Upload folder using huggingface_hub
Browse files- README.md +78 -0
- all_results.json +16 -0
- config.json +31 -0
- eval_results.json +10 -0
- generation_config.json +6 -0
- model.safetensors +3 -0
- special_tokens_map.json +30 -0
- tokenizer.json +0 -0
- tokenizer_config.json +47 -0
- train_results.json +9 -0
- trainer_state.json +1535 -0
- training_args.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,78 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
metrics:
|
| 6 |
+
- accuracy
|
| 7 |
+
model-index:
|
| 8 |
+
- name: babylm-base5M-gpt2
|
| 9 |
+
results: []
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 13 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 14 |
+
|
| 15 |
+
# babylm-base5M-gpt2
|
| 16 |
+
|
| 17 |
+
This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
|
| 18 |
+
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 3.0628
|
| 20 |
+
- Accuracy: 0.4521
|
| 21 |
+
|
| 22 |
+
## Model description
|
| 23 |
+
|
| 24 |
+
More information needed
|
| 25 |
+
|
| 26 |
+
## Intended uses & limitations
|
| 27 |
+
|
| 28 |
+
More information needed
|
| 29 |
+
|
| 30 |
+
## Training and evaluation data
|
| 31 |
+
|
| 32 |
+
More information needed
|
| 33 |
+
|
| 34 |
+
## Training procedure
|
| 35 |
+
|
| 36 |
+
### Training hyperparameters
|
| 37 |
+
|
| 38 |
+
The following hyperparameters were used during training:
|
| 39 |
+
- learning_rate: 5e-05
|
| 40 |
+
- train_batch_size: 16
|
| 41 |
+
- eval_batch_size: 16
|
| 42 |
+
- seed: 42
|
| 43 |
+
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
| 44 |
+
- lr_scheduler_type: linear
|
| 45 |
+
- lr_scheduler_warmup_steps: 190
|
| 46 |
+
- training_steps: 19000
|
| 47 |
+
- mixed_precision_training: Native AMP
|
| 48 |
+
|
| 49 |
+
### Training results
|
| 50 |
+
|
| 51 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 52 |
+
|:-------------:|:------:|:-----:|:---------------:|:--------:|
|
| 53 |
+
| 5.579 | 0.1024 | 200 | 4.7677 | 0.3189 |
|
| 54 |
+
| 4.7716 | 0.2048 | 400 | 4.3385 | 0.3544 |
|
| 55 |
+
| 4.5162 | 0.3072 | 600 | 4.1772 | 0.3593 |
|
| 56 |
+
| 4.4056 | 0.4096 | 800 | 4.0754 | 0.3693 |
|
| 57 |
+
| 4.3138 | 0.5120 | 1000 | 4.0143 | 0.3626 |
|
| 58 |
+
| 4.2148 | 0.6144 | 1200 | 3.9601 | 0.3554 |
|
| 59 |
+
| 4.1925 | 0.7168 | 1400 | 3.9019 | 0.3723 |
|
| 60 |
+
| 4.0293 | 0.8193 | 1600 | 3.8579 | 0.3749 |
|
| 61 |
+
| 3.9407 | 0.9217 | 1800 | 3.8101 | 0.3782 |
|
| 62 |
+
| 3.8371 | 1.0241 | 2000 | 3.7870 | 0.3721 |
|
| 63 |
+
| 3.0659 | 2.0481 | 4000 | 3.4672 | 0.4085 |
|
| 64 |
+
| 2.6866 | 3.0722 | 6000 | 3.2850 | 0.4316 |
|
| 65 |
+
| 2.5063 | 4.0963 | 8000 | 3.1963 | 0.4372 |
|
| 66 |
+
| 2.4139 | 5.1203 | 10000 | 3.1406 | 0.4442 |
|
| 67 |
+
| 2.3246 | 6.1444 | 12000 | 3.1152 | 0.4484 |
|
| 68 |
+
| 2.3111 | 7.1685 | 14000 | 3.0879 | 0.4489 |
|
| 69 |
+
| 2.2761 | 8.1925 | 16000 | 3.0668 | 0.4542 |
|
| 70 |
+
| 2.2231 | 9.2166 | 18000 | 3.0695 | 0.4517 |
|
| 71 |
+
|
| 72 |
+
|
| 73 |
+
### Framework versions
|
| 74 |
+
|
| 75 |
+
- Transformers 4.50.3
|
| 76 |
+
- Pytorch 2.7.1+cu126
|
| 77 |
+
- Datasets 3.6.0
|
| 78 |
+
- Tokenizers 0.21.4
|
all_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 9.728622631848438,
|
| 3 |
+
"eval_accuracy": 0.45214562944577746,
|
| 4 |
+
"eval_loss": 3.0628035068511963,
|
| 5 |
+
"eval_runtime": 95.1631,
|
| 6 |
+
"eval_samples": 33544,
|
| 7 |
+
"eval_samples_per_second": 352.49,
|
| 8 |
+
"eval_steps_per_second": 22.036,
|
| 9 |
+
"perplexity": 21.38743300660324,
|
| 10 |
+
"total_flos": 7.9413964701696e+16,
|
| 11 |
+
"train_loss": 2.775124670731394,
|
| 12 |
+
"train_runtime": 3271.669,
|
| 13 |
+
"train_samples": 31240,
|
| 14 |
+
"train_samples_per_second": 92.919,
|
| 15 |
+
"train_steps_per_second": 5.807
|
| 16 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_function": "gelu_new",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"GPT2LMHeadModel"
|
| 5 |
+
],
|
| 6 |
+
"attn_pdrop": 0.1,
|
| 7 |
+
"bos_token_id": 1,
|
| 8 |
+
"embd_pdrop": 0.1,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"initializer_range": 0.02,
|
| 11 |
+
"layer_norm_epsilon": 1e-05,
|
| 12 |
+
"model_type": "gpt2",
|
| 13 |
+
"n_embd": 768,
|
| 14 |
+
"n_head": 12,
|
| 15 |
+
"n_inner": null,
|
| 16 |
+
"n_layer": 12,
|
| 17 |
+
"n_positions": 1024,
|
| 18 |
+
"reorder_and_upcast_attn": false,
|
| 19 |
+
"resid_pdrop": 0.1,
|
| 20 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 21 |
+
"scale_attn_weights": true,
|
| 22 |
+
"summary_activation": null,
|
| 23 |
+
"summary_first_dropout": 0.1,
|
| 24 |
+
"summary_proj_to_labels": true,
|
| 25 |
+
"summary_type": "cls_index",
|
| 26 |
+
"summary_use_proj": true,
|
| 27 |
+
"torch_dtype": "float32",
|
| 28 |
+
"transformers_version": "4.50.3",
|
| 29 |
+
"use_cache": true,
|
| 30 |
+
"vocab_size": 16384
|
| 31 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 9.728622631848438,
|
| 3 |
+
"eval_accuracy": 0.45214562944577746,
|
| 4 |
+
"eval_loss": 3.0628035068511963,
|
| 5 |
+
"eval_runtime": 95.1631,
|
| 6 |
+
"eval_samples": 33544,
|
| 7 |
+
"eval_samples_per_second": 352.49,
|
| 8 |
+
"eval_steps_per_second": 22.036,
|
| 9 |
+
"perplexity": 21.38743300660324
|
| 10 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
+
"transformers_version": "4.50.3"
|
| 6 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f6ef61138a105c85ceffe6b6f43295e1227d2edb594f4b906933ab8302486203
|
| 3 |
+
size 393716352
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": {
|
| 17 |
+
"content": "<pad>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"unk_token": {
|
| 24 |
+
"content": "<unk>",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
}
|
| 30 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,47 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"3": {
|
| 28 |
+
"content": "<pad>",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
}
|
| 35 |
+
},
|
| 36 |
+
"bos_token": "<s>",
|
| 37 |
+
"clean_up_tokenization_spaces": false,
|
| 38 |
+
"cls_token": null,
|
| 39 |
+
"eos_token": "</s>",
|
| 40 |
+
"extra_special_tokens": {},
|
| 41 |
+
"mask_token": null,
|
| 42 |
+
"model_max_length": 1024,
|
| 43 |
+
"pad_token": "<pad>",
|
| 44 |
+
"sep_token": null,
|
| 45 |
+
"tokenizer_class": "PreTrainedTokenizer",
|
| 46 |
+
"unk_token": "<unk>"
|
| 47 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 9.728622631848438,
|
| 3 |
+
"total_flos": 7.9413964701696e+16,
|
| 4 |
+
"train_loss": 2.775124670731394,
|
| 5 |
+
"train_runtime": 3271.669,
|
| 6 |
+
"train_samples": 31240,
|
| 7 |
+
"train_samples_per_second": 92.919,
|
| 8 |
+
"train_steps_per_second": 5.807
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1535 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 9.728622631848438,
|
| 6 |
+
"eval_steps": 99999999,
|
| 7 |
+
"global_step": 19000,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.051203277009728626,
|
| 14 |
+
"grad_norm": 4.247892379760742,
|
| 15 |
+
"learning_rate": 2.6315789473684212e-05,
|
| 16 |
+
"loss": 7.4934,
|
| 17 |
+
"step": 100
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.10240655401945725,
|
| 21 |
+
"grad_norm": 3.517791748046875,
|
| 22 |
+
"learning_rate": 4.997341839447103e-05,
|
| 23 |
+
"loss": 5.579,
|
| 24 |
+
"step": 200
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.10240655401945725,
|
| 28 |
+
"eval_accuracy": 0.31893968280934165,
|
| 29 |
+
"eval_loss": 4.767695903778076,
|
| 30 |
+
"eval_runtime": 94.8557,
|
| 31 |
+
"eval_samples_per_second": 353.632,
|
| 32 |
+
"eval_steps_per_second": 22.107,
|
| 33 |
+
"step": 200
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.15360983102918588,
|
| 37 |
+
"grad_norm": 2.3053529262542725,
|
| 38 |
+
"learning_rate": 4.970760233918128e-05,
|
| 39 |
+
"loss": 4.9703,
|
| 40 |
+
"step": 300
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 0.2048131080389145,
|
| 44 |
+
"grad_norm": 2.0182411670684814,
|
| 45 |
+
"learning_rate": 4.944178628389155e-05,
|
| 46 |
+
"loss": 4.7716,
|
| 47 |
+
"step": 400
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 0.2048131080389145,
|
| 51 |
+
"eval_accuracy": 0.354354102424925,
|
| 52 |
+
"eval_loss": 4.33852481842041,
|
| 53 |
+
"eval_runtime": 95.6535,
|
| 54 |
+
"eval_samples_per_second": 350.682,
|
| 55 |
+
"eval_steps_per_second": 21.923,
|
| 56 |
+
"step": 400
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.2560163850486431,
|
| 60 |
+
"grad_norm": 1.6917961835861206,
|
| 61 |
+
"learning_rate": 4.917597022860181e-05,
|
| 62 |
+
"loss": 4.5491,
|
| 63 |
+
"step": 500
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.30721966205837176,
|
| 67 |
+
"grad_norm": 1.3638228178024292,
|
| 68 |
+
"learning_rate": 4.8910154173312074e-05,
|
| 69 |
+
"loss": 4.5162,
|
| 70 |
+
"step": 600
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 0.30721966205837176,
|
| 74 |
+
"eval_accuracy": 0.35930259313000934,
|
| 75 |
+
"eval_loss": 4.177200794219971,
|
| 76 |
+
"eval_runtime": 95.6652,
|
| 77 |
+
"eval_samples_per_second": 350.64,
|
| 78 |
+
"eval_steps_per_second": 21.92,
|
| 79 |
+
"step": 600
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.35842293906810035,
|
| 83 |
+
"grad_norm": 1.5296661853790283,
|
| 84 |
+
"learning_rate": 4.8644338118022334e-05,
|
| 85 |
+
"loss": 4.5039,
|
| 86 |
+
"step": 700
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.409626216077829,
|
| 90 |
+
"grad_norm": 1.1248077154159546,
|
| 91 |
+
"learning_rate": 4.837852206273259e-05,
|
| 92 |
+
"loss": 4.4056,
|
| 93 |
+
"step": 800
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.409626216077829,
|
| 97 |
+
"eval_accuracy": 0.36928340870045734,
|
| 98 |
+
"eval_loss": 4.075441837310791,
|
| 99 |
+
"eval_runtime": 96.0011,
|
| 100 |
+
"eval_samples_per_second": 349.412,
|
| 101 |
+
"eval_steps_per_second": 21.843,
|
| 102 |
+
"step": 800
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 0.4608294930875576,
|
| 106 |
+
"grad_norm": 1.6162877082824707,
|
| 107 |
+
"learning_rate": 4.811270600744285e-05,
|
| 108 |
+
"loss": 4.3569,
|
| 109 |
+
"step": 900
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.5120327700972862,
|
| 113 |
+
"grad_norm": 1.6072983741760254,
|
| 114 |
+
"learning_rate": 4.784688995215311e-05,
|
| 115 |
+
"loss": 4.3138,
|
| 116 |
+
"step": 1000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.5120327700972862,
|
| 120 |
+
"eval_accuracy": 0.3626070708659433,
|
| 121 |
+
"eval_loss": 4.014294624328613,
|
| 122 |
+
"eval_runtime": 95.7344,
|
| 123 |
+
"eval_samples_per_second": 350.386,
|
| 124 |
+
"eval_steps_per_second": 21.904,
|
| 125 |
+
"step": 1000
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 0.5632360471070148,
|
| 129 |
+
"grad_norm": 1.4564248323440552,
|
| 130 |
+
"learning_rate": 4.758107389686337e-05,
|
| 131 |
+
"loss": 4.2611,
|
| 132 |
+
"step": 1100
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 0.6144393241167435,
|
| 136 |
+
"grad_norm": 1.2788530588150024,
|
| 137 |
+
"learning_rate": 4.731525784157364e-05,
|
| 138 |
+
"loss": 4.2148,
|
| 139 |
+
"step": 1200
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 0.6144393241167435,
|
| 143 |
+
"eval_accuracy": 0.35539686636426476,
|
| 144 |
+
"eval_loss": 3.9600954055786133,
|
| 145 |
+
"eval_runtime": 95.926,
|
| 146 |
+
"eval_samples_per_second": 349.686,
|
| 147 |
+
"eval_steps_per_second": 21.861,
|
| 148 |
+
"step": 1200
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 0.6656426011264721,
|
| 152 |
+
"grad_norm": 1.3278919458389282,
|
| 153 |
+
"learning_rate": 4.7049441786283896e-05,
|
| 154 |
+
"loss": 4.2095,
|
| 155 |
+
"step": 1300
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 0.7168458781362007,
|
| 159 |
+
"grad_norm": 1.6991857290267944,
|
| 160 |
+
"learning_rate": 4.678362573099415e-05,
|
| 161 |
+
"loss": 4.1925,
|
| 162 |
+
"step": 1400
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 0.7168458781362007,
|
| 166 |
+
"eval_accuracy": 0.3723133981106336,
|
| 167 |
+
"eval_loss": 3.901859998703003,
|
| 168 |
+
"eval_runtime": 95.2539,
|
| 169 |
+
"eval_samples_per_second": 352.154,
|
| 170 |
+
"eval_steps_per_second": 22.015,
|
| 171 |
+
"step": 1400
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.7680491551459293,
|
| 175 |
+
"grad_norm": 1.7155863046646118,
|
| 176 |
+
"learning_rate": 4.6517809675704415e-05,
|
| 177 |
+
"loss": 4.1556,
|
| 178 |
+
"step": 1500
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.819252432155658,
|
| 182 |
+
"grad_norm": 1.6039901971817017,
|
| 183 |
+
"learning_rate": 4.6251993620414674e-05,
|
| 184 |
+
"loss": 4.0293,
|
| 185 |
+
"step": 1600
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.819252432155658,
|
| 189 |
+
"eval_accuracy": 0.37494784430112066,
|
| 190 |
+
"eval_loss": 3.857882499694824,
|
| 191 |
+
"eval_runtime": 95.7224,
|
| 192 |
+
"eval_samples_per_second": 350.43,
|
| 193 |
+
"eval_steps_per_second": 21.907,
|
| 194 |
+
"step": 1600
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.8704557091653866,
|
| 198 |
+
"grad_norm": 1.6892971992492676,
|
| 199 |
+
"learning_rate": 4.598617756512494e-05,
|
| 200 |
+
"loss": 3.939,
|
| 201 |
+
"step": 1700
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 0.9216589861751152,
|
| 205 |
+
"grad_norm": 1.5192146301269531,
|
| 206 |
+
"learning_rate": 4.57203615098352e-05,
|
| 207 |
+
"loss": 3.9407,
|
| 208 |
+
"step": 1800
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 0.9216589861751152,
|
| 212 |
+
"eval_accuracy": 0.37819625757774467,
|
| 213 |
+
"eval_loss": 3.8101186752319336,
|
| 214 |
+
"eval_runtime": 95.9641,
|
| 215 |
+
"eval_samples_per_second": 349.547,
|
| 216 |
+
"eval_steps_per_second": 21.852,
|
| 217 |
+
"step": 1800
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 0.9728622631848438,
|
| 221 |
+
"grad_norm": 1.6878364086151123,
|
| 222 |
+
"learning_rate": 4.545454545454546e-05,
|
| 223 |
+
"loss": 3.9125,
|
| 224 |
+
"step": 1900
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 1.0240655401945724,
|
| 228 |
+
"grad_norm": 1.6546193361282349,
|
| 229 |
+
"learning_rate": 4.518872939925572e-05,
|
| 230 |
+
"loss": 3.8371,
|
| 231 |
+
"step": 2000
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 1.0240655401945724,
|
| 235 |
+
"eval_accuracy": 0.37209030706755225,
|
| 236 |
+
"eval_loss": 3.787036180496216,
|
| 237 |
+
"eval_runtime": 95.7788,
|
| 238 |
+
"eval_samples_per_second": 350.224,
|
| 239 |
+
"eval_steps_per_second": 21.894,
|
| 240 |
+
"step": 2000
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 1.075268817204301,
|
| 244 |
+
"grad_norm": 1.744416356086731,
|
| 245 |
+
"learning_rate": 4.492291334396598e-05,
|
| 246 |
+
"loss": 3.8068,
|
| 247 |
+
"step": 2100
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 1.1264720942140296,
|
| 251 |
+
"grad_norm": 1.6214791536331177,
|
| 252 |
+
"learning_rate": 4.4657097288676236e-05,
|
| 253 |
+
"loss": 3.7642,
|
| 254 |
+
"step": 2200
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 1.1776753712237582,
|
| 258 |
+
"grad_norm": 1.5377932786941528,
|
| 259 |
+
"learning_rate": 4.43912812333865e-05,
|
| 260 |
+
"loss": 3.7599,
|
| 261 |
+
"step": 2300
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 1.228878648233487,
|
| 265 |
+
"grad_norm": 1.8464128971099854,
|
| 266 |
+
"learning_rate": 4.412546517809676e-05,
|
| 267 |
+
"loss": 3.6853,
|
| 268 |
+
"step": 2400
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 1.2800819252432156,
|
| 272 |
+
"grad_norm": 1.901739239692688,
|
| 273 |
+
"learning_rate": 4.3859649122807014e-05,
|
| 274 |
+
"loss": 3.6537,
|
| 275 |
+
"step": 2500
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 1.3312852022529442,
|
| 279 |
+
"grad_norm": 1.936339020729065,
|
| 280 |
+
"learning_rate": 4.359383306751728e-05,
|
| 281 |
+
"loss": 3.6377,
|
| 282 |
+
"step": 2600
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 1.3824884792626728,
|
| 286 |
+
"grad_norm": 1.8641879558563232,
|
| 287 |
+
"learning_rate": 4.332801701222754e-05,
|
| 288 |
+
"loss": 3.5523,
|
| 289 |
+
"step": 2700
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 1.4336917562724014,
|
| 293 |
+
"grad_norm": 2.087099075317383,
|
| 294 |
+
"learning_rate": 4.3062200956937806e-05,
|
| 295 |
+
"loss": 3.5007,
|
| 296 |
+
"step": 2800
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 1.48489503328213,
|
| 300 |
+
"grad_norm": 1.7824835777282715,
|
| 301 |
+
"learning_rate": 4.2796384901648065e-05,
|
| 302 |
+
"loss": 3.4422,
|
| 303 |
+
"step": 2900
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 1.5360983102918588,
|
| 307 |
+
"grad_norm": 2.0298142433166504,
|
| 308 |
+
"learning_rate": 4.253056884635832e-05,
|
| 309 |
+
"loss": 3.4254,
|
| 310 |
+
"step": 3000
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 1.5873015873015874,
|
| 314 |
+
"grad_norm": 2.227325439453125,
|
| 315 |
+
"learning_rate": 4.2264752791068584e-05,
|
| 316 |
+
"loss": 3.4087,
|
| 317 |
+
"step": 3100
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 1.638504864311316,
|
| 321 |
+
"grad_norm": 2.2700014114379883,
|
| 322 |
+
"learning_rate": 4.199893673577884e-05,
|
| 323 |
+
"loss": 3.4023,
|
| 324 |
+
"step": 3200
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 1.6897081413210446,
|
| 328 |
+
"grad_norm": 2.290431022644043,
|
| 329 |
+
"learning_rate": 4.17331206804891e-05,
|
| 330 |
+
"loss": 3.3579,
|
| 331 |
+
"step": 3300
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 1.7409114183307732,
|
| 335 |
+
"grad_norm": 2.5916295051574707,
|
| 336 |
+
"learning_rate": 4.146730462519937e-05,
|
| 337 |
+
"loss": 3.2602,
|
| 338 |
+
"step": 3400
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 1.7921146953405018,
|
| 342 |
+
"grad_norm": 2.0242862701416016,
|
| 343 |
+
"learning_rate": 4.120148856990963e-05,
|
| 344 |
+
"loss": 3.1941,
|
| 345 |
+
"step": 3500
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 1.8433179723502304,
|
| 349 |
+
"grad_norm": 1.8922170400619507,
|
| 350 |
+
"learning_rate": 4.093567251461988e-05,
|
| 351 |
+
"loss": 3.1774,
|
| 352 |
+
"step": 3600
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 1.894521249359959,
|
| 356 |
+
"grad_norm": 2.2697248458862305,
|
| 357 |
+
"learning_rate": 4.0669856459330146e-05,
|
| 358 |
+
"loss": 3.2001,
|
| 359 |
+
"step": 3700
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 1.9457245263696876,
|
| 363 |
+
"grad_norm": 2.68994402885437,
|
| 364 |
+
"learning_rate": 4.0404040404040405e-05,
|
| 365 |
+
"loss": 3.1232,
|
| 366 |
+
"step": 3800
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 1.9969278033794162,
|
| 370 |
+
"grad_norm": 2.16379451751709,
|
| 371 |
+
"learning_rate": 4.0138224348750665e-05,
|
| 372 |
+
"loss": 3.101,
|
| 373 |
+
"step": 3900
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 2.048131080389145,
|
| 377 |
+
"grad_norm": 1.908273458480835,
|
| 378 |
+
"learning_rate": 3.987240829346093e-05,
|
| 379 |
+
"loss": 3.0659,
|
| 380 |
+
"step": 4000
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 2.048131080389145,
|
| 384 |
+
"eval_accuracy": 0.4085060110901451,
|
| 385 |
+
"eval_loss": 3.467172384262085,
|
| 386 |
+
"eval_runtime": 95.3375,
|
| 387 |
+
"eval_samples_per_second": 351.845,
|
| 388 |
+
"eval_steps_per_second": 21.996,
|
| 389 |
+
"step": 4000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 2.0993343573988734,
|
| 393 |
+
"grad_norm": 2.217175006866455,
|
| 394 |
+
"learning_rate": 3.960659223817118e-05,
|
| 395 |
+
"loss": 3.009,
|
| 396 |
+
"step": 4100
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 2.150537634408602,
|
| 400 |
+
"grad_norm": 2.744868755340576,
|
| 401 |
+
"learning_rate": 3.934077618288145e-05,
|
| 402 |
+
"loss": 3.0265,
|
| 403 |
+
"step": 4200
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 2.2017409114183306,
|
| 407 |
+
"grad_norm": 2.3460192680358887,
|
| 408 |
+
"learning_rate": 3.907496012759171e-05,
|
| 409 |
+
"loss": 2.9379,
|
| 410 |
+
"step": 4300
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 2.252944188428059,
|
| 414 |
+
"grad_norm": 2.120131015777588,
|
| 415 |
+
"learning_rate": 3.880914407230197e-05,
|
| 416 |
+
"loss": 2.9437,
|
| 417 |
+
"step": 4400
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 2.3041474654377883,
|
| 421 |
+
"grad_norm": 2.3475334644317627,
|
| 422 |
+
"learning_rate": 3.8543328017012234e-05,
|
| 423 |
+
"loss": 2.933,
|
| 424 |
+
"step": 4500
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 2.3553507424475164,
|
| 428 |
+
"grad_norm": 2.450538396835327,
|
| 429 |
+
"learning_rate": 3.8277511961722486e-05,
|
| 430 |
+
"loss": 2.9619,
|
| 431 |
+
"step": 4600
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 2.4065540194572455,
|
| 435 |
+
"grad_norm": 2.027897596359253,
|
| 436 |
+
"learning_rate": 3.8011695906432746e-05,
|
| 437 |
+
"loss": 2.9179,
|
| 438 |
+
"step": 4700
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 2.457757296466974,
|
| 442 |
+
"grad_norm": 2.3188517093658447,
|
| 443 |
+
"learning_rate": 3.774587985114301e-05,
|
| 444 |
+
"loss": 2.8655,
|
| 445 |
+
"step": 4800
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 2.5089605734767026,
|
| 449 |
+
"grad_norm": 1.909847617149353,
|
| 450 |
+
"learning_rate": 3.748006379585327e-05,
|
| 451 |
+
"loss": 2.8649,
|
| 452 |
+
"step": 4900
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 2.5601638504864312,
|
| 456 |
+
"grad_norm": 1.8613382577896118,
|
| 457 |
+
"learning_rate": 3.721424774056353e-05,
|
| 458 |
+
"loss": 2.8555,
|
| 459 |
+
"step": 5000
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 2.61136712749616,
|
| 463 |
+
"grad_norm": 1.7689636945724487,
|
| 464 |
+
"learning_rate": 3.6948431685273796e-05,
|
| 465 |
+
"loss": 2.8957,
|
| 466 |
+
"step": 5100
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 2.6625704045058884,
|
| 470 |
+
"grad_norm": 1.8662793636322021,
|
| 471 |
+
"learning_rate": 3.668261562998405e-05,
|
| 472 |
+
"loss": 2.8132,
|
| 473 |
+
"step": 5200
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 2.713773681515617,
|
| 477 |
+
"grad_norm": 1.9725576639175415,
|
| 478 |
+
"learning_rate": 3.6416799574694315e-05,
|
| 479 |
+
"loss": 2.8121,
|
| 480 |
+
"step": 5300
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 2.7649769585253456,
|
| 484 |
+
"grad_norm": 1.7306541204452515,
|
| 485 |
+
"learning_rate": 3.6150983519404574e-05,
|
| 486 |
+
"loss": 2.8154,
|
| 487 |
+
"step": 5400
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 2.8161802355350742,
|
| 491 |
+
"grad_norm": 1.824539065361023,
|
| 492 |
+
"learning_rate": 3.5885167464114834e-05,
|
| 493 |
+
"loss": 2.776,
|
| 494 |
+
"step": 5500
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 2.867383512544803,
|
| 498 |
+
"grad_norm": 1.9677956104278564,
|
| 499 |
+
"learning_rate": 3.56193514088251e-05,
|
| 500 |
+
"loss": 2.8155,
|
| 501 |
+
"step": 5600
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 2.9185867895545314,
|
| 505 |
+
"grad_norm": 1.9770982265472412,
|
| 506 |
+
"learning_rate": 3.535353535353535e-05,
|
| 507 |
+
"loss": 2.7638,
|
| 508 |
+
"step": 5700
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 2.96979006656426,
|
| 512 |
+
"grad_norm": 2.1280829906463623,
|
| 513 |
+
"learning_rate": 3.508771929824561e-05,
|
| 514 |
+
"loss": 2.7771,
|
| 515 |
+
"step": 5800
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 3.0209933435739886,
|
| 519 |
+
"grad_norm": 1.7677674293518066,
|
| 520 |
+
"learning_rate": 3.482190324295588e-05,
|
| 521 |
+
"loss": 2.7382,
|
| 522 |
+
"step": 5900
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 3.0721966205837172,
|
| 526 |
+
"grad_norm": 1.6372385025024414,
|
| 527 |
+
"learning_rate": 3.455608718766614e-05,
|
| 528 |
+
"loss": 2.6866,
|
| 529 |
+
"step": 6000
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 3.0721966205837172,
|
| 533 |
+
"eval_accuracy": 0.4316442393272171,
|
| 534 |
+
"eval_loss": 3.2849764823913574,
|
| 535 |
+
"eval_runtime": 95.1067,
|
| 536 |
+
"eval_samples_per_second": 352.699,
|
| 537 |
+
"eval_steps_per_second": 22.049,
|
| 538 |
+
"step": 6000
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 3.123399897593446,
|
| 542 |
+
"grad_norm": 1.9155375957489014,
|
| 543 |
+
"learning_rate": 3.4290271132376396e-05,
|
| 544 |
+
"loss": 2.6794,
|
| 545 |
+
"step": 6100
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 3.1746031746031744,
|
| 549 |
+
"grad_norm": 1.5887531042099,
|
| 550 |
+
"learning_rate": 3.402445507708666e-05,
|
| 551 |
+
"loss": 2.6785,
|
| 552 |
+
"step": 6200
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 3.225806451612903,
|
| 556 |
+
"grad_norm": 1.7831392288208008,
|
| 557 |
+
"learning_rate": 3.3758639021796915e-05,
|
| 558 |
+
"loss": 2.682,
|
| 559 |
+
"step": 6300
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 3.277009728622632,
|
| 563 |
+
"grad_norm": 1.6347112655639648,
|
| 564 |
+
"learning_rate": 3.349282296650718e-05,
|
| 565 |
+
"loss": 2.6161,
|
| 566 |
+
"step": 6400
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 3.32821300563236,
|
| 570 |
+
"grad_norm": 1.774921178817749,
|
| 571 |
+
"learning_rate": 3.322700691121744e-05,
|
| 572 |
+
"loss": 2.6727,
|
| 573 |
+
"step": 6500
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 3.3794162826420893,
|
| 577 |
+
"grad_norm": 1.7162151336669922,
|
| 578 |
+
"learning_rate": 3.29611908559277e-05,
|
| 579 |
+
"loss": 2.692,
|
| 580 |
+
"step": 6600
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"epoch": 3.430619559651818,
|
| 584 |
+
"grad_norm": 1.7720268964767456,
|
| 585 |
+
"learning_rate": 3.2695374800637965e-05,
|
| 586 |
+
"loss": 2.6795,
|
| 587 |
+
"step": 6700
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 3.4818228366615465,
|
| 591 |
+
"grad_norm": 1.7839425802230835,
|
| 592 |
+
"learning_rate": 3.242955874534822e-05,
|
| 593 |
+
"loss": 2.6487,
|
| 594 |
+
"step": 6800
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 3.533026113671275,
|
| 598 |
+
"grad_norm": 1.7593103647232056,
|
| 599 |
+
"learning_rate": 3.216374269005848e-05,
|
| 600 |
+
"loss": 2.6821,
|
| 601 |
+
"step": 6900
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 3.5842293906810037,
|
| 605 |
+
"grad_norm": 1.838611125946045,
|
| 606 |
+
"learning_rate": 3.189792663476874e-05,
|
| 607 |
+
"loss": 2.6547,
|
| 608 |
+
"step": 7000
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 3.6354326676907323,
|
| 612 |
+
"grad_norm": 1.5734059810638428,
|
| 613 |
+
"learning_rate": 3.1632110579479e-05,
|
| 614 |
+
"loss": 2.6343,
|
| 615 |
+
"step": 7100
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 3.686635944700461,
|
| 619 |
+
"grad_norm": 1.7309696674346924,
|
| 620 |
+
"learning_rate": 3.136629452418926e-05,
|
| 621 |
+
"loss": 2.6379,
|
| 622 |
+
"step": 7200
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 3.7378392217101895,
|
| 626 |
+
"grad_norm": 1.837202787399292,
|
| 627 |
+
"learning_rate": 3.110047846889952e-05,
|
| 628 |
+
"loss": 2.6094,
|
| 629 |
+
"step": 7300
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 3.789042498719918,
|
| 633 |
+
"grad_norm": 1.6872186660766602,
|
| 634 |
+
"learning_rate": 3.083466241360978e-05,
|
| 635 |
+
"loss": 2.5951,
|
| 636 |
+
"step": 7400
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 3.8402457757296466,
|
| 640 |
+
"grad_norm": 1.7772648334503174,
|
| 641 |
+
"learning_rate": 3.056884635832004e-05,
|
| 642 |
+
"loss": 2.5935,
|
| 643 |
+
"step": 7500
|
| 644 |
+
},
|
| 645 |
+
{
|
| 646 |
+
"epoch": 3.8914490527393752,
|
| 647 |
+
"grad_norm": 1.7157689332962036,
|
| 648 |
+
"learning_rate": 3.0303030303030306e-05,
|
| 649 |
+
"loss": 2.6076,
|
| 650 |
+
"step": 7600
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 3.942652329749104,
|
| 654 |
+
"grad_norm": 1.7484747171401978,
|
| 655 |
+
"learning_rate": 3.0037214247740565e-05,
|
| 656 |
+
"loss": 2.6294,
|
| 657 |
+
"step": 7700
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 3.9938556067588324,
|
| 661 |
+
"grad_norm": 1.6567468643188477,
|
| 662 |
+
"learning_rate": 2.9771398192450828e-05,
|
| 663 |
+
"loss": 2.603,
|
| 664 |
+
"step": 7800
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 4.0450588837685615,
|
| 668 |
+
"grad_norm": 1.6986713409423828,
|
| 669 |
+
"learning_rate": 2.9505582137161087e-05,
|
| 670 |
+
"loss": 2.5767,
|
| 671 |
+
"step": 7900
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 4.09626216077829,
|
| 675 |
+
"grad_norm": 1.9400880336761475,
|
| 676 |
+
"learning_rate": 2.9239766081871346e-05,
|
| 677 |
+
"loss": 2.5063,
|
| 678 |
+
"step": 8000
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 4.09626216077829,
|
| 682 |
+
"eval_accuracy": 0.4372182483806064,
|
| 683 |
+
"eval_loss": 3.1963303089141846,
|
| 684 |
+
"eval_runtime": 95.395,
|
| 685 |
+
"eval_samples_per_second": 351.633,
|
| 686 |
+
"eval_steps_per_second": 21.982,
|
| 687 |
+
"step": 8000
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 4.147465437788019,
|
| 691 |
+
"grad_norm": 1.7278245687484741,
|
| 692 |
+
"learning_rate": 2.8976608187134503e-05,
|
| 693 |
+
"loss": 2.5088,
|
| 694 |
+
"step": 8100
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 4.198668714797747,
|
| 698 |
+
"grad_norm": 1.778096318244934,
|
| 699 |
+
"learning_rate": 2.8710792131844765e-05,
|
| 700 |
+
"loss": 2.5284,
|
| 701 |
+
"step": 8200
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 4.249871991807476,
|
| 705 |
+
"grad_norm": 1.990983009338379,
|
| 706 |
+
"learning_rate": 2.8444976076555025e-05,
|
| 707 |
+
"loss": 2.5181,
|
| 708 |
+
"step": 8300
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 4.301075268817204,
|
| 712 |
+
"grad_norm": 1.577682375907898,
|
| 713 |
+
"learning_rate": 2.8179160021265287e-05,
|
| 714 |
+
"loss": 2.4916,
|
| 715 |
+
"step": 8400
|
| 716 |
+
},
|
| 717 |
+
{
|
| 718 |
+
"epoch": 4.352278545826933,
|
| 719 |
+
"grad_norm": 1.6727075576782227,
|
| 720 |
+
"learning_rate": 2.7913343965975547e-05,
|
| 721 |
+
"loss": 2.5346,
|
| 722 |
+
"step": 8500
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 4.403481822836661,
|
| 726 |
+
"grad_norm": 1.6889344453811646,
|
| 727 |
+
"learning_rate": 2.7647527910685806e-05,
|
| 728 |
+
"loss": 2.5164,
|
| 729 |
+
"step": 8600
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 4.45468509984639,
|
| 733 |
+
"grad_norm": 1.763092041015625,
|
| 734 |
+
"learning_rate": 2.738171185539607e-05,
|
| 735 |
+
"loss": 2.5073,
|
| 736 |
+
"step": 8700
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 4.505888376856118,
|
| 740 |
+
"grad_norm": 1.6232457160949707,
|
| 741 |
+
"learning_rate": 2.7115895800106328e-05,
|
| 742 |
+
"loss": 2.515,
|
| 743 |
+
"step": 8800
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 4.5570916538658475,
|
| 747 |
+
"grad_norm": 1.7539464235305786,
|
| 748 |
+
"learning_rate": 2.6850079744816587e-05,
|
| 749 |
+
"loss": 2.538,
|
| 750 |
+
"step": 8900
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 4.6082949308755765,
|
| 754 |
+
"grad_norm": 1.729683518409729,
|
| 755 |
+
"learning_rate": 2.658426368952685e-05,
|
| 756 |
+
"loss": 2.5301,
|
| 757 |
+
"step": 9000
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 4.659498207885305,
|
| 761 |
+
"grad_norm": 1.753625750541687,
|
| 762 |
+
"learning_rate": 2.631844763423711e-05,
|
| 763 |
+
"loss": 2.5267,
|
| 764 |
+
"step": 9100
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 4.710701484895033,
|
| 768 |
+
"grad_norm": 1.7453761100769043,
|
| 769 |
+
"learning_rate": 2.605263157894737e-05,
|
| 770 |
+
"loss": 2.5207,
|
| 771 |
+
"step": 9200
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 4.761904761904762,
|
| 775 |
+
"grad_norm": 1.9901796579360962,
|
| 776 |
+
"learning_rate": 2.578681552365763e-05,
|
| 777 |
+
"loss": 2.4987,
|
| 778 |
+
"step": 9300
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 4.813108038914491,
|
| 782 |
+
"grad_norm": 1.6312249898910522,
|
| 783 |
+
"learning_rate": 2.552099946836789e-05,
|
| 784 |
+
"loss": 2.4943,
|
| 785 |
+
"step": 9400
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 4.864311315924219,
|
| 789 |
+
"grad_norm": 1.7069945335388184,
|
| 790 |
+
"learning_rate": 2.525518341307815e-05,
|
| 791 |
+
"loss": 2.4666,
|
| 792 |
+
"step": 9500
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 4.915514592933948,
|
| 796 |
+
"grad_norm": 1.7387930154800415,
|
| 797 |
+
"learning_rate": 2.4989367357788412e-05,
|
| 798 |
+
"loss": 2.4932,
|
| 799 |
+
"step": 9600
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 4.966717869943676,
|
| 803 |
+
"grad_norm": 1.6212061643600464,
|
| 804 |
+
"learning_rate": 2.472355130249867e-05,
|
| 805 |
+
"loss": 2.52,
|
| 806 |
+
"step": 9700
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 5.017921146953405,
|
| 810 |
+
"grad_norm": 1.7624274492263794,
|
| 811 |
+
"learning_rate": 2.445773524720893e-05,
|
| 812 |
+
"loss": 2.4938,
|
| 813 |
+
"step": 9800
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 5.0691244239631335,
|
| 817 |
+
"grad_norm": 1.824257254600525,
|
| 818 |
+
"learning_rate": 2.4191919191919194e-05,
|
| 819 |
+
"loss": 2.4275,
|
| 820 |
+
"step": 9900
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 5.1203277009728625,
|
| 824 |
+
"grad_norm": 1.616184115409851,
|
| 825 |
+
"learning_rate": 2.3926103136629456e-05,
|
| 826 |
+
"loss": 2.4139,
|
| 827 |
+
"step": 10000
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 5.1203277009728625,
|
| 831 |
+
"eval_accuracy": 0.4442466080127022,
|
| 832 |
+
"eval_loss": 3.1405584812164307,
|
| 833 |
+
"eval_runtime": 95.5986,
|
| 834 |
+
"eval_samples_per_second": 350.884,
|
| 835 |
+
"eval_steps_per_second": 21.935,
|
| 836 |
+
"step": 10000
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 5.171530977982591,
|
| 840 |
+
"grad_norm": 1.764162302017212,
|
| 841 |
+
"learning_rate": 2.3660287081339712e-05,
|
| 842 |
+
"loss": 2.4878,
|
| 843 |
+
"step": 10100
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 5.22273425499232,
|
| 847 |
+
"grad_norm": 1.6120530366897583,
|
| 848 |
+
"learning_rate": 2.3394471026049975e-05,
|
| 849 |
+
"loss": 2.4391,
|
| 850 |
+
"step": 10200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 5.273937532002048,
|
| 854 |
+
"grad_norm": 1.6949564218521118,
|
| 855 |
+
"learning_rate": 2.3128654970760234e-05,
|
| 856 |
+
"loss": 2.4131,
|
| 857 |
+
"step": 10300
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 5.325140809011777,
|
| 861 |
+
"grad_norm": 1.667853832244873,
|
| 862 |
+
"learning_rate": 2.2862838915470493e-05,
|
| 863 |
+
"loss": 2.4112,
|
| 864 |
+
"step": 10400
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 5.376344086021505,
|
| 868 |
+
"grad_norm": 1.427404522895813,
|
| 869 |
+
"learning_rate": 2.2597022860180756e-05,
|
| 870 |
+
"loss": 2.4242,
|
| 871 |
+
"step": 10500
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 5.427547363031234,
|
| 875 |
+
"grad_norm": 1.6098235845565796,
|
| 876 |
+
"learning_rate": 2.2331206804891015e-05,
|
| 877 |
+
"loss": 2.429,
|
| 878 |
+
"step": 10600
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 5.478750640040962,
|
| 882 |
+
"grad_norm": 1.8480441570281982,
|
| 883 |
+
"learning_rate": 2.2065390749601278e-05,
|
| 884 |
+
"loss": 2.4439,
|
| 885 |
+
"step": 10700
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 5.529953917050691,
|
| 889 |
+
"grad_norm": 1.8656944036483765,
|
| 890 |
+
"learning_rate": 2.1799574694311537e-05,
|
| 891 |
+
"loss": 2.4164,
|
| 892 |
+
"step": 10800
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 5.58115719406042,
|
| 896 |
+
"grad_norm": 1.9880622625350952,
|
| 897 |
+
"learning_rate": 2.1533758639021797e-05,
|
| 898 |
+
"loss": 2.465,
|
| 899 |
+
"step": 10900
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 5.6323604710701485,
|
| 903 |
+
"grad_norm": 1.8432066440582275,
|
| 904 |
+
"learning_rate": 2.126794258373206e-05,
|
| 905 |
+
"loss": 2.4216,
|
| 906 |
+
"step": 11000
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 5.683563748079877,
|
| 910 |
+
"grad_norm": 1.7552499771118164,
|
| 911 |
+
"learning_rate": 2.100212652844232e-05,
|
| 912 |
+
"loss": 2.379,
|
| 913 |
+
"step": 11100
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 5.734767025089606,
|
| 917 |
+
"grad_norm": 1.7043852806091309,
|
| 918 |
+
"learning_rate": 2.0736310473152578e-05,
|
| 919 |
+
"loss": 2.4308,
|
| 920 |
+
"step": 11200
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 5.785970302099335,
|
| 924 |
+
"grad_norm": 1.8842682838439941,
|
| 925 |
+
"learning_rate": 2.047049441786284e-05,
|
| 926 |
+
"loss": 2.412,
|
| 927 |
+
"step": 11300
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 5.837173579109063,
|
| 931 |
+
"grad_norm": 1.6765940189361572,
|
| 932 |
+
"learning_rate": 2.02046783625731e-05,
|
| 933 |
+
"loss": 2.4299,
|
| 934 |
+
"step": 11400
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 5.888376856118792,
|
| 938 |
+
"grad_norm": 1.7585973739624023,
|
| 939 |
+
"learning_rate": 1.993886230728336e-05,
|
| 940 |
+
"loss": 2.4078,
|
| 941 |
+
"step": 11500
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 5.93958013312852,
|
| 945 |
+
"grad_norm": 1.788201928138733,
|
| 946 |
+
"learning_rate": 1.9673046251993622e-05,
|
| 947 |
+
"loss": 2.4186,
|
| 948 |
+
"step": 11600
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 5.990783410138249,
|
| 952 |
+
"grad_norm": 1.6128724813461304,
|
| 953 |
+
"learning_rate": 1.940723019670388e-05,
|
| 954 |
+
"loss": 2.3844,
|
| 955 |
+
"step": 11700
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 6.041986687147977,
|
| 959 |
+
"grad_norm": 1.7053714990615845,
|
| 960 |
+
"learning_rate": 1.9141414141414144e-05,
|
| 961 |
+
"loss": 2.3299,
|
| 962 |
+
"step": 11800
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 6.093189964157706,
|
| 966 |
+
"grad_norm": 2.06257963180542,
|
| 967 |
+
"learning_rate": 1.8875598086124403e-05,
|
| 968 |
+
"loss": 2.3666,
|
| 969 |
+
"step": 11900
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 6.1443932411674345,
|
| 973 |
+
"grad_norm": 1.8419615030288696,
|
| 974 |
+
"learning_rate": 1.8609782030834662e-05,
|
| 975 |
+
"loss": 2.3246,
|
| 976 |
+
"step": 12000
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 6.1443932411674345,
|
| 980 |
+
"eval_accuracy": 0.44839520298251256,
|
| 981 |
+
"eval_loss": 3.115219831466675,
|
| 982 |
+
"eval_runtime": 95.2063,
|
| 983 |
+
"eval_samples_per_second": 352.329,
|
| 984 |
+
"eval_steps_per_second": 22.026,
|
| 985 |
+
"step": 12000
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 6.1955965181771635,
|
| 989 |
+
"grad_norm": 1.4872585535049438,
|
| 990 |
+
"learning_rate": 1.834662413609782e-05,
|
| 991 |
+
"loss": 2.3234,
|
| 992 |
+
"step": 12100
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 6.246799795186892,
|
| 996 |
+
"grad_norm": 1.6765344142913818,
|
| 997 |
+
"learning_rate": 1.808080808080808e-05,
|
| 998 |
+
"loss": 2.353,
|
| 999 |
+
"step": 12200
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 6.298003072196621,
|
| 1003 |
+
"grad_norm": 1.611560583114624,
|
| 1004 |
+
"learning_rate": 1.7814992025518344e-05,
|
| 1005 |
+
"loss": 2.3758,
|
| 1006 |
+
"step": 12300
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 6.349206349206349,
|
| 1010 |
+
"grad_norm": 1.7853387594223022,
|
| 1011 |
+
"learning_rate": 1.75491759702286e-05,
|
| 1012 |
+
"loss": 2.3232,
|
| 1013 |
+
"step": 12400
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 6.400409626216078,
|
| 1017 |
+
"grad_norm": 1.614193320274353,
|
| 1018 |
+
"learning_rate": 1.7283359914938863e-05,
|
| 1019 |
+
"loss": 2.3759,
|
| 1020 |
+
"step": 12500
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 6.451612903225806,
|
| 1024 |
+
"grad_norm": 1.8242110013961792,
|
| 1025 |
+
"learning_rate": 1.7017543859649125e-05,
|
| 1026 |
+
"loss": 2.4106,
|
| 1027 |
+
"step": 12600
|
| 1028 |
+
},
|
| 1029 |
+
{
|
| 1030 |
+
"epoch": 6.502816180235535,
|
| 1031 |
+
"grad_norm": 1.674735188484192,
|
| 1032 |
+
"learning_rate": 1.6751727804359385e-05,
|
| 1033 |
+
"loss": 2.3834,
|
| 1034 |
+
"step": 12700
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 6.554019457245264,
|
| 1038 |
+
"grad_norm": 1.8332616090774536,
|
| 1039 |
+
"learning_rate": 1.6485911749069644e-05,
|
| 1040 |
+
"loss": 2.3631,
|
| 1041 |
+
"step": 12800
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 6.605222734254992,
|
| 1045 |
+
"grad_norm": 1.8917430639266968,
|
| 1046 |
+
"learning_rate": 1.6220095693779903e-05,
|
| 1047 |
+
"loss": 2.3549,
|
| 1048 |
+
"step": 12900
|
| 1049 |
+
},
|
| 1050 |
+
{
|
| 1051 |
+
"epoch": 6.65642601126472,
|
| 1052 |
+
"grad_norm": 1.9428924322128296,
|
| 1053 |
+
"learning_rate": 1.5954279638490166e-05,
|
| 1054 |
+
"loss": 2.3205,
|
| 1055 |
+
"step": 13000
|
| 1056 |
+
},
|
| 1057 |
+
{
|
| 1058 |
+
"epoch": 6.7076292882744495,
|
| 1059 |
+
"grad_norm": 1.639167308807373,
|
| 1060 |
+
"learning_rate": 1.568846358320043e-05,
|
| 1061 |
+
"loss": 2.3646,
|
| 1062 |
+
"step": 13100
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"epoch": 6.7588325652841785,
|
| 1066 |
+
"grad_norm": 1.6185379028320312,
|
| 1067 |
+
"learning_rate": 1.5422647527910685e-05,
|
| 1068 |
+
"loss": 2.3757,
|
| 1069 |
+
"step": 13200
|
| 1070 |
+
},
|
| 1071 |
+
{
|
| 1072 |
+
"epoch": 6.810035842293907,
|
| 1073 |
+
"grad_norm": 1.6808964014053345,
|
| 1074 |
+
"learning_rate": 1.5156831472620947e-05,
|
| 1075 |
+
"loss": 2.3542,
|
| 1076 |
+
"step": 13300
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 6.861239119303636,
|
| 1080 |
+
"grad_norm": 1.4692785739898682,
|
| 1081 |
+
"learning_rate": 1.4891015417331208e-05,
|
| 1082 |
+
"loss": 2.3476,
|
| 1083 |
+
"step": 13400
|
| 1084 |
+
},
|
| 1085 |
+
{
|
| 1086 |
+
"epoch": 6.912442396313364,
|
| 1087 |
+
"grad_norm": 1.6999351978302002,
|
| 1088 |
+
"learning_rate": 1.4625199362041468e-05,
|
| 1089 |
+
"loss": 2.3441,
|
| 1090 |
+
"step": 13500
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"epoch": 6.963645673323093,
|
| 1094 |
+
"grad_norm": 1.7041444778442383,
|
| 1095 |
+
"learning_rate": 1.4359383306751729e-05,
|
| 1096 |
+
"loss": 2.3725,
|
| 1097 |
+
"step": 13600
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 7.014848950332821,
|
| 1101 |
+
"grad_norm": 1.773606538772583,
|
| 1102 |
+
"learning_rate": 1.409356725146199e-05,
|
| 1103 |
+
"loss": 2.3379,
|
| 1104 |
+
"step": 13700
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 7.06605222734255,
|
| 1108 |
+
"grad_norm": 1.7857049703598022,
|
| 1109 |
+
"learning_rate": 1.382775119617225e-05,
|
| 1110 |
+
"loss": 2.3471,
|
| 1111 |
+
"step": 13800
|
| 1112 |
+
},
|
| 1113 |
+
{
|
| 1114 |
+
"epoch": 7.117255504352278,
|
| 1115 |
+
"grad_norm": 1.7172592878341675,
|
| 1116 |
+
"learning_rate": 1.356193514088251e-05,
|
| 1117 |
+
"loss": 2.2639,
|
| 1118 |
+
"step": 13900
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 7.168458781362007,
|
| 1122 |
+
"grad_norm": 2.0828044414520264,
|
| 1123 |
+
"learning_rate": 1.329611908559277e-05,
|
| 1124 |
+
"loss": 2.3111,
|
| 1125 |
+
"step": 14000
|
| 1126 |
+
},
|
| 1127 |
+
{
|
| 1128 |
+
"epoch": 7.168458781362007,
|
| 1129 |
+
"eval_accuracy": 0.44894773835621105,
|
| 1130 |
+
"eval_loss": 3.0879344940185547,
|
| 1131 |
+
"eval_runtime": 95.1844,
|
| 1132 |
+
"eval_samples_per_second": 352.411,
|
| 1133 |
+
"eval_steps_per_second": 22.031,
|
| 1134 |
+
"step": 14000
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 7.2196620583717355,
|
| 1138 |
+
"grad_norm": 1.7171036005020142,
|
| 1139 |
+
"learning_rate": 1.3032961190855927e-05,
|
| 1140 |
+
"loss": 2.3203,
|
| 1141 |
+
"step": 14100
|
| 1142 |
+
},
|
| 1143 |
+
{
|
| 1144 |
+
"epoch": 7.2708653353814645,
|
| 1145 |
+
"grad_norm": 1.8851195573806763,
|
| 1146 |
+
"learning_rate": 1.2767145135566188e-05,
|
| 1147 |
+
"loss": 2.3111,
|
| 1148 |
+
"step": 14200
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 7.322068612391193,
|
| 1152 |
+
"grad_norm": 1.8158164024353027,
|
| 1153 |
+
"learning_rate": 1.250132908027645e-05,
|
| 1154 |
+
"loss": 2.3227,
|
| 1155 |
+
"step": 14300
|
| 1156 |
+
},
|
| 1157 |
+
{
|
| 1158 |
+
"epoch": 7.373271889400922,
|
| 1159 |
+
"grad_norm": 1.833143711090088,
|
| 1160 |
+
"learning_rate": 1.223551302498671e-05,
|
| 1161 |
+
"loss": 2.2796,
|
| 1162 |
+
"step": 14400
|
| 1163 |
+
},
|
| 1164 |
+
{
|
| 1165 |
+
"epoch": 7.42447516641065,
|
| 1166 |
+
"grad_norm": 1.5996191501617432,
|
| 1167 |
+
"learning_rate": 1.196969696969697e-05,
|
| 1168 |
+
"loss": 2.2753,
|
| 1169 |
+
"step": 14500
|
| 1170 |
+
},
|
| 1171 |
+
{
|
| 1172 |
+
"epoch": 7.475678443420379,
|
| 1173 |
+
"grad_norm": 1.7543715238571167,
|
| 1174 |
+
"learning_rate": 1.170388091440723e-05,
|
| 1175 |
+
"loss": 2.3191,
|
| 1176 |
+
"step": 14600
|
| 1177 |
+
},
|
| 1178 |
+
{
|
| 1179 |
+
"epoch": 7.526881720430108,
|
| 1180 |
+
"grad_norm": 1.8224458694458008,
|
| 1181 |
+
"learning_rate": 1.1438064859117491e-05,
|
| 1182 |
+
"loss": 2.3362,
|
| 1183 |
+
"step": 14700
|
| 1184 |
+
},
|
| 1185 |
+
{
|
| 1186 |
+
"epoch": 7.578084997439836,
|
| 1187 |
+
"grad_norm": 1.825265884399414,
|
| 1188 |
+
"learning_rate": 1.1172248803827752e-05,
|
| 1189 |
+
"loss": 2.2788,
|
| 1190 |
+
"step": 14800
|
| 1191 |
+
},
|
| 1192 |
+
{
|
| 1193 |
+
"epoch": 7.629288274449565,
|
| 1194 |
+
"grad_norm": 1.6064096689224243,
|
| 1195 |
+
"learning_rate": 1.0906432748538012e-05,
|
| 1196 |
+
"loss": 2.3287,
|
| 1197 |
+
"step": 14900
|
| 1198 |
+
},
|
| 1199 |
+
{
|
| 1200 |
+
"epoch": 7.680491551459293,
|
| 1201 |
+
"grad_norm": 1.9973961114883423,
|
| 1202 |
+
"learning_rate": 1.0640616693248273e-05,
|
| 1203 |
+
"loss": 2.3054,
|
| 1204 |
+
"step": 15000
|
| 1205 |
+
},
|
| 1206 |
+
{
|
| 1207 |
+
"epoch": 7.731694828469022,
|
| 1208 |
+
"grad_norm": 1.9129321575164795,
|
| 1209 |
+
"learning_rate": 1.0374800637958534e-05,
|
| 1210 |
+
"loss": 2.2897,
|
| 1211 |
+
"step": 15100
|
| 1212 |
+
},
|
| 1213 |
+
{
|
| 1214 |
+
"epoch": 7.7828981054787505,
|
| 1215 |
+
"grad_norm": 1.7142568826675415,
|
| 1216 |
+
"learning_rate": 1.0108984582668795e-05,
|
| 1217 |
+
"loss": 2.2763,
|
| 1218 |
+
"step": 15200
|
| 1219 |
+
},
|
| 1220 |
+
{
|
| 1221 |
+
"epoch": 7.8341013824884795,
|
| 1222 |
+
"grad_norm": 1.8948873281478882,
|
| 1223 |
+
"learning_rate": 9.843168527379054e-06,
|
| 1224 |
+
"loss": 2.2831,
|
| 1225 |
+
"step": 15300
|
| 1226 |
+
},
|
| 1227 |
+
{
|
| 1228 |
+
"epoch": 7.885304659498208,
|
| 1229 |
+
"grad_norm": 1.671781301498413,
|
| 1230 |
+
"learning_rate": 9.577352472089313e-06,
|
| 1231 |
+
"loss": 2.301,
|
| 1232 |
+
"step": 15400
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 7.936507936507937,
|
| 1236 |
+
"grad_norm": 1.8178682327270508,
|
| 1237 |
+
"learning_rate": 9.311536416799576e-06,
|
| 1238 |
+
"loss": 2.3297,
|
| 1239 |
+
"step": 15500
|
| 1240 |
+
},
|
| 1241 |
+
{
|
| 1242 |
+
"epoch": 7.987711213517665,
|
| 1243 |
+
"grad_norm": 1.7934144735336304,
|
| 1244 |
+
"learning_rate": 9.045720361509835e-06,
|
| 1245 |
+
"loss": 2.2703,
|
| 1246 |
+
"step": 15600
|
| 1247 |
+
},
|
| 1248 |
+
{
|
| 1249 |
+
"epoch": 8.038914490527393,
|
| 1250 |
+
"grad_norm": 1.7798469066619873,
|
| 1251 |
+
"learning_rate": 8.779904306220096e-06,
|
| 1252 |
+
"loss": 2.3071,
|
| 1253 |
+
"step": 15700
|
| 1254 |
+
},
|
| 1255 |
+
{
|
| 1256 |
+
"epoch": 8.090117767537123,
|
| 1257 |
+
"grad_norm": 2.005934238433838,
|
| 1258 |
+
"learning_rate": 8.514088250930355e-06,
|
| 1259 |
+
"loss": 2.2579,
|
| 1260 |
+
"step": 15800
|
| 1261 |
+
},
|
| 1262 |
+
{
|
| 1263 |
+
"epoch": 8.141321044546851,
|
| 1264 |
+
"grad_norm": 1.7841377258300781,
|
| 1265 |
+
"learning_rate": 8.248272195640618e-06,
|
| 1266 |
+
"loss": 2.2942,
|
| 1267 |
+
"step": 15900
|
| 1268 |
+
},
|
| 1269 |
+
{
|
| 1270 |
+
"epoch": 8.19252432155658,
|
| 1271 |
+
"grad_norm": 1.814175009727478,
|
| 1272 |
+
"learning_rate": 7.982456140350877e-06,
|
| 1273 |
+
"loss": 2.2761,
|
| 1274 |
+
"step": 16000
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 8.19252432155658,
|
| 1278 |
+
"eval_accuracy": 0.45419183636131977,
|
| 1279 |
+
"eval_loss": 3.0668067932128906,
|
| 1280 |
+
"eval_runtime": 95.2855,
|
| 1281 |
+
"eval_samples_per_second": 352.037,
|
| 1282 |
+
"eval_steps_per_second": 22.008,
|
| 1283 |
+
"step": 16000
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 8.243727598566307,
|
| 1287 |
+
"grad_norm": 1.7196694612503052,
|
| 1288 |
+
"learning_rate": 7.719298245614036e-06,
|
| 1289 |
+
"loss": 2.2475,
|
| 1290 |
+
"step": 16100
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 8.294930875576037,
|
| 1294 |
+
"grad_norm": 1.780168890953064,
|
| 1295 |
+
"learning_rate": 7.453482190324296e-06,
|
| 1296 |
+
"loss": 2.2623,
|
| 1297 |
+
"step": 16200
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 8.346134152585766,
|
| 1301 |
+
"grad_norm": 1.7353187799453735,
|
| 1302 |
+
"learning_rate": 7.187666135034557e-06,
|
| 1303 |
+
"loss": 2.2436,
|
| 1304 |
+
"step": 16300
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 8.397337429595494,
|
| 1308 |
+
"grad_norm": 1.8766587972640991,
|
| 1309 |
+
"learning_rate": 6.921850079744817e-06,
|
| 1310 |
+
"loss": 2.2514,
|
| 1311 |
+
"step": 16400
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 8.448540706605222,
|
| 1315 |
+
"grad_norm": 1.8156400918960571,
|
| 1316 |
+
"learning_rate": 6.656034024455077e-06,
|
| 1317 |
+
"loss": 2.2647,
|
| 1318 |
+
"step": 16500
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 8.499743983614952,
|
| 1322 |
+
"grad_norm": 1.7604610919952393,
|
| 1323 |
+
"learning_rate": 6.390217969165338e-06,
|
| 1324 |
+
"loss": 2.2633,
|
| 1325 |
+
"step": 16600
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 8.55094726062468,
|
| 1329 |
+
"grad_norm": 1.9719775915145874,
|
| 1330 |
+
"learning_rate": 6.124401913875599e-06,
|
| 1331 |
+
"loss": 2.2729,
|
| 1332 |
+
"step": 16700
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 8.602150537634408,
|
| 1336 |
+
"grad_norm": 1.7817537784576416,
|
| 1337 |
+
"learning_rate": 5.858585858585859e-06,
|
| 1338 |
+
"loss": 2.2605,
|
| 1339 |
+
"step": 16800
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 8.653353814644138,
|
| 1343 |
+
"grad_norm": 2.052873373031616,
|
| 1344 |
+
"learning_rate": 5.59276980329612e-06,
|
| 1345 |
+
"loss": 2.2612,
|
| 1346 |
+
"step": 16900
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 8.704557091653866,
|
| 1350 |
+
"grad_norm": 1.7299405336380005,
|
| 1351 |
+
"learning_rate": 5.326953748006379e-06,
|
| 1352 |
+
"loss": 2.2544,
|
| 1353 |
+
"step": 17000
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 8.755760368663594,
|
| 1357 |
+
"grad_norm": 1.68231999874115,
|
| 1358 |
+
"learning_rate": 5.06113769271664e-06,
|
| 1359 |
+
"loss": 2.2786,
|
| 1360 |
+
"step": 17100
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 8.806963645673322,
|
| 1364 |
+
"grad_norm": 1.740615725517273,
|
| 1365 |
+
"learning_rate": 4.7953216374269005e-06,
|
| 1366 |
+
"loss": 2.2514,
|
| 1367 |
+
"step": 17200
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 8.858166922683052,
|
| 1371 |
+
"grad_norm": 1.6444553136825562,
|
| 1372 |
+
"learning_rate": 4.5295055821371614e-06,
|
| 1373 |
+
"loss": 2.2666,
|
| 1374 |
+
"step": 17300
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 8.90937019969278,
|
| 1378 |
+
"grad_norm": 1.6717356443405151,
|
| 1379 |
+
"learning_rate": 4.263689526847422e-06,
|
| 1380 |
+
"loss": 2.2459,
|
| 1381 |
+
"step": 17400
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 8.960573476702509,
|
| 1385 |
+
"grad_norm": 1.9801234006881714,
|
| 1386 |
+
"learning_rate": 3.9978734715576826e-06,
|
| 1387 |
+
"loss": 2.2494,
|
| 1388 |
+
"step": 17500
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 9.011776753712237,
|
| 1392 |
+
"grad_norm": 1.8575870990753174,
|
| 1393 |
+
"learning_rate": 3.732057416267943e-06,
|
| 1394 |
+
"loss": 2.2438,
|
| 1395 |
+
"step": 17600
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 9.062980030721967,
|
| 1399 |
+
"grad_norm": 1.7382246255874634,
|
| 1400 |
+
"learning_rate": 3.4662413609782037e-06,
|
| 1401 |
+
"loss": 2.2627,
|
| 1402 |
+
"step": 17700
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 9.114183307731695,
|
| 1406 |
+
"grad_norm": 1.739556074142456,
|
| 1407 |
+
"learning_rate": 3.2004253056884642e-06,
|
| 1408 |
+
"loss": 2.249,
|
| 1409 |
+
"step": 17800
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 9.165386584741423,
|
| 1413 |
+
"grad_norm": 1.9629456996917725,
|
| 1414 |
+
"learning_rate": 2.9346092503987244e-06,
|
| 1415 |
+
"loss": 2.2321,
|
| 1416 |
+
"step": 17900
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 9.216589861751151,
|
| 1420 |
+
"grad_norm": 1.9510672092437744,
|
| 1421 |
+
"learning_rate": 2.668793195108985e-06,
|
| 1422 |
+
"loss": 2.2231,
|
| 1423 |
+
"step": 18000
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 9.216589861751151,
|
| 1427 |
+
"eval_accuracy": 0.4517146740233816,
|
| 1428 |
+
"eval_loss": 3.069502830505371,
|
| 1429 |
+
"eval_runtime": 95.8923,
|
| 1430 |
+
"eval_samples_per_second": 349.809,
|
| 1431 |
+
"eval_steps_per_second": 21.868,
|
| 1432 |
+
"step": 18000
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"epoch": 9.267793138760881,
|
| 1436 |
+
"grad_norm": 1.7088594436645508,
|
| 1437 |
+
"learning_rate": 2.4056353003721427e-06,
|
| 1438 |
+
"loss": 2.2545,
|
| 1439 |
+
"step": 18100
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 9.31899641577061,
|
| 1443 |
+
"grad_norm": 1.7437766790390015,
|
| 1444 |
+
"learning_rate": 2.1398192450824032e-06,
|
| 1445 |
+
"loss": 2.232,
|
| 1446 |
+
"step": 18200
|
| 1447 |
+
},
|
| 1448 |
+
{
|
| 1449 |
+
"epoch": 9.370199692780337,
|
| 1450 |
+
"grad_norm": 1.936566948890686,
|
| 1451 |
+
"learning_rate": 1.8740031897926634e-06,
|
| 1452 |
+
"loss": 2.2572,
|
| 1453 |
+
"step": 18300
|
| 1454 |
+
},
|
| 1455 |
+
{
|
| 1456 |
+
"epoch": 9.421402969790067,
|
| 1457 |
+
"grad_norm": 1.9586554765701294,
|
| 1458 |
+
"learning_rate": 1.608187134502924e-06,
|
| 1459 |
+
"loss": 2.2504,
|
| 1460 |
+
"step": 18400
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"epoch": 9.472606246799796,
|
| 1464 |
+
"grad_norm": 1.8036390542984009,
|
| 1465 |
+
"learning_rate": 1.3423710792131845e-06,
|
| 1466 |
+
"loss": 2.2402,
|
| 1467 |
+
"step": 18500
|
| 1468 |
+
},
|
| 1469 |
+
{
|
| 1470 |
+
"epoch": 9.523809523809524,
|
| 1471 |
+
"grad_norm": 1.866513967514038,
|
| 1472 |
+
"learning_rate": 1.076555023923445e-06,
|
| 1473 |
+
"loss": 2.2471,
|
| 1474 |
+
"step": 18600
|
| 1475 |
+
},
|
| 1476 |
+
{
|
| 1477 |
+
"epoch": 9.575012800819252,
|
| 1478 |
+
"grad_norm": 1.9675647020339966,
|
| 1479 |
+
"learning_rate": 8.107389686337055e-07,
|
| 1480 |
+
"loss": 2.2286,
|
| 1481 |
+
"step": 18700
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 9.626216077828982,
|
| 1485 |
+
"grad_norm": 1.8157950639724731,
|
| 1486 |
+
"learning_rate": 5.44922913343966e-07,
|
| 1487 |
+
"loss": 2.2222,
|
| 1488 |
+
"step": 18800
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 9.67741935483871,
|
| 1492 |
+
"grad_norm": 1.8483563661575317,
|
| 1493 |
+
"learning_rate": 2.7910685805422647e-07,
|
| 1494 |
+
"loss": 2.2518,
|
| 1495 |
+
"step": 18900
|
| 1496 |
+
},
|
| 1497 |
+
{
|
| 1498 |
+
"epoch": 9.728622631848438,
|
| 1499 |
+
"grad_norm": 1.9589660167694092,
|
| 1500 |
+
"learning_rate": 1.3290802764486977e-08,
|
| 1501 |
+
"loss": 2.2022,
|
| 1502 |
+
"step": 19000
|
| 1503 |
+
},
|
| 1504 |
+
{
|
| 1505 |
+
"epoch": 9.728622631848438,
|
| 1506 |
+
"step": 19000,
|
| 1507 |
+
"total_flos": 7.9413964701696e+16,
|
| 1508 |
+
"train_loss": 2.775124670731394,
|
| 1509 |
+
"train_runtime": 3271.669,
|
| 1510 |
+
"train_samples_per_second": 92.919,
|
| 1511 |
+
"train_steps_per_second": 5.807
|
| 1512 |
+
}
|
| 1513 |
+
],
|
| 1514 |
+
"logging_steps": 100,
|
| 1515 |
+
"max_steps": 19000,
|
| 1516 |
+
"num_input_tokens_seen": 0,
|
| 1517 |
+
"num_train_epochs": 10,
|
| 1518 |
+
"save_steps": 99999999,
|
| 1519 |
+
"stateful_callbacks": {
|
| 1520 |
+
"TrainerControl": {
|
| 1521 |
+
"args": {
|
| 1522 |
+
"should_epoch_stop": false,
|
| 1523 |
+
"should_evaluate": false,
|
| 1524 |
+
"should_log": false,
|
| 1525 |
+
"should_save": true,
|
| 1526 |
+
"should_training_stop": true
|
| 1527 |
+
},
|
| 1528 |
+
"attributes": {}
|
| 1529 |
+
}
|
| 1530 |
+
},
|
| 1531 |
+
"total_flos": 7.9413964701696e+16,
|
| 1532 |
+
"train_batch_size": 16,
|
| 1533 |
+
"trial_name": null,
|
| 1534 |
+
"trial_params": null
|
| 1535 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d250537b2585c6d06b8b0dcfae2c77def32b4b994aeb860086cca09095186307
|
| 3 |
+
size 5841
|