End of training
Browse files- README.md +106 -0
- all_results.json +9 -0
- config.json +89 -0
- model.safetensors +3 -0
- preprocessor_config.json +9 -0
- runs/Aug05_18-01-01_dengzhipeng-pc/events.out.tfevents.1754388064.dengzhipeng-pc.22996.0 +3 -0
- runs/Aug05_18-01-01_dengzhipeng-pc/events.out.tfevents.1754388350.dengzhipeng-pc.22996.1 +3 -0
- runs/Aug05_18-01-01_dengzhipeng-pc/events.out.tfevents.1754388504.dengzhipeng-pc.22996.2 +3 -0
- runs/Aug05_18-01-01_dengzhipeng-pc/events.out.tfevents.1754390203.dengzhipeng-pc.22996.3 +3 -0
- runs/Aug05_18-01-01_dengzhipeng-pc/events.out.tfevents.1754390300.dengzhipeng-pc.22996.4 +3 -0
- runs/Aug05_18-44-57_dengzhipeng-pc/events.out.tfevents.1754390714.dengzhipeng-pc.22996.5 +3 -0
- runs/Aug05_18-46-12_dengzhipeng-pc/events.out.tfevents.1754390772.dengzhipeng-pc.22996.6 +3 -0
- runs/Aug05_19-09-27_dengzhipeng-pc/events.out.tfevents.1754392167.dengzhipeng-pc.22996.7 +3 -0
- runs/Aug05_19-09-27_dengzhipeng-pc/events.out.tfevents.1754392276.dengzhipeng-pc.22996.8 +3 -0
- runs/Aug05_19-09-27_dengzhipeng-pc/events.out.tfevents.1754393216.dengzhipeng-pc.22996.9 +3 -0
- runs/Aug05_20-02-33_dengzhipeng-pc/events.out.tfevents.1754395354.dengzhipeng-pc.22996.10 +3 -0
- train_results.json +9 -0
- trainer_state.json +730 -0
- training_args.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,106 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
license: mit
|
| 4 |
+
base_model: TencentGameMate/chinese-hubert-base
|
| 5 |
+
tags:
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
metrics:
|
| 8 |
+
- accuracy
|
| 9 |
+
model-index:
|
| 10 |
+
- name: hubert-base-ser
|
| 11 |
+
results: []
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 15 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 16 |
+
|
| 17 |
+
# hubert-base-ser
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [TencentGameMate/chinese-hubert-base](https://huggingface.co/TencentGameMate/chinese-hubert-base) on an unknown dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.1466
|
| 22 |
+
- Accuracy: 0.9526
|
| 23 |
+
|
| 24 |
+
## Model description
|
| 25 |
+
|
| 26 |
+
More information needed
|
| 27 |
+
|
| 28 |
+
## Intended uses & limitations
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Training and evaluation data
|
| 33 |
+
|
| 34 |
+
More information needed
|
| 35 |
+
|
| 36 |
+
## Training procedure
|
| 37 |
+
|
| 38 |
+
### Training hyperparameters
|
| 39 |
+
|
| 40 |
+
The following hyperparameters were used during training:
|
| 41 |
+
- learning_rate: 0.0001
|
| 42 |
+
- train_batch_size: 32
|
| 43 |
+
- eval_batch_size: 4
|
| 44 |
+
- seed: 42
|
| 45 |
+
- gradient_accumulation_steps: 2
|
| 46 |
+
- total_train_batch_size: 64
|
| 47 |
+
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
| 48 |
+
- lr_scheduler_type: linear
|
| 49 |
+
- num_epochs: 1.0
|
| 50 |
+
- mixed_precision_training: Native AMP
|
| 51 |
+
|
| 52 |
+
### Training results
|
| 53 |
+
|
| 54 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 55 |
+
|:-------------:|:------:|:----:|:---------------:|:--------:|
|
| 56 |
+
| 0.9709 | 0.0229 | 10 | 0.8923 | 0.6399 |
|
| 57 |
+
| 0.9219 | 0.0457 | 20 | 0.6903 | 0.7664 |
|
| 58 |
+
| 0.7112 | 0.0686 | 30 | 0.5838 | 0.7909 |
|
| 59 |
+
| 0.567 | 0.0914 | 40 | 0.5405 | 0.8159 |
|
| 60 |
+
| 0.6184 | 0.1143 | 50 | 0.4148 | 0.8581 |
|
| 61 |
+
| 0.5291 | 0.1371 | 60 | 0.4444 | 0.8511 |
|
| 62 |
+
| 0.533 | 0.16 | 70 | 0.4643 | 0.8271 |
|
| 63 |
+
| 0.4753 | 0.1829 | 80 | 0.3560 | 0.8767 |
|
| 64 |
+
| 0.4252 | 0.2057 | 90 | 0.5889 | 0.8103 |
|
| 65 |
+
| 0.5007 | 0.2286 | 100 | 0.3882 | 0.8663 |
|
| 66 |
+
| 0.5605 | 0.2514 | 110 | 0.3221 | 0.8921 |
|
| 67 |
+
| 0.4875 | 0.2743 | 120 | 0.3639 | 0.8559 |
|
| 68 |
+
| 0.4277 | 0.2971 | 130 | 0.3571 | 0.8746 |
|
| 69 |
+
| 0.3415 | 0.32 | 140 | 0.3382 | 0.8861 |
|
| 70 |
+
| 0.413 | 0.3429 | 150 | 0.2596 | 0.9104 |
|
| 71 |
+
| 0.377 | 0.3657 | 160 | 0.3519 | 0.8711 |
|
| 72 |
+
| 0.4219 | 0.3886 | 170 | 0.2979 | 0.8947 |
|
| 73 |
+
| 0.3317 | 0.4114 | 180 | 0.2227 | 0.9226 |
|
| 74 |
+
| 0.3131 | 0.4343 | 190 | 0.3680 | 0.8693 |
|
| 75 |
+
| 0.3266 | 0.4571 | 200 | 0.2098 | 0.9309 |
|
| 76 |
+
| 0.3306 | 0.48 | 210 | 0.3849 | 0.8824 |
|
| 77 |
+
| 0.3037 | 0.5029 | 220 | 0.2852 | 0.9024 |
|
| 78 |
+
| 0.3086 | 0.5257 | 230 | 0.2725 | 0.9121 |
|
| 79 |
+
| 0.2576 | 0.5486 | 240 | 0.1869 | 0.9356 |
|
| 80 |
+
| 0.2469 | 0.5714 | 250 | 0.2262 | 0.9243 |
|
| 81 |
+
| 0.2405 | 0.5943 | 260 | 0.1963 | 0.9347 |
|
| 82 |
+
| 0.2802 | 0.6171 | 270 | 0.3680 | 0.8804 |
|
| 83 |
+
| 0.2442 | 0.64 | 280 | 0.2053 | 0.9293 |
|
| 84 |
+
| 0.2302 | 0.6629 | 290 | 0.3356 | 0.8967 |
|
| 85 |
+
| 0.2492 | 0.6857 | 300 | 0.1880 | 0.9371 |
|
| 86 |
+
| 0.2089 | 0.7086 | 310 | 0.2076 | 0.9289 |
|
| 87 |
+
| 0.2824 | 0.7314 | 320 | 0.1999 | 0.9301 |
|
| 88 |
+
| 0.2009 | 0.7543 | 330 | 0.1492 | 0.9521 |
|
| 89 |
+
| 0.2001 | 0.7771 | 340 | 0.1496 | 0.9517 |
|
| 90 |
+
| 0.2298 | 0.8 | 350 | 0.1579 | 0.9490 |
|
| 91 |
+
| 0.1802 | 0.8229 | 360 | 0.1506 | 0.9501 |
|
| 92 |
+
| 0.1914 | 0.8457 | 370 | 0.2036 | 0.9311 |
|
| 93 |
+
| 0.1897 | 0.8686 | 380 | 0.1838 | 0.9383 |
|
| 94 |
+
| 0.1203 | 0.8914 | 390 | 0.1459 | 0.9504 |
|
| 95 |
+
| 0.1372 | 0.9143 | 400 | 0.1748 | 0.9419 |
|
| 96 |
+
| 0.1942 | 0.9371 | 410 | 0.1813 | 0.9406 |
|
| 97 |
+
| 0.1886 | 0.96 | 420 | 0.1536 | 0.9510 |
|
| 98 |
+
| 0.1872 | 0.9829 | 430 | 0.1466 | 0.9526 |
|
| 99 |
+
|
| 100 |
+
|
| 101 |
+
### Framework versions
|
| 102 |
+
|
| 103 |
+
- Transformers 4.47.0
|
| 104 |
+
- Pytorch 2.4.1+cu118
|
| 105 |
+
- Datasets 3.6.0
|
| 106 |
+
- Tokenizers 0.21.0
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9988571428571429,
|
| 3 |
+
"total_flos": 1.3128537437918904e+18,
|
| 4 |
+
"train_loss": 0.3557066834218442,
|
| 5 |
+
"train_runtime": 12202.3201,
|
| 6 |
+
"train_samples": 28000,
|
| 7 |
+
"train_samples_per_second": 2.295,
|
| 8 |
+
"train_steps_per_second": 0.036
|
| 9 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,89 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "TencentGameMate/chinese-hubert-base",
|
| 3 |
+
"activation_dropout": 0.1,
|
| 4 |
+
"apply_spec_augment": true,
|
| 5 |
+
"architectures": [
|
| 6 |
+
"HubertForSpeechClassification"
|
| 7 |
+
],
|
| 8 |
+
"attention_dropout": 0.1,
|
| 9 |
+
"bos_token_id": 1,
|
| 10 |
+
"classifier_proj_size": 256,
|
| 11 |
+
"conv_bias": false,
|
| 12 |
+
"conv_dim": [
|
| 13 |
+
512,
|
| 14 |
+
512,
|
| 15 |
+
512,
|
| 16 |
+
512,
|
| 17 |
+
512,
|
| 18 |
+
512,
|
| 19 |
+
512
|
| 20 |
+
],
|
| 21 |
+
"conv_kernel": [
|
| 22 |
+
10,
|
| 23 |
+
3,
|
| 24 |
+
3,
|
| 25 |
+
3,
|
| 26 |
+
3,
|
| 27 |
+
2,
|
| 28 |
+
2
|
| 29 |
+
],
|
| 30 |
+
"conv_stride": [
|
| 31 |
+
5,
|
| 32 |
+
2,
|
| 33 |
+
2,
|
| 34 |
+
2,
|
| 35 |
+
2,
|
| 36 |
+
2,
|
| 37 |
+
2
|
| 38 |
+
],
|
| 39 |
+
"ctc_loss_reduction": "sum",
|
| 40 |
+
"ctc_zero_infinity": false,
|
| 41 |
+
"do_stable_layer_norm": false,
|
| 42 |
+
"eos_token_id": 2,
|
| 43 |
+
"feat_extract_activation": "gelu",
|
| 44 |
+
"feat_extract_norm": "group",
|
| 45 |
+
"feat_proj_dropout": 0.0,
|
| 46 |
+
"feat_proj_layer_norm": true,
|
| 47 |
+
"final_dropout": 0.1,
|
| 48 |
+
"finetuning_task": "wav2vec2_clf",
|
| 49 |
+
"hidden_act": "gelu",
|
| 50 |
+
"hidden_dropout": 0.1,
|
| 51 |
+
"hidden_size": 768,
|
| 52 |
+
"id2label": {
|
| 53 |
+
"0": "Angry",
|
| 54 |
+
"1": "Happy",
|
| 55 |
+
"2": "Neutral",
|
| 56 |
+
"3": "Sad",
|
| 57 |
+
"4": "Surprise"
|
| 58 |
+
},
|
| 59 |
+
"initializer_range": 0.02,
|
| 60 |
+
"intermediate_size": 3072,
|
| 61 |
+
"label2id": {
|
| 62 |
+
"Angry": 0,
|
| 63 |
+
"Happy": 1,
|
| 64 |
+
"Neutral": 2,
|
| 65 |
+
"Sad": 3,
|
| 66 |
+
"Surprise": 4
|
| 67 |
+
},
|
| 68 |
+
"layer_norm_eps": 1e-05,
|
| 69 |
+
"layerdrop": 0.1,
|
| 70 |
+
"mask_feature_length": 10,
|
| 71 |
+
"mask_feature_min_masks": 0,
|
| 72 |
+
"mask_feature_prob": 0.0,
|
| 73 |
+
"mask_time_length": 10,
|
| 74 |
+
"mask_time_min_masks": 2,
|
| 75 |
+
"mask_time_prob": 0.05,
|
| 76 |
+
"model_type": "hubert",
|
| 77 |
+
"num_attention_heads": 12,
|
| 78 |
+
"num_conv_pos_embedding_groups": 16,
|
| 79 |
+
"num_conv_pos_embeddings": 128,
|
| 80 |
+
"num_feat_extract_layers": 7,
|
| 81 |
+
"num_hidden_layers": 12,
|
| 82 |
+
"pad_token_id": 0,
|
| 83 |
+
"pooling_mode": "mean",
|
| 84 |
+
"problem_type": "single_label_classification",
|
| 85 |
+
"torch_dtype": "float32",
|
| 86 |
+
"transformers_version": "4.47.0",
|
| 87 |
+
"use_weighted_layer_sum": false,
|
| 88 |
+
"vocab_size": 32
|
| 89 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d57dffd9609c1029aee07cfb2f7d6a796d411ae54bd6ea5911681dfca89f2e4d
|
| 3 |
+
size 379890236
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
| 4 |
+
"feature_size": 1,
|
| 5 |
+
"padding_side": "right",
|
| 6 |
+
"padding_value": 0,
|
| 7 |
+
"return_attention_mask": false,
|
| 8 |
+
"sampling_rate": 16000
|
| 9 |
+
}
|
runs/Aug05_18-01-01_dengzhipeng-pc/events.out.tfevents.1754388064.dengzhipeng-pc.22996.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6d5aeda078dd282aeb2c10ce9b70fb8eb6cdbafc827615b025ee77e80dae37ae
|
| 3 |
+
size 6161
|
runs/Aug05_18-01-01_dengzhipeng-pc/events.out.tfevents.1754388350.dengzhipeng-pc.22996.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:70efb564af3877bab96664e175bc42e28eedf803a61377c4ab2be965171f4af7
|
| 3 |
+
size 6161
|
runs/Aug05_18-01-01_dengzhipeng-pc/events.out.tfevents.1754388504.dengzhipeng-pc.22996.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:929ef6291819fb5ea9c0bb11387abc37d167e01be481804bac0049f206b23ab2
|
| 3 |
+
size 6161
|
runs/Aug05_18-01-01_dengzhipeng-pc/events.out.tfevents.1754390203.dengzhipeng-pc.22996.3
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3b932e4c4fdcc196e288297a854b39fc0426b3cfa69c2443d5a6124de658cba4
|
| 3 |
+
size 6210
|
runs/Aug05_18-01-01_dengzhipeng-pc/events.out.tfevents.1754390300.dengzhipeng-pc.22996.4
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a13f7f5fd06a3cf5ec60bb1483293d6b7928f33dcc9d56febec1401a63fff7e5
|
| 3 |
+
size 6210
|
runs/Aug05_18-44-57_dengzhipeng-pc/events.out.tfevents.1754390714.dengzhipeng-pc.22996.5
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ea7e7d9d9726a643450a0b577939a86c5e1947fece6c0c4084cca8300ec28422
|
| 3 |
+
size 6211
|
runs/Aug05_18-46-12_dengzhipeng-pc/events.out.tfevents.1754390772.dengzhipeng-pc.22996.6
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:859cd8c2c9fcf9cee046ab92346063de05c23edd09bb4244de477447ff938423
|
| 3 |
+
size 6211
|
runs/Aug05_19-09-27_dengzhipeng-pc/events.out.tfevents.1754392167.dengzhipeng-pc.22996.7
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:47b5ad34997c78343e2065f694947c60707f9bac9bebb801f8585940e3defe0f
|
| 3 |
+
size 6210
|
runs/Aug05_19-09-27_dengzhipeng-pc/events.out.tfevents.1754392276.dengzhipeng-pc.22996.8
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:54043277e473862bfecd954d6916cc6bf6d3c5fce985b4ce9bd94f3d7fd4ea40
|
| 3 |
+
size 7989
|
runs/Aug05_19-09-27_dengzhipeng-pc/events.out.tfevents.1754393216.dengzhipeng-pc.22996.9
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a11e88fbe40ca4ac8d4668f97a81e41c16d6331c785614d2538fc765db77ca84
|
| 3 |
+
size 10587
|
runs/Aug05_20-02-33_dengzhipeng-pc/events.out.tfevents.1754395354.dengzhipeng-pc.22996.10
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e4be4563dbc6148f2f85c1677e3b813bee42047b57fba848cdcbe2e6392c757c
|
| 3 |
+
size 29385
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9988571428571429,
|
| 3 |
+
"total_flos": 1.3128537437918904e+18,
|
| 4 |
+
"train_loss": 0.3557066834218442,
|
| 5 |
+
"train_runtime": 12202.3201,
|
| 6 |
+
"train_samples": 28000,
|
| 7 |
+
"train_samples_per_second": 2.295,
|
| 8 |
+
"train_steps_per_second": 0.036
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,730 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9988571428571429,
|
| 5 |
+
"eval_steps": 10,
|
| 6 |
+
"global_step": 437,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.022857142857142857,
|
| 13 |
+
"grad_norm": 6.883157253265381,
|
| 14 |
+
"learning_rate": 9.77116704805492e-05,
|
| 15 |
+
"loss": 0.9709,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.022857142857142857,
|
| 20 |
+
"eval_accuracy": 0.6398571133613586,
|
| 21 |
+
"eval_loss": 0.8923419117927551,
|
| 22 |
+
"eval_runtime": 252.6626,
|
| 23 |
+
"eval_samples_per_second": 27.705,
|
| 24 |
+
"eval_steps_per_second": 6.926,
|
| 25 |
+
"step": 10
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.045714285714285714,
|
| 29 |
+
"grad_norm": 4.793847560882568,
|
| 30 |
+
"learning_rate": 9.542334096109841e-05,
|
| 31 |
+
"loss": 0.9219,
|
| 32 |
+
"step": 20
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.045714285714285714,
|
| 36 |
+
"eval_accuracy": 0.7664285898208618,
|
| 37 |
+
"eval_loss": 0.6903320550918579,
|
| 38 |
+
"eval_runtime": 260.5483,
|
| 39 |
+
"eval_samples_per_second": 26.866,
|
| 40 |
+
"eval_steps_per_second": 6.717,
|
| 41 |
+
"step": 20
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 0.06857142857142857,
|
| 45 |
+
"grad_norm": 6.191551685333252,
|
| 46 |
+
"learning_rate": 9.31350114416476e-05,
|
| 47 |
+
"loss": 0.7112,
|
| 48 |
+
"step": 30
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 0.06857142857142857,
|
| 52 |
+
"eval_accuracy": 0.7908571362495422,
|
| 53 |
+
"eval_loss": 0.5838488936424255,
|
| 54 |
+
"eval_runtime": 254.6091,
|
| 55 |
+
"eval_samples_per_second": 27.493,
|
| 56 |
+
"eval_steps_per_second": 6.873,
|
| 57 |
+
"step": 30
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.09142857142857143,
|
| 61 |
+
"grad_norm": 9.833272933959961,
|
| 62 |
+
"learning_rate": 9.08466819221968e-05,
|
| 63 |
+
"loss": 0.567,
|
| 64 |
+
"step": 40
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.09142857142857143,
|
| 68 |
+
"eval_accuracy": 0.8158571720123291,
|
| 69 |
+
"eval_loss": 0.5405334830284119,
|
| 70 |
+
"eval_runtime": 263.3184,
|
| 71 |
+
"eval_samples_per_second": 26.584,
|
| 72 |
+
"eval_steps_per_second": 6.646,
|
| 73 |
+
"step": 40
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.11428571428571428,
|
| 77 |
+
"grad_norm": 9.925666809082031,
|
| 78 |
+
"learning_rate": 8.878718535469108e-05,
|
| 79 |
+
"loss": 0.6184,
|
| 80 |
+
"step": 50
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.11428571428571428,
|
| 84 |
+
"eval_accuracy": 0.8581428527832031,
|
| 85 |
+
"eval_loss": 0.41476812958717346,
|
| 86 |
+
"eval_runtime": 259.1036,
|
| 87 |
+
"eval_samples_per_second": 27.016,
|
| 88 |
+
"eval_steps_per_second": 6.754,
|
| 89 |
+
"step": 50
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.13714285714285715,
|
| 93 |
+
"grad_norm": 3.723980665206909,
|
| 94 |
+
"learning_rate": 8.649885583524028e-05,
|
| 95 |
+
"loss": 0.5291,
|
| 96 |
+
"step": 60
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 0.13714285714285715,
|
| 100 |
+
"eval_accuracy": 0.8511428833007812,
|
| 101 |
+
"eval_loss": 0.44439756870269775,
|
| 102 |
+
"eval_runtime": 253.5826,
|
| 103 |
+
"eval_samples_per_second": 27.604,
|
| 104 |
+
"eval_steps_per_second": 6.901,
|
| 105 |
+
"step": 60
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.16,
|
| 109 |
+
"grad_norm": 10.508088111877441,
|
| 110 |
+
"learning_rate": 8.421052631578948e-05,
|
| 111 |
+
"loss": 0.533,
|
| 112 |
+
"step": 70
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 0.16,
|
| 116 |
+
"eval_accuracy": 0.8271428346633911,
|
| 117 |
+
"eval_loss": 0.4642958641052246,
|
| 118 |
+
"eval_runtime": 260.9488,
|
| 119 |
+
"eval_samples_per_second": 26.825,
|
| 120 |
+
"eval_steps_per_second": 6.706,
|
| 121 |
+
"step": 70
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.18285714285714286,
|
| 125 |
+
"grad_norm": 7.824756622314453,
|
| 126 |
+
"learning_rate": 8.192219679633868e-05,
|
| 127 |
+
"loss": 0.4753,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.18285714285714286,
|
| 132 |
+
"eval_accuracy": 0.876714289188385,
|
| 133 |
+
"eval_loss": 0.35598087310791016,
|
| 134 |
+
"eval_runtime": 262.7831,
|
| 135 |
+
"eval_samples_per_second": 26.638,
|
| 136 |
+
"eval_steps_per_second": 6.659,
|
| 137 |
+
"step": 80
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.2057142857142857,
|
| 141 |
+
"grad_norm": 5.332316875457764,
|
| 142 |
+
"learning_rate": 7.963386727688788e-05,
|
| 143 |
+
"loss": 0.4252,
|
| 144 |
+
"step": 90
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.2057142857142857,
|
| 148 |
+
"eval_accuracy": 0.8102856874465942,
|
| 149 |
+
"eval_loss": 0.5888535380363464,
|
| 150 |
+
"eval_runtime": 262.7552,
|
| 151 |
+
"eval_samples_per_second": 26.641,
|
| 152 |
+
"eval_steps_per_second": 6.66,
|
| 153 |
+
"step": 90
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.22857142857142856,
|
| 157 |
+
"grad_norm": 17.482688903808594,
|
| 158 |
+
"learning_rate": 7.734553775743708e-05,
|
| 159 |
+
"loss": 0.5007,
|
| 160 |
+
"step": 100
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 0.22857142857142856,
|
| 164 |
+
"eval_accuracy": 0.8662857413291931,
|
| 165 |
+
"eval_loss": 0.38821107149124146,
|
| 166 |
+
"eval_runtime": 261.4572,
|
| 167 |
+
"eval_samples_per_second": 26.773,
|
| 168 |
+
"eval_steps_per_second": 6.693,
|
| 169 |
+
"step": 100
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 0.25142857142857145,
|
| 173 |
+
"grad_norm": 8.691084861755371,
|
| 174 |
+
"learning_rate": 7.505720823798627e-05,
|
| 175 |
+
"loss": 0.5605,
|
| 176 |
+
"step": 110
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.25142857142857145,
|
| 180 |
+
"eval_accuracy": 0.8921428322792053,
|
| 181 |
+
"eval_loss": 0.32210296392440796,
|
| 182 |
+
"eval_runtime": 261.1514,
|
| 183 |
+
"eval_samples_per_second": 26.804,
|
| 184 |
+
"eval_steps_per_second": 6.701,
|
| 185 |
+
"step": 110
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.2742857142857143,
|
| 189 |
+
"grad_norm": 11.754142761230469,
|
| 190 |
+
"learning_rate": 7.276887871853547e-05,
|
| 191 |
+
"loss": 0.4875,
|
| 192 |
+
"step": 120
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.2742857142857143,
|
| 196 |
+
"eval_accuracy": 0.8558571338653564,
|
| 197 |
+
"eval_loss": 0.36388570070266724,
|
| 198 |
+
"eval_runtime": 265.2182,
|
| 199 |
+
"eval_samples_per_second": 26.393,
|
| 200 |
+
"eval_steps_per_second": 6.598,
|
| 201 |
+
"step": 120
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 0.29714285714285715,
|
| 205 |
+
"grad_norm": 7.222925662994385,
|
| 206 |
+
"learning_rate": 7.048054919908466e-05,
|
| 207 |
+
"loss": 0.4277,
|
| 208 |
+
"step": 130
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 0.29714285714285715,
|
| 212 |
+
"eval_accuracy": 0.8745714426040649,
|
| 213 |
+
"eval_loss": 0.35708051919937134,
|
| 214 |
+
"eval_runtime": 264.6016,
|
| 215 |
+
"eval_samples_per_second": 26.455,
|
| 216 |
+
"eval_steps_per_second": 6.614,
|
| 217 |
+
"step": 130
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 0.32,
|
| 221 |
+
"grad_norm": 6.181695938110352,
|
| 222 |
+
"learning_rate": 6.819221967963387e-05,
|
| 223 |
+
"loss": 0.3415,
|
| 224 |
+
"step": 140
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.32,
|
| 228 |
+
"eval_accuracy": 0.8861428499221802,
|
| 229 |
+
"eval_loss": 0.33818891644477844,
|
| 230 |
+
"eval_runtime": 262.5039,
|
| 231 |
+
"eval_samples_per_second": 26.666,
|
| 232 |
+
"eval_steps_per_second": 6.667,
|
| 233 |
+
"step": 140
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.34285714285714286,
|
| 237 |
+
"grad_norm": 8.087543487548828,
|
| 238 |
+
"learning_rate": 6.590389016018307e-05,
|
| 239 |
+
"loss": 0.413,
|
| 240 |
+
"step": 150
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.34285714285714286,
|
| 244 |
+
"eval_accuracy": 0.9104285836219788,
|
| 245 |
+
"eval_loss": 0.2596481442451477,
|
| 246 |
+
"eval_runtime": 265.6837,
|
| 247 |
+
"eval_samples_per_second": 26.347,
|
| 248 |
+
"eval_steps_per_second": 6.587,
|
| 249 |
+
"step": 150
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.3657142857142857,
|
| 253 |
+
"grad_norm": 11.313796997070312,
|
| 254 |
+
"learning_rate": 6.361556064073226e-05,
|
| 255 |
+
"loss": 0.377,
|
| 256 |
+
"step": 160
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 0.3657142857142857,
|
| 260 |
+
"eval_accuracy": 0.8711428642272949,
|
| 261 |
+
"eval_loss": 0.3518799841403961,
|
| 262 |
+
"eval_runtime": 264.3798,
|
| 263 |
+
"eval_samples_per_second": 26.477,
|
| 264 |
+
"eval_steps_per_second": 6.619,
|
| 265 |
+
"step": 160
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 0.38857142857142857,
|
| 269 |
+
"grad_norm": 7.65640115737915,
|
| 270 |
+
"learning_rate": 6.132723112128147e-05,
|
| 271 |
+
"loss": 0.4219,
|
| 272 |
+
"step": 170
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.38857142857142857,
|
| 276 |
+
"eval_accuracy": 0.8947142958641052,
|
| 277 |
+
"eval_loss": 0.2979215681552887,
|
| 278 |
+
"eval_runtime": 262.8341,
|
| 279 |
+
"eval_samples_per_second": 26.633,
|
| 280 |
+
"eval_steps_per_second": 6.658,
|
| 281 |
+
"step": 170
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 0.4114285714285714,
|
| 285 |
+
"grad_norm": 6.2714433670043945,
|
| 286 |
+
"learning_rate": 5.903890160183066e-05,
|
| 287 |
+
"loss": 0.3317,
|
| 288 |
+
"step": 180
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 0.4114285714285714,
|
| 292 |
+
"eval_accuracy": 0.9225714206695557,
|
| 293 |
+
"eval_loss": 0.22266168892383575,
|
| 294 |
+
"eval_runtime": 265.1248,
|
| 295 |
+
"eval_samples_per_second": 26.403,
|
| 296 |
+
"eval_steps_per_second": 6.601,
|
| 297 |
+
"step": 180
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.4342857142857143,
|
| 301 |
+
"grad_norm": 8.710111618041992,
|
| 302 |
+
"learning_rate": 5.675057208237986e-05,
|
| 303 |
+
"loss": 0.3131,
|
| 304 |
+
"step": 190
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.4342857142857143,
|
| 308 |
+
"eval_accuracy": 0.8692857027053833,
|
| 309 |
+
"eval_loss": 0.3680011034011841,
|
| 310 |
+
"eval_runtime": 260.0056,
|
| 311 |
+
"eval_samples_per_second": 26.923,
|
| 312 |
+
"eval_steps_per_second": 6.731,
|
| 313 |
+
"step": 190
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 0.45714285714285713,
|
| 317 |
+
"grad_norm": 4.041360378265381,
|
| 318 |
+
"learning_rate": 5.446224256292907e-05,
|
| 319 |
+
"loss": 0.3266,
|
| 320 |
+
"step": 200
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.45714285714285713,
|
| 324 |
+
"eval_accuracy": 0.9308571219444275,
|
| 325 |
+
"eval_loss": 0.20981180667877197,
|
| 326 |
+
"eval_runtime": 256.153,
|
| 327 |
+
"eval_samples_per_second": 27.327,
|
| 328 |
+
"eval_steps_per_second": 6.832,
|
| 329 |
+
"step": 200
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 0.48,
|
| 333 |
+
"grad_norm": 10.932918548583984,
|
| 334 |
+
"learning_rate": 5.217391304347826e-05,
|
| 335 |
+
"loss": 0.3306,
|
| 336 |
+
"step": 210
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 0.48,
|
| 340 |
+
"eval_accuracy": 0.8824285864830017,
|
| 341 |
+
"eval_loss": 0.3848917782306671,
|
| 342 |
+
"eval_runtime": 253.9958,
|
| 343 |
+
"eval_samples_per_second": 27.56,
|
| 344 |
+
"eval_steps_per_second": 6.89,
|
| 345 |
+
"step": 210
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.5028571428571429,
|
| 349 |
+
"grad_norm": 9.440160751342773,
|
| 350 |
+
"learning_rate": 4.9885583524027466e-05,
|
| 351 |
+
"loss": 0.3037,
|
| 352 |
+
"step": 220
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.5028571428571429,
|
| 356 |
+
"eval_accuracy": 0.9024285674095154,
|
| 357 |
+
"eval_loss": 0.28518444299697876,
|
| 358 |
+
"eval_runtime": 259.3612,
|
| 359 |
+
"eval_samples_per_second": 26.989,
|
| 360 |
+
"eval_steps_per_second": 6.747,
|
| 361 |
+
"step": 220
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.5257142857142857,
|
| 365 |
+
"grad_norm": 9.196854591369629,
|
| 366 |
+
"learning_rate": 4.759725400457666e-05,
|
| 367 |
+
"loss": 0.3086,
|
| 368 |
+
"step": 230
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 0.5257142857142857,
|
| 372 |
+
"eval_accuracy": 0.9121428728103638,
|
| 373 |
+
"eval_loss": 0.272481232881546,
|
| 374 |
+
"eval_runtime": 254.9581,
|
| 375 |
+
"eval_samples_per_second": 27.455,
|
| 376 |
+
"eval_steps_per_second": 6.864,
|
| 377 |
+
"step": 230
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.5485714285714286,
|
| 381 |
+
"grad_norm": 6.610895156860352,
|
| 382 |
+
"learning_rate": 4.530892448512586e-05,
|
| 383 |
+
"loss": 0.2576,
|
| 384 |
+
"step": 240
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.5485714285714286,
|
| 388 |
+
"eval_accuracy": 0.9355714321136475,
|
| 389 |
+
"eval_loss": 0.18688350915908813,
|
| 390 |
+
"eval_runtime": 255.2292,
|
| 391 |
+
"eval_samples_per_second": 27.426,
|
| 392 |
+
"eval_steps_per_second": 6.857,
|
| 393 |
+
"step": 240
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 0.5714285714285714,
|
| 397 |
+
"grad_norm": 15.24905014038086,
|
| 398 |
+
"learning_rate": 4.302059496567506e-05,
|
| 399 |
+
"loss": 0.2469,
|
| 400 |
+
"step": 250
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 0.5714285714285714,
|
| 404 |
+
"eval_accuracy": 0.9242857098579407,
|
| 405 |
+
"eval_loss": 0.2262311726808548,
|
| 406 |
+
"eval_runtime": 254.9064,
|
| 407 |
+
"eval_samples_per_second": 27.461,
|
| 408 |
+
"eval_steps_per_second": 6.865,
|
| 409 |
+
"step": 250
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.5942857142857143,
|
| 413 |
+
"grad_norm": 9.8357515335083,
|
| 414 |
+
"learning_rate": 4.073226544622426e-05,
|
| 415 |
+
"loss": 0.2405,
|
| 416 |
+
"step": 260
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.5942857142857143,
|
| 420 |
+
"eval_accuracy": 0.9347142577171326,
|
| 421 |
+
"eval_loss": 0.19631564617156982,
|
| 422 |
+
"eval_runtime": 271.1966,
|
| 423 |
+
"eval_samples_per_second": 25.812,
|
| 424 |
+
"eval_steps_per_second": 6.453,
|
| 425 |
+
"step": 260
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 0.6171428571428571,
|
| 429 |
+
"grad_norm": 19.872060775756836,
|
| 430 |
+
"learning_rate": 3.844393592677346e-05,
|
| 431 |
+
"loss": 0.2802,
|
| 432 |
+
"step": 270
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 0.6171428571428571,
|
| 436 |
+
"eval_accuracy": 0.8804285526275635,
|
| 437 |
+
"eval_loss": 0.3679888844490051,
|
| 438 |
+
"eval_runtime": 256.0669,
|
| 439 |
+
"eval_samples_per_second": 27.337,
|
| 440 |
+
"eval_steps_per_second": 6.834,
|
| 441 |
+
"step": 270
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 0.64,
|
| 445 |
+
"grad_norm": 3.6445915699005127,
|
| 446 |
+
"learning_rate": 3.6155606407322653e-05,
|
| 447 |
+
"loss": 0.2442,
|
| 448 |
+
"step": 280
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 0.64,
|
| 452 |
+
"eval_accuracy": 0.9292857050895691,
|
| 453 |
+
"eval_loss": 0.20533673465251923,
|
| 454 |
+
"eval_runtime": 255.7952,
|
| 455 |
+
"eval_samples_per_second": 27.366,
|
| 456 |
+
"eval_steps_per_second": 6.841,
|
| 457 |
+
"step": 280
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.6628571428571428,
|
| 461 |
+
"grad_norm": 8.114418983459473,
|
| 462 |
+
"learning_rate": 3.3867276887871856e-05,
|
| 463 |
+
"loss": 0.2302,
|
| 464 |
+
"step": 290
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.6628571428571428,
|
| 468 |
+
"eval_accuracy": 0.8967142701148987,
|
| 469 |
+
"eval_loss": 0.3355866074562073,
|
| 470 |
+
"eval_runtime": 257.891,
|
| 471 |
+
"eval_samples_per_second": 27.143,
|
| 472 |
+
"eval_steps_per_second": 6.786,
|
| 473 |
+
"step": 290
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 0.6857142857142857,
|
| 477 |
+
"grad_norm": 5.993322372436523,
|
| 478 |
+
"learning_rate": 3.157894736842105e-05,
|
| 479 |
+
"loss": 0.2492,
|
| 480 |
+
"step": 300
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 0.6857142857142857,
|
| 484 |
+
"eval_accuracy": 0.9371428489685059,
|
| 485 |
+
"eval_loss": 0.18795913457870483,
|
| 486 |
+
"eval_runtime": 254.5882,
|
| 487 |
+
"eval_samples_per_second": 27.495,
|
| 488 |
+
"eval_steps_per_second": 6.874,
|
| 489 |
+
"step": 300
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 0.7085714285714285,
|
| 493 |
+
"grad_norm": 6.529418468475342,
|
| 494 |
+
"learning_rate": 2.9290617848970254e-05,
|
| 495 |
+
"loss": 0.2089,
|
| 496 |
+
"step": 310
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 0.7085714285714285,
|
| 500 |
+
"eval_accuracy": 0.928857147693634,
|
| 501 |
+
"eval_loss": 0.2076321393251419,
|
| 502 |
+
"eval_runtime": 260.5938,
|
| 503 |
+
"eval_samples_per_second": 26.862,
|
| 504 |
+
"eval_steps_per_second": 6.715,
|
| 505 |
+
"step": 310
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 0.7314285714285714,
|
| 509 |
+
"grad_norm": 6.433741092681885,
|
| 510 |
+
"learning_rate": 2.7002288329519453e-05,
|
| 511 |
+
"loss": 0.2824,
|
| 512 |
+
"step": 320
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 0.7314285714285714,
|
| 516 |
+
"eval_accuracy": 0.930142879486084,
|
| 517 |
+
"eval_loss": 0.1999480277299881,
|
| 518 |
+
"eval_runtime": 255.2396,
|
| 519 |
+
"eval_samples_per_second": 27.425,
|
| 520 |
+
"eval_steps_per_second": 6.856,
|
| 521 |
+
"step": 320
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.7542857142857143,
|
| 525 |
+
"grad_norm": 5.394837379455566,
|
| 526 |
+
"learning_rate": 2.4713958810068652e-05,
|
| 527 |
+
"loss": 0.2009,
|
| 528 |
+
"step": 330
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.7542857142857143,
|
| 532 |
+
"eval_accuracy": 0.9521428346633911,
|
| 533 |
+
"eval_loss": 0.14918017387390137,
|
| 534 |
+
"eval_runtime": 258.1497,
|
| 535 |
+
"eval_samples_per_second": 27.116,
|
| 536 |
+
"eval_steps_per_second": 6.779,
|
| 537 |
+
"step": 330
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 0.7771428571428571,
|
| 541 |
+
"grad_norm": 5.843348503112793,
|
| 542 |
+
"learning_rate": 2.242562929061785e-05,
|
| 543 |
+
"loss": 0.2001,
|
| 544 |
+
"step": 340
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 0.7771428571428571,
|
| 548 |
+
"eval_accuracy": 0.951714277267456,
|
| 549 |
+
"eval_loss": 0.14960123598575592,
|
| 550 |
+
"eval_runtime": 253.1262,
|
| 551 |
+
"eval_samples_per_second": 27.654,
|
| 552 |
+
"eval_steps_per_second": 6.914,
|
| 553 |
+
"step": 340
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 0.8,
|
| 557 |
+
"grad_norm": 7.778473377227783,
|
| 558 |
+
"learning_rate": 2.0137299771167047e-05,
|
| 559 |
+
"loss": 0.2298,
|
| 560 |
+
"step": 350
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 0.8,
|
| 564 |
+
"eval_accuracy": 0.9490000009536743,
|
| 565 |
+
"eval_loss": 0.15794885158538818,
|
| 566 |
+
"eval_runtime": 258.4154,
|
| 567 |
+
"eval_samples_per_second": 27.088,
|
| 568 |
+
"eval_steps_per_second": 6.772,
|
| 569 |
+
"step": 350
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.8228571428571428,
|
| 573 |
+
"grad_norm": 7.672749042510986,
|
| 574 |
+
"learning_rate": 1.784897025171625e-05,
|
| 575 |
+
"loss": 0.1802,
|
| 576 |
+
"step": 360
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.8228571428571428,
|
| 580 |
+
"eval_accuracy": 0.9501428604125977,
|
| 581 |
+
"eval_loss": 0.15056686103343964,
|
| 582 |
+
"eval_runtime": 253.0586,
|
| 583 |
+
"eval_samples_per_second": 27.662,
|
| 584 |
+
"eval_steps_per_second": 6.915,
|
| 585 |
+
"step": 360
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 0.8457142857142858,
|
| 589 |
+
"grad_norm": 7.994875431060791,
|
| 590 |
+
"learning_rate": 1.5560640732265445e-05,
|
| 591 |
+
"loss": 0.1914,
|
| 592 |
+
"step": 370
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 0.8457142857142858,
|
| 596 |
+
"eval_accuracy": 0.9311428666114807,
|
| 597 |
+
"eval_loss": 0.20363783836364746,
|
| 598 |
+
"eval_runtime": 261.3379,
|
| 599 |
+
"eval_samples_per_second": 26.785,
|
| 600 |
+
"eval_steps_per_second": 6.696,
|
| 601 |
+
"step": 370
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 0.8685714285714285,
|
| 605 |
+
"grad_norm": 3.988149404525757,
|
| 606 |
+
"learning_rate": 1.3272311212814645e-05,
|
| 607 |
+
"loss": 0.1897,
|
| 608 |
+
"step": 380
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 0.8685714285714285,
|
| 612 |
+
"eval_accuracy": 0.9382857084274292,
|
| 613 |
+
"eval_loss": 0.18375040590763092,
|
| 614 |
+
"eval_runtime": 256.8539,
|
| 615 |
+
"eval_samples_per_second": 27.253,
|
| 616 |
+
"eval_steps_per_second": 6.813,
|
| 617 |
+
"step": 380
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 0.8914285714285715,
|
| 621 |
+
"grad_norm": 7.280108451843262,
|
| 622 |
+
"learning_rate": 1.0983981693363844e-05,
|
| 623 |
+
"loss": 0.1203,
|
| 624 |
+
"step": 390
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 0.8914285714285715,
|
| 628 |
+
"eval_accuracy": 0.9504285454750061,
|
| 629 |
+
"eval_loss": 0.1459112912416458,
|
| 630 |
+
"eval_runtime": 256.3941,
|
| 631 |
+
"eval_samples_per_second": 27.302,
|
| 632 |
+
"eval_steps_per_second": 6.825,
|
| 633 |
+
"step": 390
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.9142857142857143,
|
| 637 |
+
"grad_norm": 6.386229991912842,
|
| 638 |
+
"learning_rate": 8.695652173913044e-06,
|
| 639 |
+
"loss": 0.1372,
|
| 640 |
+
"step": 400
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.9142857142857143,
|
| 644 |
+
"eval_accuracy": 0.9418571591377258,
|
| 645 |
+
"eval_loss": 0.1748434156179428,
|
| 646 |
+
"eval_runtime": 266.7645,
|
| 647 |
+
"eval_samples_per_second": 26.24,
|
| 648 |
+
"eval_steps_per_second": 6.56,
|
| 649 |
+
"step": 400
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 0.9371428571428572,
|
| 653 |
+
"grad_norm": 7.714508056640625,
|
| 654 |
+
"learning_rate": 6.407322654462243e-06,
|
| 655 |
+
"loss": 0.1942,
|
| 656 |
+
"step": 410
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 0.9371428571428572,
|
| 660 |
+
"eval_accuracy": 0.9405714273452759,
|
| 661 |
+
"eval_loss": 0.18131674826145172,
|
| 662 |
+
"eval_runtime": 266.6389,
|
| 663 |
+
"eval_samples_per_second": 26.253,
|
| 664 |
+
"eval_steps_per_second": 6.563,
|
| 665 |
+
"step": 410
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 0.96,
|
| 669 |
+
"grad_norm": 4.493211269378662,
|
| 670 |
+
"learning_rate": 4.118993135011442e-06,
|
| 671 |
+
"loss": 0.1886,
|
| 672 |
+
"step": 420
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 0.96,
|
| 676 |
+
"eval_accuracy": 0.9509999752044678,
|
| 677 |
+
"eval_loss": 0.15357272326946259,
|
| 678 |
+
"eval_runtime": 273.0321,
|
| 679 |
+
"eval_samples_per_second": 25.638,
|
| 680 |
+
"eval_steps_per_second": 6.41,
|
| 681 |
+
"step": 420
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.9828571428571429,
|
| 685 |
+
"grad_norm": 4.66563606262207,
|
| 686 |
+
"learning_rate": 1.8306636155606409e-06,
|
| 687 |
+
"loss": 0.1872,
|
| 688 |
+
"step": 430
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.9828571428571429,
|
| 692 |
+
"eval_accuracy": 0.952571451663971,
|
| 693 |
+
"eval_loss": 0.1465713381767273,
|
| 694 |
+
"eval_runtime": 266.7172,
|
| 695 |
+
"eval_samples_per_second": 26.245,
|
| 696 |
+
"eval_steps_per_second": 6.561,
|
| 697 |
+
"step": 430
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 0.9988571428571429,
|
| 701 |
+
"step": 437,
|
| 702 |
+
"total_flos": 1.3128537437918904e+18,
|
| 703 |
+
"train_loss": 0.3557066834218442,
|
| 704 |
+
"train_runtime": 12202.3201,
|
| 705 |
+
"train_samples_per_second": 2.295,
|
| 706 |
+
"train_steps_per_second": 0.036
|
| 707 |
+
}
|
| 708 |
+
],
|
| 709 |
+
"logging_steps": 10,
|
| 710 |
+
"max_steps": 437,
|
| 711 |
+
"num_input_tokens_seen": 0,
|
| 712 |
+
"num_train_epochs": 1,
|
| 713 |
+
"save_steps": 10,
|
| 714 |
+
"stateful_callbacks": {
|
| 715 |
+
"TrainerControl": {
|
| 716 |
+
"args": {
|
| 717 |
+
"should_epoch_stop": false,
|
| 718 |
+
"should_evaluate": false,
|
| 719 |
+
"should_log": false,
|
| 720 |
+
"should_save": true,
|
| 721 |
+
"should_training_stop": true
|
| 722 |
+
},
|
| 723 |
+
"attributes": {}
|
| 724 |
+
}
|
| 725 |
+
},
|
| 726 |
+
"total_flos": 1.3128537437918904e+18,
|
| 727 |
+
"train_batch_size": 32,
|
| 728 |
+
"trial_name": null,
|
| 729 |
+
"trial_params": null
|
| 730 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b30299245c31111db5b268f0927632aca7ff3e92f02299a9653ecdfa84cdf28c
|
| 3 |
+
size 5368
|