
+
"step": 5
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.43,
|
| 25 |
+
"learning_rate": 4.9974091841168195e-05,
|
| 26 |
+
"loss": 1.8242,
|
| 27 |
+
"step": 10
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.55,
|
| 31 |
+
"eval_loss": 1.6627923250198364,
|
| 32 |
+
"eval_runtime": 2.2915,
|
| 33 |
+
"eval_samples_per_second": 36.22,
|
| 34 |
+
"eval_steps_per_second": 1.309,
|
| 35 |
+
"step": 13
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 1.06,
|
| 39 |
+
"learning_rate": 4.994171922976348e-05,
|
| 40 |
+
"loss": 1.7045,
|
| 41 |
+
"step": 15
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 1.28,
|
| 45 |
+
"learning_rate": 4.9896421063288286e-05,
|
| 46 |
+
"loss": 1.6023,
|
| 47 |
+
"step": 20
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 1.49,
|
| 51 |
+
"learning_rate": 4.983822081662578e-05,
|
| 52 |
+
"loss": 1.5291,
|
| 53 |
+
"step": 25
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.57,
|
| 57 |
+
"eval_loss": 1.3837387561798096,
|
| 58 |
+
"eval_runtime": 2.1494,
|
| 59 |
+
"eval_samples_per_second": 38.615,
|
| 60 |
+
"eval_steps_per_second": 1.396,
|
| 61 |
+
"step": 27
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 2.13,
|
| 65 |
+
"learning_rate": 4.976714865090827e-05,
|
| 66 |
+
"loss": 1.417,
|
| 67 |
+
"step": 30
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 2.34,
|
| 71 |
+
"learning_rate": 4.9683241397886745e-05,
|
| 72 |
+
"loss": 1.3126,
|
| 73 |
+
"step": 35
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 2.55,
|
| 77 |
+
"learning_rate": 4.958654254084355e-05,
|
| 78 |
+
"loss": 1.2362,
|
| 79 |
+
"step": 40
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 2.55,
|
| 83 |
+
"eval_loss": 1.1785995960235596,
|
| 84 |
+
"eval_runtime": 2.169,
|
| 85 |
+
"eval_samples_per_second": 38.267,
|
| 86 |
+
"eval_steps_per_second": 1.383,
|
| 87 |
+
"step": 40
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 3.19,
|
| 91 |
+
"learning_rate": 4.947710219205808e-05,
|
| 92 |
+
"loss": 1.1757,
|
| 93 |
+
"step": 45
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 3.4,
|
| 97 |
+
"learning_rate": 4.9354977066836986e-05,
|
| 98 |
+
"loss": 1.1401,
|
| 99 |
+
"step": 50
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 3.57,
|
| 103 |
+
"eval_loss": 1.097777247428894,
|
| 104 |
+
"eval_runtime": 2.1856,
|
| 105 |
+
"eval_samples_per_second": 37.976,
|
| 106 |
+
"eval_steps_per_second": 1.373,
|
| 107 |
+
"step": 54
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 4.04,
|
| 111 |
+
"learning_rate": 4.922023045412266e-05,
|
| 112 |
+
"loss": 1.1249,
|
| 113 |
+
"step": 55
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 4.26,
|
| 117 |
+
"learning_rate": 4.907293218369499e-05,
|
| 118 |
+
"loss": 1.0927,
|
| 119 |
+
"step": 60
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 4.47,
|
| 123 |
+
"learning_rate": 4.8913158589983374e-05,
|
| 124 |
+
"loss": 1.0666,
|
| 125 |
+
"step": 65
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 4.55,
|
| 129 |
+
"eval_loss": 1.042842984199524,
|
| 130 |
+
"eval_runtime": 2.1605,
|
| 131 |
+
"eval_samples_per_second": 38.418,
|
| 132 |
+
"eval_steps_per_second": 1.389,
|
| 133 |
+
"step": 67
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 5.11,
|
| 137 |
+
"learning_rate": 4.874099247250798e-05,
|
| 138 |
+
"loss": 1.0616,
|
| 139 |
+
"step": 70
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 5.32,
|
| 143 |
+
"learning_rate": 4.855652305297052e-05,
|
| 144 |
+
"loss": 1.0397,
|
| 145 |
+
"step": 75
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 5.53,
|
| 149 |
+
"learning_rate": 4.835984592901678e-05,
|
| 150 |
+
"loss": 1.021,
|
| 151 |
+
"step": 80
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 5.57,
|
| 155 |
+
"eval_loss": 1.0062576532363892,
|
| 156 |
+
"eval_runtime": 2.1747,
|
| 157 |
+
"eval_samples_per_second": 38.166,
|
| 158 |
+
"eval_steps_per_second": 1.379,
|
| 159 |
+
"step": 81
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 6.17,
|
| 163 |
+
"learning_rate": 4.815106302469513e-05,
|
| 164 |
+
"loss": 0.9995,
|
| 165 |
+
"step": 85
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 6.38,
|
| 169 |
+
"learning_rate": 4.793028253763633e-05,
|
| 170 |
+
"loss": 0.9852,
|
| 171 |
+
"step": 90
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 6.55,
|
| 175 |
+
"eval_loss": 0.9628537893295288,
|
| 176 |
+
"eval_runtime": 2.1651,
|
| 177 |
+
"eval_samples_per_second": 38.335,
|
| 178 |
+
"eval_steps_per_second": 1.386,
|
| 179 |
+
"step": 94
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 7.02,
|
| 183 |
+
"learning_rate": 4.769761888298222e-05,
|
| 184 |
+
"loss": 0.9799,
|
| 185 |
+
"step": 95
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 7.23,
|
| 189 |
+
"learning_rate": 4.74531926340924e-05,
|
| 190 |
+
"loss": 0.9521,
|
| 191 |
+
"step": 100
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 7.45,
|
| 195 |
+
"learning_rate": 4.719713046005938e-05,
|
| 196 |
+
"loss": 0.9317,
|
| 197 |
+
"step": 105
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 7.57,
|
| 201 |
+
"eval_loss": 0.9245708584785461,
|
| 202 |
+
"eval_runtime": 2.1462,
|
| 203 |
+
"eval_samples_per_second": 38.673,
|
| 204 |
+
"eval_steps_per_second": 1.398,
|
| 205 |
+
"step": 108
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 8.09,
|
| 209 |
+
"learning_rate": 4.6929565060064864e-05,
|
| 210 |
+
"loss": 0.9231,
|
| 211 |
+
"step": 110
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 8.3,
|
| 215 |
+
"learning_rate": 4.665063509461097e-05,
|
| 216 |
+
"loss": 0.8929,
|
| 217 |
+
"step": 115
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 8.51,
|
| 221 |
+
"learning_rate": 4.6360485113662216e-05,
|
| 222 |
+
"loss": 0.8735,
|
| 223 |
+
"step": 120
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 8.55,
|
| 227 |
+
"eval_loss": 0.8884902596473694,
|
| 228 |
+
"eval_runtime": 2.1658,
|
| 229 |
+
"eval_samples_per_second": 38.323,
|
| 230 |
+
"eval_steps_per_second": 1.385,
|
| 231 |
+
"step": 121
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 9.15,
|
| 235 |
+
"learning_rate": 4.605926548173529e-05,
|
| 236 |
+
"loss": 0.8606,
|
| 237 |
+
"step": 125
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 9.36,
|
| 241 |
+
"learning_rate": 4.574713229997563e-05,
|
| 242 |
+
"loss": 0.8246,
|
| 243 |
+
"step": 130
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 9.57,
|
| 247 |
+
"learning_rate": 4.542424732526105e-05,
|
| 248 |
+
"loss": 0.8147,
|
| 249 |
+
"step": 135
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 9.57,
|
| 253 |
+
"eval_loss": 0.8334853053092957,
|
| 254 |
+
"eval_runtime": 2.1468,
|
| 255 |
+
"eval_samples_per_second": 38.663,
|
| 256 |
+
"eval_steps_per_second": 1.397,
|
| 257 |
+
"step": 135
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 10.21,
|
| 261 |
+
"learning_rate": 4.509077788637446e-05,
|
| 262 |
+
"loss": 0.7726,
|
| 263 |
+
"step": 140
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 10.43,
|
| 267 |
+
"learning_rate": 4.4746896797288965e-05,
|
| 268 |
+
"loss": 0.763,
|
| 269 |
+
"step": 145
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 10.55,
|
| 273 |
+
"eval_loss": 0.7860499024391174,
|
| 274 |
+
"eval_runtime": 2.1649,
|
| 275 |
+
"eval_samples_per_second": 38.339,
|
| 276 |
+
"eval_steps_per_second": 1.386,
|
| 277 |
+
"step": 148
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 11.06,
|
| 281 |
+
"learning_rate": 4.43927822676105e-05,
|
| 282 |
+
"loss": 0.7436,
|
| 283 |
+
"step": 150
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 11.28,
|
| 287 |
+
"learning_rate": 4.402861781022412e-05,
|
| 288 |
+
"loss": 0.7065,
|
| 289 |
+
"step": 155
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 11.49,
|
| 293 |
+
"learning_rate": 4.365459214619214e-05,
|
| 294 |
+
"loss": 0.6926,
|
| 295 |
+
"step": 160
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 11.57,
|
| 299 |
+
"eval_loss": 0.7296475172042847,
|
| 300 |
+
"eval_runtime": 2.173,
|
| 301 |
+
"eval_samples_per_second": 38.196,
|
| 302 |
+
"eval_steps_per_second": 1.381,
|
| 303 |
+
"step": 162
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 12.13,
|
| 307 |
+
"learning_rate": 4.3270899106953105e-05,
|
| 308 |
+
"loss": 0.6764,
|
| 309 |
+
"step": 165
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 12.34,
|
| 313 |
+
"learning_rate": 4.2877737533872485e-05,
|
| 314 |
+
"loss": 0.6473,
|
| 315 |
+
"step": 170
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 12.55,
|
| 319 |
+
"learning_rate": 4.2475311175197056e-05,
|
| 320 |
+
"loss": 0.6332,
|
| 321 |
+
"step": 175
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 12.55,
|
| 325 |
+
"eval_loss": 0.686292827129364,
|
| 326 |
+
"eval_runtime": 2.1677,
|
| 327 |
+
"eval_samples_per_second": 38.289,
|
| 328 |
+
"eval_steps_per_second": 1.384,
|
| 329 |
+
"step": 175
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 13.19,
|
| 333 |
+
"learning_rate": 4.206382858046636e-05,
|
| 334 |
+
"loss": 0.6181,
|
| 335 |
+
"step": 180
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 13.4,
|
| 339 |
+
"learning_rate": 4.164350299243601e-05,
|
| 340 |
+
"loss": 0.5898,
|
| 341 |
+
"step": 185
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 13.57,
|
| 345 |
+
"eval_loss": 0.6491469740867615,
|
| 346 |
+
"eval_runtime": 2.1549,
|
| 347 |
+
"eval_samples_per_second": 38.517,
|
| 348 |
+
"eval_steps_per_second": 1.392,
|
| 349 |
+
"step": 189
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 14.04,
|
| 353 |
+
"learning_rate": 4.12145522365689e-05,
|
| 354 |
+
"loss": 0.5682,
|
| 355 |
+
"step": 190
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 14.26,
|
| 359 |
+
"learning_rate": 4.077719860815132e-05,
|
| 360 |
+
"loss": 0.5651,
|
| 361 |
+
"step": 195
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 14.47,
|
| 365 |
+
"learning_rate": 4.033166875709291e-05,
|
| 366 |
+
"loss": 0.536,
|
| 367 |
+
"step": 200
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 14.55,
|
| 371 |
+
"eval_loss": 0.6179705262184143,
|
| 372 |
+
"eval_runtime": 2.1686,
|
| 373 |
+
"eval_samples_per_second": 38.273,
|
| 374 |
+
"eval_steps_per_second": 1.383,
|
| 375 |
+
"step": 202
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 15.11,
|
| 379 |
+
"learning_rate": 3.9878193570469746e-05,
|
| 380 |
+
"loss": 0.5464,
|
| 381 |
+
"step": 205
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 15.32,
|
| 385 |
+
"learning_rate": 3.941700805287168e-05,
|
| 386 |
+
"loss": 0.5064,
|
| 387 |
+
"step": 210
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 15.53,
|
| 391 |
+
"learning_rate": 3.894835120461584e-05,
|
| 392 |
+
"loss": 0.5263,
|
| 393 |
+
"step": 215
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 15.57,
|
| 397 |
+
"eval_loss": 0.5763722062110901,
|
| 398 |
+
"eval_runtime": 2.1517,
|
| 399 |
+
"eval_samples_per_second": 38.575,
|
| 400 |
+
"eval_steps_per_second": 1.394,
|
| 401 |
+
"step": 216
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 16.17,
|
| 405 |
+
"learning_rate": 3.8472465897889394e-05,
|
| 406 |
+
"loss": 0.4961,
|
| 407 |
+
"step": 220
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 16.38,
|
| 411 |
+
"learning_rate": 3.798959875088584e-05,
|
| 412 |
+
"loss": 0.5071,
|
| 413 |
+
"step": 225
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 16.55,
|
| 417 |
+
"eval_loss": 0.5683054327964783,
|
| 418 |
+
"eval_runtime": 2.1772,
|
| 419 |
+
"eval_samples_per_second": 38.123,
|
| 420 |
+
"eval_steps_per_second": 1.378,
|
| 421 |
+
"step": 229
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 17.02,
|
| 425 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 426 |
+
"loss": 0.4835,
|
| 427 |
+
"step": 230
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 17.23,
|
| 431 |
+
"learning_rate": 3.700392337014789e-05,
|
| 432 |
+
"loss": 0.4797,
|
| 433 |
+
"step": 235
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 17.45,
|
| 437 |
+
"learning_rate": 3.6501625943278805e-05,
|
| 438 |
+
"loss": 0.4756,
|
| 439 |
+
"step": 240
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 17.57,
|
| 443 |
+
"eval_loss": 0.5597423315048218,
|
| 444 |
+
"eval_runtime": 2.1615,
|
| 445 |
+
"eval_samples_per_second": 38.399,
|
| 446 |
+
"eval_steps_per_second": 1.388,
|
| 447 |
+
"step": 243
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 18.09,
|
| 451 |
+
"learning_rate": 3.599336802514764e-05,
|
| 452 |
+
"loss": 0.4652,
|
| 453 |
+
"step": 245
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 18.3,
|
| 457 |
+
"learning_rate": 3.547941301041661e-05,
|
| 458 |
+
"loss": 0.4466,
|
| 459 |
+
"step": 250
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 18.51,
|
| 463 |
+
"learning_rate": 3.496002724615604e-05,
|
| 464 |
+
"loss": 0.4693,
|
| 465 |
+
"step": 255
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 18.55,
|
| 469 |
+
"eval_loss": 0.5342021584510803,
|
| 470 |
+
"eval_runtime": 2.1635,
|
| 471 |
+
"eval_samples_per_second": 38.365,
|
| 472 |
+
"eval_steps_per_second": 1.387,
|
| 473 |
+
"step": 256
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 19.15,
|
| 477 |
+
"learning_rate": 3.443547989381536e-05,
|
| 478 |
+
"loss": 0.4602,
|
| 479 |
+
"step": 260
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 19.36,
|
| 483 |
+
"learning_rate": 3.390604278973543e-05,
|
| 484 |
+
"loss": 0.4522,
|
| 485 |
+
"step": 265
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 19.57,
|
| 489 |
+
"learning_rate": 3.3371990304274656e-05,
|
| 490 |
+
"loss": 0.4414,
|
| 491 |
+
"step": 270
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 19.57,
|
| 495 |
+
"eval_loss": 0.5385740399360657,
|
| 496 |
+
"eval_runtime": 2.1714,
|
| 497 |
+
"eval_samples_per_second": 38.224,
|
| 498 |
+
"eval_steps_per_second": 1.382,
|
| 499 |
+
"step": 270
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 20.21,
|
| 503 |
+
"learning_rate": 3.283359919962206e-05,
|
| 504 |
+
"loss": 0.4492,
|
| 505 |
+
"step": 275
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 20.43,
|
| 509 |
+
"learning_rate": 3.2291148486370626e-05,
|
| 510 |
+
"loss": 0.4266,
|
| 511 |
+
"step": 280
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 20.55,
|
| 515 |
+
"eval_loss": 0.5345889925956726,
|
| 516 |
+
"eval_runtime": 2.1657,
|
| 517 |
+
"eval_samples_per_second": 38.324,
|
| 518 |
+
"eval_steps_per_second": 1.385,
|
| 519 |
+
"step": 283
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 21.06,
|
| 523 |
+
"learning_rate": 3.1744919278925605e-05,
|
| 524 |
+
"loss": 0.437,
|
| 525 |
+
"step": 285
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 21.28,
|
| 529 |
+
"learning_rate": 3.11951946498225e-05,
|
| 530 |
+
"loss": 0.4219,
|
| 531 |
+
"step": 290
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 21.49,
|
| 535 |
+
"learning_rate": 3.0642259483030266e-05,
|
| 536 |
+
"loss": 0.4286,
|
| 537 |
+
"step": 295
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 21.57,
|
| 541 |
+
"eval_loss": 0.5155086517333984,
|
| 542 |
+
"eval_runtime": 2.1707,
|
| 543 |
+
"eval_samples_per_second": 38.237,
|
| 544 |
+
"eval_steps_per_second": 1.382,
|
| 545 |
+
"step": 297
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 22.13,
|
| 549 |
+
"learning_rate": 3.008640032631585e-05,
|
| 550 |
+
"loss": 0.4214,
|
| 551 |
+
"step": 300
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 22.34,
|
| 555 |
+
"learning_rate": 2.9527905242746394e-05,
|
| 556 |
+
"loss": 0.4136,
|
| 557 |
+
"step": 305
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 22.55,
|
| 561 |
+
"learning_rate": 2.8967063661406285e-05,
|
| 562 |
+
"loss": 0.4256,
|
| 563 |
+
"step": 310
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 22.55,
|
| 567 |
+
"eval_loss": 0.5108062624931335,
|
| 568 |
+
"eval_runtime": 2.1723,
|
| 569 |
+
"eval_samples_per_second": 38.209,
|
| 570 |
+
"eval_steps_per_second": 1.381,
|
| 571 |
+
"step": 310
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 23.19,
|
| 575 |
+
"learning_rate": 2.840416622740617e-05,
|
| 576 |
+
"loss": 0.4066,
|
| 577 |
+
"step": 315
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 23.4,
|
| 581 |
+
"learning_rate": 2.7839504651261872e-05,
|
| 582 |
+
"loss": 0.418,
|
| 583 |
+
"step": 320
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 23.57,
|
| 587 |
+
"eval_loss": 0.5229755640029907,
|
| 588 |
+
"eval_runtime": 2.1749,
|
| 589 |
+
"eval_samples_per_second": 38.163,
|
| 590 |
+
"eval_steps_per_second": 1.379,
|
| 591 |
+
"step": 324
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 24.04,
|
| 595 |
+
"learning_rate": 2.7273371557721176e-05,
|
| 596 |
+
"loss": 0.4153,
|
| 597 |
+
"step": 325
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 24.26,
|
| 601 |
+
"learning_rate": 2.6706060334116777e-05,
|
| 602 |
+
"loss": 0.4139,
|
| 603 |
+
"step": 330
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 24.47,
|
| 607 |
+
"learning_rate": 2.6137864978324095e-05,
|
| 608 |
+
"loss": 0.407,
|
| 609 |
+
"step": 335
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 24.55,
|
| 613 |
+
"eval_loss": 0.5164679884910583,
|
| 614 |
+
"eval_runtime": 2.1723,
|
| 615 |
+
"eval_samples_per_second": 38.209,
|
| 616 |
+
"eval_steps_per_second": 1.381,
|
| 617 |
+
"step": 337
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 25.11,
|
| 621 |
+
"learning_rate": 2.556907994640264e-05,
|
| 622 |
+
"loss": 0.406,
|
| 623 |
+
"step": 340
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 25.32,
|
| 627 |
+
"learning_rate": 2.5e-05,
|
| 628 |
+
"loss": 0.3903,
|
| 629 |
+
"step": 345
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 25.53,
|
| 633 |
+
"learning_rate": 2.4430920053597356e-05,
|
| 634 |
+
"loss": 0.411,
|
| 635 |
+
"step": 350
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 25.57,
|
| 639 |
+
"eval_loss": 0.5128384232521057,
|
| 640 |
+
"eval_runtime": 2.1499,
|
| 641 |
+
"eval_samples_per_second": 38.606,
|
| 642 |
+
"eval_steps_per_second": 1.395,
|
| 643 |
+
"step": 351
|
| 644 |
+
},
|
| 645 |
+
{
|
| 646 |
+
"epoch": 26.17,
|
| 647 |
+
"learning_rate": 2.3862135021675918e-05,
|
| 648 |
+
"loss": 0.4,
|
| 649 |
+
"step": 355
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 26.38,
|
| 653 |
+
"learning_rate": 2.329393966588323e-05,
|
| 654 |
+
"loss": 0.3896,
|
| 655 |
+
"step": 360
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 26.55,
|
| 659 |
+
"eval_loss": 0.5026661157608032,
|
| 660 |
+
"eval_runtime": 2.1644,
|
| 661 |
+
"eval_samples_per_second": 38.349,
|
| 662 |
+
"eval_steps_per_second": 1.386,
|
| 663 |
+
"step": 364
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 27.02,
|
| 667 |
+
"learning_rate": 2.2726628442278826e-05,
|
| 668 |
+
"loss": 0.4005,
|
| 669 |
+
"step": 365
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 27.23,
|
| 673 |
+
"learning_rate": 2.2160495348738123e-05,
|
| 674 |
+
"loss": 0.4014,
|
| 675 |
+
"step": 370
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 27.45,
|
| 679 |
+
"learning_rate": 2.159583377259384e-05,
|
| 680 |
+
"loss": 0.39,
|
| 681 |
+
"step": 375
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 27.57,
|
| 685 |
+
"eval_loss": 0.5062728524208069,
|
| 686 |
+
"eval_runtime": 2.1531,
|
| 687 |
+
"eval_samples_per_second": 38.549,
|
| 688 |
+
"eval_steps_per_second": 1.393,
|
| 689 |
+
"step": 378
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 28.09,
|
| 693 |
+
"learning_rate": 2.1032936338593718e-05,
|
| 694 |
+
"loss": 0.3872,
|
| 695 |
+
"step": 380
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 28.3,
|
| 699 |
+
"learning_rate": 2.0472094757253605e-05,
|
| 700 |
+
"loss": 0.3929,
|
| 701 |
+
"step": 385
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 28.51,
|
| 705 |
+
"learning_rate": 1.991359967368416e-05,
|
| 706 |
+
"loss": 0.3928,
|
| 707 |
+
"step": 390
|
| 708 |
+
},
|
| 709 |
+
{
|
| 710 |
+
"epoch": 28.55,
|
| 711 |
+
"eval_loss": 0.494553804397583,
|
| 712 |
+
"eval_runtime": 2.1731,
|
| 713 |
+
"eval_samples_per_second": 38.195,
|
| 714 |
+
"eval_steps_per_second": 1.381,
|
| 715 |
+
"step": 391
|
| 716 |
+
},
|
| 717 |
+
{
|
| 718 |
+
"epoch": 29.15,
|
| 719 |
+
"learning_rate": 1.9357740516969736e-05,
|
| 720 |
+
"loss": 0.386,
|
| 721 |
+
"step": 395
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 29.36,
|
| 725 |
+
"learning_rate": 1.8804805350177505e-05,
|
| 726 |
+
"loss": 0.3948,
|
| 727 |
+
"step": 400
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 29.57,
|
| 731 |
+
"learning_rate": 1.825508072107439e-05,
|
| 732 |
+
"loss": 0.3818,
|
| 733 |
+
"step": 405
|
| 734 |
+
},
|
| 735 |
+
{
|
| 736 |
+
"epoch": 29.57,
|
| 737 |
+
"eval_loss": 0.5063261985778809,
|
| 738 |
+
"eval_runtime": 2.1689,
|
| 739 |
+
"eval_samples_per_second": 38.269,
|
| 740 |
+
"eval_steps_per_second": 1.383,
|
| 741 |
+
"step": 405
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 29.57,
|
| 745 |
+
"step": 405,
|
| 746 |
+
"total_flos": 2750244861247488.0,
|
| 747 |
+
"train_loss": 0.7121148815861454,
|
| 748 |
+
"train_runtime": 1989.4524,
|
| 749 |
+
"train_samples_per_second": 11.249,
|
| 750 |
+
"train_steps_per_second": 0.347
|
| 751 |
+
}
|
| 752 |
+
],
|
| 753 |
+
"logging_steps": 5,
|
| 754 |
+
"max_steps": 690,
|
| 755 |
+
"num_train_epochs": 30,
|
| 756 |
+
"save_steps": 500,
|
| 757 |
+
"total_flos": 2750244861247488.0,
|
| 758 |
+
"trial_name": null,
|
| 759 |
+
"trial_params": null
|
| 760 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c25bedbdfd3b15bc69ceca6d97309ac48e07b9fdce718c2979627799ccf94e44
|
| 3 |
+
size 5688
|