Commit
•
053fce4
1
Parent(s):
9b897af
End of training
Browse files- README.md +166 -0
- all_results.json +13 -0
- config.json +57 -0
- eval_results.json +8 -0
- model.safetensors +3 -0
- preprocessor_config.json +22 -0
- train_results.json +8 -0
- trainer_state.json +1263 -0
- training_args.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,166 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: facebook/convnextv2-tiny-1k-224
|
| 4 |
+
tags:
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
datasets:
|
| 7 |
+
- imagefolder
|
| 8 |
+
metrics:
|
| 9 |
+
- accuracy
|
| 10 |
+
model-index:
|
| 11 |
+
- name: convnextv2-tiny-1k-224-finetuned-print
|
| 12 |
+
results:
|
| 13 |
+
- task:
|
| 14 |
+
name: Image Classification
|
| 15 |
+
type: image-classification
|
| 16 |
+
dataset:
|
| 17 |
+
name: imagefolder
|
| 18 |
+
type: imagefolder
|
| 19 |
+
config: default
|
| 20 |
+
split: train
|
| 21 |
+
args: default
|
| 22 |
+
metrics:
|
| 23 |
+
- name: Accuracy
|
| 24 |
+
type: accuracy
|
| 25 |
+
value: 0.9263157894736842
|
| 26 |
+
---
|
| 27 |
+
|
| 28 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 29 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 30 |
+
|
| 31 |
+
# convnextv2-tiny-1k-224-finetuned-print
|
| 32 |
+
|
| 33 |
+
This model is a fine-tuned version of [facebook/convnextv2-tiny-1k-224](https://huggingface.co/facebook/convnextv2-tiny-1k-224) on the imagefolder dataset.
|
| 34 |
+
It achieves the following results on the evaluation set:
|
| 35 |
+
- Loss: 0.1811
|
| 36 |
+
- Accuracy: 0.9263
|
| 37 |
+
|
| 38 |
+
## Model description
|
| 39 |
+
|
| 40 |
+
More information needed
|
| 41 |
+
|
| 42 |
+
## Intended uses & limitations
|
| 43 |
+
|
| 44 |
+
More information needed
|
| 45 |
+
|
| 46 |
+
## Training and evaluation data
|
| 47 |
+
|
| 48 |
+
More information needed
|
| 49 |
+
|
| 50 |
+
## Training procedure
|
| 51 |
+
|
| 52 |
+
### Training hyperparameters
|
| 53 |
+
|
| 54 |
+
The following hyperparameters were used during training:
|
| 55 |
+
- learning_rate: 5e-05
|
| 56 |
+
- train_batch_size: 32
|
| 57 |
+
- eval_batch_size: 32
|
| 58 |
+
- seed: 42
|
| 59 |
+
- gradient_accumulation_steps: 4
|
| 60 |
+
- total_train_batch_size: 128
|
| 61 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 62 |
+
- lr_scheduler_type: linear
|
| 63 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 64 |
+
- num_epochs: 100
|
| 65 |
+
|
| 66 |
+
### Training results
|
| 67 |
+
|
| 68 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 69 |
+
|:-------------:|:-------:|:----:|:---------------:|:--------:|
|
| 70 |
+
| No log | 0.8889 | 6 | 1.6653 | 0.1158 |
|
| 71 |
+
| 1.6915 | 1.9259 | 13 | 1.5942 | 0.2316 |
|
| 72 |
+
| 1.5895 | 2.9630 | 20 | 1.4918 | 0.3895 |
|
| 73 |
+
| 1.5895 | 4.0 | 27 | 1.3637 | 0.5053 |
|
| 74 |
+
| 1.433 | 4.8889 | 33 | 1.2241 | 0.6737 |
|
| 75 |
+
| 1.2158 | 5.9259 | 40 | 1.0279 | 0.7474 |
|
| 76 |
+
| 1.2158 | 6.9630 | 47 | 0.8303 | 0.8105 |
|
| 77 |
+
| 0.9438 | 8.0 | 54 | 0.6756 | 0.8211 |
|
| 78 |
+
| 0.727 | 8.8889 | 60 | 0.5751 | 0.8421 |
|
| 79 |
+
| 0.727 | 9.9259 | 67 | 0.4679 | 0.8632 |
|
| 80 |
+
| 0.5516 | 10.9630 | 74 | 0.4432 | 0.8526 |
|
| 81 |
+
| 0.4337 | 12.0 | 81 | 0.3815 | 0.8526 |
|
| 82 |
+
| 0.4337 | 12.8889 | 87 | 0.3641 | 0.8842 |
|
| 83 |
+
| 0.3757 | 13.9259 | 94 | 0.3233 | 0.8737 |
|
| 84 |
+
| 0.3017 | 14.9630 | 101 | 0.3230 | 0.9158 |
|
| 85 |
+
| 0.3017 | 16.0 | 108 | 0.3018 | 0.8842 |
|
| 86 |
+
| 0.2495 | 16.8889 | 114 | 0.3445 | 0.9053 |
|
| 87 |
+
| 0.2177 | 17.9259 | 121 | 0.2987 | 0.8947 |
|
| 88 |
+
| 0.2177 | 18.9630 | 128 | 0.2727 | 0.8947 |
|
| 89 |
+
| 0.1738 | 20.0 | 135 | 0.2865 | 0.8842 |
|
| 90 |
+
| 0.1572 | 20.8889 | 141 | 0.2646 | 0.9263 |
|
| 91 |
+
| 0.1572 | 21.9259 | 148 | 0.3100 | 0.9053 |
|
| 92 |
+
| 0.1165 | 22.9630 | 155 | 0.3039 | 0.9263 |
|
| 93 |
+
| 0.1057 | 24.0 | 162 | 0.3023 | 0.9053 |
|
| 94 |
+
| 0.1057 | 24.8889 | 168 | 0.2254 | 0.9158 |
|
| 95 |
+
| 0.0825 | 25.9259 | 175 | 0.3308 | 0.8737 |
|
| 96 |
+
| 0.0795 | 26.9630 | 182 | 0.2040 | 0.9368 |
|
| 97 |
+
| 0.0795 | 28.0 | 189 | 0.2148 | 0.9263 |
|
| 98 |
+
| 0.072 | 28.8889 | 195 | 0.3450 | 0.8632 |
|
| 99 |
+
| 0.0701 | 29.9259 | 202 | 0.2418 | 0.9263 |
|
| 100 |
+
| 0.0701 | 30.9630 | 209 | 0.2495 | 0.9263 |
|
| 101 |
+
| 0.0635 | 32.0 | 216 | 0.3267 | 0.8947 |
|
| 102 |
+
| 0.0537 | 32.8889 | 222 | 0.3728 | 0.9158 |
|
| 103 |
+
| 0.0537 | 33.9259 | 229 | 0.2852 | 0.9053 |
|
| 104 |
+
| 0.0607 | 34.9630 | 236 | 0.2386 | 0.9474 |
|
| 105 |
+
| 0.052 | 36.0 | 243 | 0.2070 | 0.9158 |
|
| 106 |
+
| 0.052 | 36.8889 | 249 | 0.1860 | 0.9474 |
|
| 107 |
+
| 0.049 | 37.9259 | 256 | 0.3069 | 0.8947 |
|
| 108 |
+
| 0.0578 | 38.9630 | 263 | 0.4477 | 0.8737 |
|
| 109 |
+
| 0.0533 | 40.0 | 270 | 0.2612 | 0.8947 |
|
| 110 |
+
| 0.0533 | 40.8889 | 276 | 0.2649 | 0.8842 |
|
| 111 |
+
| 0.0505 | 41.9259 | 283 | 0.1950 | 0.9263 |
|
| 112 |
+
| 0.0433 | 42.9630 | 290 | 0.2903 | 0.8842 |
|
| 113 |
+
| 0.0433 | 44.0 | 297 | 0.2526 | 0.9368 |
|
| 114 |
+
| 0.0395 | 44.8889 | 303 | 0.3016 | 0.8842 |
|
| 115 |
+
| 0.035 | 45.9259 | 310 | 0.3509 | 0.8947 |
|
| 116 |
+
| 0.035 | 46.9630 | 317 | 0.2943 | 0.8842 |
|
| 117 |
+
| 0.0335 | 48.0 | 324 | 0.2613 | 0.8842 |
|
| 118 |
+
| 0.0408 | 48.8889 | 330 | 0.2165 | 0.9158 |
|
| 119 |
+
| 0.0408 | 49.9259 | 337 | 0.2872 | 0.9263 |
|
| 120 |
+
| 0.0244 | 50.9630 | 344 | 0.3134 | 0.8842 |
|
| 121 |
+
| 0.0323 | 52.0 | 351 | 0.3006 | 0.9158 |
|
| 122 |
+
| 0.0323 | 52.8889 | 357 | 0.3758 | 0.8737 |
|
| 123 |
+
| 0.0241 | 53.9259 | 364 | 0.3033 | 0.9263 |
|
| 124 |
+
| 0.0193 | 54.9630 | 371 | 0.2741 | 0.9368 |
|
| 125 |
+
| 0.0193 | 56.0 | 378 | 0.1684 | 0.9368 |
|
| 126 |
+
| 0.0273 | 56.8889 | 384 | 0.2403 | 0.9474 |
|
| 127 |
+
| 0.0244 | 57.9259 | 391 | 0.1500 | 0.9474 |
|
| 128 |
+
| 0.0244 | 58.9630 | 398 | 0.1377 | 0.9368 |
|
| 129 |
+
| 0.0268 | 60.0 | 405 | 0.1898 | 0.9158 |
|
| 130 |
+
| 0.0405 | 60.8889 | 411 | 0.1756 | 0.9053 |
|
| 131 |
+
| 0.0405 | 61.9259 | 418 | 0.1907 | 0.9263 |
|
| 132 |
+
| 0.0219 | 62.9630 | 425 | 0.1790 | 0.9053 |
|
| 133 |
+
| 0.0329 | 64.0 | 432 | 0.1885 | 0.9368 |
|
| 134 |
+
| 0.0329 | 64.8889 | 438 | 0.1550 | 0.9368 |
|
| 135 |
+
| 0.019 | 65.9259 | 445 | 0.1811 | 0.9158 |
|
| 136 |
+
| 0.0205 | 66.9630 | 452 | 0.2166 | 0.9263 |
|
| 137 |
+
| 0.0205 | 68.0 | 459 | 0.1701 | 0.9053 |
|
| 138 |
+
| 0.0232 | 68.8889 | 465 | 0.2153 | 0.9158 |
|
| 139 |
+
| 0.0269 | 69.9259 | 472 | 0.2229 | 0.9263 |
|
| 140 |
+
| 0.0269 | 70.9630 | 479 | 0.2237 | 0.9263 |
|
| 141 |
+
| 0.0306 | 72.0 | 486 | 0.1828 | 0.9368 |
|
| 142 |
+
| 0.0298 | 72.8889 | 492 | 0.1448 | 0.9368 |
|
| 143 |
+
| 0.0298 | 73.9259 | 499 | 0.1948 | 0.9158 |
|
| 144 |
+
| 0.0154 | 74.9630 | 506 | 0.2570 | 0.9158 |
|
| 145 |
+
| 0.0193 | 76.0 | 513 | 0.2462 | 0.9263 |
|
| 146 |
+
| 0.0193 | 76.8889 | 519 | 0.2194 | 0.9158 |
|
| 147 |
+
| 0.0188 | 77.9259 | 526 | 0.2254 | 0.9158 |
|
| 148 |
+
| 0.0198 | 78.9630 | 533 | 0.1924 | 0.9263 |
|
| 149 |
+
| 0.0147 | 80.0 | 540 | 0.1525 | 0.9368 |
|
| 150 |
+
| 0.0147 | 80.8889 | 546 | 0.1314 | 0.9474 |
|
| 151 |
+
| 0.0282 | 81.9259 | 553 | 0.1381 | 0.9368 |
|
| 152 |
+
| 0.0168 | 82.9630 | 560 | 0.1496 | 0.9158 |
|
| 153 |
+
| 0.0168 | 84.0 | 567 | 0.1806 | 0.9263 |
|
| 154 |
+
| 0.018 | 84.8889 | 573 | 0.2080 | 0.9263 |
|
| 155 |
+
| 0.0172 | 85.9259 | 580 | 0.2199 | 0.9158 |
|
| 156 |
+
| 0.0172 | 86.9630 | 587 | 0.1939 | 0.9263 |
|
| 157 |
+
| 0.0117 | 88.0 | 594 | 0.1815 | 0.9263 |
|
| 158 |
+
| 0.0149 | 88.8889 | 600 | 0.1811 | 0.9263 |
|
| 159 |
+
|
| 160 |
+
|
| 161 |
+
### Framework versions
|
| 162 |
+
|
| 163 |
+
- Transformers 4.44.0
|
| 164 |
+
- Pytorch 2.4.0
|
| 165 |
+
- Datasets 2.21.0
|
| 166 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 88.88888888888889,
|
| 3 |
+
"eval_accuracy": 0.9263157894736842,
|
| 4 |
+
"eval_loss": 0.1811211258172989,
|
| 5 |
+
"eval_runtime": 1.544,
|
| 6 |
+
"eval_samples_per_second": 61.527,
|
| 7 |
+
"eval_steps_per_second": 1.943,
|
| 8 |
+
"total_flos": 1.9132429834630103e+18,
|
| 9 |
+
"train_loss": 0.19732174752900997,
|
| 10 |
+
"train_runtime": 1871.4873,
|
| 11 |
+
"train_samples_per_second": 45.686,
|
| 12 |
+
"train_steps_per_second": 0.321
|
| 13 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "facebook/convnextv2-tiny-1k-224",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"ConvNextV2ForImageClassification"
|
| 5 |
+
],
|
| 6 |
+
"depths": [
|
| 7 |
+
3,
|
| 8 |
+
3,
|
| 9 |
+
9,
|
| 10 |
+
3
|
| 11 |
+
],
|
| 12 |
+
"drop_path_rate": 0.0,
|
| 13 |
+
"hidden_act": "gelu",
|
| 14 |
+
"hidden_sizes": [
|
| 15 |
+
96,
|
| 16 |
+
192,
|
| 17 |
+
384,
|
| 18 |
+
768
|
| 19 |
+
],
|
| 20 |
+
"id2label": {
|
| 21 |
+
"0": "Floral",
|
| 22 |
+
"1": "Leopard",
|
| 23 |
+
"2": "Sequined",
|
| 24 |
+
"3": "Tie_and_Dye",
|
| 25 |
+
"4": "Zebra"
|
| 26 |
+
},
|
| 27 |
+
"image_size": 224,
|
| 28 |
+
"initializer_range": 0.02,
|
| 29 |
+
"label2id": {
|
| 30 |
+
"Floral": 0,
|
| 31 |
+
"Leopard": 1,
|
| 32 |
+
"Sequined": 2,
|
| 33 |
+
"Tie_and_Dye": 3,
|
| 34 |
+
"Zebra": 4
|
| 35 |
+
},
|
| 36 |
+
"layer_norm_eps": 1e-12,
|
| 37 |
+
"model_type": "convnextv2",
|
| 38 |
+
"num_channels": 3,
|
| 39 |
+
"num_stages": 4,
|
| 40 |
+
"out_features": [
|
| 41 |
+
"stage4"
|
| 42 |
+
],
|
| 43 |
+
"out_indices": [
|
| 44 |
+
4
|
| 45 |
+
],
|
| 46 |
+
"patch_size": 4,
|
| 47 |
+
"problem_type": "single_label_classification",
|
| 48 |
+
"stage_names": [
|
| 49 |
+
"stem",
|
| 50 |
+
"stage1",
|
| 51 |
+
"stage2",
|
| 52 |
+
"stage3",
|
| 53 |
+
"stage4"
|
| 54 |
+
],
|
| 55 |
+
"torch_dtype": "float32",
|
| 56 |
+
"transformers_version": "4.44.0"
|
| 57 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 88.88888888888889,
|
| 3 |
+
"eval_accuracy": 0.9263157894736842,
|
| 4 |
+
"eval_loss": 0.1811211258172989,
|
| 5 |
+
"eval_runtime": 1.544,
|
| 6 |
+
"eval_samples_per_second": 61.527,
|
| 7 |
+
"eval_steps_per_second": 1.943
|
| 8 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d0143674d58f7b45de93c4477981c0c96a74f7f7fdebef703066160b9c3003f3
|
| 3 |
+
size 111505052
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"crop_pct": 0.875,
|
| 3 |
+
"do_normalize": true,
|
| 4 |
+
"do_rescale": true,
|
| 5 |
+
"do_resize": true,
|
| 6 |
+
"image_mean": [
|
| 7 |
+
0.485,
|
| 8 |
+
0.456,
|
| 9 |
+
0.406
|
| 10 |
+
],
|
| 11 |
+
"image_processor_type": "ConvNextImageProcessor",
|
| 12 |
+
"image_std": [
|
| 13 |
+
0.229,
|
| 14 |
+
0.224,
|
| 15 |
+
0.225
|
| 16 |
+
],
|
| 17 |
+
"resample": 3,
|
| 18 |
+
"rescale_factor": 0.00392156862745098,
|
| 19 |
+
"size": {
|
| 20 |
+
"shortest_edge": 224
|
| 21 |
+
}
|
| 22 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 88.88888888888889,
|
| 3 |
+
"total_flos": 1.9132429834630103e+18,
|
| 4 |
+
"train_loss": 0.19732174752900997,
|
| 5 |
+
"train_runtime": 1871.4873,
|
| 6 |
+
"train_samples_per_second": 45.686,
|
| 7 |
+
"train_steps_per_second": 0.321
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1263 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 88.88888888888889,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 600,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.8888888888888888,
|
| 13 |
+
"eval_accuracy": 0.11578947368421053,
|
| 14 |
+
"eval_loss": 1.6652907133102417,
|
| 15 |
+
"eval_runtime": 1.8033,
|
| 16 |
+
"eval_samples_per_second": 52.682,
|
| 17 |
+
"eval_steps_per_second": 1.664,
|
| 18 |
+
"step": 6
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 1.4814814814814814,
|
| 22 |
+
"grad_norm": 5.346019268035889,
|
| 23 |
+
"learning_rate": 8.333333333333334e-06,
|
| 24 |
+
"loss": 1.6915,
|
| 25 |
+
"step": 10
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 1.925925925925926,
|
| 29 |
+
"eval_accuracy": 0.23157894736842105,
|
| 30 |
+
"eval_loss": 1.5941598415374756,
|
| 31 |
+
"eval_runtime": 1.5205,
|
| 32 |
+
"eval_samples_per_second": 62.481,
|
| 33 |
+
"eval_steps_per_second": 1.973,
|
| 34 |
+
"step": 13
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 2.962962962962963,
|
| 38 |
+
"grad_norm": 8.961028099060059,
|
| 39 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 40 |
+
"loss": 1.5895,
|
| 41 |
+
"step": 20
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 2.962962962962963,
|
| 45 |
+
"eval_accuracy": 0.3894736842105263,
|
| 46 |
+
"eval_loss": 1.4918299913406372,
|
| 47 |
+
"eval_runtime": 1.5278,
|
| 48 |
+
"eval_samples_per_second": 62.179,
|
| 49 |
+
"eval_steps_per_second": 1.964,
|
| 50 |
+
"step": 20
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 4.0,
|
| 54 |
+
"eval_accuracy": 0.5052631578947369,
|
| 55 |
+
"eval_loss": 1.363681674003601,
|
| 56 |
+
"eval_runtime": 1.5355,
|
| 57 |
+
"eval_samples_per_second": 61.868,
|
| 58 |
+
"eval_steps_per_second": 1.954,
|
| 59 |
+
"step": 27
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 4.444444444444445,
|
| 63 |
+
"grad_norm": 12.23503303527832,
|
| 64 |
+
"learning_rate": 2.5e-05,
|
| 65 |
+
"loss": 1.433,
|
| 66 |
+
"step": 30
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 4.888888888888889,
|
| 70 |
+
"eval_accuracy": 0.6736842105263158,
|
| 71 |
+
"eval_loss": 1.2241352796554565,
|
| 72 |
+
"eval_runtime": 1.595,
|
| 73 |
+
"eval_samples_per_second": 59.559,
|
| 74 |
+
"eval_steps_per_second": 1.881,
|
| 75 |
+
"step": 33
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 5.925925925925926,
|
| 79 |
+
"grad_norm": 29.141326904296875,
|
| 80 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 81 |
+
"loss": 1.2158,
|
| 82 |
+
"step": 40
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 5.925925925925926,
|
| 86 |
+
"eval_accuracy": 0.7473684210526316,
|
| 87 |
+
"eval_loss": 1.0278701782226562,
|
| 88 |
+
"eval_runtime": 1.5753,
|
| 89 |
+
"eval_samples_per_second": 60.307,
|
| 90 |
+
"eval_steps_per_second": 1.904,
|
| 91 |
+
"step": 40
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 6.962962962962963,
|
| 95 |
+
"eval_accuracy": 0.8105263157894737,
|
| 96 |
+
"eval_loss": 0.8302664756774902,
|
| 97 |
+
"eval_runtime": 1.6713,
|
| 98 |
+
"eval_samples_per_second": 56.842,
|
| 99 |
+
"eval_steps_per_second": 1.795,
|
| 100 |
+
"step": 47
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 7.407407407407407,
|
| 104 |
+
"grad_norm": 28.541772842407227,
|
| 105 |
+
"learning_rate": 4.166666666666667e-05,
|
| 106 |
+
"loss": 0.9438,
|
| 107 |
+
"step": 50
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 8.0,
|
| 111 |
+
"eval_accuracy": 0.8210526315789474,
|
| 112 |
+
"eval_loss": 0.6755695343017578,
|
| 113 |
+
"eval_runtime": 1.5959,
|
| 114 |
+
"eval_samples_per_second": 59.527,
|
| 115 |
+
"eval_steps_per_second": 1.88,
|
| 116 |
+
"step": 54
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 8.88888888888889,
|
| 120 |
+
"grad_norm": 14.663153648376465,
|
| 121 |
+
"learning_rate": 5e-05,
|
| 122 |
+
"loss": 0.727,
|
| 123 |
+
"step": 60
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 8.88888888888889,
|
| 127 |
+
"eval_accuracy": 0.8421052631578947,
|
| 128 |
+
"eval_loss": 0.5751134157180786,
|
| 129 |
+
"eval_runtime": 1.7131,
|
| 130 |
+
"eval_samples_per_second": 55.455,
|
| 131 |
+
"eval_steps_per_second": 1.751,
|
| 132 |
+
"step": 60
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 9.925925925925926,
|
| 136 |
+
"eval_accuracy": 0.8631578947368421,
|
| 137 |
+
"eval_loss": 0.4678628146648407,
|
| 138 |
+
"eval_runtime": 1.5414,
|
| 139 |
+
"eval_samples_per_second": 61.633,
|
| 140 |
+
"eval_steps_per_second": 1.946,
|
| 141 |
+
"step": 67
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 10.37037037037037,
|
| 145 |
+
"grad_norm": 36.71305847167969,
|
| 146 |
+
"learning_rate": 4.9074074074074075e-05,
|
| 147 |
+
"loss": 0.5516,
|
| 148 |
+
"step": 70
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 10.962962962962964,
|
| 152 |
+
"eval_accuracy": 0.8526315789473684,
|
| 153 |
+
"eval_loss": 0.4432494640350342,
|
| 154 |
+
"eval_runtime": 1.5533,
|
| 155 |
+
"eval_samples_per_second": 61.159,
|
| 156 |
+
"eval_steps_per_second": 1.931,
|
| 157 |
+
"step": 74
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 11.851851851851851,
|
| 161 |
+
"grad_norm": 25.495798110961914,
|
| 162 |
+
"learning_rate": 4.814814814814815e-05,
|
| 163 |
+
"loss": 0.4337,
|
| 164 |
+
"step": 80
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 12.0,
|
| 168 |
+
"eval_accuracy": 0.8526315789473684,
|
| 169 |
+
"eval_loss": 0.38154730200767517,
|
| 170 |
+
"eval_runtime": 1.7941,
|
| 171 |
+
"eval_samples_per_second": 52.95,
|
| 172 |
+
"eval_steps_per_second": 1.672,
|
| 173 |
+
"step": 81
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 12.88888888888889,
|
| 177 |
+
"eval_accuracy": 0.8842105263157894,
|
| 178 |
+
"eval_loss": 0.36410775780677795,
|
| 179 |
+
"eval_runtime": 1.5444,
|
| 180 |
+
"eval_samples_per_second": 61.514,
|
| 181 |
+
"eval_steps_per_second": 1.943,
|
| 182 |
+
"step": 87
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 13.333333333333334,
|
| 186 |
+
"grad_norm": 20.939546585083008,
|
| 187 |
+
"learning_rate": 4.722222222222222e-05,
|
| 188 |
+
"loss": 0.3757,
|
| 189 |
+
"step": 90
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 13.925925925925926,
|
| 193 |
+
"eval_accuracy": 0.8736842105263158,
|
| 194 |
+
"eval_loss": 0.3233407735824585,
|
| 195 |
+
"eval_runtime": 1.5363,
|
| 196 |
+
"eval_samples_per_second": 61.836,
|
| 197 |
+
"eval_steps_per_second": 1.953,
|
| 198 |
+
"step": 94
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 14.814814814814815,
|
| 202 |
+
"grad_norm": 87.37303924560547,
|
| 203 |
+
"learning_rate": 4.62962962962963e-05,
|
| 204 |
+
"loss": 0.3017,
|
| 205 |
+
"step": 100
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 14.962962962962964,
|
| 209 |
+
"eval_accuracy": 0.9157894736842105,
|
| 210 |
+
"eval_loss": 0.32304683327674866,
|
| 211 |
+
"eval_runtime": 1.5993,
|
| 212 |
+
"eval_samples_per_second": 59.403,
|
| 213 |
+
"eval_steps_per_second": 1.876,
|
| 214 |
+
"step": 101
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 16.0,
|
| 218 |
+
"eval_accuracy": 0.8842105263157894,
|
| 219 |
+
"eval_loss": 0.30177825689315796,
|
| 220 |
+
"eval_runtime": 1.5164,
|
| 221 |
+
"eval_samples_per_second": 62.65,
|
| 222 |
+
"eval_steps_per_second": 1.978,
|
| 223 |
+
"step": 108
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 16.296296296296298,
|
| 227 |
+
"grad_norm": 25.734861373901367,
|
| 228 |
+
"learning_rate": 4.5370370370370374e-05,
|
| 229 |
+
"loss": 0.2495,
|
| 230 |
+
"step": 110
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 16.88888888888889,
|
| 234 |
+
"eval_accuracy": 0.9052631578947369,
|
| 235 |
+
"eval_loss": 0.34445926547050476,
|
| 236 |
+
"eval_runtime": 1.5124,
|
| 237 |
+
"eval_samples_per_second": 62.812,
|
| 238 |
+
"eval_steps_per_second": 1.984,
|
| 239 |
+
"step": 114
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 17.77777777777778,
|
| 243 |
+
"grad_norm": 22.723478317260742,
|
| 244 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 245 |
+
"loss": 0.2177,
|
| 246 |
+
"step": 120
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 17.925925925925927,
|
| 250 |
+
"eval_accuracy": 0.8947368421052632,
|
| 251 |
+
"eval_loss": 0.29870516061782837,
|
| 252 |
+
"eval_runtime": 1.6176,
|
| 253 |
+
"eval_samples_per_second": 58.728,
|
| 254 |
+
"eval_steps_per_second": 1.855,
|
| 255 |
+
"step": 121
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 18.962962962962962,
|
| 259 |
+
"eval_accuracy": 0.8947368421052632,
|
| 260 |
+
"eval_loss": 0.27269816398620605,
|
| 261 |
+
"eval_runtime": 1.5486,
|
| 262 |
+
"eval_samples_per_second": 61.345,
|
| 263 |
+
"eval_steps_per_second": 1.937,
|
| 264 |
+
"step": 128
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 19.25925925925926,
|
| 268 |
+
"grad_norm": 21.11945343017578,
|
| 269 |
+
"learning_rate": 4.351851851851852e-05,
|
| 270 |
+
"loss": 0.1738,
|
| 271 |
+
"step": 130
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 20.0,
|
| 275 |
+
"eval_accuracy": 0.8842105263157894,
|
| 276 |
+
"eval_loss": 0.28645142912864685,
|
| 277 |
+
"eval_runtime": 1.7081,
|
| 278 |
+
"eval_samples_per_second": 55.618,
|
| 279 |
+
"eval_steps_per_second": 1.756,
|
| 280 |
+
"step": 135
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 20.74074074074074,
|
| 284 |
+
"grad_norm": 55.72514343261719,
|
| 285 |
+
"learning_rate": 4.259259259259259e-05,
|
| 286 |
+
"loss": 0.1572,
|
| 287 |
+
"step": 140
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 20.88888888888889,
|
| 291 |
+
"eval_accuracy": 0.9263157894736842,
|
| 292 |
+
"eval_loss": 0.2645561993122101,
|
| 293 |
+
"eval_runtime": 1.8502,
|
| 294 |
+
"eval_samples_per_second": 51.345,
|
| 295 |
+
"eval_steps_per_second": 1.621,
|
| 296 |
+
"step": 141
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 21.925925925925927,
|
| 300 |
+
"eval_accuracy": 0.9052631578947369,
|
| 301 |
+
"eval_loss": 0.31001701951026917,
|
| 302 |
+
"eval_runtime": 1.5502,
|
| 303 |
+
"eval_samples_per_second": 61.282,
|
| 304 |
+
"eval_steps_per_second": 1.935,
|
| 305 |
+
"step": 148
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 22.22222222222222,
|
| 309 |
+
"grad_norm": 11.173165321350098,
|
| 310 |
+
"learning_rate": 4.166666666666667e-05,
|
| 311 |
+
"loss": 0.1165,
|
| 312 |
+
"step": 150
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 22.962962962962962,
|
| 316 |
+
"eval_accuracy": 0.9263157894736842,
|
| 317 |
+
"eval_loss": 0.30389800667762756,
|
| 318 |
+
"eval_runtime": 1.5599,
|
| 319 |
+
"eval_samples_per_second": 60.903,
|
| 320 |
+
"eval_steps_per_second": 1.923,
|
| 321 |
+
"step": 155
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 23.703703703703702,
|
| 325 |
+
"grad_norm": 13.963062286376953,
|
| 326 |
+
"learning_rate": 4.074074074074074e-05,
|
| 327 |
+
"loss": 0.1057,
|
| 328 |
+
"step": 160
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 24.0,
|
| 332 |
+
"eval_accuracy": 0.9052631578947369,
|
| 333 |
+
"eval_loss": 0.3022925555706024,
|
| 334 |
+
"eval_runtime": 1.682,
|
| 335 |
+
"eval_samples_per_second": 56.482,
|
| 336 |
+
"eval_steps_per_second": 1.784,
|
| 337 |
+
"step": 162
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 24.88888888888889,
|
| 341 |
+
"eval_accuracy": 0.9157894736842105,
|
| 342 |
+
"eval_loss": 0.2254202663898468,
|
| 343 |
+
"eval_runtime": 1.5031,
|
| 344 |
+
"eval_samples_per_second": 63.201,
|
| 345 |
+
"eval_steps_per_second": 1.996,
|
| 346 |
+
"step": 168
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 25.185185185185187,
|
| 350 |
+
"grad_norm": 22.49197769165039,
|
| 351 |
+
"learning_rate": 3.981481481481482e-05,
|
| 352 |
+
"loss": 0.0825,
|
| 353 |
+
"step": 170
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 25.925925925925927,
|
| 357 |
+
"eval_accuracy": 0.8736842105263158,
|
| 358 |
+
"eval_loss": 0.3308357298374176,
|
| 359 |
+
"eval_runtime": 1.5096,
|
| 360 |
+
"eval_samples_per_second": 62.929,
|
| 361 |
+
"eval_steps_per_second": 1.987,
|
| 362 |
+
"step": 175
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 26.666666666666668,
|
| 366 |
+
"grad_norm": 17.941728591918945,
|
| 367 |
+
"learning_rate": 3.888888888888889e-05,
|
| 368 |
+
"loss": 0.0795,
|
| 369 |
+
"step": 180
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 26.962962962962962,
|
| 373 |
+
"eval_accuracy": 0.9368421052631579,
|
| 374 |
+
"eval_loss": 0.20397017896175385,
|
| 375 |
+
"eval_runtime": 1.5389,
|
| 376 |
+
"eval_samples_per_second": 61.731,
|
| 377 |
+
"eval_steps_per_second": 1.949,
|
| 378 |
+
"step": 182
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 28.0,
|
| 382 |
+
"eval_accuracy": 0.9263157894736842,
|
| 383 |
+
"eval_loss": 0.21477854251861572,
|
| 384 |
+
"eval_runtime": 1.5127,
|
| 385 |
+
"eval_samples_per_second": 62.8,
|
| 386 |
+
"eval_steps_per_second": 1.983,
|
| 387 |
+
"step": 189
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 28.14814814814815,
|
| 391 |
+
"grad_norm": 11.859210014343262,
|
| 392 |
+
"learning_rate": 3.7962962962962964e-05,
|
| 393 |
+
"loss": 0.072,
|
| 394 |
+
"step": 190
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 28.88888888888889,
|
| 398 |
+
"eval_accuracy": 0.8631578947368421,
|
| 399 |
+
"eval_loss": 0.3449535667896271,
|
| 400 |
+
"eval_runtime": 1.5393,
|
| 401 |
+
"eval_samples_per_second": 61.716,
|
| 402 |
+
"eval_steps_per_second": 1.949,
|
| 403 |
+
"step": 195
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 29.62962962962963,
|
| 407 |
+
"grad_norm": 17.80461883544922,
|
| 408 |
+
"learning_rate": 3.7037037037037037e-05,
|
| 409 |
+
"loss": 0.0701,
|
| 410 |
+
"step": 200
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 29.925925925925927,
|
| 414 |
+
"eval_accuracy": 0.9263157894736842,
|
| 415 |
+
"eval_loss": 0.24177835881710052,
|
| 416 |
+
"eval_runtime": 1.7581,
|
| 417 |
+
"eval_samples_per_second": 54.036,
|
| 418 |
+
"eval_steps_per_second": 1.706,
|
| 419 |
+
"step": 202
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 30.962962962962962,
|
| 423 |
+
"eval_accuracy": 0.9263157894736842,
|
| 424 |
+
"eval_loss": 0.24954235553741455,
|
| 425 |
+
"eval_runtime": 1.5375,
|
| 426 |
+
"eval_samples_per_second": 61.787,
|
| 427 |
+
"eval_steps_per_second": 1.951,
|
| 428 |
+
"step": 209
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 31.11111111111111,
|
| 432 |
+
"grad_norm": 11.903740882873535,
|
| 433 |
+
"learning_rate": 3.611111111111111e-05,
|
| 434 |
+
"loss": 0.0635,
|
| 435 |
+
"step": 210
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 32.0,
|
| 439 |
+
"eval_accuracy": 0.8947368421052632,
|
| 440 |
+
"eval_loss": 0.3266756236553192,
|
| 441 |
+
"eval_runtime": 1.5287,
|
| 442 |
+
"eval_samples_per_second": 62.145,
|
| 443 |
+
"eval_steps_per_second": 1.962,
|
| 444 |
+
"step": 216
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 32.592592592592595,
|
| 448 |
+
"grad_norm": 9.013089179992676,
|
| 449 |
+
"learning_rate": 3.518518518518519e-05,
|
| 450 |
+
"loss": 0.0537,
|
| 451 |
+
"step": 220
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 32.888888888888886,
|
| 455 |
+
"eval_accuracy": 0.9157894736842105,
|
| 456 |
+
"eval_loss": 0.3727841377258301,
|
| 457 |
+
"eval_runtime": 1.7008,
|
| 458 |
+
"eval_samples_per_second": 55.857,
|
| 459 |
+
"eval_steps_per_second": 1.764,
|
| 460 |
+
"step": 222
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 33.925925925925924,
|
| 464 |
+
"eval_accuracy": 0.9052631578947369,
|
| 465 |
+
"eval_loss": 0.28518009185791016,
|
| 466 |
+
"eval_runtime": 1.5355,
|
| 467 |
+
"eval_samples_per_second": 61.868,
|
| 468 |
+
"eval_steps_per_second": 1.954,
|
| 469 |
+
"step": 229
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 34.074074074074076,
|
| 473 |
+
"grad_norm": 37.26246643066406,
|
| 474 |
+
"learning_rate": 3.425925925925926e-05,
|
| 475 |
+
"loss": 0.0607,
|
| 476 |
+
"step": 230
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 34.96296296296296,
|
| 480 |
+
"eval_accuracy": 0.9473684210526315,
|
| 481 |
+
"eval_loss": 0.23858819901943207,
|
| 482 |
+
"eval_runtime": 1.5204,
|
| 483 |
+
"eval_samples_per_second": 62.485,
|
| 484 |
+
"eval_steps_per_second": 1.973,
|
| 485 |
+
"step": 236
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 35.55555555555556,
|
| 489 |
+
"grad_norm": 14.775269508361816,
|
| 490 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 491 |
+
"loss": 0.052,
|
| 492 |
+
"step": 240
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 36.0,
|
| 496 |
+
"eval_accuracy": 0.9157894736842105,
|
| 497 |
+
"eval_loss": 0.20699043571949005,
|
| 498 |
+
"eval_runtime": 1.6355,
|
| 499 |
+
"eval_samples_per_second": 58.086,
|
| 500 |
+
"eval_steps_per_second": 1.834,
|
| 501 |
+
"step": 243
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 36.888888888888886,
|
| 505 |
+
"eval_accuracy": 0.9473684210526315,
|
| 506 |
+
"eval_loss": 0.18596884608268738,
|
| 507 |
+
"eval_runtime": 1.5409,
|
| 508 |
+
"eval_samples_per_second": 61.654,
|
| 509 |
+
"eval_steps_per_second": 1.947,
|
| 510 |
+
"step": 249
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 37.03703703703704,
|
| 514 |
+
"grad_norm": 7.913994312286377,
|
| 515 |
+
"learning_rate": 3.240740740740741e-05,
|
| 516 |
+
"loss": 0.049,
|
| 517 |
+
"step": 250
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 37.925925925925924,
|
| 521 |
+
"eval_accuracy": 0.8947368421052632,
|
| 522 |
+
"eval_loss": 0.3068939745426178,
|
| 523 |
+
"eval_runtime": 1.5267,
|
| 524 |
+
"eval_samples_per_second": 62.227,
|
| 525 |
+
"eval_steps_per_second": 1.965,
|
| 526 |
+
"step": 256
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 38.51851851851852,
|
| 530 |
+
"grad_norm": 28.389638900756836,
|
| 531 |
+
"learning_rate": 3.148148148148148e-05,
|
| 532 |
+
"loss": 0.0578,
|
| 533 |
+
"step": 260
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 38.96296296296296,
|
| 537 |
+
"eval_accuracy": 0.8736842105263158,
|
| 538 |
+
"eval_loss": 0.4477124810218811,
|
| 539 |
+
"eval_runtime": 1.6335,
|
| 540 |
+
"eval_samples_per_second": 58.158,
|
| 541 |
+
"eval_steps_per_second": 1.837,
|
| 542 |
+
"step": 263
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 40.0,
|
| 546 |
+
"grad_norm": 66.57148742675781,
|
| 547 |
+
"learning_rate": 3.055555555555556e-05,
|
| 548 |
+
"loss": 0.0533,
|
| 549 |
+
"step": 270
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 40.0,
|
| 553 |
+
"eval_accuracy": 0.8947368421052632,
|
| 554 |
+
"eval_loss": 0.26121658086776733,
|
| 555 |
+
"eval_runtime": 1.5219,
|
| 556 |
+
"eval_samples_per_second": 62.422,
|
| 557 |
+
"eval_steps_per_second": 1.971,
|
| 558 |
+
"step": 270
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 40.888888888888886,
|
| 562 |
+
"eval_accuracy": 0.8842105263157894,
|
| 563 |
+
"eval_loss": 0.264914333820343,
|
| 564 |
+
"eval_runtime": 1.509,
|
| 565 |
+
"eval_samples_per_second": 62.957,
|
| 566 |
+
"eval_steps_per_second": 1.988,
|
| 567 |
+
"step": 276
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 41.48148148148148,
|
| 571 |
+
"grad_norm": 5.600070953369141,
|
| 572 |
+
"learning_rate": 2.962962962962963e-05,
|
| 573 |
+
"loss": 0.0505,
|
| 574 |
+
"step": 280
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 41.925925925925924,
|
| 578 |
+
"eval_accuracy": 0.9263157894736842,
|
| 579 |
+
"eval_loss": 0.19498933851718903,
|
| 580 |
+
"eval_runtime": 1.6324,
|
| 581 |
+
"eval_samples_per_second": 58.195,
|
| 582 |
+
"eval_steps_per_second": 1.838,
|
| 583 |
+
"step": 283
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 42.96296296296296,
|
| 587 |
+
"grad_norm": 5.095949172973633,
|
| 588 |
+
"learning_rate": 2.8703703703703706e-05,
|
| 589 |
+
"loss": 0.0433,
|
| 590 |
+
"step": 290
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 42.96296296296296,
|
| 594 |
+
"eval_accuracy": 0.8842105263157894,
|
| 595 |
+
"eval_loss": 0.29025542736053467,
|
| 596 |
+
"eval_runtime": 1.5094,
|
| 597 |
+
"eval_samples_per_second": 62.938,
|
| 598 |
+
"eval_steps_per_second": 1.988,
|
| 599 |
+
"step": 290
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 44.0,
|
| 603 |
+
"eval_accuracy": 0.9368421052631579,
|
| 604 |
+
"eval_loss": 0.2526479661464691,
|
| 605 |
+
"eval_runtime": 1.523,
|
| 606 |
+
"eval_samples_per_second": 62.376,
|
| 607 |
+
"eval_steps_per_second": 1.97,
|
| 608 |
+
"step": 297
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 44.44444444444444,
|
| 612 |
+
"grad_norm": 9.107303619384766,
|
| 613 |
+
"learning_rate": 2.777777777777778e-05,
|
| 614 |
+
"loss": 0.0395,
|
| 615 |
+
"step": 300
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 44.888888888888886,
|
| 619 |
+
"eval_accuracy": 0.8842105263157894,
|
| 620 |
+
"eval_loss": 0.30155882239341736,
|
| 621 |
+
"eval_runtime": 1.7275,
|
| 622 |
+
"eval_samples_per_second": 54.991,
|
| 623 |
+
"eval_steps_per_second": 1.737,
|
| 624 |
+
"step": 303
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 45.925925925925924,
|
| 628 |
+
"grad_norm": 10.148141860961914,
|
| 629 |
+
"learning_rate": 2.6851851851851855e-05,
|
| 630 |
+
"loss": 0.035,
|
| 631 |
+
"step": 310
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 45.925925925925924,
|
| 635 |
+
"eval_accuracy": 0.8947368421052632,
|
| 636 |
+
"eval_loss": 0.3509025275707245,
|
| 637 |
+
"eval_runtime": 1.5108,
|
| 638 |
+
"eval_samples_per_second": 62.88,
|
| 639 |
+
"eval_steps_per_second": 1.986,
|
| 640 |
+
"step": 310
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 46.96296296296296,
|
| 644 |
+
"eval_accuracy": 0.8842105263157894,
|
| 645 |
+
"eval_loss": 0.29430651664733887,
|
| 646 |
+
"eval_runtime": 1.5062,
|
| 647 |
+
"eval_samples_per_second": 63.071,
|
| 648 |
+
"eval_steps_per_second": 1.992,
|
| 649 |
+
"step": 317
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 47.407407407407405,
|
| 653 |
+
"grad_norm": 4.681646347045898,
|
| 654 |
+
"learning_rate": 2.5925925925925925e-05,
|
| 655 |
+
"loss": 0.0335,
|
| 656 |
+
"step": 320
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 48.0,
|
| 660 |
+
"eval_accuracy": 0.8842105263157894,
|
| 661 |
+
"eval_loss": 0.2613106667995453,
|
| 662 |
+
"eval_runtime": 1.6847,
|
| 663 |
+
"eval_samples_per_second": 56.39,
|
| 664 |
+
"eval_steps_per_second": 1.781,
|
| 665 |
+
"step": 324
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 48.888888888888886,
|
| 669 |
+
"grad_norm": 10.262737274169922,
|
| 670 |
+
"learning_rate": 2.5e-05,
|
| 671 |
+
"loss": 0.0408,
|
| 672 |
+
"step": 330
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 48.888888888888886,
|
| 676 |
+
"eval_accuracy": 0.9157894736842105,
|
| 677 |
+
"eval_loss": 0.21650759875774384,
|
| 678 |
+
"eval_runtime": 1.5245,
|
| 679 |
+
"eval_samples_per_second": 62.315,
|
| 680 |
+
"eval_steps_per_second": 1.968,
|
| 681 |
+
"step": 330
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 49.925925925925924,
|
| 685 |
+
"eval_accuracy": 0.9263157894736842,
|
| 686 |
+
"eval_loss": 0.28715887665748596,
|
| 687 |
+
"eval_runtime": 1.5241,
|
| 688 |
+
"eval_samples_per_second": 62.331,
|
| 689 |
+
"eval_steps_per_second": 1.968,
|
| 690 |
+
"step": 337
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 50.370370370370374,
|
| 694 |
+
"grad_norm": 9.619718551635742,
|
| 695 |
+
"learning_rate": 2.4074074074074074e-05,
|
| 696 |
+
"loss": 0.0244,
|
| 697 |
+
"step": 340
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 50.96296296296296,
|
| 701 |
+
"eval_accuracy": 0.8842105263157894,
|
| 702 |
+
"eval_loss": 0.31339362263679504,
|
| 703 |
+
"eval_runtime": 1.6247,
|
| 704 |
+
"eval_samples_per_second": 58.472,
|
| 705 |
+
"eval_steps_per_second": 1.846,
|
| 706 |
+
"step": 344
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 51.851851851851855,
|
| 710 |
+
"grad_norm": 6.175246715545654,
|
| 711 |
+
"learning_rate": 2.314814814814815e-05,
|
| 712 |
+
"loss": 0.0323,
|
| 713 |
+
"step": 350
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 52.0,
|
| 717 |
+
"eval_accuracy": 0.9157894736842105,
|
| 718 |
+
"eval_loss": 0.3006380796432495,
|
| 719 |
+
"eval_runtime": 1.5132,
|
| 720 |
+
"eval_samples_per_second": 62.78,
|
| 721 |
+
"eval_steps_per_second": 1.983,
|
| 722 |
+
"step": 351
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 52.888888888888886,
|
| 726 |
+
"eval_accuracy": 0.8736842105263158,
|
| 727 |
+
"eval_loss": 0.37583690881729126,
|
| 728 |
+
"eval_runtime": 1.5262,
|
| 729 |
+
"eval_samples_per_second": 62.247,
|
| 730 |
+
"eval_steps_per_second": 1.966,
|
| 731 |
+
"step": 357
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 53.333333333333336,
|
| 735 |
+
"grad_norm": 11.230114936828613,
|
| 736 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 737 |
+
"loss": 0.0241,
|
| 738 |
+
"step": 360
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 53.925925925925924,
|
| 742 |
+
"eval_accuracy": 0.9263157894736842,
|
| 743 |
+
"eval_loss": 0.3033463954925537,
|
| 744 |
+
"eval_runtime": 1.6639,
|
| 745 |
+
"eval_samples_per_second": 57.093,
|
| 746 |
+
"eval_steps_per_second": 1.803,
|
| 747 |
+
"step": 364
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 54.81481481481482,
|
| 751 |
+
"grad_norm": 6.497807502746582,
|
| 752 |
+
"learning_rate": 2.1296296296296296e-05,
|
| 753 |
+
"loss": 0.0193,
|
| 754 |
+
"step": 370
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 54.96296296296296,
|
| 758 |
+
"eval_accuracy": 0.9368421052631579,
|
| 759 |
+
"eval_loss": 0.27406617999076843,
|
| 760 |
+
"eval_runtime": 1.5407,
|
| 761 |
+
"eval_samples_per_second": 61.662,
|
| 762 |
+
"eval_steps_per_second": 1.947,
|
| 763 |
+
"step": 371
|
| 764 |
+
},
|
| 765 |
+
{
|
| 766 |
+
"epoch": 56.0,
|
| 767 |
+
"eval_accuracy": 0.9368421052631579,
|
| 768 |
+
"eval_loss": 0.1684454083442688,
|
| 769 |
+
"eval_runtime": 1.5222,
|
| 770 |
+
"eval_samples_per_second": 62.409,
|
| 771 |
+
"eval_steps_per_second": 1.971,
|
| 772 |
+
"step": 378
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 56.2962962962963,
|
| 776 |
+
"grad_norm": 1.369285225868225,
|
| 777 |
+
"learning_rate": 2.037037037037037e-05,
|
| 778 |
+
"loss": 0.0273,
|
| 779 |
+
"step": 380
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 56.888888888888886,
|
| 783 |
+
"eval_accuracy": 0.9473684210526315,
|
| 784 |
+
"eval_loss": 0.240325465798378,
|
| 785 |
+
"eval_runtime": 1.6193,
|
| 786 |
+
"eval_samples_per_second": 58.666,
|
| 787 |
+
"eval_steps_per_second": 1.853,
|
| 788 |
+
"step": 384
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 57.77777777777778,
|
| 792 |
+
"grad_norm": 16.219879150390625,
|
| 793 |
+
"learning_rate": 1.9444444444444445e-05,
|
| 794 |
+
"loss": 0.0244,
|
| 795 |
+
"step": 390
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 57.925925925925924,
|
| 799 |
+
"eval_accuracy": 0.9473684210526315,
|
| 800 |
+
"eval_loss": 0.14995306730270386,
|
| 801 |
+
"eval_runtime": 1.509,
|
| 802 |
+
"eval_samples_per_second": 62.958,
|
| 803 |
+
"eval_steps_per_second": 1.988,
|
| 804 |
+
"step": 391
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 58.96296296296296,
|
| 808 |
+
"eval_accuracy": 0.9368421052631579,
|
| 809 |
+
"eval_loss": 0.13768525421619415,
|
| 810 |
+
"eval_runtime": 1.5002,
|
| 811 |
+
"eval_samples_per_second": 63.323,
|
| 812 |
+
"eval_steps_per_second": 2.0,
|
| 813 |
+
"step": 398
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 59.25925925925926,
|
| 817 |
+
"grad_norm": 17.92609214782715,
|
| 818 |
+
"learning_rate": 1.8518518518518518e-05,
|
| 819 |
+
"loss": 0.0268,
|
| 820 |
+
"step": 400
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 60.0,
|
| 824 |
+
"eval_accuracy": 0.9157894736842105,
|
| 825 |
+
"eval_loss": 0.18984580039978027,
|
| 826 |
+
"eval_runtime": 1.5627,
|
| 827 |
+
"eval_samples_per_second": 60.793,
|
| 828 |
+
"eval_steps_per_second": 1.92,
|
| 829 |
+
"step": 405
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 60.74074074074074,
|
| 833 |
+
"grad_norm": 4.754770278930664,
|
| 834 |
+
"learning_rate": 1.7592592592592595e-05,
|
| 835 |
+
"loss": 0.0405,
|
| 836 |
+
"step": 410
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 60.888888888888886,
|
| 840 |
+
"eval_accuracy": 0.9052631578947369,
|
| 841 |
+
"eval_loss": 0.17558255791664124,
|
| 842 |
+
"eval_runtime": 1.5023,
|
| 843 |
+
"eval_samples_per_second": 63.237,
|
| 844 |
+
"eval_steps_per_second": 1.997,
|
| 845 |
+
"step": 411
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 61.925925925925924,
|
| 849 |
+
"eval_accuracy": 0.9263157894736842,
|
| 850 |
+
"eval_loss": 0.1907454878091812,
|
| 851 |
+
"eval_runtime": 1.5123,
|
| 852 |
+
"eval_samples_per_second": 62.817,
|
| 853 |
+
"eval_steps_per_second": 1.984,
|
| 854 |
+
"step": 418
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 62.22222222222222,
|
| 858 |
+
"grad_norm": 6.852460861206055,
|
| 859 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 860 |
+
"loss": 0.0219,
|
| 861 |
+
"step": 420
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 62.96296296296296,
|
| 865 |
+
"eval_accuracy": 0.9052631578947369,
|
| 866 |
+
"eval_loss": 0.17904597520828247,
|
| 867 |
+
"eval_runtime": 1.5634,
|
| 868 |
+
"eval_samples_per_second": 60.763,
|
| 869 |
+
"eval_steps_per_second": 1.919,
|
| 870 |
+
"step": 425
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 63.7037037037037,
|
| 874 |
+
"grad_norm": 17.433446884155273,
|
| 875 |
+
"learning_rate": 1.574074074074074e-05,
|
| 876 |
+
"loss": 0.0329,
|
| 877 |
+
"step": 430
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 64.0,
|
| 881 |
+
"eval_accuracy": 0.9368421052631579,
|
| 882 |
+
"eval_loss": 0.18854853510856628,
|
| 883 |
+
"eval_runtime": 1.5466,
|
| 884 |
+
"eval_samples_per_second": 61.427,
|
| 885 |
+
"eval_steps_per_second": 1.94,
|
| 886 |
+
"step": 432
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 64.88888888888889,
|
| 890 |
+
"eval_accuracy": 0.9368421052631579,
|
| 891 |
+
"eval_loss": 0.15500715374946594,
|
| 892 |
+
"eval_runtime": 1.5091,
|
| 893 |
+
"eval_samples_per_second": 62.951,
|
| 894 |
+
"eval_steps_per_second": 1.988,
|
| 895 |
+
"step": 438
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 65.18518518518519,
|
| 899 |
+
"grad_norm": 12.199813842773438,
|
| 900 |
+
"learning_rate": 1.4814814814814815e-05,
|
| 901 |
+
"loss": 0.019,
|
| 902 |
+
"step": 440
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 65.92592592592592,
|
| 906 |
+
"eval_accuracy": 0.9157894736842105,
|
| 907 |
+
"eval_loss": 0.18106068670749664,
|
| 908 |
+
"eval_runtime": 1.5434,
|
| 909 |
+
"eval_samples_per_second": 61.553,
|
| 910 |
+
"eval_steps_per_second": 1.944,
|
| 911 |
+
"step": 445
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 66.66666666666667,
|
| 915 |
+
"grad_norm": 5.085737228393555,
|
| 916 |
+
"learning_rate": 1.388888888888889e-05,
|
| 917 |
+
"loss": 0.0205,
|
| 918 |
+
"step": 450
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 66.96296296296296,
|
| 922 |
+
"eval_accuracy": 0.9263157894736842,
|
| 923 |
+
"eval_loss": 0.2165663242340088,
|
| 924 |
+
"eval_runtime": 1.5356,
|
| 925 |
+
"eval_samples_per_second": 61.864,
|
| 926 |
+
"eval_steps_per_second": 1.954,
|
| 927 |
+
"step": 452
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 68.0,
|
| 931 |
+
"eval_accuracy": 0.9052631578947369,
|
| 932 |
+
"eval_loss": 0.17012225091457367,
|
| 933 |
+
"eval_runtime": 1.505,
|
| 934 |
+
"eval_samples_per_second": 63.122,
|
| 935 |
+
"eval_steps_per_second": 1.993,
|
| 936 |
+
"step": 459
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 68.14814814814815,
|
| 940 |
+
"grad_norm": 4.042739391326904,
|
| 941 |
+
"learning_rate": 1.2962962962962962e-05,
|
| 942 |
+
"loss": 0.0232,
|
| 943 |
+
"step": 460
|
| 944 |
+
},
|
| 945 |
+
{
|
| 946 |
+
"epoch": 68.88888888888889,
|
| 947 |
+
"eval_accuracy": 0.9157894736842105,
|
| 948 |
+
"eval_loss": 0.21532098948955536,
|
| 949 |
+
"eval_runtime": 1.4982,
|
| 950 |
+
"eval_samples_per_second": 63.41,
|
| 951 |
+
"eval_steps_per_second": 2.002,
|
| 952 |
+
"step": 465
|
| 953 |
+
},
|
| 954 |
+
{
|
| 955 |
+
"epoch": 69.62962962962963,
|
| 956 |
+
"grad_norm": 3.4432425498962402,
|
| 957 |
+
"learning_rate": 1.2037037037037037e-05,
|
| 958 |
+
"loss": 0.0269,
|
| 959 |
+
"step": 470
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 69.92592592592592,
|
| 963 |
+
"eval_accuracy": 0.9263157894736842,
|
| 964 |
+
"eval_loss": 0.22287000715732574,
|
| 965 |
+
"eval_runtime": 1.5307,
|
| 966 |
+
"eval_samples_per_second": 62.063,
|
| 967 |
+
"eval_steps_per_second": 1.96,
|
| 968 |
+
"step": 472
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 70.96296296296296,
|
| 972 |
+
"eval_accuracy": 0.9263157894736842,
|
| 973 |
+
"eval_loss": 0.2237367182970047,
|
| 974 |
+
"eval_runtime": 1.5031,
|
| 975 |
+
"eval_samples_per_second": 63.204,
|
| 976 |
+
"eval_steps_per_second": 1.996,
|
| 977 |
+
"step": 479
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 71.11111111111111,
|
| 981 |
+
"grad_norm": 9.640032768249512,
|
| 982 |
+
"learning_rate": 1.1111111111111112e-05,
|
| 983 |
+
"loss": 0.0306,
|
| 984 |
+
"step": 480
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 72.0,
|
| 988 |
+
"eval_accuracy": 0.9368421052631579,
|
| 989 |
+
"eval_loss": 0.18282432854175568,
|
| 990 |
+
"eval_runtime": 1.5,
|
| 991 |
+
"eval_samples_per_second": 63.334,
|
| 992 |
+
"eval_steps_per_second": 2.0,
|
| 993 |
+
"step": 486
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 72.5925925925926,
|
| 997 |
+
"grad_norm": 9.28294849395752,
|
| 998 |
+
"learning_rate": 1.0185185185185185e-05,
|
| 999 |
+
"loss": 0.0298,
|
| 1000 |
+
"step": 490
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 72.88888888888889,
|
| 1004 |
+
"eval_accuracy": 0.9368421052631579,
|
| 1005 |
+
"eval_loss": 0.14476627111434937,
|
| 1006 |
+
"eval_runtime": 1.5255,
|
| 1007 |
+
"eval_samples_per_second": 62.276,
|
| 1008 |
+
"eval_steps_per_second": 1.967,
|
| 1009 |
+
"step": 492
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 73.92592592592592,
|
| 1013 |
+
"eval_accuracy": 0.9157894736842105,
|
| 1014 |
+
"eval_loss": 0.19477160274982452,
|
| 1015 |
+
"eval_runtime": 1.5341,
|
| 1016 |
+
"eval_samples_per_second": 61.925,
|
| 1017 |
+
"eval_steps_per_second": 1.956,
|
| 1018 |
+
"step": 499
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 74.07407407407408,
|
| 1022 |
+
"grad_norm": 0.5367471575737,
|
| 1023 |
+
"learning_rate": 9.259259259259259e-06,
|
| 1024 |
+
"loss": 0.0154,
|
| 1025 |
+
"step": 500
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 74.96296296296296,
|
| 1029 |
+
"eval_accuracy": 0.9157894736842105,
|
| 1030 |
+
"eval_loss": 0.25699615478515625,
|
| 1031 |
+
"eval_runtime": 1.5191,
|
| 1032 |
+
"eval_samples_per_second": 62.538,
|
| 1033 |
+
"eval_steps_per_second": 1.975,
|
| 1034 |
+
"step": 506
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 75.55555555555556,
|
| 1038 |
+
"grad_norm": 16.879281997680664,
|
| 1039 |
+
"learning_rate": 8.333333333333334e-06,
|
| 1040 |
+
"loss": 0.0193,
|
| 1041 |
+
"step": 510
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 76.0,
|
| 1045 |
+
"eval_accuracy": 0.9263157894736842,
|
| 1046 |
+
"eval_loss": 0.24617379903793335,
|
| 1047 |
+
"eval_runtime": 1.5485,
|
| 1048 |
+
"eval_samples_per_second": 61.351,
|
| 1049 |
+
"eval_steps_per_second": 1.937,
|
| 1050 |
+
"step": 513
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 76.88888888888889,
|
| 1054 |
+
"eval_accuracy": 0.9157894736842105,
|
| 1055 |
+
"eval_loss": 0.21939770877361298,
|
| 1056 |
+
"eval_runtime": 1.5513,
|
| 1057 |
+
"eval_samples_per_second": 61.24,
|
| 1058 |
+
"eval_steps_per_second": 1.934,
|
| 1059 |
+
"step": 519
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 77.03703703703704,
|
| 1063 |
+
"grad_norm": 9.813831329345703,
|
| 1064 |
+
"learning_rate": 7.4074074074074075e-06,
|
| 1065 |
+
"loss": 0.0188,
|
| 1066 |
+
"step": 520
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 77.92592592592592,
|
| 1070 |
+
"eval_accuracy": 0.9157894736842105,
|
| 1071 |
+
"eval_loss": 0.22538267076015472,
|
| 1072 |
+
"eval_runtime": 1.5394,
|
| 1073 |
+
"eval_samples_per_second": 61.71,
|
| 1074 |
+
"eval_steps_per_second": 1.949,
|
| 1075 |
+
"step": 526
|
| 1076 |
+
},
|
| 1077 |
+
{
|
| 1078 |
+
"epoch": 78.51851851851852,
|
| 1079 |
+
"grad_norm": 6.672292709350586,
|
| 1080 |
+
"learning_rate": 6.481481481481481e-06,
|
| 1081 |
+
"loss": 0.0198,
|
| 1082 |
+
"step": 530
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 78.96296296296296,
|
| 1086 |
+
"eval_accuracy": 0.9263157894736842,
|
| 1087 |
+
"eval_loss": 0.192366823554039,
|
| 1088 |
+
"eval_runtime": 1.511,
|
| 1089 |
+
"eval_samples_per_second": 62.873,
|
| 1090 |
+
"eval_steps_per_second": 1.985,
|
| 1091 |
+
"step": 533
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 80.0,
|
| 1095 |
+
"grad_norm": 12.444893836975098,
|
| 1096 |
+
"learning_rate": 5.555555555555556e-06,
|
| 1097 |
+
"loss": 0.0147,
|
| 1098 |
+
"step": 540
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 80.0,
|
| 1102 |
+
"eval_accuracy": 0.9368421052631579,
|
| 1103 |
+
"eval_loss": 0.15250863134860992,
|
| 1104 |
+
"eval_runtime": 1.5101,
|
| 1105 |
+
"eval_samples_per_second": 62.91,
|
| 1106 |
+
"eval_steps_per_second": 1.987,
|
| 1107 |
+
"step": 540
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 80.88888888888889,
|
| 1111 |
+
"eval_accuracy": 0.9473684210526315,
|
| 1112 |
+
"eval_loss": 0.13136789202690125,
|
| 1113 |
+
"eval_runtime": 1.5108,
|
| 1114 |
+
"eval_samples_per_second": 62.88,
|
| 1115 |
+
"eval_steps_per_second": 1.986,
|
| 1116 |
+
"step": 546
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 81.48148148148148,
|
| 1120 |
+
"grad_norm": 5.455392837524414,
|
| 1121 |
+
"learning_rate": 4.6296296296296296e-06,
|
| 1122 |
+
"loss": 0.0282,
|
| 1123 |
+
"step": 550
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 81.92592592592592,
|
| 1127 |
+
"eval_accuracy": 0.9368421052631579,
|
| 1128 |
+
"eval_loss": 0.13807313144207,
|
| 1129 |
+
"eval_runtime": 1.5108,
|
| 1130 |
+
"eval_samples_per_second": 62.881,
|
| 1131 |
+
"eval_steps_per_second": 1.986,
|
| 1132 |
+
"step": 553
|
| 1133 |
+
},
|
| 1134 |
+
{
|
| 1135 |
+
"epoch": 82.96296296296296,
|
| 1136 |
+
"grad_norm": 0.557128369808197,
|
| 1137 |
+
"learning_rate": 3.7037037037037037e-06,
|
| 1138 |
+
"loss": 0.0168,
|
| 1139 |
+
"step": 560
|
| 1140 |
+
},
|
| 1141 |
+
{
|
| 1142 |
+
"epoch": 82.96296296296296,
|
| 1143 |
+
"eval_accuracy": 0.9157894736842105,
|
| 1144 |
+
"eval_loss": 0.14955918490886688,
|
| 1145 |
+
"eval_runtime": 1.5176,
|
| 1146 |
+
"eval_samples_per_second": 62.599,
|
| 1147 |
+
"eval_steps_per_second": 1.977,
|
| 1148 |
+
"step": 560
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 84.0,
|
| 1152 |
+
"eval_accuracy": 0.9263157894736842,
|
| 1153 |
+
"eval_loss": 0.1806280016899109,
|
| 1154 |
+
"eval_runtime": 1.5082,
|
| 1155 |
+
"eval_samples_per_second": 62.991,
|
| 1156 |
+
"eval_steps_per_second": 1.989,
|
| 1157 |
+
"step": 567
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 84.44444444444444,
|
| 1161 |
+
"grad_norm": 6.701883316040039,
|
| 1162 |
+
"learning_rate": 2.777777777777778e-06,
|
| 1163 |
+
"loss": 0.018,
|
| 1164 |
+
"step": 570
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 84.88888888888889,
|
| 1168 |
+
"eval_accuracy": 0.9263157894736842,
|
| 1169 |
+
"eval_loss": 0.20803439617156982,
|
| 1170 |
+
"eval_runtime": 1.5233,
|
| 1171 |
+
"eval_samples_per_second": 62.366,
|
| 1172 |
+
"eval_steps_per_second": 1.969,
|
| 1173 |
+
"step": 573
|
| 1174 |
+
},
|
| 1175 |
+
{
|
| 1176 |
+
"epoch": 85.92592592592592,
|
| 1177 |
+
"grad_norm": 5.107237339019775,
|
| 1178 |
+
"learning_rate": 1.8518518518518519e-06,
|
| 1179 |
+
"loss": 0.0172,
|
| 1180 |
+
"step": 580
|
| 1181 |
+
},
|
| 1182 |
+
{
|
| 1183 |
+
"epoch": 85.92592592592592,
|
| 1184 |
+
"eval_accuracy": 0.9157894736842105,
|
| 1185 |
+
"eval_loss": 0.21987786889076233,
|
| 1186 |
+
"eval_runtime": 1.5032,
|
| 1187 |
+
"eval_samples_per_second": 63.198,
|
| 1188 |
+
"eval_steps_per_second": 1.996,
|
| 1189 |
+
"step": 580
|
| 1190 |
+
},
|
| 1191 |
+
{
|
| 1192 |
+
"epoch": 86.96296296296296,
|
| 1193 |
+
"eval_accuracy": 0.9263157894736842,
|
| 1194 |
+
"eval_loss": 0.19394682347774506,
|
| 1195 |
+
"eval_runtime": 1.5704,
|
| 1196 |
+
"eval_samples_per_second": 60.494,
|
| 1197 |
+
"eval_steps_per_second": 1.91,
|
| 1198 |
+
"step": 587
|
| 1199 |
+
},
|
| 1200 |
+
{
|
| 1201 |
+
"epoch": 87.4074074074074,
|
| 1202 |
+
"grad_norm": 13.376716613769531,
|
| 1203 |
+
"learning_rate": 9.259259259259259e-07,
|
| 1204 |
+
"loss": 0.0117,
|
| 1205 |
+
"step": 590
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"epoch": 88.0,
|
| 1209 |
+
"eval_accuracy": 0.9263157894736842,
|
| 1210 |
+
"eval_loss": 0.18152017891407013,
|
| 1211 |
+
"eval_runtime": 1.5369,
|
| 1212 |
+
"eval_samples_per_second": 61.812,
|
| 1213 |
+
"eval_steps_per_second": 1.952,
|
| 1214 |
+
"step": 594
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 88.88888888888889,
|
| 1218 |
+
"grad_norm": 8.48166561126709,
|
| 1219 |
+
"learning_rate": 0.0,
|
| 1220 |
+
"loss": 0.0149,
|
| 1221 |
+
"step": 600
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 88.88888888888889,
|
| 1225 |
+
"eval_accuracy": 0.9263157894736842,
|
| 1226 |
+
"eval_loss": 0.1811211258172989,
|
| 1227 |
+
"eval_runtime": 1.529,
|
| 1228 |
+
"eval_samples_per_second": 62.13,
|
| 1229 |
+
"eval_steps_per_second": 1.962,
|
| 1230 |
+
"step": 600
|
| 1231 |
+
},
|
| 1232 |
+
{
|
| 1233 |
+
"epoch": 88.88888888888889,
|
| 1234 |
+
"step": 600,
|
| 1235 |
+
"total_flos": 1.9132429834630103e+18,
|
| 1236 |
+
"train_loss": 0.19732174752900997,
|
| 1237 |
+
"train_runtime": 1871.4873,
|
| 1238 |
+
"train_samples_per_second": 45.686,
|
| 1239 |
+
"train_steps_per_second": 0.321
|
| 1240 |
+
}
|
| 1241 |
+
],
|
| 1242 |
+
"logging_steps": 10,
|
| 1243 |
+
"max_steps": 600,
|
| 1244 |
+
"num_input_tokens_seen": 0,
|
| 1245 |
+
"num_train_epochs": 100,
|
| 1246 |
+
"save_steps": 500,
|
| 1247 |
+
"stateful_callbacks": {
|
| 1248 |
+
"TrainerControl": {
|
| 1249 |
+
"args": {
|
| 1250 |
+
"should_epoch_stop": false,
|
| 1251 |
+
"should_evaluate": false,
|
| 1252 |
+
"should_log": false,
|
| 1253 |
+
"should_save": false,
|
| 1254 |
+
"should_training_stop": false
|
| 1255 |
+
},
|
| 1256 |
+
"attributes": {}
|
| 1257 |
+
}
|
| 1258 |
+
},
|
| 1259 |
+
"total_flos": 1.9132429834630103e+18,
|
| 1260 |
+
"train_batch_size": 32,
|
| 1261 |
+
"trial_name": null,
|
| 1262 |
+
"trial_params": null
|
| 1263 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:18be3493fba34b5ff2a06754d242143c7cc13c3f17fadae41c1895a702cd0069
|
| 3 |
+
size 5240
|