Commit
•
b813155
1
Parent(s):
c5a24b6
End of training
Browse files- README.md +154 -0
- all_results.json +13 -0
- config.json +59 -0
- eval_results.json +8 -0
- model.safetensors +3 -0
- preprocessor_config.json +22 -0
- train_results.json +8 -0
- trainer_state.json +1071 -0
- training_args.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,154 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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-sleeve-length
|
| 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: 0.8620689655172413
|
| 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-sleeve-length
|
| 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: 0.5496
|
| 36 |
+
- Accuracy: 0.8621
|
| 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: 80
|
| 65 |
+
|
| 66 |
+
### Training results
|
| 67 |
+
|
| 68 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 69 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 70 |
+
| No log | 0.96 | 6 | 1.7957 | 0.2299 |
|
| 71 |
+
| 1.8656 | 1.92 | 12 | 1.7704 | 0.2759 |
|
| 72 |
+
| 1.8656 | 2.88 | 18 | 1.7382 | 0.3218 |
|
| 73 |
+
| 1.7835 | 4.0 | 25 | 1.6674 | 0.3793 |
|
| 74 |
+
| 1.664 | 4.96 | 31 | 1.5982 | 0.4253 |
|
| 75 |
+
| 1.664 | 5.92 | 37 | 1.4861 | 0.4368 |
|
| 76 |
+
| 1.5072 | 6.88 | 43 | 1.3645 | 0.4713 |
|
| 77 |
+
| 1.3304 | 8.0 | 50 | 1.2859 | 0.4598 |
|
| 78 |
+
| 1.3304 | 8.96 | 56 | 1.2796 | 0.4713 |
|
| 79 |
+
| 1.1651 | 9.92 | 62 | 1.2456 | 0.5172 |
|
| 80 |
+
| 1.1651 | 10.88 | 68 | 1.1667 | 0.5402 |
|
| 81 |
+
| 1.0876 | 12.0 | 75 | 1.1510 | 0.5632 |
|
| 82 |
+
| 1.0046 | 12.96 | 81 | 1.0510 | 0.6092 |
|
| 83 |
+
| 1.0046 | 13.92 | 87 | 1.0338 | 0.5862 |
|
| 84 |
+
| 0.9465 | 14.88 | 93 | 0.9883 | 0.5862 |
|
| 85 |
+
| 0.8699 | 16.0 | 100 | 0.9882 | 0.5632 |
|
| 86 |
+
| 0.8699 | 16.96 | 106 | 0.9276 | 0.5747 |
|
| 87 |
+
| 0.7969 | 17.92 | 112 | 0.9145 | 0.5862 |
|
| 88 |
+
| 0.7969 | 18.88 | 118 | 0.8144 | 0.6667 |
|
| 89 |
+
| 0.7254 | 20.0 | 125 | 0.7587 | 0.6667 |
|
| 90 |
+
| 0.6447 | 20.96 | 131 | 0.6990 | 0.7471 |
|
| 91 |
+
| 0.6447 | 21.92 | 137 | 0.7042 | 0.7241 |
|
| 92 |
+
| 0.6021 | 22.88 | 143 | 0.6526 | 0.7701 |
|
| 93 |
+
| 0.516 | 24.0 | 150 | 0.6485 | 0.8046 |
|
| 94 |
+
| 0.516 | 24.96 | 156 | 0.5803 | 0.8161 |
|
| 95 |
+
| 0.4497 | 25.92 | 162 | 0.6085 | 0.8046 |
|
| 96 |
+
| 0.4497 | 26.88 | 168 | 0.6095 | 0.8046 |
|
| 97 |
+
| 0.3935 | 28.0 | 175 | 0.5372 | 0.8276 |
|
| 98 |
+
| 0.3321 | 28.96 | 181 | 0.5829 | 0.8161 |
|
| 99 |
+
| 0.3321 | 29.92 | 187 | 0.6205 | 0.8161 |
|
| 100 |
+
| 0.3007 | 30.88 | 193 | 0.5150 | 0.8276 |
|
| 101 |
+
| 0.2618 | 32.0 | 200 | 0.6069 | 0.8391 |
|
| 102 |
+
| 0.2618 | 32.96 | 206 | 0.5273 | 0.8391 |
|
| 103 |
+
| 0.2411 | 33.92 | 212 | 0.4727 | 0.8621 |
|
| 104 |
+
| 0.2411 | 34.88 | 218 | 0.4611 | 0.8736 |
|
| 105 |
+
| 0.2108 | 36.0 | 225 | 0.5696 | 0.8506 |
|
| 106 |
+
| 0.2143 | 36.96 | 231 | 0.4944 | 0.8621 |
|
| 107 |
+
| 0.2143 | 37.92 | 237 | 0.5628 | 0.8161 |
|
| 108 |
+
| 0.1663 | 38.88 | 243 | 0.6131 | 0.8046 |
|
| 109 |
+
| 0.1714 | 40.0 | 250 | 0.4962 | 0.8506 |
|
| 110 |
+
| 0.1714 | 40.96 | 256 | 0.5023 | 0.8391 |
|
| 111 |
+
| 0.174 | 41.92 | 262 | 0.4842 | 0.8276 |
|
| 112 |
+
| 0.174 | 42.88 | 268 | 0.4679 | 0.8276 |
|
| 113 |
+
| 0.138 | 44.0 | 275 | 0.6271 | 0.8161 |
|
| 114 |
+
| 0.1437 | 44.96 | 281 | 0.5326 | 0.8506 |
|
| 115 |
+
| 0.1437 | 45.92 | 287 | 0.5655 | 0.8161 |
|
| 116 |
+
| 0.136 | 46.88 | 293 | 0.4672 | 0.8391 |
|
| 117 |
+
| 0.1401 | 48.0 | 300 | 0.4990 | 0.8621 |
|
| 118 |
+
| 0.1401 | 48.96 | 306 | 0.5445 | 0.8276 |
|
| 119 |
+
| 0.1281 | 49.92 | 312 | 0.4761 | 0.8736 |
|
| 120 |
+
| 0.1281 | 50.88 | 318 | 0.5665 | 0.8506 |
|
| 121 |
+
| 0.1156 | 52.0 | 325 | 0.5090 | 0.8506 |
|
| 122 |
+
| 0.0981 | 52.96 | 331 | 0.5152 | 0.8506 |
|
| 123 |
+
| 0.0981 | 53.92 | 337 | 0.5466 | 0.8161 |
|
| 124 |
+
| 0.1055 | 54.88 | 343 | 0.5390 | 0.8276 |
|
| 125 |
+
| 0.112 | 56.0 | 350 | 0.5574 | 0.8506 |
|
| 126 |
+
| 0.112 | 56.96 | 356 | 0.5449 | 0.8506 |
|
| 127 |
+
| 0.0855 | 57.92 | 362 | 0.5390 | 0.8506 |
|
| 128 |
+
| 0.0855 | 58.88 | 368 | 0.5206 | 0.8506 |
|
| 129 |
+
| 0.0899 | 60.0 | 375 | 0.5476 | 0.8621 |
|
| 130 |
+
| 0.1026 | 60.96 | 381 | 0.5344 | 0.8506 |
|
| 131 |
+
| 0.1026 | 61.92 | 387 | 0.5531 | 0.8391 |
|
| 132 |
+
| 0.0799 | 62.88 | 393 | 0.5723 | 0.8276 |
|
| 133 |
+
| 0.0844 | 64.0 | 400 | 0.5340 | 0.8161 |
|
| 134 |
+
| 0.0844 | 64.96 | 406 | 0.5236 | 0.8736 |
|
| 135 |
+
| 0.0724 | 65.92 | 412 | 0.6137 | 0.8391 |
|
| 136 |
+
| 0.0724 | 66.88 | 418 | 0.5825 | 0.8276 |
|
| 137 |
+
| 0.0867 | 68.0 | 425 | 0.5105 | 0.8621 |
|
| 138 |
+
| 0.071 | 68.96 | 431 | 0.5272 | 0.8506 |
|
| 139 |
+
| 0.071 | 69.92 | 437 | 0.5524 | 0.8506 |
|
| 140 |
+
| 0.0723 | 70.88 | 443 | 0.5508 | 0.8391 |
|
| 141 |
+
| 0.0748 | 72.0 | 450 | 0.5689 | 0.8161 |
|
| 142 |
+
| 0.0748 | 72.96 | 456 | 0.5556 | 0.8506 |
|
| 143 |
+
| 0.0589 | 73.92 | 462 | 0.5452 | 0.8506 |
|
| 144 |
+
| 0.0589 | 74.88 | 468 | 0.5475 | 0.8621 |
|
| 145 |
+
| 0.0719 | 76.0 | 475 | 0.5484 | 0.8621 |
|
| 146 |
+
| 0.0801 | 76.8 | 480 | 0.5496 | 0.8621 |
|
| 147 |
+
|
| 148 |
+
|
| 149 |
+
### Framework versions
|
| 150 |
+
|
| 151 |
+
- Transformers 4.44.0
|
| 152 |
+
- Pytorch 2.4.0
|
| 153 |
+
- Datasets 2.21.0
|
| 154 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 76.8,
|
| 3 |
+
"eval_accuracy": 0.8620689655172413,
|
| 4 |
+
"eval_loss": 0.5496163368225098,
|
| 5 |
+
"eval_runtime": 2.0578,
|
| 6 |
+
"eval_samples_per_second": 42.279,
|
| 7 |
+
"eval_steps_per_second": 1.458,
|
| 8 |
+
"total_flos": 1.514063180200919e+18,
|
| 9 |
+
"train_loss": 0.449434948215882,
|
| 10 |
+
"train_runtime": 1985.4684,
|
| 11 |
+
"train_samples_per_second": 31.549,
|
| 12 |
+
"train_steps_per_second": 0.242
|
| 13 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,59 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": "bell sleeve",
|
| 22 |
+
"1": "half sleeve",
|
| 23 |
+
"2": "long sleeve",
|
| 24 |
+
"3": "short sleeve",
|
| 25 |
+
"4": "sleeveless",
|
| 26 |
+
"5": "three-quarter sleeves"
|
| 27 |
+
},
|
| 28 |
+
"image_size": 224,
|
| 29 |
+
"initializer_range": 0.02,
|
| 30 |
+
"label2id": {
|
| 31 |
+
"bell sleeve": 0,
|
| 32 |
+
"half sleeve": 1,
|
| 33 |
+
"long sleeve": 2,
|
| 34 |
+
"short sleeve": 3,
|
| 35 |
+
"sleeveless": 4,
|
| 36 |
+
"three-quarter sleeves": 5
|
| 37 |
+
},
|
| 38 |
+
"layer_norm_eps": 1e-12,
|
| 39 |
+
"model_type": "convnextv2",
|
| 40 |
+
"num_channels": 3,
|
| 41 |
+
"num_stages": 4,
|
| 42 |
+
"out_features": [
|
| 43 |
+
"stage4"
|
| 44 |
+
],
|
| 45 |
+
"out_indices": [
|
| 46 |
+
4
|
| 47 |
+
],
|
| 48 |
+
"patch_size": 4,
|
| 49 |
+
"problem_type": "single_label_classification",
|
| 50 |
+
"stage_names": [
|
| 51 |
+
"stem",
|
| 52 |
+
"stage1",
|
| 53 |
+
"stage2",
|
| 54 |
+
"stage3",
|
| 55 |
+
"stage4"
|
| 56 |
+
],
|
| 57 |
+
"torch_dtype": "float32",
|
| 58 |
+
"transformers_version": "4.44.0"
|
| 59 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 76.8,
|
| 3 |
+
"eval_accuracy": 0.8620689655172413,
|
| 4 |
+
"eval_loss": 0.5496163368225098,
|
| 5 |
+
"eval_runtime": 2.0578,
|
| 6 |
+
"eval_samples_per_second": 42.279,
|
| 7 |
+
"eval_steps_per_second": 1.458
|
| 8 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7e280ffcf375e022f942b34fd32ddf23b0251721c089464e933379b60366f542
|
| 3 |
+
size 111508128
|
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": 76.8,
|
| 3 |
+
"total_flos": 1.514063180200919e+18,
|
| 4 |
+
"train_loss": 0.449434948215882,
|
| 5 |
+
"train_runtime": 1985.4684,
|
| 6 |
+
"train_samples_per_second": 31.549,
|
| 7 |
+
"train_steps_per_second": 0.242
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1071 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 76.8,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 480,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.96,
|
| 13 |
+
"eval_accuracy": 0.22988505747126436,
|
| 14 |
+
"eval_loss": 1.7957335710525513,
|
| 15 |
+
"eval_runtime": 2.3499,
|
| 16 |
+
"eval_samples_per_second": 37.022,
|
| 17 |
+
"eval_steps_per_second": 1.277,
|
| 18 |
+
"step": 6
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 1.6,
|
| 22 |
+
"grad_norm": 5.402665615081787,
|
| 23 |
+
"learning_rate": 1.0416666666666668e-05,
|
| 24 |
+
"loss": 1.8656,
|
| 25 |
+
"step": 10
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 1.92,
|
| 29 |
+
"eval_accuracy": 0.27586206896551724,
|
| 30 |
+
"eval_loss": 1.7703895568847656,
|
| 31 |
+
"eval_runtime": 2.2621,
|
| 32 |
+
"eval_samples_per_second": 38.46,
|
| 33 |
+
"eval_steps_per_second": 1.326,
|
| 34 |
+
"step": 12
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 2.88,
|
| 38 |
+
"eval_accuracy": 0.3218390804597701,
|
| 39 |
+
"eval_loss": 1.738166332244873,
|
| 40 |
+
"eval_runtime": 2.047,
|
| 41 |
+
"eval_samples_per_second": 42.502,
|
| 42 |
+
"eval_steps_per_second": 1.466,
|
| 43 |
+
"step": 18
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 3.2,
|
| 47 |
+
"grad_norm": 9.074636459350586,
|
| 48 |
+
"learning_rate": 2.0833333333333336e-05,
|
| 49 |
+
"loss": 1.7835,
|
| 50 |
+
"step": 20
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 4.0,
|
| 54 |
+
"eval_accuracy": 0.3793103448275862,
|
| 55 |
+
"eval_loss": 1.6673917770385742,
|
| 56 |
+
"eval_runtime": 2.0721,
|
| 57 |
+
"eval_samples_per_second": 41.987,
|
| 58 |
+
"eval_steps_per_second": 1.448,
|
| 59 |
+
"step": 25
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 4.8,
|
| 63 |
+
"grad_norm": 23.296255111694336,
|
| 64 |
+
"learning_rate": 3.125e-05,
|
| 65 |
+
"loss": 1.664,
|
| 66 |
+
"step": 30
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 4.96,
|
| 70 |
+
"eval_accuracy": 0.42528735632183906,
|
| 71 |
+
"eval_loss": 1.5981522798538208,
|
| 72 |
+
"eval_runtime": 2.0749,
|
| 73 |
+
"eval_samples_per_second": 41.93,
|
| 74 |
+
"eval_steps_per_second": 1.446,
|
| 75 |
+
"step": 31
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 5.92,
|
| 79 |
+
"eval_accuracy": 0.4367816091954023,
|
| 80 |
+
"eval_loss": 1.4861106872558594,
|
| 81 |
+
"eval_runtime": 2.0842,
|
| 82 |
+
"eval_samples_per_second": 41.743,
|
| 83 |
+
"eval_steps_per_second": 1.439,
|
| 84 |
+
"step": 37
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 6.4,
|
| 88 |
+
"grad_norm": 36.56019592285156,
|
| 89 |
+
"learning_rate": 4.166666666666667e-05,
|
| 90 |
+
"loss": 1.5072,
|
| 91 |
+
"step": 40
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 6.88,
|
| 95 |
+
"eval_accuracy": 0.47126436781609193,
|
| 96 |
+
"eval_loss": 1.3644713163375854,
|
| 97 |
+
"eval_runtime": 2.0344,
|
| 98 |
+
"eval_samples_per_second": 42.765,
|
| 99 |
+
"eval_steps_per_second": 1.475,
|
| 100 |
+
"step": 43
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 8.0,
|
| 104 |
+
"grad_norm": 70.99100494384766,
|
| 105 |
+
"learning_rate": 4.976851851851852e-05,
|
| 106 |
+
"loss": 1.3304,
|
| 107 |
+
"step": 50
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 8.0,
|
| 111 |
+
"eval_accuracy": 0.45977011494252873,
|
| 112 |
+
"eval_loss": 1.285918116569519,
|
| 113 |
+
"eval_runtime": 2.0671,
|
| 114 |
+
"eval_samples_per_second": 42.088,
|
| 115 |
+
"eval_steps_per_second": 1.451,
|
| 116 |
+
"step": 50
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 8.96,
|
| 120 |
+
"eval_accuracy": 0.47126436781609193,
|
| 121 |
+
"eval_loss": 1.2795610427856445,
|
| 122 |
+
"eval_runtime": 2.0462,
|
| 123 |
+
"eval_samples_per_second": 42.519,
|
| 124 |
+
"eval_steps_per_second": 1.466,
|
| 125 |
+
"step": 56
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 9.6,
|
| 129 |
+
"grad_norm": 60.155181884765625,
|
| 130 |
+
"learning_rate": 4.8611111111111115e-05,
|
| 131 |
+
"loss": 1.1651,
|
| 132 |
+
"step": 60
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 9.92,
|
| 136 |
+
"eval_accuracy": 0.5172413793103449,
|
| 137 |
+
"eval_loss": 1.2455964088439941,
|
| 138 |
+
"eval_runtime": 2.0479,
|
| 139 |
+
"eval_samples_per_second": 42.483,
|
| 140 |
+
"eval_steps_per_second": 1.465,
|
| 141 |
+
"step": 62
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 10.88,
|
| 145 |
+
"eval_accuracy": 0.5402298850574713,
|
| 146 |
+
"eval_loss": 1.1666686534881592,
|
| 147 |
+
"eval_runtime": 2.0486,
|
| 148 |
+
"eval_samples_per_second": 42.468,
|
| 149 |
+
"eval_steps_per_second": 1.464,
|
| 150 |
+
"step": 68
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 11.2,
|
| 154 |
+
"grad_norm": 17.20172119140625,
|
| 155 |
+
"learning_rate": 4.745370370370371e-05,
|
| 156 |
+
"loss": 1.0876,
|
| 157 |
+
"step": 70
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 12.0,
|
| 161 |
+
"eval_accuracy": 0.5632183908045977,
|
| 162 |
+
"eval_loss": 1.1510032415390015,
|
| 163 |
+
"eval_runtime": 2.0486,
|
| 164 |
+
"eval_samples_per_second": 42.468,
|
| 165 |
+
"eval_steps_per_second": 1.464,
|
| 166 |
+
"step": 75
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 12.8,
|
| 170 |
+
"grad_norm": 98.55331420898438,
|
| 171 |
+
"learning_rate": 4.62962962962963e-05,
|
| 172 |
+
"loss": 1.0046,
|
| 173 |
+
"step": 80
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 12.96,
|
| 177 |
+
"eval_accuracy": 0.6091954022988506,
|
| 178 |
+
"eval_loss": 1.0509852170944214,
|
| 179 |
+
"eval_runtime": 2.2,
|
| 180 |
+
"eval_samples_per_second": 39.546,
|
| 181 |
+
"eval_steps_per_second": 1.364,
|
| 182 |
+
"step": 81
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 13.92,
|
| 186 |
+
"eval_accuracy": 0.5862068965517241,
|
| 187 |
+
"eval_loss": 1.033838152885437,
|
| 188 |
+
"eval_runtime": 2.0133,
|
| 189 |
+
"eval_samples_per_second": 43.212,
|
| 190 |
+
"eval_steps_per_second": 1.49,
|
| 191 |
+
"step": 87
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 14.4,
|
| 195 |
+
"grad_norm": 53.443302154541016,
|
| 196 |
+
"learning_rate": 4.5138888888888894e-05,
|
| 197 |
+
"loss": 0.9465,
|
| 198 |
+
"step": 90
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 14.88,
|
| 202 |
+
"eval_accuracy": 0.5862068965517241,
|
| 203 |
+
"eval_loss": 0.9883113503456116,
|
| 204 |
+
"eval_runtime": 2.05,
|
| 205 |
+
"eval_samples_per_second": 42.439,
|
| 206 |
+
"eval_steps_per_second": 1.463,
|
| 207 |
+
"step": 93
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 16.0,
|
| 211 |
+
"grad_norm": 30.475088119506836,
|
| 212 |
+
"learning_rate": 4.3981481481481486e-05,
|
| 213 |
+
"loss": 0.8699,
|
| 214 |
+
"step": 100
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 16.0,
|
| 218 |
+
"eval_accuracy": 0.5632183908045977,
|
| 219 |
+
"eval_loss": 0.9881502389907837,
|
| 220 |
+
"eval_runtime": 2.0881,
|
| 221 |
+
"eval_samples_per_second": 41.664,
|
| 222 |
+
"eval_steps_per_second": 1.437,
|
| 223 |
+
"step": 100
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 16.96,
|
| 227 |
+
"eval_accuracy": 0.5747126436781609,
|
| 228 |
+
"eval_loss": 0.9276102781295776,
|
| 229 |
+
"eval_runtime": 2.0889,
|
| 230 |
+
"eval_samples_per_second": 41.648,
|
| 231 |
+
"eval_steps_per_second": 1.436,
|
| 232 |
+
"step": 106
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 17.6,
|
| 236 |
+
"grad_norm": 21.802074432373047,
|
| 237 |
+
"learning_rate": 4.282407407407408e-05,
|
| 238 |
+
"loss": 0.7969,
|
| 239 |
+
"step": 110
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 17.92,
|
| 243 |
+
"eval_accuracy": 0.5862068965517241,
|
| 244 |
+
"eval_loss": 0.9144545197486877,
|
| 245 |
+
"eval_runtime": 2.0314,
|
| 246 |
+
"eval_samples_per_second": 42.828,
|
| 247 |
+
"eval_steps_per_second": 1.477,
|
| 248 |
+
"step": 112
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 18.88,
|
| 252 |
+
"eval_accuracy": 0.6666666666666666,
|
| 253 |
+
"eval_loss": 0.8143898844718933,
|
| 254 |
+
"eval_runtime": 2.0134,
|
| 255 |
+
"eval_samples_per_second": 43.21,
|
| 256 |
+
"eval_steps_per_second": 1.49,
|
| 257 |
+
"step": 118
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 19.2,
|
| 261 |
+
"grad_norm": 58.785552978515625,
|
| 262 |
+
"learning_rate": 4.166666666666667e-05,
|
| 263 |
+
"loss": 0.7254,
|
| 264 |
+
"step": 120
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 20.0,
|
| 268 |
+
"eval_accuracy": 0.6666666666666666,
|
| 269 |
+
"eval_loss": 0.7586901187896729,
|
| 270 |
+
"eval_runtime": 2.0526,
|
| 271 |
+
"eval_samples_per_second": 42.386,
|
| 272 |
+
"eval_steps_per_second": 1.462,
|
| 273 |
+
"step": 125
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 20.8,
|
| 277 |
+
"grad_norm": 24.079566955566406,
|
| 278 |
+
"learning_rate": 4.0509259259259265e-05,
|
| 279 |
+
"loss": 0.6447,
|
| 280 |
+
"step": 130
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 20.96,
|
| 284 |
+
"eval_accuracy": 0.7471264367816092,
|
| 285 |
+
"eval_loss": 0.6990374326705933,
|
| 286 |
+
"eval_runtime": 2.0625,
|
| 287 |
+
"eval_samples_per_second": 42.182,
|
| 288 |
+
"eval_steps_per_second": 1.455,
|
| 289 |
+
"step": 131
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 21.92,
|
| 293 |
+
"eval_accuracy": 0.7241379310344828,
|
| 294 |
+
"eval_loss": 0.7041503190994263,
|
| 295 |
+
"eval_runtime": 2.0267,
|
| 296 |
+
"eval_samples_per_second": 42.926,
|
| 297 |
+
"eval_steps_per_second": 1.48,
|
| 298 |
+
"step": 137
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 22.4,
|
| 302 |
+
"grad_norm": 24.671459197998047,
|
| 303 |
+
"learning_rate": 3.935185185185186e-05,
|
| 304 |
+
"loss": 0.6021,
|
| 305 |
+
"step": 140
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 22.88,
|
| 309 |
+
"eval_accuracy": 0.7701149425287356,
|
| 310 |
+
"eval_loss": 0.6526182293891907,
|
| 311 |
+
"eval_runtime": 2.1122,
|
| 312 |
+
"eval_samples_per_second": 41.189,
|
| 313 |
+
"eval_steps_per_second": 1.42,
|
| 314 |
+
"step": 143
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 24.0,
|
| 318 |
+
"grad_norm": 55.466121673583984,
|
| 319 |
+
"learning_rate": 3.8194444444444444e-05,
|
| 320 |
+
"loss": 0.516,
|
| 321 |
+
"step": 150
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 24.0,
|
| 325 |
+
"eval_accuracy": 0.8045977011494253,
|
| 326 |
+
"eval_loss": 0.6485260128974915,
|
| 327 |
+
"eval_runtime": 2.0692,
|
| 328 |
+
"eval_samples_per_second": 42.046,
|
| 329 |
+
"eval_steps_per_second": 1.45,
|
| 330 |
+
"step": 150
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 24.96,
|
| 334 |
+
"eval_accuracy": 0.8160919540229885,
|
| 335 |
+
"eval_loss": 0.5802629590034485,
|
| 336 |
+
"eval_runtime": 2.0421,
|
| 337 |
+
"eval_samples_per_second": 42.603,
|
| 338 |
+
"eval_steps_per_second": 1.469,
|
| 339 |
+
"step": 156
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 25.6,
|
| 343 |
+
"grad_norm": 17.66895294189453,
|
| 344 |
+
"learning_rate": 3.7037037037037037e-05,
|
| 345 |
+
"loss": 0.4497,
|
| 346 |
+
"step": 160
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 25.92,
|
| 350 |
+
"eval_accuracy": 0.8045977011494253,
|
| 351 |
+
"eval_loss": 0.6084781289100647,
|
| 352 |
+
"eval_runtime": 2.0191,
|
| 353 |
+
"eval_samples_per_second": 43.088,
|
| 354 |
+
"eval_steps_per_second": 1.486,
|
| 355 |
+
"step": 162
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 26.88,
|
| 359 |
+
"eval_accuracy": 0.8045977011494253,
|
| 360 |
+
"eval_loss": 0.6094852685928345,
|
| 361 |
+
"eval_runtime": 1.9897,
|
| 362 |
+
"eval_samples_per_second": 43.724,
|
| 363 |
+
"eval_steps_per_second": 1.508,
|
| 364 |
+
"step": 168
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 27.2,
|
| 368 |
+
"grad_norm": 31.39649200439453,
|
| 369 |
+
"learning_rate": 3.587962962962963e-05,
|
| 370 |
+
"loss": 0.3935,
|
| 371 |
+
"step": 170
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 28.0,
|
| 375 |
+
"eval_accuracy": 0.8275862068965517,
|
| 376 |
+
"eval_loss": 0.5372287034988403,
|
| 377 |
+
"eval_runtime": 2.0637,
|
| 378 |
+
"eval_samples_per_second": 42.158,
|
| 379 |
+
"eval_steps_per_second": 1.454,
|
| 380 |
+
"step": 175
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 28.8,
|
| 384 |
+
"grad_norm": 31.86827278137207,
|
| 385 |
+
"learning_rate": 3.472222222222222e-05,
|
| 386 |
+
"loss": 0.3321,
|
| 387 |
+
"step": 180
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 28.96,
|
| 391 |
+
"eval_accuracy": 0.8160919540229885,
|
| 392 |
+
"eval_loss": 0.5828755497932434,
|
| 393 |
+
"eval_runtime": 2.1428,
|
| 394 |
+
"eval_samples_per_second": 40.6,
|
| 395 |
+
"eval_steps_per_second": 1.4,
|
| 396 |
+
"step": 181
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 29.92,
|
| 400 |
+
"eval_accuracy": 0.8160919540229885,
|
| 401 |
+
"eval_loss": 0.6204901337623596,
|
| 402 |
+
"eval_runtime": 2.0154,
|
| 403 |
+
"eval_samples_per_second": 43.168,
|
| 404 |
+
"eval_steps_per_second": 1.489,
|
| 405 |
+
"step": 187
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 30.4,
|
| 409 |
+
"grad_norm": 42.88612747192383,
|
| 410 |
+
"learning_rate": 3.3564814814814815e-05,
|
| 411 |
+
"loss": 0.3007,
|
| 412 |
+
"step": 190
|
| 413 |
+
},
|
| 414 |
+
{
|
| 415 |
+
"epoch": 30.88,
|
| 416 |
+
"eval_accuracy": 0.8275862068965517,
|
| 417 |
+
"eval_loss": 0.5149825811386108,
|
| 418 |
+
"eval_runtime": 2.0492,
|
| 419 |
+
"eval_samples_per_second": 42.456,
|
| 420 |
+
"eval_steps_per_second": 1.464,
|
| 421 |
+
"step": 193
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 32.0,
|
| 425 |
+
"grad_norm": 30.13237190246582,
|
| 426 |
+
"learning_rate": 3.240740740740741e-05,
|
| 427 |
+
"loss": 0.2618,
|
| 428 |
+
"step": 200
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 32.0,
|
| 432 |
+
"eval_accuracy": 0.8390804597701149,
|
| 433 |
+
"eval_loss": 0.6068965196609497,
|
| 434 |
+
"eval_runtime": 2.0657,
|
| 435 |
+
"eval_samples_per_second": 42.117,
|
| 436 |
+
"eval_steps_per_second": 1.452,
|
| 437 |
+
"step": 200
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 32.96,
|
| 441 |
+
"eval_accuracy": 0.8390804597701149,
|
| 442 |
+
"eval_loss": 0.5272508859634399,
|
| 443 |
+
"eval_runtime": 2.0395,
|
| 444 |
+
"eval_samples_per_second": 42.657,
|
| 445 |
+
"eval_steps_per_second": 1.471,
|
| 446 |
+
"step": 206
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 33.6,
|
| 450 |
+
"grad_norm": 24.97075080871582,
|
| 451 |
+
"learning_rate": 3.125e-05,
|
| 452 |
+
"loss": 0.2411,
|
| 453 |
+
"step": 210
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 33.92,
|
| 457 |
+
"eval_accuracy": 0.8620689655172413,
|
| 458 |
+
"eval_loss": 0.4726714789867401,
|
| 459 |
+
"eval_runtime": 2.0758,
|
| 460 |
+
"eval_samples_per_second": 41.912,
|
| 461 |
+
"eval_steps_per_second": 1.445,
|
| 462 |
+
"step": 212
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 34.88,
|
| 466 |
+
"eval_accuracy": 0.8735632183908046,
|
| 467 |
+
"eval_loss": 0.4611084461212158,
|
| 468 |
+
"eval_runtime": 2.0264,
|
| 469 |
+
"eval_samples_per_second": 42.934,
|
| 470 |
+
"eval_steps_per_second": 1.48,
|
| 471 |
+
"step": 218
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 35.2,
|
| 475 |
+
"grad_norm": 60.3193359375,
|
| 476 |
+
"learning_rate": 3.0092592592592593e-05,
|
| 477 |
+
"loss": 0.2108,
|
| 478 |
+
"step": 220
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 36.0,
|
| 482 |
+
"eval_accuracy": 0.8505747126436781,
|
| 483 |
+
"eval_loss": 0.5696073770523071,
|
| 484 |
+
"eval_runtime": 2.0919,
|
| 485 |
+
"eval_samples_per_second": 41.589,
|
| 486 |
+
"eval_steps_per_second": 1.434,
|
| 487 |
+
"step": 225
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 36.8,
|
| 491 |
+
"grad_norm": 16.915546417236328,
|
| 492 |
+
"learning_rate": 2.8935185185185186e-05,
|
| 493 |
+
"loss": 0.2143,
|
| 494 |
+
"step": 230
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 36.96,
|
| 498 |
+
"eval_accuracy": 0.8620689655172413,
|
| 499 |
+
"eval_loss": 0.49439194798469543,
|
| 500 |
+
"eval_runtime": 2.0923,
|
| 501 |
+
"eval_samples_per_second": 41.58,
|
| 502 |
+
"eval_steps_per_second": 1.434,
|
| 503 |
+
"step": 231
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 37.92,
|
| 507 |
+
"eval_accuracy": 0.8160919540229885,
|
| 508 |
+
"eval_loss": 0.5627816915512085,
|
| 509 |
+
"eval_runtime": 2.0503,
|
| 510 |
+
"eval_samples_per_second": 42.432,
|
| 511 |
+
"eval_steps_per_second": 1.463,
|
| 512 |
+
"step": 237
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 38.4,
|
| 516 |
+
"grad_norm": 14.699493408203125,
|
| 517 |
+
"learning_rate": 2.777777777777778e-05,
|
| 518 |
+
"loss": 0.1663,
|
| 519 |
+
"step": 240
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 38.88,
|
| 523 |
+
"eval_accuracy": 0.8045977011494253,
|
| 524 |
+
"eval_loss": 0.6131365895271301,
|
| 525 |
+
"eval_runtime": 2.0693,
|
| 526 |
+
"eval_samples_per_second": 42.044,
|
| 527 |
+
"eval_steps_per_second": 1.45,
|
| 528 |
+
"step": 243
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 40.0,
|
| 532 |
+
"grad_norm": 25.7874755859375,
|
| 533 |
+
"learning_rate": 2.6620370370370372e-05,
|
| 534 |
+
"loss": 0.1714,
|
| 535 |
+
"step": 250
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 40.0,
|
| 539 |
+
"eval_accuracy": 0.8505747126436781,
|
| 540 |
+
"eval_loss": 0.4961901605129242,
|
| 541 |
+
"eval_runtime": 2.0252,
|
| 542 |
+
"eval_samples_per_second": 42.959,
|
| 543 |
+
"eval_steps_per_second": 1.481,
|
| 544 |
+
"step": 250
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 40.96,
|
| 548 |
+
"eval_accuracy": 0.8390804597701149,
|
| 549 |
+
"eval_loss": 0.5022612810134888,
|
| 550 |
+
"eval_runtime": 2.127,
|
| 551 |
+
"eval_samples_per_second": 40.904,
|
| 552 |
+
"eval_steps_per_second": 1.41,
|
| 553 |
+
"step": 256
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 41.6,
|
| 557 |
+
"grad_norm": 24.087005615234375,
|
| 558 |
+
"learning_rate": 2.5462962962962965e-05,
|
| 559 |
+
"loss": 0.174,
|
| 560 |
+
"step": 260
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 41.92,
|
| 564 |
+
"eval_accuracy": 0.8275862068965517,
|
| 565 |
+
"eval_loss": 0.48418501019477844,
|
| 566 |
+
"eval_runtime": 2.0168,
|
| 567 |
+
"eval_samples_per_second": 43.137,
|
| 568 |
+
"eval_steps_per_second": 1.487,
|
| 569 |
+
"step": 262
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 42.88,
|
| 573 |
+
"eval_accuracy": 0.8275862068965517,
|
| 574 |
+
"eval_loss": 0.46790340542793274,
|
| 575 |
+
"eval_runtime": 2.0909,
|
| 576 |
+
"eval_samples_per_second": 41.609,
|
| 577 |
+
"eval_steps_per_second": 1.435,
|
| 578 |
+
"step": 268
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 43.2,
|
| 582 |
+
"grad_norm": 13.284588813781738,
|
| 583 |
+
"learning_rate": 2.4305555555555558e-05,
|
| 584 |
+
"loss": 0.138,
|
| 585 |
+
"step": 270
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 44.0,
|
| 589 |
+
"eval_accuracy": 0.8160919540229885,
|
| 590 |
+
"eval_loss": 0.6270841956138611,
|
| 591 |
+
"eval_runtime": 2.1069,
|
| 592 |
+
"eval_samples_per_second": 41.294,
|
| 593 |
+
"eval_steps_per_second": 1.424,
|
| 594 |
+
"step": 275
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 44.8,
|
| 598 |
+
"grad_norm": 14.41830062866211,
|
| 599 |
+
"learning_rate": 2.314814814814815e-05,
|
| 600 |
+
"loss": 0.1437,
|
| 601 |
+
"step": 280
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 44.96,
|
| 605 |
+
"eval_accuracy": 0.8505747126436781,
|
| 606 |
+
"eval_loss": 0.5325595736503601,
|
| 607 |
+
"eval_runtime": 2.1982,
|
| 608 |
+
"eval_samples_per_second": 39.578,
|
| 609 |
+
"eval_steps_per_second": 1.365,
|
| 610 |
+
"step": 281
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 45.92,
|
| 614 |
+
"eval_accuracy": 0.8160919540229885,
|
| 615 |
+
"eval_loss": 0.5655315518379211,
|
| 616 |
+
"eval_runtime": 2.0683,
|
| 617 |
+
"eval_samples_per_second": 42.063,
|
| 618 |
+
"eval_steps_per_second": 1.45,
|
| 619 |
+
"step": 287
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 46.4,
|
| 623 |
+
"grad_norm": 17.588279724121094,
|
| 624 |
+
"learning_rate": 2.1990740740740743e-05,
|
| 625 |
+
"loss": 0.136,
|
| 626 |
+
"step": 290
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 46.88,
|
| 630 |
+
"eval_accuracy": 0.8390804597701149,
|
| 631 |
+
"eval_loss": 0.46718767285346985,
|
| 632 |
+
"eval_runtime": 2.0892,
|
| 633 |
+
"eval_samples_per_second": 41.643,
|
| 634 |
+
"eval_steps_per_second": 1.436,
|
| 635 |
+
"step": 293
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 48.0,
|
| 639 |
+
"grad_norm": 22.864524841308594,
|
| 640 |
+
"learning_rate": 2.0833333333333336e-05,
|
| 641 |
+
"loss": 0.1401,
|
| 642 |
+
"step": 300
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 48.0,
|
| 646 |
+
"eval_accuracy": 0.8620689655172413,
|
| 647 |
+
"eval_loss": 0.498960942029953,
|
| 648 |
+
"eval_runtime": 2.0484,
|
| 649 |
+
"eval_samples_per_second": 42.471,
|
| 650 |
+
"eval_steps_per_second": 1.465,
|
| 651 |
+
"step": 300
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 48.96,
|
| 655 |
+
"eval_accuracy": 0.8275862068965517,
|
| 656 |
+
"eval_loss": 0.5445386171340942,
|
| 657 |
+
"eval_runtime": 2.0365,
|
| 658 |
+
"eval_samples_per_second": 42.721,
|
| 659 |
+
"eval_steps_per_second": 1.473,
|
| 660 |
+
"step": 306
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 49.6,
|
| 664 |
+
"grad_norm": 22.651620864868164,
|
| 665 |
+
"learning_rate": 1.967592592592593e-05,
|
| 666 |
+
"loss": 0.1281,
|
| 667 |
+
"step": 310
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 49.92,
|
| 671 |
+
"eval_accuracy": 0.8735632183908046,
|
| 672 |
+
"eval_loss": 0.47610902786254883,
|
| 673 |
+
"eval_runtime": 2.1166,
|
| 674 |
+
"eval_samples_per_second": 41.104,
|
| 675 |
+
"eval_steps_per_second": 1.417,
|
| 676 |
+
"step": 312
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 50.88,
|
| 680 |
+
"eval_accuracy": 0.8505747126436781,
|
| 681 |
+
"eval_loss": 0.5665103793144226,
|
| 682 |
+
"eval_runtime": 2.1168,
|
| 683 |
+
"eval_samples_per_second": 41.1,
|
| 684 |
+
"eval_steps_per_second": 1.417,
|
| 685 |
+
"step": 318
|
| 686 |
+
},
|
| 687 |
+
{
|
| 688 |
+
"epoch": 51.2,
|
| 689 |
+
"grad_norm": 26.539594650268555,
|
| 690 |
+
"learning_rate": 1.8518518518518518e-05,
|
| 691 |
+
"loss": 0.1156,
|
| 692 |
+
"step": 320
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 52.0,
|
| 696 |
+
"eval_accuracy": 0.8505747126436781,
|
| 697 |
+
"eval_loss": 0.5089926719665527,
|
| 698 |
+
"eval_runtime": 2.0775,
|
| 699 |
+
"eval_samples_per_second": 41.877,
|
| 700 |
+
"eval_steps_per_second": 1.444,
|
| 701 |
+
"step": 325
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 52.8,
|
| 705 |
+
"grad_norm": 23.464221954345703,
|
| 706 |
+
"learning_rate": 1.736111111111111e-05,
|
| 707 |
+
"loss": 0.0981,
|
| 708 |
+
"step": 330
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 52.96,
|
| 712 |
+
"eval_accuracy": 0.8505747126436781,
|
| 713 |
+
"eval_loss": 0.5152259469032288,
|
| 714 |
+
"eval_runtime": 2.0607,
|
| 715 |
+
"eval_samples_per_second": 42.219,
|
| 716 |
+
"eval_steps_per_second": 1.456,
|
| 717 |
+
"step": 331
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 53.92,
|
| 721 |
+
"eval_accuracy": 0.8160919540229885,
|
| 722 |
+
"eval_loss": 0.5466004610061646,
|
| 723 |
+
"eval_runtime": 2.0591,
|
| 724 |
+
"eval_samples_per_second": 42.251,
|
| 725 |
+
"eval_steps_per_second": 1.457,
|
| 726 |
+
"step": 337
|
| 727 |
+
},
|
| 728 |
+
{
|
| 729 |
+
"epoch": 54.4,
|
| 730 |
+
"grad_norm": 14.581974983215332,
|
| 731 |
+
"learning_rate": 1.6203703703703704e-05,
|
| 732 |
+
"loss": 0.1055,
|
| 733 |
+
"step": 340
|
| 734 |
+
},
|
| 735 |
+
{
|
| 736 |
+
"epoch": 54.88,
|
| 737 |
+
"eval_accuracy": 0.8275862068965517,
|
| 738 |
+
"eval_loss": 0.5390048623085022,
|
| 739 |
+
"eval_runtime": 2.0443,
|
| 740 |
+
"eval_samples_per_second": 42.558,
|
| 741 |
+
"eval_steps_per_second": 1.468,
|
| 742 |
+
"step": 343
|
| 743 |
+
},
|
| 744 |
+
{
|
| 745 |
+
"epoch": 56.0,
|
| 746 |
+
"grad_norm": 14.774139404296875,
|
| 747 |
+
"learning_rate": 1.5046296296296297e-05,
|
| 748 |
+
"loss": 0.112,
|
| 749 |
+
"step": 350
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 56.0,
|
| 753 |
+
"eval_accuracy": 0.8505747126436781,
|
| 754 |
+
"eval_loss": 0.5574498176574707,
|
| 755 |
+
"eval_runtime": 2.0874,
|
| 756 |
+
"eval_samples_per_second": 41.679,
|
| 757 |
+
"eval_steps_per_second": 1.437,
|
| 758 |
+
"step": 350
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 56.96,
|
| 762 |
+
"eval_accuracy": 0.8505747126436781,
|
| 763 |
+
"eval_loss": 0.5448784828186035,
|
| 764 |
+
"eval_runtime": 2.0514,
|
| 765 |
+
"eval_samples_per_second": 42.41,
|
| 766 |
+
"eval_steps_per_second": 1.462,
|
| 767 |
+
"step": 356
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 57.6,
|
| 771 |
+
"grad_norm": 18.17756462097168,
|
| 772 |
+
"learning_rate": 1.388888888888889e-05,
|
| 773 |
+
"loss": 0.0855,
|
| 774 |
+
"step": 360
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 57.92,
|
| 778 |
+
"eval_accuracy": 0.8505747126436781,
|
| 779 |
+
"eval_loss": 0.5390240550041199,
|
| 780 |
+
"eval_runtime": 2.077,
|
| 781 |
+
"eval_samples_per_second": 41.888,
|
| 782 |
+
"eval_steps_per_second": 1.444,
|
| 783 |
+
"step": 362
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 58.88,
|
| 787 |
+
"eval_accuracy": 0.8505747126436781,
|
| 788 |
+
"eval_loss": 0.5206344723701477,
|
| 789 |
+
"eval_runtime": 2.0568,
|
| 790 |
+
"eval_samples_per_second": 42.299,
|
| 791 |
+
"eval_steps_per_second": 1.459,
|
| 792 |
+
"step": 368
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 59.2,
|
| 796 |
+
"grad_norm": 40.29678726196289,
|
| 797 |
+
"learning_rate": 1.2731481481481482e-05,
|
| 798 |
+
"loss": 0.0899,
|
| 799 |
+
"step": 370
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 60.0,
|
| 803 |
+
"eval_accuracy": 0.8620689655172413,
|
| 804 |
+
"eval_loss": 0.5475941300392151,
|
| 805 |
+
"eval_runtime": 2.063,
|
| 806 |
+
"eval_samples_per_second": 42.172,
|
| 807 |
+
"eval_steps_per_second": 1.454,
|
| 808 |
+
"step": 375
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 60.8,
|
| 812 |
+
"grad_norm": 33.910377502441406,
|
| 813 |
+
"learning_rate": 1.1574074074074075e-05,
|
| 814 |
+
"loss": 0.1026,
|
| 815 |
+
"step": 380
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 60.96,
|
| 819 |
+
"eval_accuracy": 0.8505747126436781,
|
| 820 |
+
"eval_loss": 0.5344437956809998,
|
| 821 |
+
"eval_runtime": 2.298,
|
| 822 |
+
"eval_samples_per_second": 37.858,
|
| 823 |
+
"eval_steps_per_second": 1.305,
|
| 824 |
+
"step": 381
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 61.92,
|
| 828 |
+
"eval_accuracy": 0.8390804597701149,
|
| 829 |
+
"eval_loss": 0.553070068359375,
|
| 830 |
+
"eval_runtime": 2.1032,
|
| 831 |
+
"eval_samples_per_second": 41.366,
|
| 832 |
+
"eval_steps_per_second": 1.426,
|
| 833 |
+
"step": 387
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 62.4,
|
| 837 |
+
"grad_norm": 13.71580982208252,
|
| 838 |
+
"learning_rate": 1.0416666666666668e-05,
|
| 839 |
+
"loss": 0.0799,
|
| 840 |
+
"step": 390
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 62.88,
|
| 844 |
+
"eval_accuracy": 0.8275862068965517,
|
| 845 |
+
"eval_loss": 0.57228684425354,
|
| 846 |
+
"eval_runtime": 2.0779,
|
| 847 |
+
"eval_samples_per_second": 41.868,
|
| 848 |
+
"eval_steps_per_second": 1.444,
|
| 849 |
+
"step": 393
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 64.0,
|
| 853 |
+
"grad_norm": 28.238468170166016,
|
| 854 |
+
"learning_rate": 9.259259259259259e-06,
|
| 855 |
+
"loss": 0.0844,
|
| 856 |
+
"step": 400
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 64.0,
|
| 860 |
+
"eval_accuracy": 0.8160919540229885,
|
| 861 |
+
"eval_loss": 0.5339850783348083,
|
| 862 |
+
"eval_runtime": 2.0258,
|
| 863 |
+
"eval_samples_per_second": 42.946,
|
| 864 |
+
"eval_steps_per_second": 1.481,
|
| 865 |
+
"step": 400
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 64.96,
|
| 869 |
+
"eval_accuracy": 0.8735632183908046,
|
| 870 |
+
"eval_loss": 0.52364581823349,
|
| 871 |
+
"eval_runtime": 2.0251,
|
| 872 |
+
"eval_samples_per_second": 42.961,
|
| 873 |
+
"eval_steps_per_second": 1.481,
|
| 874 |
+
"step": 406
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 65.6,
|
| 878 |
+
"grad_norm": 10.24560832977295,
|
| 879 |
+
"learning_rate": 8.101851851851852e-06,
|
| 880 |
+
"loss": 0.0724,
|
| 881 |
+
"step": 410
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 65.92,
|
| 885 |
+
"eval_accuracy": 0.8390804597701149,
|
| 886 |
+
"eval_loss": 0.6136645674705505,
|
| 887 |
+
"eval_runtime": 2.03,
|
| 888 |
+
"eval_samples_per_second": 42.858,
|
| 889 |
+
"eval_steps_per_second": 1.478,
|
| 890 |
+
"step": 412
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 66.88,
|
| 894 |
+
"eval_accuracy": 0.8275862068965517,
|
| 895 |
+
"eval_loss": 0.5824962854385376,
|
| 896 |
+
"eval_runtime": 2.0787,
|
| 897 |
+
"eval_samples_per_second": 41.854,
|
| 898 |
+
"eval_steps_per_second": 1.443,
|
| 899 |
+
"step": 418
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 67.2,
|
| 903 |
+
"grad_norm": 20.803382873535156,
|
| 904 |
+
"learning_rate": 6.944444444444445e-06,
|
| 905 |
+
"loss": 0.0867,
|
| 906 |
+
"step": 420
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 68.0,
|
| 910 |
+
"eval_accuracy": 0.8620689655172413,
|
| 911 |
+
"eval_loss": 0.510515034198761,
|
| 912 |
+
"eval_runtime": 2.0565,
|
| 913 |
+
"eval_samples_per_second": 42.305,
|
| 914 |
+
"eval_steps_per_second": 1.459,
|
| 915 |
+
"step": 425
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 68.8,
|
| 919 |
+
"grad_norm": 15.880162239074707,
|
| 920 |
+
"learning_rate": 5.787037037037038e-06,
|
| 921 |
+
"loss": 0.071,
|
| 922 |
+
"step": 430
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 68.96,
|
| 926 |
+
"eval_accuracy": 0.8505747126436781,
|
| 927 |
+
"eval_loss": 0.5272470116615295,
|
| 928 |
+
"eval_runtime": 2.0378,
|
| 929 |
+
"eval_samples_per_second": 42.693,
|
| 930 |
+
"eval_steps_per_second": 1.472,
|
| 931 |
+
"step": 431
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 69.92,
|
| 935 |
+
"eval_accuracy": 0.8505747126436781,
|
| 936 |
+
"eval_loss": 0.5523571372032166,
|
| 937 |
+
"eval_runtime": 2.0569,
|
| 938 |
+
"eval_samples_per_second": 42.297,
|
| 939 |
+
"eval_steps_per_second": 1.459,
|
| 940 |
+
"step": 437
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 70.4,
|
| 944 |
+
"grad_norm": 14.639904975891113,
|
| 945 |
+
"learning_rate": 4.6296296296296296e-06,
|
| 946 |
+
"loss": 0.0723,
|
| 947 |
+
"step": 440
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 70.88,
|
| 951 |
+
"eval_accuracy": 0.8390804597701149,
|
| 952 |
+
"eval_loss": 0.5507646799087524,
|
| 953 |
+
"eval_runtime": 2.1114,
|
| 954 |
+
"eval_samples_per_second": 41.205,
|
| 955 |
+
"eval_steps_per_second": 1.421,
|
| 956 |
+
"step": 443
|
| 957 |
+
},
|
| 958 |
+
{
|
| 959 |
+
"epoch": 72.0,
|
| 960 |
+
"grad_norm": 6.164122104644775,
|
| 961 |
+
"learning_rate": 3.4722222222222224e-06,
|
| 962 |
+
"loss": 0.0748,
|
| 963 |
+
"step": 450
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 72.0,
|
| 967 |
+
"eval_accuracy": 0.8160919540229885,
|
| 968 |
+
"eval_loss": 0.568942129611969,
|
| 969 |
+
"eval_runtime": 2.0852,
|
| 970 |
+
"eval_samples_per_second": 41.723,
|
| 971 |
+
"eval_steps_per_second": 1.439,
|
| 972 |
+
"step": 450
|
| 973 |
+
},
|
| 974 |
+
{
|
| 975 |
+
"epoch": 72.96,
|
| 976 |
+
"eval_accuracy": 0.8505747126436781,
|
| 977 |
+
"eval_loss": 0.555583119392395,
|
| 978 |
+
"eval_runtime": 2.0316,
|
| 979 |
+
"eval_samples_per_second": 42.824,
|
| 980 |
+
"eval_steps_per_second": 1.477,
|
| 981 |
+
"step": 456
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 73.6,
|
| 985 |
+
"grad_norm": 11.653559684753418,
|
| 986 |
+
"learning_rate": 2.3148148148148148e-06,
|
| 987 |
+
"loss": 0.0589,
|
| 988 |
+
"step": 460
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 73.92,
|
| 992 |
+
"eval_accuracy": 0.8505747126436781,
|
| 993 |
+
"eval_loss": 0.5452274084091187,
|
| 994 |
+
"eval_runtime": 2.0938,
|
| 995 |
+
"eval_samples_per_second": 41.551,
|
| 996 |
+
"eval_steps_per_second": 1.433,
|
| 997 |
+
"step": 462
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 74.88,
|
| 1001 |
+
"eval_accuracy": 0.8620689655172413,
|
| 1002 |
+
"eval_loss": 0.5475078225135803,
|
| 1003 |
+
"eval_runtime": 2.0547,
|
| 1004 |
+
"eval_samples_per_second": 42.342,
|
| 1005 |
+
"eval_steps_per_second": 1.46,
|
| 1006 |
+
"step": 468
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 75.2,
|
| 1010 |
+
"grad_norm": 21.146989822387695,
|
| 1011 |
+
"learning_rate": 1.1574074074074074e-06,
|
| 1012 |
+
"loss": 0.0719,
|
| 1013 |
+
"step": 470
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 76.0,
|
| 1017 |
+
"eval_accuracy": 0.8620689655172413,
|
| 1018 |
+
"eval_loss": 0.5483731031417847,
|
| 1019 |
+
"eval_runtime": 2.1022,
|
| 1020 |
+
"eval_samples_per_second": 41.386,
|
| 1021 |
+
"eval_steps_per_second": 1.427,
|
| 1022 |
+
"step": 475
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 76.8,
|
| 1026 |
+
"grad_norm": 12.87066650390625,
|
| 1027 |
+
"learning_rate": 0.0,
|
| 1028 |
+
"loss": 0.0801,
|
| 1029 |
+
"step": 480
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 76.8,
|
| 1033 |
+
"eval_accuracy": 0.8620689655172413,
|
| 1034 |
+
"eval_loss": 0.5496163368225098,
|
| 1035 |
+
"eval_runtime": 2.0924,
|
| 1036 |
+
"eval_samples_per_second": 41.58,
|
| 1037 |
+
"eval_steps_per_second": 1.434,
|
| 1038 |
+
"step": 480
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 76.8,
|
| 1042 |
+
"step": 480,
|
| 1043 |
+
"total_flos": 1.514063180200919e+18,
|
| 1044 |
+
"train_loss": 0.449434948215882,
|
| 1045 |
+
"train_runtime": 1985.4684,
|
| 1046 |
+
"train_samples_per_second": 31.549,
|
| 1047 |
+
"train_steps_per_second": 0.242
|
| 1048 |
+
}
|
| 1049 |
+
],
|
| 1050 |
+
"logging_steps": 10,
|
| 1051 |
+
"max_steps": 480,
|
| 1052 |
+
"num_input_tokens_seen": 0,
|
| 1053 |
+
"num_train_epochs": 80,
|
| 1054 |
+
"save_steps": 500,
|
| 1055 |
+
"stateful_callbacks": {
|
| 1056 |
+
"TrainerControl": {
|
| 1057 |
+
"args": {
|
| 1058 |
+
"should_epoch_stop": false,
|
| 1059 |
+
"should_evaluate": false,
|
| 1060 |
+
"should_log": false,
|
| 1061 |
+
"should_save": false,
|
| 1062 |
+
"should_training_stop": false
|
| 1063 |
+
},
|
| 1064 |
+
"attributes": {}
|
| 1065 |
+
}
|
| 1066 |
+
},
|
| 1067 |
+
"total_flos": 1.514063180200919e+18,
|
| 1068 |
+
"train_batch_size": 32,
|
| 1069 |
+
"trial_name": null,
|
| 1070 |
+
"trial_params": null
|
| 1071 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3035e41ea7baa5c7336ad050a3b81dcc016e5c5dc831d5e355c15b8a865ce2b3
|
| 3 |
+
size 5240
|