Anders L|hr commited on
Commit ·
eef99e5
1
Parent(s): 0b3df6d
Best 5 class model
Browse files- README.md +301 -0
- all_results.json +12 -0
- checkpoint-2640/config.json +124 -0
- checkpoint-2640/model.safetensors +3 -0
- checkpoint-2640/optimizer.pt +3 -0
- checkpoint-2640/preprocessor_config.json +9 -0
- checkpoint-2640/rng_state.pth +3 -0
- checkpoint-2640/scheduler.pt +3 -0
- checkpoint-2640/trainer_state.json +2885 -0
- checkpoint-2640/training_args.bin +3 -0
- checkpoint-3500/config.json +124 -0
- checkpoint-3500/model.safetensors +3 -0
- checkpoint-3500/optimizer.pt +3 -0
- checkpoint-3500/preprocessor_config.json +9 -0
- checkpoint-3500/rng_state.pth +3 -0
- checkpoint-3500/scheduler.pt +3 -0
- checkpoint-3500/trainer_state.json +0 -0
- checkpoint-3500/training_args.bin +3 -0
- config.json +124 -0
- eval_results.json +8 -0
- model.safetensors +3 -0
- preprocessor_config.json +9 -0
- runs/May04_12-45-57_ml6.hpc.uio.no/events.out.tfevents.1714819603.ml6.hpc.uio.no.743903.0 +3 -0
- runs/May04_12-54-19_ml6.hpc.uio.no/events.out.tfevents.1714820073.ml6.hpc.uio.no.746334.0 +3 -0
- runs/May04_12-54-19_ml6.hpc.uio.no/events.out.tfevents.1714823730.ml6.hpc.uio.no.746334.1 +3 -0
- train_results.json +7 -0
- trainer_state.json +0 -0
- training_args.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,301 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: anderloh/Hugginhface-master-wav2vec-pretreined-5-class-train-test
|
| 3 |
+
tags:
|
| 4 |
+
- audio-classification
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
metrics:
|
| 7 |
+
- accuracy
|
| 8 |
+
model-index:
|
| 9 |
+
- name: wav2vec2-5Class-train-test-finetune-Medium
|
| 10 |
+
results: []
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 14 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 15 |
+
|
| 16 |
+
# wav2vec2-5Class-train-test-finetune-Medium
|
| 17 |
+
|
| 18 |
+
This model is a fine-tuned version of [anderloh/Hugginhface-master-wav2vec-pretreined-5-class-train-test](https://huggingface.co/anderloh/Hugginhface-master-wav2vec-pretreined-5-class-train-test) on the anderloh/Master5Class dataset.
|
| 19 |
+
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 0.7855
|
| 21 |
+
- Accuracy: 0.7448
|
| 22 |
+
|
| 23 |
+
## Model description
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Intended uses & limitations
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training and evaluation data
|
| 32 |
+
|
| 33 |
+
More information needed
|
| 34 |
+
|
| 35 |
+
## Training procedure
|
| 36 |
+
|
| 37 |
+
### Training hyperparameters
|
| 38 |
+
|
| 39 |
+
The following hyperparameters were used during training:
|
| 40 |
+
- learning_rate: 3e-05
|
| 41 |
+
- train_batch_size: 16
|
| 42 |
+
- eval_batch_size: 16
|
| 43 |
+
- seed: 0
|
| 44 |
+
- gradient_accumulation_steps: 4
|
| 45 |
+
- total_train_batch_size: 64
|
| 46 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 47 |
+
- lr_scheduler_type: linear
|
| 48 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 49 |
+
- num_epochs: 250.0
|
| 50 |
+
- mixed_precision_training: Native AMP
|
| 51 |
+
|
| 52 |
+
### Training results
|
| 53 |
+
|
| 54 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 55 |
+
|:-------------:|:------:|:----:|:---------------:|:--------:|
|
| 56 |
+
| 1.6852 | 0.95 | 14 | 1.5987 | 0.3427 |
|
| 57 |
+
| 1.5721 | 1.97 | 29 | 1.5976 | 0.3427 |
|
| 58 |
+
| 1.5696 | 2.98 | 44 | 1.5957 | 0.3427 |
|
| 59 |
+
| 1.5671 | 4.0 | 59 | 1.5932 | 0.3357 |
|
| 60 |
+
| 1.6757 | 4.95 | 73 | 1.5902 | 0.3252 |
|
| 61 |
+
| 1.5595 | 5.97 | 88 | 1.5864 | 0.3217 |
|
| 62 |
+
| 1.5536 | 6.98 | 103 | 1.5818 | 0.3182 |
|
| 63 |
+
| 1.5484 | 8.0 | 118 | 1.5765 | 0.3112 |
|
| 64 |
+
| 1.6506 | 8.95 | 132 | 1.5708 | 0.3077 |
|
| 65 |
+
| 1.5317 | 9.97 | 147 | 1.5641 | 0.3007 |
|
| 66 |
+
| 1.5226 | 10.98 | 162 | 1.5564 | 0.2867 |
|
| 67 |
+
| 1.5116 | 12.0 | 177 | 1.5478 | 0.2692 |
|
| 68 |
+
| 1.6046 | 12.95 | 191 | 1.5385 | 0.2622 |
|
| 69 |
+
| 1.4822 | 13.97 | 206 | 1.5285 | 0.2552 |
|
| 70 |
+
| 1.4614 | 14.98 | 221 | 1.5184 | 0.2552 |
|
| 71 |
+
| 1.4396 | 16.0 | 236 | 1.5101 | 0.2517 |
|
| 72 |
+
| 1.5047 | 16.95 | 250 | 1.5104 | 0.2448 |
|
| 73 |
+
| 1.3741 | 17.97 | 265 | 1.5250 | 0.2517 |
|
| 74 |
+
| 1.3512 | 18.98 | 280 | 1.5329 | 0.2657 |
|
| 75 |
+
| 1.3286 | 20.0 | 295 | 1.5095 | 0.3252 |
|
| 76 |
+
| 1.3967 | 20.95 | 309 | 1.4829 | 0.3497 |
|
| 77 |
+
| 1.2779 | 21.97 | 324 | 1.4598 | 0.3846 |
|
| 78 |
+
| 1.2449 | 22.98 | 339 | 1.4284 | 0.4161 |
|
| 79 |
+
| 1.2118 | 24.0 | 354 | 1.4162 | 0.4231 |
|
| 80 |
+
| 1.2521 | 24.95 | 368 | 1.3798 | 0.4476 |
|
| 81 |
+
| 1.1183 | 25.97 | 383 | 1.3246 | 0.4790 |
|
| 82 |
+
| 1.0778 | 26.98 | 398 | 1.2761 | 0.4930 |
|
| 83 |
+
| 1.0306 | 28.0 | 413 | 1.2300 | 0.5105 |
|
| 84 |
+
| 1.0808 | 28.95 | 427 | 1.2064 | 0.5140 |
|
| 85 |
+
| 0.9767 | 29.97 | 442 | 1.1670 | 0.5524 |
|
| 86 |
+
| 0.9589 | 30.98 | 457 | 1.1264 | 0.5734 |
|
| 87 |
+
| 0.9193 | 32.0 | 472 | 1.1013 | 0.5874 |
|
| 88 |
+
| 0.9462 | 32.95 | 486 | 1.0736 | 0.6049 |
|
| 89 |
+
| 0.85 | 33.97 | 501 | 1.0628 | 0.6049 |
|
| 90 |
+
| 0.8294 | 34.98 | 516 | 1.0473 | 0.6189 |
|
| 91 |
+
| 0.8025 | 36.0 | 531 | 1.0030 | 0.6329 |
|
| 92 |
+
| 0.8206 | 36.95 | 545 | 0.9964 | 0.6399 |
|
| 93 |
+
| 0.7541 | 37.97 | 560 | 0.9605 | 0.6608 |
|
| 94 |
+
| 0.7413 | 38.98 | 575 | 0.9467 | 0.6643 |
|
| 95 |
+
| 0.709 | 40.0 | 590 | 0.9348 | 0.6678 |
|
| 96 |
+
| 0.7817 | 40.95 | 604 | 0.9366 | 0.6678 |
|
| 97 |
+
| 0.7034 | 41.97 | 619 | 0.9109 | 0.6818 |
|
| 98 |
+
| 0.6856 | 42.98 | 634 | 0.9277 | 0.6573 |
|
| 99 |
+
| 0.6625 | 44.0 | 649 | 0.8980 | 0.6783 |
|
| 100 |
+
| 0.7207 | 44.95 | 663 | 0.9050 | 0.6713 |
|
| 101 |
+
| 0.6684 | 45.97 | 678 | 0.8973 | 0.6748 |
|
| 102 |
+
| 0.6651 | 46.98 | 693 | 0.8935 | 0.6783 |
|
| 103 |
+
| 0.6451 | 48.0 | 708 | 0.8945 | 0.6748 |
|
| 104 |
+
| 0.6774 | 48.95 | 722 | 0.8879 | 0.6818 |
|
| 105 |
+
| 0.6308 | 49.97 | 737 | 0.8883 | 0.6818 |
|
| 106 |
+
| 0.6199 | 50.98 | 752 | 0.8826 | 0.6818 |
|
| 107 |
+
| 0.6379 | 52.0 | 767 | 0.8582 | 0.6923 |
|
| 108 |
+
| 0.6588 | 52.95 | 781 | 0.8825 | 0.6818 |
|
| 109 |
+
| 0.5857 | 53.97 | 796 | 0.8808 | 0.6748 |
|
| 110 |
+
| 0.6076 | 54.98 | 811 | 0.8555 | 0.6958 |
|
| 111 |
+
| 0.5934 | 56.0 | 826 | 0.8654 | 0.6888 |
|
| 112 |
+
| 0.6427 | 56.95 | 840 | 0.8616 | 0.6853 |
|
| 113 |
+
| 0.5782 | 57.97 | 855 | 0.8711 | 0.6678 |
|
| 114 |
+
| 0.5819 | 58.98 | 870 | 0.8689 | 0.6748 |
|
| 115 |
+
| 0.5918 | 60.0 | 885 | 0.8602 | 0.6923 |
|
| 116 |
+
| 0.5845 | 60.95 | 899 | 0.8459 | 0.6993 |
|
| 117 |
+
| 0.5667 | 61.97 | 914 | 0.8467 | 0.7028 |
|
| 118 |
+
| 0.5327 | 62.98 | 929 | 0.8540 | 0.6923 |
|
| 119 |
+
| 0.523 | 64.0 | 944 | 0.8323 | 0.7063 |
|
| 120 |
+
| 0.548 | 64.95 | 958 | 0.8407 | 0.6993 |
|
| 121 |
+
| 0.5399 | 65.97 | 973 | 0.8379 | 0.6993 |
|
| 122 |
+
| 0.5324 | 66.98 | 988 | 0.8119 | 0.7028 |
|
| 123 |
+
| 0.5171 | 68.0 | 1003 | 0.8445 | 0.6923 |
|
| 124 |
+
| 0.538 | 68.95 | 1017 | 0.8196 | 0.7098 |
|
| 125 |
+
| 0.5312 | 69.97 | 1032 | 0.8415 | 0.6853 |
|
| 126 |
+
| 0.4914 | 70.98 | 1047 | 0.8184 | 0.6958 |
|
| 127 |
+
| 0.5055 | 72.0 | 1062 | 0.8218 | 0.6923 |
|
| 128 |
+
| 0.5401 | 72.95 | 1076 | 0.8160 | 0.7028 |
|
| 129 |
+
| 0.4966 | 73.97 | 1091 | 0.8238 | 0.6888 |
|
| 130 |
+
| 0.4768 | 74.98 | 1106 | 0.8185 | 0.6993 |
|
| 131 |
+
| 0.4789 | 76.0 | 1121 | 0.8261 | 0.7028 |
|
| 132 |
+
| 0.5176 | 76.95 | 1135 | 0.8110 | 0.7098 |
|
| 133 |
+
| 0.466 | 77.97 | 1150 | 0.8141 | 0.6993 |
|
| 134 |
+
| 0.4736 | 78.98 | 1165 | 0.7970 | 0.7168 |
|
| 135 |
+
| 0.4785 | 80.0 | 1180 | 0.8062 | 0.7098 |
|
| 136 |
+
| 0.5309 | 80.95 | 1194 | 0.8051 | 0.6958 |
|
| 137 |
+
| 0.4571 | 81.97 | 1209 | 0.8024 | 0.7098 |
|
| 138 |
+
| 0.47 | 82.98 | 1224 | 0.8031 | 0.7168 |
|
| 139 |
+
| 0.4525 | 84.0 | 1239 | 0.7911 | 0.7133 |
|
| 140 |
+
| 0.5058 | 84.95 | 1253 | 0.7877 | 0.7133 |
|
| 141 |
+
| 0.4627 | 85.97 | 1268 | 0.7918 | 0.7063 |
|
| 142 |
+
| 0.4343 | 86.98 | 1283 | 0.7882 | 0.7168 |
|
| 143 |
+
| 0.4442 | 88.0 | 1298 | 0.8058 | 0.7133 |
|
| 144 |
+
| 0.4745 | 88.95 | 1312 | 0.7810 | 0.7238 |
|
| 145 |
+
| 0.4282 | 89.97 | 1327 | 0.7951 | 0.7098 |
|
| 146 |
+
| 0.4307 | 90.98 | 1342 | 0.7739 | 0.7168 |
|
| 147 |
+
| 0.4403 | 92.0 | 1357 | 0.7788 | 0.7203 |
|
| 148 |
+
| 0.4567 | 92.95 | 1371 | 0.7927 | 0.7168 |
|
| 149 |
+
| 0.4233 | 93.97 | 1386 | 0.7885 | 0.7203 |
|
| 150 |
+
| 0.4347 | 94.98 | 1401 | 0.7849 | 0.7203 |
|
| 151 |
+
| 0.4167 | 96.0 | 1416 | 0.7880 | 0.7238 |
|
| 152 |
+
| 0.4394 | 96.95 | 1430 | 0.7889 | 0.7203 |
|
| 153 |
+
| 0.4359 | 97.97 | 1445 | 0.7785 | 0.7203 |
|
| 154 |
+
| 0.4085 | 98.98 | 1460 | 0.7852 | 0.7133 |
|
| 155 |
+
| 0.3965 | 100.0 | 1475 | 0.7785 | 0.7273 |
|
| 156 |
+
| 0.445 | 100.95 | 1489 | 0.7826 | 0.7203 |
|
| 157 |
+
| 0.3988 | 101.97 | 1504 | 0.8045 | 0.7098 |
|
| 158 |
+
| 0.4129 | 102.98 | 1519 | 0.7686 | 0.7273 |
|
| 159 |
+
| 0.3937 | 104.0 | 1534 | 0.7912 | 0.7133 |
|
| 160 |
+
| 0.4356 | 104.95 | 1548 | 0.7922 | 0.7133 |
|
| 161 |
+
| 0.3969 | 105.97 | 1563 | 0.7752 | 0.7203 |
|
| 162 |
+
| 0.4051 | 106.98 | 1578 | 0.7917 | 0.7133 |
|
| 163 |
+
| 0.3982 | 108.0 | 1593 | 0.7917 | 0.7098 |
|
| 164 |
+
| 0.4117 | 108.95 | 1607 | 0.8071 | 0.7063 |
|
| 165 |
+
| 0.3666 | 109.97 | 1622 | 0.7840 | 0.7203 |
|
| 166 |
+
| 0.3894 | 110.98 | 1637 | 0.7790 | 0.7238 |
|
| 167 |
+
| 0.3858 | 112.0 | 1652 | 0.7961 | 0.7098 |
|
| 168 |
+
| 0.4037 | 112.95 | 1666 | 0.7822 | 0.7203 |
|
| 169 |
+
| 0.3886 | 113.97 | 1681 | 0.7748 | 0.7238 |
|
| 170 |
+
| 0.3762 | 114.98 | 1696 | 0.7782 | 0.7168 |
|
| 171 |
+
| 0.3444 | 116.0 | 1711 | 0.7746 | 0.7308 |
|
| 172 |
+
| 0.3961 | 116.95 | 1725 | 0.7842 | 0.7203 |
|
| 173 |
+
| 0.3578 | 117.97 | 1740 | 0.7819 | 0.7203 |
|
| 174 |
+
| 0.3578 | 118.98 | 1755 | 0.7806 | 0.7203 |
|
| 175 |
+
| 0.3489 | 120.0 | 1770 | 0.7809 | 0.7238 |
|
| 176 |
+
| 0.3622 | 120.95 | 1784 | 0.7947 | 0.7098 |
|
| 177 |
+
| 0.3545 | 121.97 | 1799 | 0.7878 | 0.7168 |
|
| 178 |
+
| 0.3361 | 122.98 | 1814 | 0.7855 | 0.7168 |
|
| 179 |
+
| 0.3618 | 124.0 | 1829 | 0.7890 | 0.7133 |
|
| 180 |
+
| 0.3472 | 124.95 | 1843 | 0.7810 | 0.7168 |
|
| 181 |
+
| 0.3511 | 125.97 | 1858 | 0.7897 | 0.7133 |
|
| 182 |
+
| 0.3389 | 126.98 | 1873 | 0.7923 | 0.7133 |
|
| 183 |
+
| 0.3391 | 128.0 | 1888 | 0.7782 | 0.7273 |
|
| 184 |
+
| 0.3746 | 128.95 | 1902 | 0.7838 | 0.7203 |
|
| 185 |
+
| 0.3238 | 129.97 | 1917 | 0.7943 | 0.7168 |
|
| 186 |
+
| 0.3601 | 130.98 | 1932 | 0.7863 | 0.7168 |
|
| 187 |
+
| 0.3339 | 132.0 | 1947 | 0.7949 | 0.7133 |
|
| 188 |
+
| 0.3805 | 132.95 | 1961 | 0.7823 | 0.7238 |
|
| 189 |
+
| 0.3524 | 133.97 | 1976 | 0.8052 | 0.7098 |
|
| 190 |
+
| 0.3103 | 134.98 | 1991 | 0.7809 | 0.7238 |
|
| 191 |
+
| 0.3484 | 136.0 | 2006 | 0.7879 | 0.7203 |
|
| 192 |
+
| 0.3424 | 136.95 | 2020 | 0.7875 | 0.7273 |
|
| 193 |
+
| 0.316 | 137.97 | 2035 | 0.7829 | 0.7273 |
|
| 194 |
+
| 0.3171 | 138.98 | 2050 | 0.7882 | 0.7203 |
|
| 195 |
+
| 0.3155 | 140.0 | 2065 | 0.7830 | 0.7168 |
|
| 196 |
+
| 0.3382 | 140.95 | 2079 | 0.7826 | 0.7273 |
|
| 197 |
+
| 0.3175 | 141.97 | 2094 | 0.7964 | 0.7203 |
|
| 198 |
+
| 0.3444 | 142.98 | 2109 | 0.7859 | 0.7238 |
|
| 199 |
+
| 0.3208 | 144.0 | 2124 | 0.7860 | 0.7273 |
|
| 200 |
+
| 0.3286 | 144.95 | 2138 | 0.7869 | 0.7273 |
|
| 201 |
+
| 0.3319 | 145.97 | 2153 | 0.7916 | 0.7168 |
|
| 202 |
+
| 0.2954 | 146.98 | 2168 | 0.7938 | 0.7238 |
|
| 203 |
+
| 0.3283 | 148.0 | 2183 | 0.7974 | 0.7168 |
|
| 204 |
+
| 0.3306 | 148.95 | 2197 | 0.7795 | 0.7308 |
|
| 205 |
+
| 0.3073 | 149.97 | 2212 | 0.7910 | 0.7203 |
|
| 206 |
+
| 0.3089 | 150.98 | 2227 | 0.7942 | 0.7203 |
|
| 207 |
+
| 0.2915 | 152.0 | 2242 | 0.7934 | 0.7168 |
|
| 208 |
+
| 0.3286 | 152.95 | 2256 | 0.7808 | 0.7308 |
|
| 209 |
+
| 0.2817 | 153.97 | 2271 | 0.7788 | 0.7308 |
|
| 210 |
+
| 0.3118 | 154.98 | 2286 | 0.7898 | 0.7273 |
|
| 211 |
+
| 0.3155 | 156.0 | 2301 | 0.7966 | 0.7203 |
|
| 212 |
+
| 0.3156 | 156.95 | 2315 | 0.7947 | 0.7203 |
|
| 213 |
+
| 0.2936 | 157.97 | 2330 | 0.7917 | 0.7168 |
|
| 214 |
+
| 0.3049 | 158.98 | 2345 | 0.7780 | 0.7308 |
|
| 215 |
+
| 0.2896 | 160.0 | 2360 | 0.7926 | 0.7273 |
|
| 216 |
+
| 0.3194 | 160.95 | 2374 | 0.8023 | 0.7203 |
|
| 217 |
+
| 0.2918 | 161.97 | 2389 | 0.7933 | 0.7273 |
|
| 218 |
+
| 0.2992 | 162.98 | 2404 | 0.7829 | 0.7413 |
|
| 219 |
+
| 0.3 | 164.0 | 2419 | 0.7946 | 0.7203 |
|
| 220 |
+
| 0.322 | 164.95 | 2433 | 0.7969 | 0.7273 |
|
| 221 |
+
| 0.2994 | 165.97 | 2448 | 0.8076 | 0.7238 |
|
| 222 |
+
| 0.2849 | 166.98 | 2463 | 0.7951 | 0.7308 |
|
| 223 |
+
| 0.2745 | 168.0 | 2478 | 0.7892 | 0.7343 |
|
| 224 |
+
| 0.2974 | 168.95 | 2492 | 0.7916 | 0.7308 |
|
| 225 |
+
| 0.2656 | 169.97 | 2507 | 0.7995 | 0.7343 |
|
| 226 |
+
| 0.295 | 170.98 | 2522 | 0.8026 | 0.7238 |
|
| 227 |
+
| 0.2791 | 172.0 | 2537 | 0.7973 | 0.7343 |
|
| 228 |
+
| 0.2836 | 172.95 | 2551 | 0.8023 | 0.7308 |
|
| 229 |
+
| 0.2806 | 173.97 | 2566 | 0.8013 | 0.7308 |
|
| 230 |
+
| 0.2661 | 174.98 | 2581 | 0.7965 | 0.7308 |
|
| 231 |
+
| 0.2695 | 176.0 | 2596 | 0.8063 | 0.7273 |
|
| 232 |
+
| 0.286 | 176.95 | 2610 | 0.7963 | 0.7238 |
|
| 233 |
+
| 0.2743 | 177.97 | 2625 | 0.7929 | 0.7413 |
|
| 234 |
+
| 0.2775 | 178.98 | 2640 | 0.7855 | 0.7448 |
|
| 235 |
+
| 0.2878 | 180.0 | 2655 | 0.7894 | 0.7378 |
|
| 236 |
+
| 0.2757 | 180.95 | 2669 | 0.8013 | 0.7273 |
|
| 237 |
+
| 0.3067 | 181.97 | 2684 | 0.8015 | 0.7308 |
|
| 238 |
+
| 0.2412 | 182.98 | 2699 | 0.7975 | 0.7273 |
|
| 239 |
+
| 0.2686 | 184.0 | 2714 | 0.8037 | 0.7238 |
|
| 240 |
+
| 0.3176 | 184.95 | 2728 | 0.8017 | 0.7308 |
|
| 241 |
+
| 0.269 | 185.97 | 2743 | 0.8060 | 0.7308 |
|
| 242 |
+
| 0.2668 | 186.98 | 2758 | 0.8030 | 0.7308 |
|
| 243 |
+
| 0.2761 | 188.0 | 2773 | 0.8007 | 0.7413 |
|
| 244 |
+
| 0.2731 | 188.95 | 2787 | 0.8039 | 0.7343 |
|
| 245 |
+
| 0.2678 | 189.97 | 2802 | 0.8054 | 0.7308 |
|
| 246 |
+
| 0.2686 | 190.98 | 2817 | 0.8042 | 0.7343 |
|
| 247 |
+
| 0.2721 | 192.0 | 2832 | 0.8063 | 0.7308 |
|
| 248 |
+
| 0.3111 | 192.95 | 2846 | 0.8075 | 0.7378 |
|
| 249 |
+
| 0.2525 | 193.97 | 2861 | 0.8114 | 0.7273 |
|
| 250 |
+
| 0.2589 | 194.98 | 2876 | 0.8071 | 0.7413 |
|
| 251 |
+
| 0.2597 | 196.0 | 2891 | 0.8161 | 0.7378 |
|
| 252 |
+
| 0.3034 | 196.95 | 2905 | 0.8163 | 0.7378 |
|
| 253 |
+
| 0.2721 | 197.97 | 2920 | 0.8151 | 0.7308 |
|
| 254 |
+
| 0.2783 | 198.98 | 2935 | 0.8093 | 0.7448 |
|
| 255 |
+
| 0.259 | 200.0 | 2950 | 0.8026 | 0.7413 |
|
| 256 |
+
| 0.2896 | 200.95 | 2964 | 0.8070 | 0.7308 |
|
| 257 |
+
| 0.2584 | 201.97 | 2979 | 0.8113 | 0.7308 |
|
| 258 |
+
| 0.2672 | 202.98 | 2994 | 0.8096 | 0.7343 |
|
| 259 |
+
| 0.2622 | 204.0 | 3009 | 0.8060 | 0.7413 |
|
| 260 |
+
| 0.2677 | 204.95 | 3023 | 0.8084 | 0.7343 |
|
| 261 |
+
| 0.263 | 205.97 | 3038 | 0.8010 | 0.7378 |
|
| 262 |
+
| 0.2608 | 206.98 | 3053 | 0.7989 | 0.7448 |
|
| 263 |
+
| 0.2528 | 208.0 | 3068 | 0.7954 | 0.7448 |
|
| 264 |
+
| 0.2553 | 208.95 | 3082 | 0.7965 | 0.7413 |
|
| 265 |
+
| 0.2652 | 209.97 | 3097 | 0.7995 | 0.7413 |
|
| 266 |
+
| 0.246 | 210.98 | 3112 | 0.8026 | 0.7378 |
|
| 267 |
+
| 0.2665 | 212.0 | 3127 | 0.8049 | 0.7378 |
|
| 268 |
+
| 0.2731 | 212.95 | 3141 | 0.8052 | 0.7378 |
|
| 269 |
+
| 0.2539 | 213.97 | 3156 | 0.8089 | 0.7378 |
|
| 270 |
+
| 0.2376 | 214.98 | 3171 | 0.8116 | 0.7378 |
|
| 271 |
+
| 0.2667 | 216.0 | 3186 | 0.8099 | 0.7413 |
|
| 272 |
+
| 0.2768 | 216.95 | 3200 | 0.8079 | 0.7413 |
|
| 273 |
+
| 0.2551 | 217.97 | 3215 | 0.8065 | 0.7448 |
|
| 274 |
+
| 0.256 | 218.98 | 3230 | 0.8080 | 0.7413 |
|
| 275 |
+
| 0.2387 | 220.0 | 3245 | 0.8076 | 0.7448 |
|
| 276 |
+
| 0.2736 | 220.95 | 3259 | 0.8081 | 0.7448 |
|
| 277 |
+
| 0.2551 | 221.97 | 3274 | 0.8092 | 0.7448 |
|
| 278 |
+
| 0.2482 | 222.98 | 3289 | 0.8085 | 0.7448 |
|
| 279 |
+
| 0.2432 | 224.0 | 3304 | 0.8085 | 0.7448 |
|
| 280 |
+
| 0.2857 | 224.95 | 3318 | 0.8079 | 0.7448 |
|
| 281 |
+
| 0.2434 | 225.97 | 3333 | 0.8079 | 0.7448 |
|
| 282 |
+
| 0.2531 | 226.98 | 3348 | 0.8078 | 0.7448 |
|
| 283 |
+
| 0.222 | 228.0 | 3363 | 0.8093 | 0.7448 |
|
| 284 |
+
| 0.2651 | 228.95 | 3377 | 0.8105 | 0.7448 |
|
| 285 |
+
| 0.2886 | 229.97 | 3392 | 0.8116 | 0.7448 |
|
| 286 |
+
| 0.2534 | 230.98 | 3407 | 0.8116 | 0.7448 |
|
| 287 |
+
| 0.2483 | 232.0 | 3422 | 0.8116 | 0.7448 |
|
| 288 |
+
| 0.272 | 232.95 | 3436 | 0.8118 | 0.7448 |
|
| 289 |
+
| 0.2562 | 233.97 | 3451 | 0.8120 | 0.7448 |
|
| 290 |
+
| 0.2531 | 234.98 | 3466 | 0.8122 | 0.7448 |
|
| 291 |
+
| 0.2397 | 236.0 | 3481 | 0.8123 | 0.7448 |
|
| 292 |
+
| 0.2573 | 236.95 | 3495 | 0.8123 | 0.7448 |
|
| 293 |
+
| 0.2023 | 237.29 | 3500 | 0.8122 | 0.7448 |
|
| 294 |
+
|
| 295 |
+
|
| 296 |
+
### Framework versions
|
| 297 |
+
|
| 298 |
+
- Transformers 4.39.0.dev0
|
| 299 |
+
- Pytorch 2.2.1+cu121
|
| 300 |
+
- Datasets 2.17.1
|
| 301 |
+
- Tokenizers 0.15.2
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 237.29,
|
| 3 |
+
"eval_accuracy": 0.7447552447552448,
|
| 4 |
+
"eval_loss": 0.7855107188224792,
|
| 5 |
+
"eval_runtime": 4.8875,
|
| 6 |
+
"eval_samples_per_second": 58.516,
|
| 7 |
+
"eval_steps_per_second": 3.683,
|
| 8 |
+
"train_loss": 0.5242921064240592,
|
| 9 |
+
"train_runtime": 3651.9597,
|
| 10 |
+
"train_samples_per_second": 63.87,
|
| 11 |
+
"train_steps_per_second": 0.958
|
| 12 |
+
}
|
checkpoint-2640/config.json
ADDED
|
@@ -0,0 +1,124 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "anderloh/Hugginhface-master-wav2vec-pretreined-5-class-train-test",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"adapter_attn_dim": null,
|
| 5 |
+
"adapter_kernel_size": 3,
|
| 6 |
+
"adapter_stride": 2,
|
| 7 |
+
"add_adapter": false,
|
| 8 |
+
"apply_spec_augment": true,
|
| 9 |
+
"architectures": [
|
| 10 |
+
"Wav2Vec2ForSequenceClassification"
|
| 11 |
+
],
|
| 12 |
+
"attention_dropout": 0.0,
|
| 13 |
+
"bos_token_id": 1,
|
| 14 |
+
"classifier_proj_size": 128,
|
| 15 |
+
"codevector_dim": 128,
|
| 16 |
+
"contrastive_logits_temperature": 0.1,
|
| 17 |
+
"conv_bias": true,
|
| 18 |
+
"conv_dim": [
|
| 19 |
+
256,
|
| 20 |
+
256,
|
| 21 |
+
256,
|
| 22 |
+
256,
|
| 23 |
+
256,
|
| 24 |
+
256,
|
| 25 |
+
256
|
| 26 |
+
],
|
| 27 |
+
"conv_kernel": [
|
| 28 |
+
10,
|
| 29 |
+
3,
|
| 30 |
+
3,
|
| 31 |
+
3,
|
| 32 |
+
3,
|
| 33 |
+
2,
|
| 34 |
+
2
|
| 35 |
+
],
|
| 36 |
+
"conv_stride": [
|
| 37 |
+
5,
|
| 38 |
+
2,
|
| 39 |
+
2,
|
| 40 |
+
2,
|
| 41 |
+
2,
|
| 42 |
+
2,
|
| 43 |
+
2
|
| 44 |
+
],
|
| 45 |
+
"ctc_loss_reduction": "sum",
|
| 46 |
+
"ctc_zero_infinity": false,
|
| 47 |
+
"diversity_loss_weight": 0.1,
|
| 48 |
+
"do_stable_layer_norm": true,
|
| 49 |
+
"eos_token_id": 2,
|
| 50 |
+
"feat_extract_activation": "gelu",
|
| 51 |
+
"feat_extract_dropout": 0.0,
|
| 52 |
+
"feat_extract_norm": "layer",
|
| 53 |
+
"feat_proj_dropout": 0.0,
|
| 54 |
+
"feat_quantizer_dropout": 0.0,
|
| 55 |
+
"final_dropout": 0.0,
|
| 56 |
+
"finetuning_task": "audio-classification",
|
| 57 |
+
"hidden_act": "gelu",
|
| 58 |
+
"hidden_dropout": 0.0,
|
| 59 |
+
"hidden_dropout_prob": 0.0,
|
| 60 |
+
"hidden_size": 384,
|
| 61 |
+
"id2label": {
|
| 62 |
+
"0": "Helicopter",
|
| 63 |
+
"1": "Jet",
|
| 64 |
+
"2": "Racecar",
|
| 65 |
+
"3": "Rail",
|
| 66 |
+
"4": "Truck"
|
| 67 |
+
},
|
| 68 |
+
"initializer_range": 0.02,
|
| 69 |
+
"intermediate_size": 1536,
|
| 70 |
+
"label2id": {
|
| 71 |
+
"Helicopter": "0",
|
| 72 |
+
"Jet": "1",
|
| 73 |
+
"Racecar": "2",
|
| 74 |
+
"Rail": "3",
|
| 75 |
+
"Truck": "4"
|
| 76 |
+
},
|
| 77 |
+
"layer_norm_eps": 1e-05,
|
| 78 |
+
"layerdrop": 0.0,
|
| 79 |
+
"mask_feature_length": 10,
|
| 80 |
+
"mask_feature_min_masks": 0,
|
| 81 |
+
"mask_feature_prob": 0.0,
|
| 82 |
+
"mask_time_length": 10,
|
| 83 |
+
"mask_time_min_masks": 2,
|
| 84 |
+
"mask_time_prob": 0.65,
|
| 85 |
+
"model_type": "wav2vec2",
|
| 86 |
+
"num_adapter_layers": 3,
|
| 87 |
+
"num_attention_heads": 6,
|
| 88 |
+
"num_codevector_groups": 2,
|
| 89 |
+
"num_codevectors_per_group": 320,
|
| 90 |
+
"num_conv_pos_embedding_groups": 16,
|
| 91 |
+
"num_conv_pos_embeddings": 128,
|
| 92 |
+
"num_feat_extract_layers": 7,
|
| 93 |
+
"num_hidden_layers": 6,
|
| 94 |
+
"num_negatives": 100,
|
| 95 |
+
"output_hidden_size": 384,
|
| 96 |
+
"pad_token_id": 0,
|
| 97 |
+
"proj_codevector_dim": 128,
|
| 98 |
+
"tdnn_dilation": [
|
| 99 |
+
1,
|
| 100 |
+
2,
|
| 101 |
+
3,
|
| 102 |
+
1,
|
| 103 |
+
1
|
| 104 |
+
],
|
| 105 |
+
"tdnn_dim": [
|
| 106 |
+
512,
|
| 107 |
+
512,
|
| 108 |
+
512,
|
| 109 |
+
512,
|
| 110 |
+
1500
|
| 111 |
+
],
|
| 112 |
+
"tdnn_kernel": [
|
| 113 |
+
5,
|
| 114 |
+
3,
|
| 115 |
+
3,
|
| 116 |
+
1,
|
| 117 |
+
1
|
| 118 |
+
],
|
| 119 |
+
"torch_dtype": "float32",
|
| 120 |
+
"transformers_version": "4.39.0.dev0",
|
| 121 |
+
"use_weighted_layer_sum": false,
|
| 122 |
+
"vocab_size": 32,
|
| 123 |
+
"xvector_output_dim": 512
|
| 124 |
+
}
|
checkpoint-2640/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:51779c90f182773c0eccdebb31d61e1a81936b73024a5badb5719741abedadfe
|
| 3 |
+
size 52151348
|
checkpoint-2640/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6dfe3189b9404e4429422dc6d24f6e84c7a87369f53c917c0bb7caf39428465c
|
| 3 |
+
size 95909946
|
checkpoint-2640/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": 0.0,
|
| 7 |
+
"return_attention_mask": true,
|
| 8 |
+
"sampling_rate": 16000
|
| 9 |
+
}
|
checkpoint-2640/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8be732c71052de6c38eca0ad16a1f1458800d818c8bbfe8e3d49a037336aa529
|
| 3 |
+
size 14308
|
checkpoint-2640/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fbf1576243ff11e40b5d376b44f6b9492d3567a305c5801472b8ce907b2cdffa
|
| 3 |
+
size 1064
|
checkpoint-2640/trainer_state.json
ADDED
|
@@ -0,0 +1,2885 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.7447552447552448,
|
| 3 |
+
"best_model_checkpoint": "wav2vec2-5Class-train-test-finetune-Medium/checkpoint-2640",
|
| 4 |
+
"epoch": 178.98305084745763,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 2640,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.95,
|
| 13 |
+
"grad_norm": 0.8927629590034485,
|
| 14 |
+
"learning_rate": 1.2000000000000002e-06,
|
| 15 |
+
"loss": 1.6852,
|
| 16 |
+
"step": 14
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.95,
|
| 20 |
+
"eval_accuracy": 0.34265734265734266,
|
| 21 |
+
"eval_loss": 1.5986738204956055,
|
| 22 |
+
"eval_runtime": 4.6543,
|
| 23 |
+
"eval_samples_per_second": 61.448,
|
| 24 |
+
"eval_steps_per_second": 3.867,
|
| 25 |
+
"step": 14
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 1.97,
|
| 29 |
+
"grad_norm": 0.6501776576042175,
|
| 30 |
+
"learning_rate": 2.4857142857142858e-06,
|
| 31 |
+
"loss": 1.5721,
|
| 32 |
+
"step": 29
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.97,
|
| 36 |
+
"eval_accuracy": 0.34265734265734266,
|
| 37 |
+
"eval_loss": 1.5975812673568726,
|
| 38 |
+
"eval_runtime": 4.4155,
|
| 39 |
+
"eval_samples_per_second": 64.772,
|
| 40 |
+
"eval_steps_per_second": 4.077,
|
| 41 |
+
"step": 29
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 2.98,
|
| 45 |
+
"grad_norm": 0.6221457719802856,
|
| 46 |
+
"learning_rate": 3.771428571428572e-06,
|
| 47 |
+
"loss": 1.5696,
|
| 48 |
+
"step": 44
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 2.98,
|
| 52 |
+
"eval_accuracy": 0.34265734265734266,
|
| 53 |
+
"eval_loss": 1.5957201719284058,
|
| 54 |
+
"eval_runtime": 4.5928,
|
| 55 |
+
"eval_samples_per_second": 62.272,
|
| 56 |
+
"eval_steps_per_second": 3.919,
|
| 57 |
+
"step": 44
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 4.0,
|
| 61 |
+
"grad_norm": 1.563844919204712,
|
| 62 |
+
"learning_rate": 5.057142857142857e-06,
|
| 63 |
+
"loss": 1.5671,
|
| 64 |
+
"step": 59
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 4.0,
|
| 68 |
+
"eval_accuracy": 0.3356643356643357,
|
| 69 |
+
"eval_loss": 1.5932137966156006,
|
| 70 |
+
"eval_runtime": 4.5483,
|
| 71 |
+
"eval_samples_per_second": 62.881,
|
| 72 |
+
"eval_steps_per_second": 3.958,
|
| 73 |
+
"step": 59
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 4.95,
|
| 77 |
+
"grad_norm": 0.7324579358100891,
|
| 78 |
+
"learning_rate": 6.257142857142858e-06,
|
| 79 |
+
"loss": 1.6757,
|
| 80 |
+
"step": 73
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 4.95,
|
| 84 |
+
"eval_accuracy": 0.32517482517482516,
|
| 85 |
+
"eval_loss": 1.5901715755462646,
|
| 86 |
+
"eval_runtime": 4.9986,
|
| 87 |
+
"eval_samples_per_second": 57.216,
|
| 88 |
+
"eval_steps_per_second": 3.601,
|
| 89 |
+
"step": 73
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 5.97,
|
| 93 |
+
"grad_norm": 0.878238320350647,
|
| 94 |
+
"learning_rate": 7.542857142857144e-06,
|
| 95 |
+
"loss": 1.5595,
|
| 96 |
+
"step": 88
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 5.97,
|
| 100 |
+
"eval_accuracy": 0.32167832167832167,
|
| 101 |
+
"eval_loss": 1.5863642692565918,
|
| 102 |
+
"eval_runtime": 4.9711,
|
| 103 |
+
"eval_samples_per_second": 57.532,
|
| 104 |
+
"eval_steps_per_second": 3.621,
|
| 105 |
+
"step": 88
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 6.98,
|
| 109 |
+
"grad_norm": 1.4210667610168457,
|
| 110 |
+
"learning_rate": 8.828571428571429e-06,
|
| 111 |
+
"loss": 1.5536,
|
| 112 |
+
"step": 103
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 6.98,
|
| 116 |
+
"eval_accuracy": 0.3181818181818182,
|
| 117 |
+
"eval_loss": 1.5817956924438477,
|
| 118 |
+
"eval_runtime": 5.5676,
|
| 119 |
+
"eval_samples_per_second": 51.368,
|
| 120 |
+
"eval_steps_per_second": 3.233,
|
| 121 |
+
"step": 103
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 8.0,
|
| 125 |
+
"grad_norm": 0.49261531233787537,
|
| 126 |
+
"learning_rate": 1.0114285714285715e-05,
|
| 127 |
+
"loss": 1.5484,
|
| 128 |
+
"step": 118
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 8.0,
|
| 132 |
+
"eval_accuracy": 0.3111888111888112,
|
| 133 |
+
"eval_loss": 1.576475739479065,
|
| 134 |
+
"eval_runtime": 5.5008,
|
| 135 |
+
"eval_samples_per_second": 51.992,
|
| 136 |
+
"eval_steps_per_second": 3.272,
|
| 137 |
+
"step": 118
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 8.95,
|
| 141 |
+
"grad_norm": 0.9810589551925659,
|
| 142 |
+
"learning_rate": 1.1314285714285715e-05,
|
| 143 |
+
"loss": 1.6506,
|
| 144 |
+
"step": 132
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 8.95,
|
| 148 |
+
"eval_accuracy": 0.3076923076923077,
|
| 149 |
+
"eval_loss": 1.570804238319397,
|
| 150 |
+
"eval_runtime": 4.0121,
|
| 151 |
+
"eval_samples_per_second": 71.285,
|
| 152 |
+
"eval_steps_per_second": 4.486,
|
| 153 |
+
"step": 132
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 9.97,
|
| 157 |
+
"grad_norm": 0.6308433413505554,
|
| 158 |
+
"learning_rate": 1.26e-05,
|
| 159 |
+
"loss": 1.5317,
|
| 160 |
+
"step": 147
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 9.97,
|
| 164 |
+
"eval_accuracy": 0.3006993006993007,
|
| 165 |
+
"eval_loss": 1.5640877485275269,
|
| 166 |
+
"eval_runtime": 5.5443,
|
| 167 |
+
"eval_samples_per_second": 51.585,
|
| 168 |
+
"eval_steps_per_second": 3.247,
|
| 169 |
+
"step": 147
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 10.98,
|
| 173 |
+
"grad_norm": 1.381785273551941,
|
| 174 |
+
"learning_rate": 1.3885714285714286e-05,
|
| 175 |
+
"loss": 1.5226,
|
| 176 |
+
"step": 162
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 10.98,
|
| 180 |
+
"eval_accuracy": 0.2867132867132867,
|
| 181 |
+
"eval_loss": 1.5564289093017578,
|
| 182 |
+
"eval_runtime": 5.0151,
|
| 183 |
+
"eval_samples_per_second": 57.027,
|
| 184 |
+
"eval_steps_per_second": 3.589,
|
| 185 |
+
"step": 162
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 12.0,
|
| 189 |
+
"grad_norm": 0.552884578704834,
|
| 190 |
+
"learning_rate": 1.517142857142857e-05,
|
| 191 |
+
"loss": 1.5116,
|
| 192 |
+
"step": 177
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 12.0,
|
| 196 |
+
"eval_accuracy": 0.2692307692307692,
|
| 197 |
+
"eval_loss": 1.5478310585021973,
|
| 198 |
+
"eval_runtime": 5.4183,
|
| 199 |
+
"eval_samples_per_second": 52.784,
|
| 200 |
+
"eval_steps_per_second": 3.322,
|
| 201 |
+
"step": 177
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 12.95,
|
| 205 |
+
"grad_norm": 0.8657445907592773,
|
| 206 |
+
"learning_rate": 1.6371428571428572e-05,
|
| 207 |
+
"loss": 1.6046,
|
| 208 |
+
"step": 191
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 12.95,
|
| 212 |
+
"eval_accuracy": 0.26223776223776224,
|
| 213 |
+
"eval_loss": 1.5384888648986816,
|
| 214 |
+
"eval_runtime": 5.1185,
|
| 215 |
+
"eval_samples_per_second": 55.875,
|
| 216 |
+
"eval_steps_per_second": 3.517,
|
| 217 |
+
"step": 191
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 13.97,
|
| 221 |
+
"grad_norm": 0.8375154733657837,
|
| 222 |
+
"learning_rate": 1.7657142857142857e-05,
|
| 223 |
+
"loss": 1.4822,
|
| 224 |
+
"step": 206
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 13.97,
|
| 228 |
+
"eval_accuracy": 0.25524475524475526,
|
| 229 |
+
"eval_loss": 1.5284570455551147,
|
| 230 |
+
"eval_runtime": 5.5939,
|
| 231 |
+
"eval_samples_per_second": 51.127,
|
| 232 |
+
"eval_steps_per_second": 3.218,
|
| 233 |
+
"step": 206
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 14.98,
|
| 237 |
+
"grad_norm": 0.7776028513908386,
|
| 238 |
+
"learning_rate": 1.8942857142857145e-05,
|
| 239 |
+
"loss": 1.4614,
|
| 240 |
+
"step": 221
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 14.98,
|
| 244 |
+
"eval_accuracy": 0.25524475524475526,
|
| 245 |
+
"eval_loss": 1.5183566808700562,
|
| 246 |
+
"eval_runtime": 5.2849,
|
| 247 |
+
"eval_samples_per_second": 54.116,
|
| 248 |
+
"eval_steps_per_second": 3.406,
|
| 249 |
+
"step": 221
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 16.0,
|
| 253 |
+
"grad_norm": 0.3529152572154999,
|
| 254 |
+
"learning_rate": 2.022857142857143e-05,
|
| 255 |
+
"loss": 1.4396,
|
| 256 |
+
"step": 236
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 16.0,
|
| 260 |
+
"eval_accuracy": 0.2517482517482518,
|
| 261 |
+
"eval_loss": 1.510148048400879,
|
| 262 |
+
"eval_runtime": 6.3848,
|
| 263 |
+
"eval_samples_per_second": 44.794,
|
| 264 |
+
"eval_steps_per_second": 2.819,
|
| 265 |
+
"step": 236
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 16.95,
|
| 269 |
+
"grad_norm": 0.5623555779457092,
|
| 270 |
+
"learning_rate": 2.1428571428571428e-05,
|
| 271 |
+
"loss": 1.5047,
|
| 272 |
+
"step": 250
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 16.95,
|
| 276 |
+
"eval_accuracy": 0.24475524475524477,
|
| 277 |
+
"eval_loss": 1.510407567024231,
|
| 278 |
+
"eval_runtime": 5.762,
|
| 279 |
+
"eval_samples_per_second": 49.635,
|
| 280 |
+
"eval_steps_per_second": 3.124,
|
| 281 |
+
"step": 250
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 17.97,
|
| 285 |
+
"grad_norm": 0.889561116695404,
|
| 286 |
+
"learning_rate": 2.2714285714285713e-05,
|
| 287 |
+
"loss": 1.3741,
|
| 288 |
+
"step": 265
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 17.97,
|
| 292 |
+
"eval_accuracy": 0.2517482517482518,
|
| 293 |
+
"eval_loss": 1.524996280670166,
|
| 294 |
+
"eval_runtime": 5.3825,
|
| 295 |
+
"eval_samples_per_second": 53.135,
|
| 296 |
+
"eval_steps_per_second": 3.344,
|
| 297 |
+
"step": 265
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 18.98,
|
| 301 |
+
"grad_norm": 0.837506890296936,
|
| 302 |
+
"learning_rate": 2.4e-05,
|
| 303 |
+
"loss": 1.3512,
|
| 304 |
+
"step": 280
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 18.98,
|
| 308 |
+
"eval_accuracy": 0.26573426573426573,
|
| 309 |
+
"eval_loss": 1.5328779220581055,
|
| 310 |
+
"eval_runtime": 4.8496,
|
| 311 |
+
"eval_samples_per_second": 58.974,
|
| 312 |
+
"eval_steps_per_second": 3.712,
|
| 313 |
+
"step": 280
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 20.0,
|
| 317 |
+
"grad_norm": 0.6774041056632996,
|
| 318 |
+
"learning_rate": 2.5285714285714285e-05,
|
| 319 |
+
"loss": 1.3286,
|
| 320 |
+
"step": 295
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 20.0,
|
| 324 |
+
"eval_accuracy": 0.32517482517482516,
|
| 325 |
+
"eval_loss": 1.509504795074463,
|
| 326 |
+
"eval_runtime": 4.7142,
|
| 327 |
+
"eval_samples_per_second": 60.667,
|
| 328 |
+
"eval_steps_per_second": 3.818,
|
| 329 |
+
"step": 295
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 20.95,
|
| 333 |
+
"grad_norm": 0.5629450678825378,
|
| 334 |
+
"learning_rate": 2.6485714285714287e-05,
|
| 335 |
+
"loss": 1.3967,
|
| 336 |
+
"step": 309
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 20.95,
|
| 340 |
+
"eval_accuracy": 0.34965034965034963,
|
| 341 |
+
"eval_loss": 1.482858419418335,
|
| 342 |
+
"eval_runtime": 5.0574,
|
| 343 |
+
"eval_samples_per_second": 56.551,
|
| 344 |
+
"eval_steps_per_second": 3.559,
|
| 345 |
+
"step": 309
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 21.97,
|
| 349 |
+
"grad_norm": 0.5265232920646667,
|
| 350 |
+
"learning_rate": 2.7771428571428572e-05,
|
| 351 |
+
"loss": 1.2779,
|
| 352 |
+
"step": 324
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 21.97,
|
| 356 |
+
"eval_accuracy": 0.38461538461538464,
|
| 357 |
+
"eval_loss": 1.4598273038864136,
|
| 358 |
+
"eval_runtime": 5.2206,
|
| 359 |
+
"eval_samples_per_second": 54.783,
|
| 360 |
+
"eval_steps_per_second": 3.448,
|
| 361 |
+
"step": 324
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 22.98,
|
| 365 |
+
"grad_norm": 0.4762378931045532,
|
| 366 |
+
"learning_rate": 2.9057142857142856e-05,
|
| 367 |
+
"loss": 1.2449,
|
| 368 |
+
"step": 339
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 22.98,
|
| 372 |
+
"eval_accuracy": 0.4160839160839161,
|
| 373 |
+
"eval_loss": 1.4284018278121948,
|
| 374 |
+
"eval_runtime": 5.3164,
|
| 375 |
+
"eval_samples_per_second": 53.796,
|
| 376 |
+
"eval_steps_per_second": 3.386,
|
| 377 |
+
"step": 339
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 24.0,
|
| 381 |
+
"grad_norm": 0.598441481590271,
|
| 382 |
+
"learning_rate": 2.996190476190476e-05,
|
| 383 |
+
"loss": 1.2118,
|
| 384 |
+
"step": 354
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 24.0,
|
| 388 |
+
"eval_accuracy": 0.4230769230769231,
|
| 389 |
+
"eval_loss": 1.4161577224731445,
|
| 390 |
+
"eval_runtime": 4.9587,
|
| 391 |
+
"eval_samples_per_second": 57.677,
|
| 392 |
+
"eval_steps_per_second": 3.63,
|
| 393 |
+
"step": 354
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 24.95,
|
| 397 |
+
"grad_norm": 1.1748360395431519,
|
| 398 |
+
"learning_rate": 2.982857142857143e-05,
|
| 399 |
+
"loss": 1.2521,
|
| 400 |
+
"step": 368
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 24.95,
|
| 404 |
+
"eval_accuracy": 0.44755244755244755,
|
| 405 |
+
"eval_loss": 1.3797944784164429,
|
| 406 |
+
"eval_runtime": 7.057,
|
| 407 |
+
"eval_samples_per_second": 40.527,
|
| 408 |
+
"eval_steps_per_second": 2.551,
|
| 409 |
+
"step": 368
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 25.97,
|
| 413 |
+
"grad_norm": 0.7063133716583252,
|
| 414 |
+
"learning_rate": 2.9685714285714284e-05,
|
| 415 |
+
"loss": 1.1183,
|
| 416 |
+
"step": 383
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 25.97,
|
| 420 |
+
"eval_accuracy": 0.479020979020979,
|
| 421 |
+
"eval_loss": 1.324568271636963,
|
| 422 |
+
"eval_runtime": 5.5598,
|
| 423 |
+
"eval_samples_per_second": 51.44,
|
| 424 |
+
"eval_steps_per_second": 3.238,
|
| 425 |
+
"step": 383
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 26.98,
|
| 429 |
+
"grad_norm": 0.91399085521698,
|
| 430 |
+
"learning_rate": 2.9542857142857142e-05,
|
| 431 |
+
"loss": 1.0778,
|
| 432 |
+
"step": 398
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 26.98,
|
| 436 |
+
"eval_accuracy": 0.493006993006993,
|
| 437 |
+
"eval_loss": 1.276139259338379,
|
| 438 |
+
"eval_runtime": 4.8059,
|
| 439 |
+
"eval_samples_per_second": 59.51,
|
| 440 |
+
"eval_steps_per_second": 3.745,
|
| 441 |
+
"step": 398
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 28.0,
|
| 445 |
+
"grad_norm": 1.1466203927993774,
|
| 446 |
+
"learning_rate": 2.94e-05,
|
| 447 |
+
"loss": 1.0306,
|
| 448 |
+
"step": 413
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 28.0,
|
| 452 |
+
"eval_accuracy": 0.5104895104895105,
|
| 453 |
+
"eval_loss": 1.2299922704696655,
|
| 454 |
+
"eval_runtime": 6.0199,
|
| 455 |
+
"eval_samples_per_second": 47.509,
|
| 456 |
+
"eval_steps_per_second": 2.99,
|
| 457 |
+
"step": 413
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 28.95,
|
| 461 |
+
"grad_norm": 1.1782015562057495,
|
| 462 |
+
"learning_rate": 2.9266666666666665e-05,
|
| 463 |
+
"loss": 1.0808,
|
| 464 |
+
"step": 427
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 28.95,
|
| 468 |
+
"eval_accuracy": 0.513986013986014,
|
| 469 |
+
"eval_loss": 1.2064085006713867,
|
| 470 |
+
"eval_runtime": 6.0946,
|
| 471 |
+
"eval_samples_per_second": 46.927,
|
| 472 |
+
"eval_steps_per_second": 2.953,
|
| 473 |
+
"step": 427
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 29.97,
|
| 477 |
+
"grad_norm": 0.9410634636878967,
|
| 478 |
+
"learning_rate": 2.9123809523809523e-05,
|
| 479 |
+
"loss": 0.9767,
|
| 480 |
+
"step": 442
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 29.97,
|
| 484 |
+
"eval_accuracy": 0.5524475524475524,
|
| 485 |
+
"eval_loss": 1.1669567823410034,
|
| 486 |
+
"eval_runtime": 4.8236,
|
| 487 |
+
"eval_samples_per_second": 59.292,
|
| 488 |
+
"eval_steps_per_second": 3.732,
|
| 489 |
+
"step": 442
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 30.98,
|
| 493 |
+
"grad_norm": 1.2911593914031982,
|
| 494 |
+
"learning_rate": 2.898095238095238e-05,
|
| 495 |
+
"loss": 0.9589,
|
| 496 |
+
"step": 457
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 30.98,
|
| 500 |
+
"eval_accuracy": 0.5734265734265734,
|
| 501 |
+
"eval_loss": 1.126379132270813,
|
| 502 |
+
"eval_runtime": 5.0702,
|
| 503 |
+
"eval_samples_per_second": 56.408,
|
| 504 |
+
"eval_steps_per_second": 3.55,
|
| 505 |
+
"step": 457
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 32.0,
|
| 509 |
+
"grad_norm": 1.2354660034179688,
|
| 510 |
+
"learning_rate": 2.883809523809524e-05,
|
| 511 |
+
"loss": 0.9193,
|
| 512 |
+
"step": 472
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 32.0,
|
| 516 |
+
"eval_accuracy": 0.5874125874125874,
|
| 517 |
+
"eval_loss": 1.1012728214263916,
|
| 518 |
+
"eval_runtime": 5.9275,
|
| 519 |
+
"eval_samples_per_second": 48.249,
|
| 520 |
+
"eval_steps_per_second": 3.037,
|
| 521 |
+
"step": 472
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 32.95,
|
| 525 |
+
"grad_norm": 0.8961493968963623,
|
| 526 |
+
"learning_rate": 2.8704761904761905e-05,
|
| 527 |
+
"loss": 0.9462,
|
| 528 |
+
"step": 486
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 32.95,
|
| 532 |
+
"eval_accuracy": 0.6048951048951049,
|
| 533 |
+
"eval_loss": 1.0736197233200073,
|
| 534 |
+
"eval_runtime": 5.7691,
|
| 535 |
+
"eval_samples_per_second": 49.575,
|
| 536 |
+
"eval_steps_per_second": 3.12,
|
| 537 |
+
"step": 486
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 33.97,
|
| 541 |
+
"grad_norm": 0.9503061175346375,
|
| 542 |
+
"learning_rate": 2.8561904761904763e-05,
|
| 543 |
+
"loss": 0.85,
|
| 544 |
+
"step": 501
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 33.97,
|
| 548 |
+
"eval_accuracy": 0.6048951048951049,
|
| 549 |
+
"eval_loss": 1.0628284215927124,
|
| 550 |
+
"eval_runtime": 6.1054,
|
| 551 |
+
"eval_samples_per_second": 46.844,
|
| 552 |
+
"eval_steps_per_second": 2.948,
|
| 553 |
+
"step": 501
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 34.98,
|
| 557 |
+
"grad_norm": 1.0706520080566406,
|
| 558 |
+
"learning_rate": 2.841904761904762e-05,
|
| 559 |
+
"loss": 0.8294,
|
| 560 |
+
"step": 516
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 34.98,
|
| 564 |
+
"eval_accuracy": 0.6188811188811189,
|
| 565 |
+
"eval_loss": 1.0472813844680786,
|
| 566 |
+
"eval_runtime": 4.7734,
|
| 567 |
+
"eval_samples_per_second": 59.916,
|
| 568 |
+
"eval_steps_per_second": 3.771,
|
| 569 |
+
"step": 516
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 36.0,
|
| 573 |
+
"grad_norm": 0.9784532189369202,
|
| 574 |
+
"learning_rate": 2.827619047619048e-05,
|
| 575 |
+
"loss": 0.8025,
|
| 576 |
+
"step": 531
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 36.0,
|
| 580 |
+
"eval_accuracy": 0.6328671328671329,
|
| 581 |
+
"eval_loss": 1.0030184984207153,
|
| 582 |
+
"eval_runtime": 5.5291,
|
| 583 |
+
"eval_samples_per_second": 51.726,
|
| 584 |
+
"eval_steps_per_second": 3.255,
|
| 585 |
+
"step": 531
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 36.95,
|
| 589 |
+
"grad_norm": 0.8138810396194458,
|
| 590 |
+
"learning_rate": 2.8142857142857145e-05,
|
| 591 |
+
"loss": 0.8206,
|
| 592 |
+
"step": 545
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 36.95,
|
| 596 |
+
"eval_accuracy": 0.6398601398601399,
|
| 597 |
+
"eval_loss": 0.9963561296463013,
|
| 598 |
+
"eval_runtime": 5.8221,
|
| 599 |
+
"eval_samples_per_second": 49.123,
|
| 600 |
+
"eval_steps_per_second": 3.092,
|
| 601 |
+
"step": 545
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 37.97,
|
| 605 |
+
"grad_norm": 0.8114917278289795,
|
| 606 |
+
"learning_rate": 2.8e-05,
|
| 607 |
+
"loss": 0.7541,
|
| 608 |
+
"step": 560
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 37.97,
|
| 612 |
+
"eval_accuracy": 0.6608391608391608,
|
| 613 |
+
"eval_loss": 0.9604987502098083,
|
| 614 |
+
"eval_runtime": 4.9473,
|
| 615 |
+
"eval_samples_per_second": 57.809,
|
| 616 |
+
"eval_steps_per_second": 3.638,
|
| 617 |
+
"step": 560
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 38.98,
|
| 621 |
+
"grad_norm": 0.9221667051315308,
|
| 622 |
+
"learning_rate": 2.7857142857142858e-05,
|
| 623 |
+
"loss": 0.7413,
|
| 624 |
+
"step": 575
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 38.98,
|
| 628 |
+
"eval_accuracy": 0.6643356643356644,
|
| 629 |
+
"eval_loss": 0.946722686290741,
|
| 630 |
+
"eval_runtime": 4.7955,
|
| 631 |
+
"eval_samples_per_second": 59.64,
|
| 632 |
+
"eval_steps_per_second": 3.754,
|
| 633 |
+
"step": 575
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 40.0,
|
| 637 |
+
"grad_norm": 1.2665307521820068,
|
| 638 |
+
"learning_rate": 2.7714285714285716e-05,
|
| 639 |
+
"loss": 0.709,
|
| 640 |
+
"step": 590
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 40.0,
|
| 644 |
+
"eval_accuracy": 0.6678321678321678,
|
| 645 |
+
"eval_loss": 0.9348079562187195,
|
| 646 |
+
"eval_runtime": 5.3252,
|
| 647 |
+
"eval_samples_per_second": 53.707,
|
| 648 |
+
"eval_steps_per_second": 3.38,
|
| 649 |
+
"step": 590
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 40.95,
|
| 653 |
+
"grad_norm": 1.0893254280090332,
|
| 654 |
+
"learning_rate": 2.758095238095238e-05,
|
| 655 |
+
"loss": 0.7817,
|
| 656 |
+
"step": 604
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 40.95,
|
| 660 |
+
"eval_accuracy": 0.6678321678321678,
|
| 661 |
+
"eval_loss": 0.9366168975830078,
|
| 662 |
+
"eval_runtime": 6.4862,
|
| 663 |
+
"eval_samples_per_second": 44.094,
|
| 664 |
+
"eval_steps_per_second": 2.775,
|
| 665 |
+
"step": 604
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 41.97,
|
| 669 |
+
"grad_norm": 0.7136903405189514,
|
| 670 |
+
"learning_rate": 2.743809523809524e-05,
|
| 671 |
+
"loss": 0.7034,
|
| 672 |
+
"step": 619
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 41.97,
|
| 676 |
+
"eval_accuracy": 0.6818181818181818,
|
| 677 |
+
"eval_loss": 0.9109411835670471,
|
| 678 |
+
"eval_runtime": 6.1724,
|
| 679 |
+
"eval_samples_per_second": 46.335,
|
| 680 |
+
"eval_steps_per_second": 2.916,
|
| 681 |
+
"step": 619
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 42.98,
|
| 685 |
+
"grad_norm": 1.142288327217102,
|
| 686 |
+
"learning_rate": 2.7295238095238097e-05,
|
| 687 |
+
"loss": 0.6856,
|
| 688 |
+
"step": 634
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 42.98,
|
| 692 |
+
"eval_accuracy": 0.6573426573426573,
|
| 693 |
+
"eval_loss": 0.9276965856552124,
|
| 694 |
+
"eval_runtime": 4.6805,
|
| 695 |
+
"eval_samples_per_second": 61.105,
|
| 696 |
+
"eval_steps_per_second": 3.846,
|
| 697 |
+
"step": 634
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 44.0,
|
| 701 |
+
"grad_norm": 1.3661341667175293,
|
| 702 |
+
"learning_rate": 2.7152380952380952e-05,
|
| 703 |
+
"loss": 0.6625,
|
| 704 |
+
"step": 649
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 44.0,
|
| 708 |
+
"eval_accuracy": 0.6783216783216783,
|
| 709 |
+
"eval_loss": 0.8979520201683044,
|
| 710 |
+
"eval_runtime": 5.8084,
|
| 711 |
+
"eval_samples_per_second": 49.239,
|
| 712 |
+
"eval_steps_per_second": 3.099,
|
| 713 |
+
"step": 649
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 44.95,
|
| 717 |
+
"grad_norm": 0.7071289420127869,
|
| 718 |
+
"learning_rate": 2.701904761904762e-05,
|
| 719 |
+
"loss": 0.7207,
|
| 720 |
+
"step": 663
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 44.95,
|
| 724 |
+
"eval_accuracy": 0.6713286713286714,
|
| 725 |
+
"eval_loss": 0.9049975275993347,
|
| 726 |
+
"eval_runtime": 5.5071,
|
| 727 |
+
"eval_samples_per_second": 51.933,
|
| 728 |
+
"eval_steps_per_second": 3.268,
|
| 729 |
+
"step": 663
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 45.97,
|
| 733 |
+
"grad_norm": 0.9930222034454346,
|
| 734 |
+
"learning_rate": 2.6876190476190476e-05,
|
| 735 |
+
"loss": 0.6684,
|
| 736 |
+
"step": 678
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 45.97,
|
| 740 |
+
"eval_accuracy": 0.6748251748251748,
|
| 741 |
+
"eval_loss": 0.8973459005355835,
|
| 742 |
+
"eval_runtime": 4.8744,
|
| 743 |
+
"eval_samples_per_second": 58.674,
|
| 744 |
+
"eval_steps_per_second": 3.693,
|
| 745 |
+
"step": 678
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 46.98,
|
| 749 |
+
"grad_norm": 1.2905802726745605,
|
| 750 |
+
"learning_rate": 2.6733333333333334e-05,
|
| 751 |
+
"loss": 0.6651,
|
| 752 |
+
"step": 693
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 46.98,
|
| 756 |
+
"eval_accuracy": 0.6783216783216783,
|
| 757 |
+
"eval_loss": 0.8934686779975891,
|
| 758 |
+
"eval_runtime": 4.9699,
|
| 759 |
+
"eval_samples_per_second": 57.546,
|
| 760 |
+
"eval_steps_per_second": 3.622,
|
| 761 |
+
"step": 693
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 48.0,
|
| 765 |
+
"grad_norm": 1.5243291854858398,
|
| 766 |
+
"learning_rate": 2.6590476190476192e-05,
|
| 767 |
+
"loss": 0.6451,
|
| 768 |
+
"step": 708
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 48.0,
|
| 772 |
+
"eval_accuracy": 0.6748251748251748,
|
| 773 |
+
"eval_loss": 0.8944538831710815,
|
| 774 |
+
"eval_runtime": 4.9181,
|
| 775 |
+
"eval_samples_per_second": 58.153,
|
| 776 |
+
"eval_steps_per_second": 3.66,
|
| 777 |
+
"step": 708
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 48.95,
|
| 781 |
+
"grad_norm": 1.4421076774597168,
|
| 782 |
+
"learning_rate": 2.6457142857142857e-05,
|
| 783 |
+
"loss": 0.6774,
|
| 784 |
+
"step": 722
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 48.95,
|
| 788 |
+
"eval_accuracy": 0.6818181818181818,
|
| 789 |
+
"eval_loss": 0.8879284858703613,
|
| 790 |
+
"eval_runtime": 5.1997,
|
| 791 |
+
"eval_samples_per_second": 55.003,
|
| 792 |
+
"eval_steps_per_second": 3.462,
|
| 793 |
+
"step": 722
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 49.97,
|
| 797 |
+
"grad_norm": 1.3786752223968506,
|
| 798 |
+
"learning_rate": 2.6314285714285715e-05,
|
| 799 |
+
"loss": 0.6308,
|
| 800 |
+
"step": 737
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 49.97,
|
| 804 |
+
"eval_accuracy": 0.6818181818181818,
|
| 805 |
+
"eval_loss": 0.8882840275764465,
|
| 806 |
+
"eval_runtime": 5.1707,
|
| 807 |
+
"eval_samples_per_second": 55.312,
|
| 808 |
+
"eval_steps_per_second": 3.481,
|
| 809 |
+
"step": 737
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 50.98,
|
| 813 |
+
"grad_norm": 1.9291437864303589,
|
| 814 |
+
"learning_rate": 2.617142857142857e-05,
|
| 815 |
+
"loss": 0.6199,
|
| 816 |
+
"step": 752
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 50.98,
|
| 820 |
+
"eval_accuracy": 0.6818181818181818,
|
| 821 |
+
"eval_loss": 0.8825680017471313,
|
| 822 |
+
"eval_runtime": 4.9861,
|
| 823 |
+
"eval_samples_per_second": 57.359,
|
| 824 |
+
"eval_steps_per_second": 3.61,
|
| 825 |
+
"step": 752
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 52.0,
|
| 829 |
+
"grad_norm": 1.2054837942123413,
|
| 830 |
+
"learning_rate": 2.6028571428571428e-05,
|
| 831 |
+
"loss": 0.6379,
|
| 832 |
+
"step": 767
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 52.0,
|
| 836 |
+
"eval_accuracy": 0.6923076923076923,
|
| 837 |
+
"eval_loss": 0.8582118153572083,
|
| 838 |
+
"eval_runtime": 5.287,
|
| 839 |
+
"eval_samples_per_second": 54.095,
|
| 840 |
+
"eval_steps_per_second": 3.405,
|
| 841 |
+
"step": 767
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 52.95,
|
| 845 |
+
"grad_norm": 1.3063265085220337,
|
| 846 |
+
"learning_rate": 2.5895238095238094e-05,
|
| 847 |
+
"loss": 0.6588,
|
| 848 |
+
"step": 781
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 52.95,
|
| 852 |
+
"eval_accuracy": 0.6818181818181818,
|
| 853 |
+
"eval_loss": 0.8825483918190002,
|
| 854 |
+
"eval_runtime": 4.6024,
|
| 855 |
+
"eval_samples_per_second": 62.141,
|
| 856 |
+
"eval_steps_per_second": 3.911,
|
| 857 |
+
"step": 781
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 53.97,
|
| 861 |
+
"grad_norm": 1.2957918643951416,
|
| 862 |
+
"learning_rate": 2.5752380952380952e-05,
|
| 863 |
+
"loss": 0.5857,
|
| 864 |
+
"step": 796
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 53.97,
|
| 868 |
+
"eval_accuracy": 0.6748251748251748,
|
| 869 |
+
"eval_loss": 0.8808070421218872,
|
| 870 |
+
"eval_runtime": 5.2003,
|
| 871 |
+
"eval_samples_per_second": 54.997,
|
| 872 |
+
"eval_steps_per_second": 3.461,
|
| 873 |
+
"step": 796
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 54.98,
|
| 877 |
+
"grad_norm": 0.9337932467460632,
|
| 878 |
+
"learning_rate": 2.560952380952381e-05,
|
| 879 |
+
"loss": 0.6076,
|
| 880 |
+
"step": 811
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 54.98,
|
| 884 |
+
"eval_accuracy": 0.6958041958041958,
|
| 885 |
+
"eval_loss": 0.8554547429084778,
|
| 886 |
+
"eval_runtime": 5.108,
|
| 887 |
+
"eval_samples_per_second": 55.991,
|
| 888 |
+
"eval_steps_per_second": 3.524,
|
| 889 |
+
"step": 811
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 56.0,
|
| 893 |
+
"grad_norm": 1.4797887802124023,
|
| 894 |
+
"learning_rate": 2.5466666666666668e-05,
|
| 895 |
+
"loss": 0.5934,
|
| 896 |
+
"step": 826
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 56.0,
|
| 900 |
+
"eval_accuracy": 0.6888111888111889,
|
| 901 |
+
"eval_loss": 0.8653700351715088,
|
| 902 |
+
"eval_runtime": 4.9201,
|
| 903 |
+
"eval_samples_per_second": 58.128,
|
| 904 |
+
"eval_steps_per_second": 3.658,
|
| 905 |
+
"step": 826
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 56.95,
|
| 909 |
+
"grad_norm": 0.9749770164489746,
|
| 910 |
+
"learning_rate": 2.5333333333333334e-05,
|
| 911 |
+
"loss": 0.6427,
|
| 912 |
+
"step": 840
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 56.95,
|
| 916 |
+
"eval_accuracy": 0.6853146853146853,
|
| 917 |
+
"eval_loss": 0.861566424369812,
|
| 918 |
+
"eval_runtime": 5.5361,
|
| 919 |
+
"eval_samples_per_second": 51.661,
|
| 920 |
+
"eval_steps_per_second": 3.251,
|
| 921 |
+
"step": 840
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 57.97,
|
| 925 |
+
"grad_norm": 1.1588115692138672,
|
| 926 |
+
"learning_rate": 2.519047619047619e-05,
|
| 927 |
+
"loss": 0.5782,
|
| 928 |
+
"step": 855
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 57.97,
|
| 932 |
+
"eval_accuracy": 0.6678321678321678,
|
| 933 |
+
"eval_loss": 0.8711130619049072,
|
| 934 |
+
"eval_runtime": 5.1513,
|
| 935 |
+
"eval_samples_per_second": 55.52,
|
| 936 |
+
"eval_steps_per_second": 3.494,
|
| 937 |
+
"step": 855
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 58.98,
|
| 941 |
+
"grad_norm": 2.290128707885742,
|
| 942 |
+
"learning_rate": 2.504761904761905e-05,
|
| 943 |
+
"loss": 0.5819,
|
| 944 |
+
"step": 870
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 58.98,
|
| 948 |
+
"eval_accuracy": 0.6748251748251748,
|
| 949 |
+
"eval_loss": 0.8689377903938293,
|
| 950 |
+
"eval_runtime": 5.3476,
|
| 951 |
+
"eval_samples_per_second": 53.482,
|
| 952 |
+
"eval_steps_per_second": 3.366,
|
| 953 |
+
"step": 870
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 60.0,
|
| 957 |
+
"grad_norm": 1.8434489965438843,
|
| 958 |
+
"learning_rate": 2.4904761904761908e-05,
|
| 959 |
+
"loss": 0.5918,
|
| 960 |
+
"step": 885
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 60.0,
|
| 964 |
+
"eval_accuracy": 0.6923076923076923,
|
| 965 |
+
"eval_loss": 0.8602246642112732,
|
| 966 |
+
"eval_runtime": 5.1182,
|
| 967 |
+
"eval_samples_per_second": 55.879,
|
| 968 |
+
"eval_steps_per_second": 3.517,
|
| 969 |
+
"step": 885
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 60.95,
|
| 973 |
+
"grad_norm": 1.1310744285583496,
|
| 974 |
+
"learning_rate": 2.4771428571428573e-05,
|
| 975 |
+
"loss": 0.5845,
|
| 976 |
+
"step": 899
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 60.95,
|
| 980 |
+
"eval_accuracy": 0.6993006993006993,
|
| 981 |
+
"eval_loss": 0.8458691835403442,
|
| 982 |
+
"eval_runtime": 4.8028,
|
| 983 |
+
"eval_samples_per_second": 59.549,
|
| 984 |
+
"eval_steps_per_second": 3.748,
|
| 985 |
+
"step": 899
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 61.97,
|
| 989 |
+
"grad_norm": 1.319384217262268,
|
| 990 |
+
"learning_rate": 2.462857142857143e-05,
|
| 991 |
+
"loss": 0.5667,
|
| 992 |
+
"step": 914
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 61.97,
|
| 996 |
+
"eval_accuracy": 0.7027972027972028,
|
| 997 |
+
"eval_loss": 0.8466892838478088,
|
| 998 |
+
"eval_runtime": 4.9074,
|
| 999 |
+
"eval_samples_per_second": 58.279,
|
| 1000 |
+
"eval_steps_per_second": 3.668,
|
| 1001 |
+
"step": 914
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 62.98,
|
| 1005 |
+
"grad_norm": 1.2371602058410645,
|
| 1006 |
+
"learning_rate": 2.448571428571429e-05,
|
| 1007 |
+
"loss": 0.5327,
|
| 1008 |
+
"step": 929
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 62.98,
|
| 1012 |
+
"eval_accuracy": 0.6923076923076923,
|
| 1013 |
+
"eval_loss": 0.8539786338806152,
|
| 1014 |
+
"eval_runtime": 5.32,
|
| 1015 |
+
"eval_samples_per_second": 53.759,
|
| 1016 |
+
"eval_steps_per_second": 3.383,
|
| 1017 |
+
"step": 929
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 64.0,
|
| 1021 |
+
"grad_norm": 2.1778481006622314,
|
| 1022 |
+
"learning_rate": 2.434285714285714e-05,
|
| 1023 |
+
"loss": 0.523,
|
| 1024 |
+
"step": 944
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 64.0,
|
| 1028 |
+
"eval_accuracy": 0.7062937062937062,
|
| 1029 |
+
"eval_loss": 0.8323072791099548,
|
| 1030 |
+
"eval_runtime": 5.1152,
|
| 1031 |
+
"eval_samples_per_second": 55.912,
|
| 1032 |
+
"eval_steps_per_second": 3.519,
|
| 1033 |
+
"step": 944
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 64.95,
|
| 1037 |
+
"grad_norm": 0.8219490051269531,
|
| 1038 |
+
"learning_rate": 2.420952380952381e-05,
|
| 1039 |
+
"loss": 0.548,
|
| 1040 |
+
"step": 958
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 64.95,
|
| 1044 |
+
"eval_accuracy": 0.6993006993006993,
|
| 1045 |
+
"eval_loss": 0.8406782746315002,
|
| 1046 |
+
"eval_runtime": 5.325,
|
| 1047 |
+
"eval_samples_per_second": 53.709,
|
| 1048 |
+
"eval_steps_per_second": 3.38,
|
| 1049 |
+
"step": 958
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 65.97,
|
| 1053 |
+
"grad_norm": 2.0205609798431396,
|
| 1054 |
+
"learning_rate": 2.4066666666666664e-05,
|
| 1055 |
+
"loss": 0.5399,
|
| 1056 |
+
"step": 973
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 65.97,
|
| 1060 |
+
"eval_accuracy": 0.6993006993006993,
|
| 1061 |
+
"eval_loss": 0.837882936000824,
|
| 1062 |
+
"eval_runtime": 5.6808,
|
| 1063 |
+
"eval_samples_per_second": 50.345,
|
| 1064 |
+
"eval_steps_per_second": 3.169,
|
| 1065 |
+
"step": 973
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 66.98,
|
| 1069 |
+
"grad_norm": 1.527024507522583,
|
| 1070 |
+
"learning_rate": 2.3923809523809522e-05,
|
| 1071 |
+
"loss": 0.5324,
|
| 1072 |
+
"step": 988
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 66.98,
|
| 1076 |
+
"eval_accuracy": 0.7027972027972028,
|
| 1077 |
+
"eval_loss": 0.8118743896484375,
|
| 1078 |
+
"eval_runtime": 5.543,
|
| 1079 |
+
"eval_samples_per_second": 51.596,
|
| 1080 |
+
"eval_steps_per_second": 3.247,
|
| 1081 |
+
"step": 988
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 68.0,
|
| 1085 |
+
"grad_norm": 1.3294800519943237,
|
| 1086 |
+
"learning_rate": 2.378095238095238e-05,
|
| 1087 |
+
"loss": 0.5171,
|
| 1088 |
+
"step": 1003
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 68.0,
|
| 1092 |
+
"eval_accuracy": 0.6923076923076923,
|
| 1093 |
+
"eval_loss": 0.8444966673851013,
|
| 1094 |
+
"eval_runtime": 5.5571,
|
| 1095 |
+
"eval_samples_per_second": 51.465,
|
| 1096 |
+
"eval_steps_per_second": 3.239,
|
| 1097 |
+
"step": 1003
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 68.95,
|
| 1101 |
+
"grad_norm": 1.6269547939300537,
|
| 1102 |
+
"learning_rate": 2.3647619047619046e-05,
|
| 1103 |
+
"loss": 0.538,
|
| 1104 |
+
"step": 1017
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 68.95,
|
| 1108 |
+
"eval_accuracy": 0.7097902097902098,
|
| 1109 |
+
"eval_loss": 0.8196296095848083,
|
| 1110 |
+
"eval_runtime": 4.8636,
|
| 1111 |
+
"eval_samples_per_second": 58.805,
|
| 1112 |
+
"eval_steps_per_second": 3.701,
|
| 1113 |
+
"step": 1017
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 69.97,
|
| 1117 |
+
"grad_norm": 1.0726577043533325,
|
| 1118 |
+
"learning_rate": 2.3504761904761904e-05,
|
| 1119 |
+
"loss": 0.5312,
|
| 1120 |
+
"step": 1032
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 69.97,
|
| 1124 |
+
"eval_accuracy": 0.6853146853146853,
|
| 1125 |
+
"eval_loss": 0.8414965271949768,
|
| 1126 |
+
"eval_runtime": 5.2322,
|
| 1127 |
+
"eval_samples_per_second": 54.661,
|
| 1128 |
+
"eval_steps_per_second": 3.44,
|
| 1129 |
+
"step": 1032
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 70.98,
|
| 1133 |
+
"grad_norm": 1.0595276355743408,
|
| 1134 |
+
"learning_rate": 2.3361904761904762e-05,
|
| 1135 |
+
"loss": 0.4914,
|
| 1136 |
+
"step": 1047
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 70.98,
|
| 1140 |
+
"eval_accuracy": 0.6958041958041958,
|
| 1141 |
+
"eval_loss": 0.8184240460395813,
|
| 1142 |
+
"eval_runtime": 4.6109,
|
| 1143 |
+
"eval_samples_per_second": 62.027,
|
| 1144 |
+
"eval_steps_per_second": 3.904,
|
| 1145 |
+
"step": 1047
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 72.0,
|
| 1149 |
+
"grad_norm": 1.3313194513320923,
|
| 1150 |
+
"learning_rate": 2.321904761904762e-05,
|
| 1151 |
+
"loss": 0.5055,
|
| 1152 |
+
"step": 1062
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 72.0,
|
| 1156 |
+
"eval_accuracy": 0.6923076923076923,
|
| 1157 |
+
"eval_loss": 0.8218348622322083,
|
| 1158 |
+
"eval_runtime": 5.5535,
|
| 1159 |
+
"eval_samples_per_second": 51.499,
|
| 1160 |
+
"eval_steps_per_second": 3.241,
|
| 1161 |
+
"step": 1062
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 72.95,
|
| 1165 |
+
"grad_norm": 1.9882231950759888,
|
| 1166 |
+
"learning_rate": 2.3085714285714286e-05,
|
| 1167 |
+
"loss": 0.5401,
|
| 1168 |
+
"step": 1076
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 72.95,
|
| 1172 |
+
"eval_accuracy": 0.7027972027972028,
|
| 1173 |
+
"eval_loss": 0.815979540348053,
|
| 1174 |
+
"eval_runtime": 6.0512,
|
| 1175 |
+
"eval_samples_per_second": 47.264,
|
| 1176 |
+
"eval_steps_per_second": 2.975,
|
| 1177 |
+
"step": 1076
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"epoch": 73.97,
|
| 1181 |
+
"grad_norm": 1.8936933279037476,
|
| 1182 |
+
"learning_rate": 2.2942857142857144e-05,
|
| 1183 |
+
"loss": 0.4966,
|
| 1184 |
+
"step": 1091
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 73.97,
|
| 1188 |
+
"eval_accuracy": 0.6888111888111889,
|
| 1189 |
+
"eval_loss": 0.8237566351890564,
|
| 1190 |
+
"eval_runtime": 4.6622,
|
| 1191 |
+
"eval_samples_per_second": 61.344,
|
| 1192 |
+
"eval_steps_per_second": 3.861,
|
| 1193 |
+
"step": 1091
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 74.98,
|
| 1197 |
+
"grad_norm": 2.0124337673187256,
|
| 1198 |
+
"learning_rate": 2.2800000000000002e-05,
|
| 1199 |
+
"loss": 0.4768,
|
| 1200 |
+
"step": 1106
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 74.98,
|
| 1204 |
+
"eval_accuracy": 0.6993006993006993,
|
| 1205 |
+
"eval_loss": 0.8185241222381592,
|
| 1206 |
+
"eval_runtime": 4.2442,
|
| 1207 |
+
"eval_samples_per_second": 67.386,
|
| 1208 |
+
"eval_steps_per_second": 4.241,
|
| 1209 |
+
"step": 1106
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 76.0,
|
| 1213 |
+
"grad_norm": 1.7522495985031128,
|
| 1214 |
+
"learning_rate": 2.265714285714286e-05,
|
| 1215 |
+
"loss": 0.4789,
|
| 1216 |
+
"step": 1121
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 76.0,
|
| 1220 |
+
"eval_accuracy": 0.7027972027972028,
|
| 1221 |
+
"eval_loss": 0.8260769844055176,
|
| 1222 |
+
"eval_runtime": 6.128,
|
| 1223 |
+
"eval_samples_per_second": 46.671,
|
| 1224 |
+
"eval_steps_per_second": 2.937,
|
| 1225 |
+
"step": 1121
|
| 1226 |
+
},
|
| 1227 |
+
{
|
| 1228 |
+
"epoch": 76.95,
|
| 1229 |
+
"grad_norm": 1.8254278898239136,
|
| 1230 |
+
"learning_rate": 2.2523809523809526e-05,
|
| 1231 |
+
"loss": 0.5176,
|
| 1232 |
+
"step": 1135
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 76.95,
|
| 1236 |
+
"eval_accuracy": 0.7097902097902098,
|
| 1237 |
+
"eval_loss": 0.8109525442123413,
|
| 1238 |
+
"eval_runtime": 6.0147,
|
| 1239 |
+
"eval_samples_per_second": 47.55,
|
| 1240 |
+
"eval_steps_per_second": 2.993,
|
| 1241 |
+
"step": 1135
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 77.97,
|
| 1245 |
+
"grad_norm": 1.4586073160171509,
|
| 1246 |
+
"learning_rate": 2.238095238095238e-05,
|
| 1247 |
+
"loss": 0.466,
|
| 1248 |
+
"step": 1150
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 77.97,
|
| 1252 |
+
"eval_accuracy": 0.6993006993006993,
|
| 1253 |
+
"eval_loss": 0.814057469367981,
|
| 1254 |
+
"eval_runtime": 4.782,
|
| 1255 |
+
"eval_samples_per_second": 59.808,
|
| 1256 |
+
"eval_steps_per_second": 3.764,
|
| 1257 |
+
"step": 1150
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 78.98,
|
| 1261 |
+
"grad_norm": 1.7543085813522339,
|
| 1262 |
+
"learning_rate": 2.223809523809524e-05,
|
| 1263 |
+
"loss": 0.4736,
|
| 1264 |
+
"step": 1165
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 78.98,
|
| 1268 |
+
"eval_accuracy": 0.7167832167832168,
|
| 1269 |
+
"eval_loss": 0.7970029711723328,
|
| 1270 |
+
"eval_runtime": 5.8124,
|
| 1271 |
+
"eval_samples_per_second": 49.205,
|
| 1272 |
+
"eval_steps_per_second": 3.097,
|
| 1273 |
+
"step": 1165
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 80.0,
|
| 1277 |
+
"grad_norm": 1.6676491498947144,
|
| 1278 |
+
"learning_rate": 2.2095238095238096e-05,
|
| 1279 |
+
"loss": 0.4785,
|
| 1280 |
+
"step": 1180
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 80.0,
|
| 1284 |
+
"eval_accuracy": 0.7097902097902098,
|
| 1285 |
+
"eval_loss": 0.8062326312065125,
|
| 1286 |
+
"eval_runtime": 4.5174,
|
| 1287 |
+
"eval_samples_per_second": 63.31,
|
| 1288 |
+
"eval_steps_per_second": 3.985,
|
| 1289 |
+
"step": 1180
|
| 1290 |
+
},
|
| 1291 |
+
{
|
| 1292 |
+
"epoch": 80.95,
|
| 1293 |
+
"grad_norm": 1.827528476715088,
|
| 1294 |
+
"learning_rate": 2.1961904761904762e-05,
|
| 1295 |
+
"loss": 0.5309,
|
| 1296 |
+
"step": 1194
|
| 1297 |
+
},
|
| 1298 |
+
{
|
| 1299 |
+
"epoch": 80.95,
|
| 1300 |
+
"eval_accuracy": 0.6958041958041958,
|
| 1301 |
+
"eval_loss": 0.8051398396492004,
|
| 1302 |
+
"eval_runtime": 4.9397,
|
| 1303 |
+
"eval_samples_per_second": 57.899,
|
| 1304 |
+
"eval_steps_per_second": 3.644,
|
| 1305 |
+
"step": 1194
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 81.97,
|
| 1309 |
+
"grad_norm": 2.3700757026672363,
|
| 1310 |
+
"learning_rate": 2.181904761904762e-05,
|
| 1311 |
+
"loss": 0.4571,
|
| 1312 |
+
"step": 1209
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 81.97,
|
| 1316 |
+
"eval_accuracy": 0.7097902097902098,
|
| 1317 |
+
"eval_loss": 0.8024002909660339,
|
| 1318 |
+
"eval_runtime": 6.3277,
|
| 1319 |
+
"eval_samples_per_second": 45.198,
|
| 1320 |
+
"eval_steps_per_second": 2.845,
|
| 1321 |
+
"step": 1209
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 82.98,
|
| 1325 |
+
"grad_norm": 1.5626410245895386,
|
| 1326 |
+
"learning_rate": 2.1676190476190478e-05,
|
| 1327 |
+
"loss": 0.47,
|
| 1328 |
+
"step": 1224
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 82.98,
|
| 1332 |
+
"eval_accuracy": 0.7167832167832168,
|
| 1333 |
+
"eval_loss": 0.8030509948730469,
|
| 1334 |
+
"eval_runtime": 5.4575,
|
| 1335 |
+
"eval_samples_per_second": 52.405,
|
| 1336 |
+
"eval_steps_per_second": 3.298,
|
| 1337 |
+
"step": 1224
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 84.0,
|
| 1341 |
+
"grad_norm": 1.4918785095214844,
|
| 1342 |
+
"learning_rate": 2.1533333333333333e-05,
|
| 1343 |
+
"loss": 0.4525,
|
| 1344 |
+
"step": 1239
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 84.0,
|
| 1348 |
+
"eval_accuracy": 0.7132867132867133,
|
| 1349 |
+
"eval_loss": 0.7910680174827576,
|
| 1350 |
+
"eval_runtime": 4.6183,
|
| 1351 |
+
"eval_samples_per_second": 61.928,
|
| 1352 |
+
"eval_steps_per_second": 3.898,
|
| 1353 |
+
"step": 1239
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 84.95,
|
| 1357 |
+
"grad_norm": 1.4631191492080688,
|
| 1358 |
+
"learning_rate": 2.1400000000000002e-05,
|
| 1359 |
+
"loss": 0.5058,
|
| 1360 |
+
"step": 1253
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 84.95,
|
| 1364 |
+
"eval_accuracy": 0.7132867132867133,
|
| 1365 |
+
"eval_loss": 0.7877373695373535,
|
| 1366 |
+
"eval_runtime": 4.6086,
|
| 1367 |
+
"eval_samples_per_second": 62.058,
|
| 1368 |
+
"eval_steps_per_second": 3.906,
|
| 1369 |
+
"step": 1253
|
| 1370 |
+
},
|
| 1371 |
+
{
|
| 1372 |
+
"epoch": 85.97,
|
| 1373 |
+
"grad_norm": 1.463442087173462,
|
| 1374 |
+
"learning_rate": 2.1257142857142856e-05,
|
| 1375 |
+
"loss": 0.4627,
|
| 1376 |
+
"step": 1268
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 85.97,
|
| 1380 |
+
"eval_accuracy": 0.7062937062937062,
|
| 1381 |
+
"eval_loss": 0.7917687296867371,
|
| 1382 |
+
"eval_runtime": 6.4645,
|
| 1383 |
+
"eval_samples_per_second": 44.241,
|
| 1384 |
+
"eval_steps_per_second": 2.784,
|
| 1385 |
+
"step": 1268
|
| 1386 |
+
},
|
| 1387 |
+
{
|
| 1388 |
+
"epoch": 86.98,
|
| 1389 |
+
"grad_norm": 1.5172946453094482,
|
| 1390 |
+
"learning_rate": 2.1114285714285714e-05,
|
| 1391 |
+
"loss": 0.4343,
|
| 1392 |
+
"step": 1283
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"epoch": 86.98,
|
| 1396 |
+
"eval_accuracy": 0.7167832167832168,
|
| 1397 |
+
"eval_loss": 0.78819739818573,
|
| 1398 |
+
"eval_runtime": 5.3121,
|
| 1399 |
+
"eval_samples_per_second": 53.839,
|
| 1400 |
+
"eval_steps_per_second": 3.388,
|
| 1401 |
+
"step": 1283
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"epoch": 88.0,
|
| 1405 |
+
"grad_norm": 1.8876270055770874,
|
| 1406 |
+
"learning_rate": 2.0971428571428572e-05,
|
| 1407 |
+
"loss": 0.4442,
|
| 1408 |
+
"step": 1298
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 88.0,
|
| 1412 |
+
"eval_accuracy": 0.7132867132867133,
|
| 1413 |
+
"eval_loss": 0.805809736251831,
|
| 1414 |
+
"eval_runtime": 5.4992,
|
| 1415 |
+
"eval_samples_per_second": 52.007,
|
| 1416 |
+
"eval_steps_per_second": 3.273,
|
| 1417 |
+
"step": 1298
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 88.95,
|
| 1421 |
+
"grad_norm": 1.4128847122192383,
|
| 1422 |
+
"learning_rate": 2.0838095238095238e-05,
|
| 1423 |
+
"loss": 0.4745,
|
| 1424 |
+
"step": 1312
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 88.95,
|
| 1428 |
+
"eval_accuracy": 0.7237762237762237,
|
| 1429 |
+
"eval_loss": 0.7810028791427612,
|
| 1430 |
+
"eval_runtime": 5.1967,
|
| 1431 |
+
"eval_samples_per_second": 55.035,
|
| 1432 |
+
"eval_steps_per_second": 3.464,
|
| 1433 |
+
"step": 1312
|
| 1434 |
+
},
|
| 1435 |
+
{
|
| 1436 |
+
"epoch": 89.97,
|
| 1437 |
+
"grad_norm": 1.3974703550338745,
|
| 1438 |
+
"learning_rate": 2.0695238095238096e-05,
|
| 1439 |
+
"loss": 0.4282,
|
| 1440 |
+
"step": 1327
|
| 1441 |
+
},
|
| 1442 |
+
{
|
| 1443 |
+
"epoch": 89.97,
|
| 1444 |
+
"eval_accuracy": 0.7097902097902098,
|
| 1445 |
+
"eval_loss": 0.7951435446739197,
|
| 1446 |
+
"eval_runtime": 5.9497,
|
| 1447 |
+
"eval_samples_per_second": 48.07,
|
| 1448 |
+
"eval_steps_per_second": 3.025,
|
| 1449 |
+
"step": 1327
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 90.98,
|
| 1453 |
+
"grad_norm": 1.8035740852355957,
|
| 1454 |
+
"learning_rate": 2.055238095238095e-05,
|
| 1455 |
+
"loss": 0.4307,
|
| 1456 |
+
"step": 1342
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 90.98,
|
| 1460 |
+
"eval_accuracy": 0.7167832167832168,
|
| 1461 |
+
"eval_loss": 0.7739275097846985,
|
| 1462 |
+
"eval_runtime": 5.0539,
|
| 1463 |
+
"eval_samples_per_second": 56.59,
|
| 1464 |
+
"eval_steps_per_second": 3.562,
|
| 1465 |
+
"step": 1342
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 92.0,
|
| 1469 |
+
"grad_norm": 2.104257822036743,
|
| 1470 |
+
"learning_rate": 2.040952380952381e-05,
|
| 1471 |
+
"loss": 0.4403,
|
| 1472 |
+
"step": 1357
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 92.0,
|
| 1476 |
+
"eval_accuracy": 0.7202797202797203,
|
| 1477 |
+
"eval_loss": 0.7788061499595642,
|
| 1478 |
+
"eval_runtime": 4.9104,
|
| 1479 |
+
"eval_samples_per_second": 58.244,
|
| 1480 |
+
"eval_steps_per_second": 3.666,
|
| 1481 |
+
"step": 1357
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 92.95,
|
| 1485 |
+
"grad_norm": 1.6670126914978027,
|
| 1486 |
+
"learning_rate": 2.0276190476190475e-05,
|
| 1487 |
+
"loss": 0.4567,
|
| 1488 |
+
"step": 1371
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 92.95,
|
| 1492 |
+
"eval_accuracy": 0.7167832167832168,
|
| 1493 |
+
"eval_loss": 0.7926862835884094,
|
| 1494 |
+
"eval_runtime": 5.6039,
|
| 1495 |
+
"eval_samples_per_second": 51.036,
|
| 1496 |
+
"eval_steps_per_second": 3.212,
|
| 1497 |
+
"step": 1371
|
| 1498 |
+
},
|
| 1499 |
+
{
|
| 1500 |
+
"epoch": 93.97,
|
| 1501 |
+
"grad_norm": 0.9627218842506409,
|
| 1502 |
+
"learning_rate": 2.0133333333333333e-05,
|
| 1503 |
+
"loss": 0.4233,
|
| 1504 |
+
"step": 1386
|
| 1505 |
+
},
|
| 1506 |
+
{
|
| 1507 |
+
"epoch": 93.97,
|
| 1508 |
+
"eval_accuracy": 0.7202797202797203,
|
| 1509 |
+
"eval_loss": 0.7885140180587769,
|
| 1510 |
+
"eval_runtime": 4.3051,
|
| 1511 |
+
"eval_samples_per_second": 66.433,
|
| 1512 |
+
"eval_steps_per_second": 4.181,
|
| 1513 |
+
"step": 1386
|
| 1514 |
+
},
|
| 1515 |
+
{
|
| 1516 |
+
"epoch": 94.98,
|
| 1517 |
+
"grad_norm": 1.5669583082199097,
|
| 1518 |
+
"learning_rate": 1.999047619047619e-05,
|
| 1519 |
+
"loss": 0.4347,
|
| 1520 |
+
"step": 1401
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 94.98,
|
| 1524 |
+
"eval_accuracy": 0.7202797202797203,
|
| 1525 |
+
"eval_loss": 0.7849389314651489,
|
| 1526 |
+
"eval_runtime": 4.7453,
|
| 1527 |
+
"eval_samples_per_second": 60.27,
|
| 1528 |
+
"eval_steps_per_second": 3.793,
|
| 1529 |
+
"step": 1401
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 96.0,
|
| 1533 |
+
"grad_norm": 1.5593161582946777,
|
| 1534 |
+
"learning_rate": 1.984761904761905e-05,
|
| 1535 |
+
"loss": 0.4167,
|
| 1536 |
+
"step": 1416
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 96.0,
|
| 1540 |
+
"eval_accuracy": 0.7237762237762237,
|
| 1541 |
+
"eval_loss": 0.7880135774612427,
|
| 1542 |
+
"eval_runtime": 5.155,
|
| 1543 |
+
"eval_samples_per_second": 55.48,
|
| 1544 |
+
"eval_steps_per_second": 3.492,
|
| 1545 |
+
"step": 1416
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 96.95,
|
| 1549 |
+
"grad_norm": 1.337389349937439,
|
| 1550 |
+
"learning_rate": 1.9714285714285714e-05,
|
| 1551 |
+
"loss": 0.4394,
|
| 1552 |
+
"step": 1430
|
| 1553 |
+
},
|
| 1554 |
+
{
|
| 1555 |
+
"epoch": 96.95,
|
| 1556 |
+
"eval_accuracy": 0.7202797202797203,
|
| 1557 |
+
"eval_loss": 0.7888504862785339,
|
| 1558 |
+
"eval_runtime": 4.6698,
|
| 1559 |
+
"eval_samples_per_second": 61.244,
|
| 1560 |
+
"eval_steps_per_second": 3.855,
|
| 1561 |
+
"step": 1430
|
| 1562 |
+
},
|
| 1563 |
+
{
|
| 1564 |
+
"epoch": 97.97,
|
| 1565 |
+
"grad_norm": 1.8546737432479858,
|
| 1566 |
+
"learning_rate": 1.9571428571428572e-05,
|
| 1567 |
+
"loss": 0.4359,
|
| 1568 |
+
"step": 1445
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 97.97,
|
| 1572 |
+
"eval_accuracy": 0.7202797202797203,
|
| 1573 |
+
"eval_loss": 0.778459370136261,
|
| 1574 |
+
"eval_runtime": 4.7578,
|
| 1575 |
+
"eval_samples_per_second": 60.112,
|
| 1576 |
+
"eval_steps_per_second": 3.783,
|
| 1577 |
+
"step": 1445
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 98.98,
|
| 1581 |
+
"grad_norm": 1.3580291271209717,
|
| 1582 |
+
"learning_rate": 1.942857142857143e-05,
|
| 1583 |
+
"loss": 0.4085,
|
| 1584 |
+
"step": 1460
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 98.98,
|
| 1588 |
+
"eval_accuracy": 0.7132867132867133,
|
| 1589 |
+
"eval_loss": 0.7851797342300415,
|
| 1590 |
+
"eval_runtime": 5.8892,
|
| 1591 |
+
"eval_samples_per_second": 48.564,
|
| 1592 |
+
"eval_steps_per_second": 3.056,
|
| 1593 |
+
"step": 1460
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 100.0,
|
| 1597 |
+
"grad_norm": 1.4461547136306763,
|
| 1598 |
+
"learning_rate": 1.928571428571429e-05,
|
| 1599 |
+
"loss": 0.3965,
|
| 1600 |
+
"step": 1475
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 100.0,
|
| 1604 |
+
"eval_accuracy": 0.7272727272727273,
|
| 1605 |
+
"eval_loss": 0.7784654498100281,
|
| 1606 |
+
"eval_runtime": 5.3557,
|
| 1607 |
+
"eval_samples_per_second": 53.401,
|
| 1608 |
+
"eval_steps_per_second": 3.361,
|
| 1609 |
+
"step": 1475
|
| 1610 |
+
},
|
| 1611 |
+
{
|
| 1612 |
+
"epoch": 100.95,
|
| 1613 |
+
"grad_norm": 1.765655755996704,
|
| 1614 |
+
"learning_rate": 1.9152380952380954e-05,
|
| 1615 |
+
"loss": 0.445,
|
| 1616 |
+
"step": 1489
|
| 1617 |
+
},
|
| 1618 |
+
{
|
| 1619 |
+
"epoch": 100.95,
|
| 1620 |
+
"eval_accuracy": 0.7202797202797203,
|
| 1621 |
+
"eval_loss": 0.7826104760169983,
|
| 1622 |
+
"eval_runtime": 5.628,
|
| 1623 |
+
"eval_samples_per_second": 50.818,
|
| 1624 |
+
"eval_steps_per_second": 3.198,
|
| 1625 |
+
"step": 1489
|
| 1626 |
+
},
|
| 1627 |
+
{
|
| 1628 |
+
"epoch": 101.97,
|
| 1629 |
+
"grad_norm": 1.7069743871688843,
|
| 1630 |
+
"learning_rate": 1.9009523809523812e-05,
|
| 1631 |
+
"loss": 0.3988,
|
| 1632 |
+
"step": 1504
|
| 1633 |
+
},
|
| 1634 |
+
{
|
| 1635 |
+
"epoch": 101.97,
|
| 1636 |
+
"eval_accuracy": 0.7097902097902098,
|
| 1637 |
+
"eval_loss": 0.8045200109481812,
|
| 1638 |
+
"eval_runtime": 5.2204,
|
| 1639 |
+
"eval_samples_per_second": 54.785,
|
| 1640 |
+
"eval_steps_per_second": 3.448,
|
| 1641 |
+
"step": 1504
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 102.98,
|
| 1645 |
+
"grad_norm": 1.2846873998641968,
|
| 1646 |
+
"learning_rate": 1.886666666666667e-05,
|
| 1647 |
+
"loss": 0.4129,
|
| 1648 |
+
"step": 1519
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 102.98,
|
| 1652 |
+
"eval_accuracy": 0.7272727272727273,
|
| 1653 |
+
"eval_loss": 0.7685949206352234,
|
| 1654 |
+
"eval_runtime": 5.3375,
|
| 1655 |
+
"eval_samples_per_second": 53.583,
|
| 1656 |
+
"eval_steps_per_second": 3.372,
|
| 1657 |
+
"step": 1519
|
| 1658 |
+
},
|
| 1659 |
+
{
|
| 1660 |
+
"epoch": 104.0,
|
| 1661 |
+
"grad_norm": 1.9040664434432983,
|
| 1662 |
+
"learning_rate": 1.872380952380952e-05,
|
| 1663 |
+
"loss": 0.3937,
|
| 1664 |
+
"step": 1534
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"epoch": 104.0,
|
| 1668 |
+
"eval_accuracy": 0.7132867132867133,
|
| 1669 |
+
"eval_loss": 0.7911521196365356,
|
| 1670 |
+
"eval_runtime": 5.5547,
|
| 1671 |
+
"eval_samples_per_second": 51.488,
|
| 1672 |
+
"eval_steps_per_second": 3.24,
|
| 1673 |
+
"step": 1534
|
| 1674 |
+
},
|
| 1675 |
+
{
|
| 1676 |
+
"epoch": 104.95,
|
| 1677 |
+
"grad_norm": 1.998403549194336,
|
| 1678 |
+
"learning_rate": 1.8590476190476194e-05,
|
| 1679 |
+
"loss": 0.4356,
|
| 1680 |
+
"step": 1548
|
| 1681 |
+
},
|
| 1682 |
+
{
|
| 1683 |
+
"epoch": 104.95,
|
| 1684 |
+
"eval_accuracy": 0.7132867132867133,
|
| 1685 |
+
"eval_loss": 0.7922284007072449,
|
| 1686 |
+
"eval_runtime": 5.247,
|
| 1687 |
+
"eval_samples_per_second": 54.507,
|
| 1688 |
+
"eval_steps_per_second": 3.431,
|
| 1689 |
+
"step": 1548
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 105.97,
|
| 1693 |
+
"grad_norm": 1.3670754432678223,
|
| 1694 |
+
"learning_rate": 1.8447619047619045e-05,
|
| 1695 |
+
"loss": 0.3969,
|
| 1696 |
+
"step": 1563
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 105.97,
|
| 1700 |
+
"eval_accuracy": 0.7202797202797203,
|
| 1701 |
+
"eval_loss": 0.7751501202583313,
|
| 1702 |
+
"eval_runtime": 4.5814,
|
| 1703 |
+
"eval_samples_per_second": 62.427,
|
| 1704 |
+
"eval_steps_per_second": 3.929,
|
| 1705 |
+
"step": 1563
|
| 1706 |
+
},
|
| 1707 |
+
{
|
| 1708 |
+
"epoch": 106.98,
|
| 1709 |
+
"grad_norm": 1.0057995319366455,
|
| 1710 |
+
"learning_rate": 1.8304761904761903e-05,
|
| 1711 |
+
"loss": 0.4051,
|
| 1712 |
+
"step": 1578
|
| 1713 |
+
},
|
| 1714 |
+
{
|
| 1715 |
+
"epoch": 106.98,
|
| 1716 |
+
"eval_accuracy": 0.7132867132867133,
|
| 1717 |
+
"eval_loss": 0.7917311787605286,
|
| 1718 |
+
"eval_runtime": 5.1261,
|
| 1719 |
+
"eval_samples_per_second": 55.793,
|
| 1720 |
+
"eval_steps_per_second": 3.511,
|
| 1721 |
+
"step": 1578
|
| 1722 |
+
},
|
| 1723 |
+
{
|
| 1724 |
+
"epoch": 108.0,
|
| 1725 |
+
"grad_norm": 1.7176306247711182,
|
| 1726 |
+
"learning_rate": 1.816190476190476e-05,
|
| 1727 |
+
"loss": 0.3982,
|
| 1728 |
+
"step": 1593
|
| 1729 |
+
},
|
| 1730 |
+
{
|
| 1731 |
+
"epoch": 108.0,
|
| 1732 |
+
"eval_accuracy": 0.7097902097902098,
|
| 1733 |
+
"eval_loss": 0.7916986346244812,
|
| 1734 |
+
"eval_runtime": 6.2925,
|
| 1735 |
+
"eval_samples_per_second": 45.451,
|
| 1736 |
+
"eval_steps_per_second": 2.861,
|
| 1737 |
+
"step": 1593
|
| 1738 |
+
},
|
| 1739 |
+
{
|
| 1740 |
+
"epoch": 108.95,
|
| 1741 |
+
"grad_norm": 1.6695342063903809,
|
| 1742 |
+
"learning_rate": 1.8028571428571427e-05,
|
| 1743 |
+
"loss": 0.4117,
|
| 1744 |
+
"step": 1607
|
| 1745 |
+
},
|
| 1746 |
+
{
|
| 1747 |
+
"epoch": 108.95,
|
| 1748 |
+
"eval_accuracy": 0.7062937062937062,
|
| 1749 |
+
"eval_loss": 0.8070589303970337,
|
| 1750 |
+
"eval_runtime": 4.8482,
|
| 1751 |
+
"eval_samples_per_second": 58.99,
|
| 1752 |
+
"eval_steps_per_second": 3.713,
|
| 1753 |
+
"step": 1607
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 109.97,
|
| 1757 |
+
"grad_norm": 1.1871509552001953,
|
| 1758 |
+
"learning_rate": 1.7885714285714285e-05,
|
| 1759 |
+
"loss": 0.3666,
|
| 1760 |
+
"step": 1622
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 109.97,
|
| 1764 |
+
"eval_accuracy": 0.7202797202797203,
|
| 1765 |
+
"eval_loss": 0.7839590907096863,
|
| 1766 |
+
"eval_runtime": 5.4971,
|
| 1767 |
+
"eval_samples_per_second": 52.027,
|
| 1768 |
+
"eval_steps_per_second": 3.274,
|
| 1769 |
+
"step": 1622
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"epoch": 110.98,
|
| 1773 |
+
"grad_norm": 2.196869134902954,
|
| 1774 |
+
"learning_rate": 1.7742857142857143e-05,
|
| 1775 |
+
"loss": 0.3894,
|
| 1776 |
+
"step": 1637
|
| 1777 |
+
},
|
| 1778 |
+
{
|
| 1779 |
+
"epoch": 110.98,
|
| 1780 |
+
"eval_accuracy": 0.7237762237762237,
|
| 1781 |
+
"eval_loss": 0.7790002226829529,
|
| 1782 |
+
"eval_runtime": 6.5509,
|
| 1783 |
+
"eval_samples_per_second": 43.658,
|
| 1784 |
+
"eval_steps_per_second": 2.748,
|
| 1785 |
+
"step": 1637
|
| 1786 |
+
},
|
| 1787 |
+
{
|
| 1788 |
+
"epoch": 112.0,
|
| 1789 |
+
"grad_norm": 2.341435432434082,
|
| 1790 |
+
"learning_rate": 1.76e-05,
|
| 1791 |
+
"loss": 0.3858,
|
| 1792 |
+
"step": 1652
|
| 1793 |
+
},
|
| 1794 |
+
{
|
| 1795 |
+
"epoch": 112.0,
|
| 1796 |
+
"eval_accuracy": 0.7097902097902098,
|
| 1797 |
+
"eval_loss": 0.7961041927337646,
|
| 1798 |
+
"eval_runtime": 4.8263,
|
| 1799 |
+
"eval_samples_per_second": 59.258,
|
| 1800 |
+
"eval_steps_per_second": 3.73,
|
| 1801 |
+
"step": 1652
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 112.95,
|
| 1805 |
+
"grad_norm": 1.5795267820358276,
|
| 1806 |
+
"learning_rate": 1.7466666666666667e-05,
|
| 1807 |
+
"loss": 0.4037,
|
| 1808 |
+
"step": 1666
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 112.95,
|
| 1812 |
+
"eval_accuracy": 0.7202797202797203,
|
| 1813 |
+
"eval_loss": 0.7822238802909851,
|
| 1814 |
+
"eval_runtime": 5.6008,
|
| 1815 |
+
"eval_samples_per_second": 51.064,
|
| 1816 |
+
"eval_steps_per_second": 3.214,
|
| 1817 |
+
"step": 1666
|
| 1818 |
+
},
|
| 1819 |
+
{
|
| 1820 |
+
"epoch": 113.97,
|
| 1821 |
+
"grad_norm": 1.5341583490371704,
|
| 1822 |
+
"learning_rate": 1.7333333333333332e-05,
|
| 1823 |
+
"loss": 0.3886,
|
| 1824 |
+
"step": 1681
|
| 1825 |
+
},
|
| 1826 |
+
{
|
| 1827 |
+
"epoch": 113.97,
|
| 1828 |
+
"eval_accuracy": 0.7237762237762237,
|
| 1829 |
+
"eval_loss": 0.7748440504074097,
|
| 1830 |
+
"eval_runtime": 4.7211,
|
| 1831 |
+
"eval_samples_per_second": 60.579,
|
| 1832 |
+
"eval_steps_per_second": 3.813,
|
| 1833 |
+
"step": 1681
|
| 1834 |
+
},
|
| 1835 |
+
{
|
| 1836 |
+
"epoch": 114.98,
|
| 1837 |
+
"grad_norm": 1.056776523590088,
|
| 1838 |
+
"learning_rate": 1.719047619047619e-05,
|
| 1839 |
+
"loss": 0.3762,
|
| 1840 |
+
"step": 1696
|
| 1841 |
+
},
|
| 1842 |
+
{
|
| 1843 |
+
"epoch": 114.98,
|
| 1844 |
+
"eval_accuracy": 0.7167832167832168,
|
| 1845 |
+
"eval_loss": 0.7781729102134705,
|
| 1846 |
+
"eval_runtime": 6.5039,
|
| 1847 |
+
"eval_samples_per_second": 43.974,
|
| 1848 |
+
"eval_steps_per_second": 2.768,
|
| 1849 |
+
"step": 1696
|
| 1850 |
+
},
|
| 1851 |
+
{
|
| 1852 |
+
"epoch": 116.0,
|
| 1853 |
+
"grad_norm": 2.7518413066864014,
|
| 1854 |
+
"learning_rate": 1.704761904761905e-05,
|
| 1855 |
+
"loss": 0.3444,
|
| 1856 |
+
"step": 1711
|
| 1857 |
+
},
|
| 1858 |
+
{
|
| 1859 |
+
"epoch": 116.0,
|
| 1860 |
+
"eval_accuracy": 0.7307692307692307,
|
| 1861 |
+
"eval_loss": 0.7745847702026367,
|
| 1862 |
+
"eval_runtime": 6.1348,
|
| 1863 |
+
"eval_samples_per_second": 46.619,
|
| 1864 |
+
"eval_steps_per_second": 2.934,
|
| 1865 |
+
"step": 1711
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 116.95,
|
| 1869 |
+
"grad_norm": 1.8518308401107788,
|
| 1870 |
+
"learning_rate": 1.6914285714285714e-05,
|
| 1871 |
+
"loss": 0.3961,
|
| 1872 |
+
"step": 1725
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 116.95,
|
| 1876 |
+
"eval_accuracy": 0.7202797202797203,
|
| 1877 |
+
"eval_loss": 0.7842003703117371,
|
| 1878 |
+
"eval_runtime": 4.8863,
|
| 1879 |
+
"eval_samples_per_second": 58.531,
|
| 1880 |
+
"eval_steps_per_second": 3.684,
|
| 1881 |
+
"step": 1725
|
| 1882 |
+
},
|
| 1883 |
+
{
|
| 1884 |
+
"epoch": 117.97,
|
| 1885 |
+
"grad_norm": 1.487108588218689,
|
| 1886 |
+
"learning_rate": 1.6771428571428572e-05,
|
| 1887 |
+
"loss": 0.3578,
|
| 1888 |
+
"step": 1740
|
| 1889 |
+
},
|
| 1890 |
+
{
|
| 1891 |
+
"epoch": 117.97,
|
| 1892 |
+
"eval_accuracy": 0.7202797202797203,
|
| 1893 |
+
"eval_loss": 0.7819164991378784,
|
| 1894 |
+
"eval_runtime": 5.6099,
|
| 1895 |
+
"eval_samples_per_second": 50.981,
|
| 1896 |
+
"eval_steps_per_second": 3.209,
|
| 1897 |
+
"step": 1740
|
| 1898 |
+
},
|
| 1899 |
+
{
|
| 1900 |
+
"epoch": 118.98,
|
| 1901 |
+
"grad_norm": 1.7816615104675293,
|
| 1902 |
+
"learning_rate": 1.662857142857143e-05,
|
| 1903 |
+
"loss": 0.3578,
|
| 1904 |
+
"step": 1755
|
| 1905 |
+
},
|
| 1906 |
+
{
|
| 1907 |
+
"epoch": 118.98,
|
| 1908 |
+
"eval_accuracy": 0.7202797202797203,
|
| 1909 |
+
"eval_loss": 0.7805832624435425,
|
| 1910 |
+
"eval_runtime": 5.5595,
|
| 1911 |
+
"eval_samples_per_second": 51.443,
|
| 1912 |
+
"eval_steps_per_second": 3.238,
|
| 1913 |
+
"step": 1755
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 120.0,
|
| 1917 |
+
"grad_norm": 2.4715042114257812,
|
| 1918 |
+
"learning_rate": 1.6485714285714285e-05,
|
| 1919 |
+
"loss": 0.3489,
|
| 1920 |
+
"step": 1770
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 120.0,
|
| 1924 |
+
"eval_accuracy": 0.7237762237762237,
|
| 1925 |
+
"eval_loss": 0.7809211015701294,
|
| 1926 |
+
"eval_runtime": 6.1205,
|
| 1927 |
+
"eval_samples_per_second": 46.728,
|
| 1928 |
+
"eval_steps_per_second": 2.941,
|
| 1929 |
+
"step": 1770
|
| 1930 |
+
},
|
| 1931 |
+
{
|
| 1932 |
+
"epoch": 120.95,
|
| 1933 |
+
"grad_norm": 1.1066619157791138,
|
| 1934 |
+
"learning_rate": 1.635238095238095e-05,
|
| 1935 |
+
"loss": 0.3622,
|
| 1936 |
+
"step": 1784
|
| 1937 |
+
},
|
| 1938 |
+
{
|
| 1939 |
+
"epoch": 120.95,
|
| 1940 |
+
"eval_accuracy": 0.7097902097902098,
|
| 1941 |
+
"eval_loss": 0.7947035431861877,
|
| 1942 |
+
"eval_runtime": 5.0722,
|
| 1943 |
+
"eval_samples_per_second": 56.386,
|
| 1944 |
+
"eval_steps_per_second": 3.549,
|
| 1945 |
+
"step": 1784
|
| 1946 |
+
},
|
| 1947 |
+
{
|
| 1948 |
+
"epoch": 121.97,
|
| 1949 |
+
"grad_norm": 1.3460161685943604,
|
| 1950 |
+
"learning_rate": 1.620952380952381e-05,
|
| 1951 |
+
"loss": 0.3545,
|
| 1952 |
+
"step": 1799
|
| 1953 |
+
},
|
| 1954 |
+
{
|
| 1955 |
+
"epoch": 121.97,
|
| 1956 |
+
"eval_accuracy": 0.7167832167832168,
|
| 1957 |
+
"eval_loss": 0.7877638339996338,
|
| 1958 |
+
"eval_runtime": 5.4395,
|
| 1959 |
+
"eval_samples_per_second": 52.578,
|
| 1960 |
+
"eval_steps_per_second": 3.309,
|
| 1961 |
+
"step": 1799
|
| 1962 |
+
},
|
| 1963 |
+
{
|
| 1964 |
+
"epoch": 122.98,
|
| 1965 |
+
"grad_norm": 1.7303364276885986,
|
| 1966 |
+
"learning_rate": 1.6066666666666666e-05,
|
| 1967 |
+
"loss": 0.3361,
|
| 1968 |
+
"step": 1814
|
| 1969 |
+
},
|
| 1970 |
+
{
|
| 1971 |
+
"epoch": 122.98,
|
| 1972 |
+
"eval_accuracy": 0.7167832167832168,
|
| 1973 |
+
"eval_loss": 0.7854802012443542,
|
| 1974 |
+
"eval_runtime": 4.9544,
|
| 1975 |
+
"eval_samples_per_second": 57.726,
|
| 1976 |
+
"eval_steps_per_second": 3.633,
|
| 1977 |
+
"step": 1814
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 124.0,
|
| 1981 |
+
"grad_norm": 1.5440623760223389,
|
| 1982 |
+
"learning_rate": 1.5923809523809524e-05,
|
| 1983 |
+
"loss": 0.3618,
|
| 1984 |
+
"step": 1829
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 124.0,
|
| 1988 |
+
"eval_accuracy": 0.7132867132867133,
|
| 1989 |
+
"eval_loss": 0.7890083193778992,
|
| 1990 |
+
"eval_runtime": 4.6202,
|
| 1991 |
+
"eval_samples_per_second": 61.902,
|
| 1992 |
+
"eval_steps_per_second": 3.896,
|
| 1993 |
+
"step": 1829
|
| 1994 |
+
},
|
| 1995 |
+
{
|
| 1996 |
+
"epoch": 124.95,
|
| 1997 |
+
"grad_norm": 1.2985795736312866,
|
| 1998 |
+
"learning_rate": 1.579047619047619e-05,
|
| 1999 |
+
"loss": 0.3472,
|
| 2000 |
+
"step": 1843
|
| 2001 |
+
},
|
| 2002 |
+
{
|
| 2003 |
+
"epoch": 124.95,
|
| 2004 |
+
"eval_accuracy": 0.7167832167832168,
|
| 2005 |
+
"eval_loss": 0.7809839844703674,
|
| 2006 |
+
"eval_runtime": 4.5623,
|
| 2007 |
+
"eval_samples_per_second": 62.687,
|
| 2008 |
+
"eval_steps_per_second": 3.945,
|
| 2009 |
+
"step": 1843
|
| 2010 |
+
},
|
| 2011 |
+
{
|
| 2012 |
+
"epoch": 125.97,
|
| 2013 |
+
"grad_norm": 1.3930552005767822,
|
| 2014 |
+
"learning_rate": 1.5647619047619048e-05,
|
| 2015 |
+
"loss": 0.3511,
|
| 2016 |
+
"step": 1858
|
| 2017 |
+
},
|
| 2018 |
+
{
|
| 2019 |
+
"epoch": 125.97,
|
| 2020 |
+
"eval_accuracy": 0.7132867132867133,
|
| 2021 |
+
"eval_loss": 0.7897189259529114,
|
| 2022 |
+
"eval_runtime": 5.1961,
|
| 2023 |
+
"eval_samples_per_second": 55.041,
|
| 2024 |
+
"eval_steps_per_second": 3.464,
|
| 2025 |
+
"step": 1858
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 126.98,
|
| 2029 |
+
"grad_norm": 2.0258781909942627,
|
| 2030 |
+
"learning_rate": 1.5504761904761906e-05,
|
| 2031 |
+
"loss": 0.3389,
|
| 2032 |
+
"step": 1873
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 126.98,
|
| 2036 |
+
"eval_accuracy": 0.7132867132867133,
|
| 2037 |
+
"eval_loss": 0.7922906875610352,
|
| 2038 |
+
"eval_runtime": 5.0606,
|
| 2039 |
+
"eval_samples_per_second": 56.515,
|
| 2040 |
+
"eval_steps_per_second": 3.557,
|
| 2041 |
+
"step": 1873
|
| 2042 |
+
},
|
| 2043 |
+
{
|
| 2044 |
+
"epoch": 128.0,
|
| 2045 |
+
"grad_norm": 1.0623548030853271,
|
| 2046 |
+
"learning_rate": 1.5361904761904764e-05,
|
| 2047 |
+
"loss": 0.3391,
|
| 2048 |
+
"step": 1888
|
| 2049 |
+
},
|
| 2050 |
+
{
|
| 2051 |
+
"epoch": 128.0,
|
| 2052 |
+
"eval_accuracy": 0.7272727272727273,
|
| 2053 |
+
"eval_loss": 0.7781790494918823,
|
| 2054 |
+
"eval_runtime": 4.5848,
|
| 2055 |
+
"eval_samples_per_second": 62.38,
|
| 2056 |
+
"eval_steps_per_second": 3.926,
|
| 2057 |
+
"step": 1888
|
| 2058 |
+
},
|
| 2059 |
+
{
|
| 2060 |
+
"epoch": 128.95,
|
| 2061 |
+
"grad_norm": 1.2839761972427368,
|
| 2062 |
+
"learning_rate": 1.522857142857143e-05,
|
| 2063 |
+
"loss": 0.3746,
|
| 2064 |
+
"step": 1902
|
| 2065 |
+
},
|
| 2066 |
+
{
|
| 2067 |
+
"epoch": 128.95,
|
| 2068 |
+
"eval_accuracy": 0.7202797202797203,
|
| 2069 |
+
"eval_loss": 0.7838397026062012,
|
| 2070 |
+
"eval_runtime": 6.371,
|
| 2071 |
+
"eval_samples_per_second": 44.891,
|
| 2072 |
+
"eval_steps_per_second": 2.825,
|
| 2073 |
+
"step": 1902
|
| 2074 |
+
},
|
| 2075 |
+
{
|
| 2076 |
+
"epoch": 129.97,
|
| 2077 |
+
"grad_norm": 1.7424538135528564,
|
| 2078 |
+
"learning_rate": 1.5085714285714288e-05,
|
| 2079 |
+
"loss": 0.3238,
|
| 2080 |
+
"step": 1917
|
| 2081 |
+
},
|
| 2082 |
+
{
|
| 2083 |
+
"epoch": 129.97,
|
| 2084 |
+
"eval_accuracy": 0.7167832167832168,
|
| 2085 |
+
"eval_loss": 0.7943305969238281,
|
| 2086 |
+
"eval_runtime": 5.0078,
|
| 2087 |
+
"eval_samples_per_second": 57.111,
|
| 2088 |
+
"eval_steps_per_second": 3.594,
|
| 2089 |
+
"step": 1917
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 130.98,
|
| 2093 |
+
"grad_norm": 1.9641584157943726,
|
| 2094 |
+
"learning_rate": 1.4942857142857143e-05,
|
| 2095 |
+
"loss": 0.3601,
|
| 2096 |
+
"step": 1932
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 130.98,
|
| 2100 |
+
"eval_accuracy": 0.7167832167832168,
|
| 2101 |
+
"eval_loss": 0.786338746547699,
|
| 2102 |
+
"eval_runtime": 5.7221,
|
| 2103 |
+
"eval_samples_per_second": 49.982,
|
| 2104 |
+
"eval_steps_per_second": 3.146,
|
| 2105 |
+
"step": 1932
|
| 2106 |
+
},
|
| 2107 |
+
{
|
| 2108 |
+
"epoch": 132.0,
|
| 2109 |
+
"grad_norm": 1.6700012683868408,
|
| 2110 |
+
"learning_rate": 1.48e-05,
|
| 2111 |
+
"loss": 0.3339,
|
| 2112 |
+
"step": 1947
|
| 2113 |
+
},
|
| 2114 |
+
{
|
| 2115 |
+
"epoch": 132.0,
|
| 2116 |
+
"eval_accuracy": 0.7132867132867133,
|
| 2117 |
+
"eval_loss": 0.7948570847511292,
|
| 2118 |
+
"eval_runtime": 4.5956,
|
| 2119 |
+
"eval_samples_per_second": 62.234,
|
| 2120 |
+
"eval_steps_per_second": 3.917,
|
| 2121 |
+
"step": 1947
|
| 2122 |
+
},
|
| 2123 |
+
{
|
| 2124 |
+
"epoch": 132.95,
|
| 2125 |
+
"grad_norm": 1.4314067363739014,
|
| 2126 |
+
"learning_rate": 1.4666666666666666e-05,
|
| 2127 |
+
"loss": 0.3805,
|
| 2128 |
+
"step": 1961
|
| 2129 |
+
},
|
| 2130 |
+
{
|
| 2131 |
+
"epoch": 132.95,
|
| 2132 |
+
"eval_accuracy": 0.7237762237762237,
|
| 2133 |
+
"eval_loss": 0.7823219299316406,
|
| 2134 |
+
"eval_runtime": 5.3022,
|
| 2135 |
+
"eval_samples_per_second": 53.94,
|
| 2136 |
+
"eval_steps_per_second": 3.395,
|
| 2137 |
+
"step": 1961
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 133.97,
|
| 2141 |
+
"grad_norm": 1.9961072206497192,
|
| 2142 |
+
"learning_rate": 1.4523809523809524e-05,
|
| 2143 |
+
"loss": 0.3524,
|
| 2144 |
+
"step": 1976
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 133.97,
|
| 2148 |
+
"eval_accuracy": 0.7097902097902098,
|
| 2149 |
+
"eval_loss": 0.8052372932434082,
|
| 2150 |
+
"eval_runtime": 4.8993,
|
| 2151 |
+
"eval_samples_per_second": 58.375,
|
| 2152 |
+
"eval_steps_per_second": 3.674,
|
| 2153 |
+
"step": 1976
|
| 2154 |
+
},
|
| 2155 |
+
{
|
| 2156 |
+
"epoch": 134.98,
|
| 2157 |
+
"grad_norm": 1.1320672035217285,
|
| 2158 |
+
"learning_rate": 1.4380952380952382e-05,
|
| 2159 |
+
"loss": 0.3103,
|
| 2160 |
+
"step": 1991
|
| 2161 |
+
},
|
| 2162 |
+
{
|
| 2163 |
+
"epoch": 134.98,
|
| 2164 |
+
"eval_accuracy": 0.7237762237762237,
|
| 2165 |
+
"eval_loss": 0.7809023261070251,
|
| 2166 |
+
"eval_runtime": 5.8892,
|
| 2167 |
+
"eval_samples_per_second": 48.563,
|
| 2168 |
+
"eval_steps_per_second": 3.056,
|
| 2169 |
+
"step": 1991
|
| 2170 |
+
},
|
| 2171 |
+
{
|
| 2172 |
+
"epoch": 136.0,
|
| 2173 |
+
"grad_norm": 2.298990249633789,
|
| 2174 |
+
"learning_rate": 1.4238095238095239e-05,
|
| 2175 |
+
"loss": 0.3484,
|
| 2176 |
+
"step": 2006
|
| 2177 |
+
},
|
| 2178 |
+
{
|
| 2179 |
+
"epoch": 136.0,
|
| 2180 |
+
"eval_accuracy": 0.7202797202797203,
|
| 2181 |
+
"eval_loss": 0.787878692150116,
|
| 2182 |
+
"eval_runtime": 5.2761,
|
| 2183 |
+
"eval_samples_per_second": 54.207,
|
| 2184 |
+
"eval_steps_per_second": 3.412,
|
| 2185 |
+
"step": 2006
|
| 2186 |
+
},
|
| 2187 |
+
{
|
| 2188 |
+
"epoch": 136.95,
|
| 2189 |
+
"grad_norm": 2.3017499446868896,
|
| 2190 |
+
"learning_rate": 1.4104761904761906e-05,
|
| 2191 |
+
"loss": 0.3424,
|
| 2192 |
+
"step": 2020
|
| 2193 |
+
},
|
| 2194 |
+
{
|
| 2195 |
+
"epoch": 136.95,
|
| 2196 |
+
"eval_accuracy": 0.7272727272727273,
|
| 2197 |
+
"eval_loss": 0.7875267267227173,
|
| 2198 |
+
"eval_runtime": 4.7804,
|
| 2199 |
+
"eval_samples_per_second": 59.828,
|
| 2200 |
+
"eval_steps_per_second": 3.765,
|
| 2201 |
+
"step": 2020
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 137.97,
|
| 2205 |
+
"grad_norm": 1.4992213249206543,
|
| 2206 |
+
"learning_rate": 1.3961904761904762e-05,
|
| 2207 |
+
"loss": 0.316,
|
| 2208 |
+
"step": 2035
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 137.97,
|
| 2212 |
+
"eval_accuracy": 0.7272727272727273,
|
| 2213 |
+
"eval_loss": 0.7829337120056152,
|
| 2214 |
+
"eval_runtime": 4.9522,
|
| 2215 |
+
"eval_samples_per_second": 57.752,
|
| 2216 |
+
"eval_steps_per_second": 3.635,
|
| 2217 |
+
"step": 2035
|
| 2218 |
+
},
|
| 2219 |
+
{
|
| 2220 |
+
"epoch": 138.98,
|
| 2221 |
+
"grad_norm": 1.634419560432434,
|
| 2222 |
+
"learning_rate": 1.3819047619047619e-05,
|
| 2223 |
+
"loss": 0.3171,
|
| 2224 |
+
"step": 2050
|
| 2225 |
+
},
|
| 2226 |
+
{
|
| 2227 |
+
"epoch": 138.98,
|
| 2228 |
+
"eval_accuracy": 0.7202797202797203,
|
| 2229 |
+
"eval_loss": 0.7882408499717712,
|
| 2230 |
+
"eval_runtime": 4.8523,
|
| 2231 |
+
"eval_samples_per_second": 58.941,
|
| 2232 |
+
"eval_steps_per_second": 3.71,
|
| 2233 |
+
"step": 2050
|
| 2234 |
+
},
|
| 2235 |
+
{
|
| 2236 |
+
"epoch": 140.0,
|
| 2237 |
+
"grad_norm": 1.2980549335479736,
|
| 2238 |
+
"learning_rate": 1.3676190476190477e-05,
|
| 2239 |
+
"loss": 0.3155,
|
| 2240 |
+
"step": 2065
|
| 2241 |
+
},
|
| 2242 |
+
{
|
| 2243 |
+
"epoch": 140.0,
|
| 2244 |
+
"eval_accuracy": 0.7167832167832168,
|
| 2245 |
+
"eval_loss": 0.7830457091331482,
|
| 2246 |
+
"eval_runtime": 4.9611,
|
| 2247 |
+
"eval_samples_per_second": 57.649,
|
| 2248 |
+
"eval_steps_per_second": 3.628,
|
| 2249 |
+
"step": 2065
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 140.95,
|
| 2253 |
+
"grad_norm": 1.5925828218460083,
|
| 2254 |
+
"learning_rate": 1.3542857142857142e-05,
|
| 2255 |
+
"loss": 0.3382,
|
| 2256 |
+
"step": 2079
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 140.95,
|
| 2260 |
+
"eval_accuracy": 0.7272727272727273,
|
| 2261 |
+
"eval_loss": 0.7825762033462524,
|
| 2262 |
+
"eval_runtime": 4.6558,
|
| 2263 |
+
"eval_samples_per_second": 61.428,
|
| 2264 |
+
"eval_steps_per_second": 3.866,
|
| 2265 |
+
"step": 2079
|
| 2266 |
+
},
|
| 2267 |
+
{
|
| 2268 |
+
"epoch": 141.97,
|
| 2269 |
+
"grad_norm": 1.9007666110992432,
|
| 2270 |
+
"learning_rate": 1.34e-05,
|
| 2271 |
+
"loss": 0.3175,
|
| 2272 |
+
"step": 2094
|
| 2273 |
+
},
|
| 2274 |
+
{
|
| 2275 |
+
"epoch": 141.97,
|
| 2276 |
+
"eval_accuracy": 0.7202797202797203,
|
| 2277 |
+
"eval_loss": 0.7963916659355164,
|
| 2278 |
+
"eval_runtime": 5.2803,
|
| 2279 |
+
"eval_samples_per_second": 54.164,
|
| 2280 |
+
"eval_steps_per_second": 3.409,
|
| 2281 |
+
"step": 2094
|
| 2282 |
+
},
|
| 2283 |
+
{
|
| 2284 |
+
"epoch": 142.98,
|
| 2285 |
+
"grad_norm": 1.8862413167953491,
|
| 2286 |
+
"learning_rate": 1.3257142857142858e-05,
|
| 2287 |
+
"loss": 0.3444,
|
| 2288 |
+
"step": 2109
|
| 2289 |
+
},
|
| 2290 |
+
{
|
| 2291 |
+
"epoch": 142.98,
|
| 2292 |
+
"eval_accuracy": 0.7237762237762237,
|
| 2293 |
+
"eval_loss": 0.7858501672744751,
|
| 2294 |
+
"eval_runtime": 5.4501,
|
| 2295 |
+
"eval_samples_per_second": 52.476,
|
| 2296 |
+
"eval_steps_per_second": 3.303,
|
| 2297 |
+
"step": 2109
|
| 2298 |
+
},
|
| 2299 |
+
{
|
| 2300 |
+
"epoch": 144.0,
|
| 2301 |
+
"grad_norm": 2.4451475143432617,
|
| 2302 |
+
"learning_rate": 1.3114285714285715e-05,
|
| 2303 |
+
"loss": 0.3208,
|
| 2304 |
+
"step": 2124
|
| 2305 |
+
},
|
| 2306 |
+
{
|
| 2307 |
+
"epoch": 144.0,
|
| 2308 |
+
"eval_accuracy": 0.7272727272727273,
|
| 2309 |
+
"eval_loss": 0.7859659194946289,
|
| 2310 |
+
"eval_runtime": 4.5157,
|
| 2311 |
+
"eval_samples_per_second": 63.334,
|
| 2312 |
+
"eval_steps_per_second": 3.986,
|
| 2313 |
+
"step": 2124
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 144.95,
|
| 2317 |
+
"grad_norm": 1.6911462545394897,
|
| 2318 |
+
"learning_rate": 1.298095238095238e-05,
|
| 2319 |
+
"loss": 0.3286,
|
| 2320 |
+
"step": 2138
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 144.95,
|
| 2324 |
+
"eval_accuracy": 0.7272727272727273,
|
| 2325 |
+
"eval_loss": 0.7869133353233337,
|
| 2326 |
+
"eval_runtime": 5.6207,
|
| 2327 |
+
"eval_samples_per_second": 50.883,
|
| 2328 |
+
"eval_steps_per_second": 3.202,
|
| 2329 |
+
"step": 2138
|
| 2330 |
+
},
|
| 2331 |
+
{
|
| 2332 |
+
"epoch": 145.97,
|
| 2333 |
+
"grad_norm": 1.9792907238006592,
|
| 2334 |
+
"learning_rate": 1.2838095238095239e-05,
|
| 2335 |
+
"loss": 0.3319,
|
| 2336 |
+
"step": 2153
|
| 2337 |
+
},
|
| 2338 |
+
{
|
| 2339 |
+
"epoch": 145.97,
|
| 2340 |
+
"eval_accuracy": 0.7167832167832168,
|
| 2341 |
+
"eval_loss": 0.7916193604469299,
|
| 2342 |
+
"eval_runtime": 5.434,
|
| 2343 |
+
"eval_samples_per_second": 52.631,
|
| 2344 |
+
"eval_steps_per_second": 3.312,
|
| 2345 |
+
"step": 2153
|
| 2346 |
+
},
|
| 2347 |
+
{
|
| 2348 |
+
"epoch": 146.98,
|
| 2349 |
+
"grad_norm": 1.3403879404067993,
|
| 2350 |
+
"learning_rate": 1.2695238095238095e-05,
|
| 2351 |
+
"loss": 0.2954,
|
| 2352 |
+
"step": 2168
|
| 2353 |
+
},
|
| 2354 |
+
{
|
| 2355 |
+
"epoch": 146.98,
|
| 2356 |
+
"eval_accuracy": 0.7237762237762237,
|
| 2357 |
+
"eval_loss": 0.7937904000282288,
|
| 2358 |
+
"eval_runtime": 4.5934,
|
| 2359 |
+
"eval_samples_per_second": 62.264,
|
| 2360 |
+
"eval_steps_per_second": 3.919,
|
| 2361 |
+
"step": 2168
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 148.0,
|
| 2365 |
+
"grad_norm": 2.6414599418640137,
|
| 2366 |
+
"learning_rate": 1.2552380952380953e-05,
|
| 2367 |
+
"loss": 0.3283,
|
| 2368 |
+
"step": 2183
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 148.0,
|
| 2372 |
+
"eval_accuracy": 0.7167832167832168,
|
| 2373 |
+
"eval_loss": 0.7974053025245667,
|
| 2374 |
+
"eval_runtime": 4.6622,
|
| 2375 |
+
"eval_samples_per_second": 61.344,
|
| 2376 |
+
"eval_steps_per_second": 3.861,
|
| 2377 |
+
"step": 2183
|
| 2378 |
+
},
|
| 2379 |
+
{
|
| 2380 |
+
"epoch": 148.95,
|
| 2381 |
+
"grad_norm": 1.7404941320419312,
|
| 2382 |
+
"learning_rate": 1.241904761904762e-05,
|
| 2383 |
+
"loss": 0.3306,
|
| 2384 |
+
"step": 2197
|
| 2385 |
+
},
|
| 2386 |
+
{
|
| 2387 |
+
"epoch": 148.95,
|
| 2388 |
+
"eval_accuracy": 0.7307692307692307,
|
| 2389 |
+
"eval_loss": 0.7795438766479492,
|
| 2390 |
+
"eval_runtime": 4.8202,
|
| 2391 |
+
"eval_samples_per_second": 59.334,
|
| 2392 |
+
"eval_steps_per_second": 3.734,
|
| 2393 |
+
"step": 2197
|
| 2394 |
+
},
|
| 2395 |
+
{
|
| 2396 |
+
"epoch": 149.97,
|
| 2397 |
+
"grad_norm": 1.3558377027511597,
|
| 2398 |
+
"learning_rate": 1.2276190476190477e-05,
|
| 2399 |
+
"loss": 0.3073,
|
| 2400 |
+
"step": 2212
|
| 2401 |
+
},
|
| 2402 |
+
{
|
| 2403 |
+
"epoch": 149.97,
|
| 2404 |
+
"eval_accuracy": 0.7202797202797203,
|
| 2405 |
+
"eval_loss": 0.7910019755363464,
|
| 2406 |
+
"eval_runtime": 5.1476,
|
| 2407 |
+
"eval_samples_per_second": 55.56,
|
| 2408 |
+
"eval_steps_per_second": 3.497,
|
| 2409 |
+
"step": 2212
|
| 2410 |
+
},
|
| 2411 |
+
{
|
| 2412 |
+
"epoch": 150.98,
|
| 2413 |
+
"grad_norm": 1.2379992008209229,
|
| 2414 |
+
"learning_rate": 1.2133333333333333e-05,
|
| 2415 |
+
"loss": 0.3089,
|
| 2416 |
+
"step": 2227
|
| 2417 |
+
},
|
| 2418 |
+
{
|
| 2419 |
+
"epoch": 150.98,
|
| 2420 |
+
"eval_accuracy": 0.7202797202797203,
|
| 2421 |
+
"eval_loss": 0.7942177653312683,
|
| 2422 |
+
"eval_runtime": 5.4133,
|
| 2423 |
+
"eval_samples_per_second": 52.833,
|
| 2424 |
+
"eval_steps_per_second": 3.325,
|
| 2425 |
+
"step": 2227
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 152.0,
|
| 2429 |
+
"grad_norm": 1.3452534675598145,
|
| 2430 |
+
"learning_rate": 1.1990476190476191e-05,
|
| 2431 |
+
"loss": 0.2915,
|
| 2432 |
+
"step": 2242
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 152.0,
|
| 2436 |
+
"eval_accuracy": 0.7167832167832168,
|
| 2437 |
+
"eval_loss": 0.7933531403541565,
|
| 2438 |
+
"eval_runtime": 6.3553,
|
| 2439 |
+
"eval_samples_per_second": 45.002,
|
| 2440 |
+
"eval_steps_per_second": 2.832,
|
| 2441 |
+
"step": 2242
|
| 2442 |
+
},
|
| 2443 |
+
{
|
| 2444 |
+
"epoch": 152.95,
|
| 2445 |
+
"grad_norm": 0.9324106574058533,
|
| 2446 |
+
"learning_rate": 1.1857142857142857e-05,
|
| 2447 |
+
"loss": 0.3286,
|
| 2448 |
+
"step": 2256
|
| 2449 |
+
},
|
| 2450 |
+
{
|
| 2451 |
+
"epoch": 152.95,
|
| 2452 |
+
"eval_accuracy": 0.7307692307692307,
|
| 2453 |
+
"eval_loss": 0.7807846069335938,
|
| 2454 |
+
"eval_runtime": 5.094,
|
| 2455 |
+
"eval_samples_per_second": 56.144,
|
| 2456 |
+
"eval_steps_per_second": 3.534,
|
| 2457 |
+
"step": 2256
|
| 2458 |
+
},
|
| 2459 |
+
{
|
| 2460 |
+
"epoch": 153.97,
|
| 2461 |
+
"grad_norm": 1.5133529901504517,
|
| 2462 |
+
"learning_rate": 1.1714285714285715e-05,
|
| 2463 |
+
"loss": 0.2817,
|
| 2464 |
+
"step": 2271
|
| 2465 |
+
},
|
| 2466 |
+
{
|
| 2467 |
+
"epoch": 153.97,
|
| 2468 |
+
"eval_accuracy": 0.7307692307692307,
|
| 2469 |
+
"eval_loss": 0.7788205742835999,
|
| 2470 |
+
"eval_runtime": 5.0531,
|
| 2471 |
+
"eval_samples_per_second": 56.599,
|
| 2472 |
+
"eval_steps_per_second": 3.562,
|
| 2473 |
+
"step": 2271
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 154.98,
|
| 2477 |
+
"grad_norm": 2.8459701538085938,
|
| 2478 |
+
"learning_rate": 1.1571428571428573e-05,
|
| 2479 |
+
"loss": 0.3118,
|
| 2480 |
+
"step": 2286
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 154.98,
|
| 2484 |
+
"eval_accuracy": 0.7272727272727273,
|
| 2485 |
+
"eval_loss": 0.7898407578468323,
|
| 2486 |
+
"eval_runtime": 4.8393,
|
| 2487 |
+
"eval_samples_per_second": 59.1,
|
| 2488 |
+
"eval_steps_per_second": 3.72,
|
| 2489 |
+
"step": 2286
|
| 2490 |
+
},
|
| 2491 |
+
{
|
| 2492 |
+
"epoch": 156.0,
|
| 2493 |
+
"grad_norm": 1.8526194095611572,
|
| 2494 |
+
"learning_rate": 1.1428571428571429e-05,
|
| 2495 |
+
"loss": 0.3155,
|
| 2496 |
+
"step": 2301
|
| 2497 |
+
},
|
| 2498 |
+
{
|
| 2499 |
+
"epoch": 156.0,
|
| 2500 |
+
"eval_accuracy": 0.7202797202797203,
|
| 2501 |
+
"eval_loss": 0.796588122844696,
|
| 2502 |
+
"eval_runtime": 5.8687,
|
| 2503 |
+
"eval_samples_per_second": 48.733,
|
| 2504 |
+
"eval_steps_per_second": 3.067,
|
| 2505 |
+
"step": 2301
|
| 2506 |
+
},
|
| 2507 |
+
{
|
| 2508 |
+
"epoch": 156.95,
|
| 2509 |
+
"grad_norm": 1.130321741104126,
|
| 2510 |
+
"learning_rate": 1.1295238095238096e-05,
|
| 2511 |
+
"loss": 0.3156,
|
| 2512 |
+
"step": 2315
|
| 2513 |
+
},
|
| 2514 |
+
{
|
| 2515 |
+
"epoch": 156.95,
|
| 2516 |
+
"eval_accuracy": 0.7202797202797203,
|
| 2517 |
+
"eval_loss": 0.7947112321853638,
|
| 2518 |
+
"eval_runtime": 4.5714,
|
| 2519 |
+
"eval_samples_per_second": 62.564,
|
| 2520 |
+
"eval_steps_per_second": 3.938,
|
| 2521 |
+
"step": 2315
|
| 2522 |
+
},
|
| 2523 |
+
{
|
| 2524 |
+
"epoch": 157.97,
|
| 2525 |
+
"grad_norm": 1.3919442892074585,
|
| 2526 |
+
"learning_rate": 1.1152380952380953e-05,
|
| 2527 |
+
"loss": 0.2936,
|
| 2528 |
+
"step": 2330
|
| 2529 |
+
},
|
| 2530 |
+
{
|
| 2531 |
+
"epoch": 157.97,
|
| 2532 |
+
"eval_accuracy": 0.7167832167832168,
|
| 2533 |
+
"eval_loss": 0.7916660904884338,
|
| 2534 |
+
"eval_runtime": 4.554,
|
| 2535 |
+
"eval_samples_per_second": 62.802,
|
| 2536 |
+
"eval_steps_per_second": 3.953,
|
| 2537 |
+
"step": 2330
|
| 2538 |
+
},
|
| 2539 |
+
{
|
| 2540 |
+
"epoch": 158.98,
|
| 2541 |
+
"grad_norm": 2.1628525257110596,
|
| 2542 |
+
"learning_rate": 1.1009523809523809e-05,
|
| 2543 |
+
"loss": 0.3049,
|
| 2544 |
+
"step": 2345
|
| 2545 |
+
},
|
| 2546 |
+
{
|
| 2547 |
+
"epoch": 158.98,
|
| 2548 |
+
"eval_accuracy": 0.7307692307692307,
|
| 2549 |
+
"eval_loss": 0.7779849171638489,
|
| 2550 |
+
"eval_runtime": 5.5631,
|
| 2551 |
+
"eval_samples_per_second": 51.41,
|
| 2552 |
+
"eval_steps_per_second": 3.236,
|
| 2553 |
+
"step": 2345
|
| 2554 |
+
},
|
| 2555 |
+
{
|
| 2556 |
+
"epoch": 160.0,
|
| 2557 |
+
"grad_norm": 1.0748810768127441,
|
| 2558 |
+
"learning_rate": 1.0866666666666667e-05,
|
| 2559 |
+
"loss": 0.2896,
|
| 2560 |
+
"step": 2360
|
| 2561 |
+
},
|
| 2562 |
+
{
|
| 2563 |
+
"epoch": 160.0,
|
| 2564 |
+
"eval_accuracy": 0.7272727272727273,
|
| 2565 |
+
"eval_loss": 0.792616605758667,
|
| 2566 |
+
"eval_runtime": 5.3985,
|
| 2567 |
+
"eval_samples_per_second": 52.978,
|
| 2568 |
+
"eval_steps_per_second": 3.334,
|
| 2569 |
+
"step": 2360
|
| 2570 |
+
},
|
| 2571 |
+
{
|
| 2572 |
+
"epoch": 160.95,
|
| 2573 |
+
"grad_norm": 1.9327268600463867,
|
| 2574 |
+
"learning_rate": 1.0733333333333333e-05,
|
| 2575 |
+
"loss": 0.3194,
|
| 2576 |
+
"step": 2374
|
| 2577 |
+
},
|
| 2578 |
+
{
|
| 2579 |
+
"epoch": 160.95,
|
| 2580 |
+
"eval_accuracy": 0.7202797202797203,
|
| 2581 |
+
"eval_loss": 0.8022683262825012,
|
| 2582 |
+
"eval_runtime": 5.5802,
|
| 2583 |
+
"eval_samples_per_second": 51.253,
|
| 2584 |
+
"eval_steps_per_second": 3.226,
|
| 2585 |
+
"step": 2374
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 161.97,
|
| 2589 |
+
"grad_norm": 1.610187292098999,
|
| 2590 |
+
"learning_rate": 1.059047619047619e-05,
|
| 2591 |
+
"loss": 0.2918,
|
| 2592 |
+
"step": 2389
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 161.97,
|
| 2596 |
+
"eval_accuracy": 0.7272727272727273,
|
| 2597 |
+
"eval_loss": 0.7933264374732971,
|
| 2598 |
+
"eval_runtime": 5.0333,
|
| 2599 |
+
"eval_samples_per_second": 56.821,
|
| 2600 |
+
"eval_steps_per_second": 3.576,
|
| 2601 |
+
"step": 2389
|
| 2602 |
+
},
|
| 2603 |
+
{
|
| 2604 |
+
"epoch": 162.98,
|
| 2605 |
+
"grad_norm": 1.961053729057312,
|
| 2606 |
+
"learning_rate": 1.0447619047619049e-05,
|
| 2607 |
+
"loss": 0.2992,
|
| 2608 |
+
"step": 2404
|
| 2609 |
+
},
|
| 2610 |
+
{
|
| 2611 |
+
"epoch": 162.98,
|
| 2612 |
+
"eval_accuracy": 0.7412587412587412,
|
| 2613 |
+
"eval_loss": 0.7828559875488281,
|
| 2614 |
+
"eval_runtime": 5.4812,
|
| 2615 |
+
"eval_samples_per_second": 52.178,
|
| 2616 |
+
"eval_steps_per_second": 3.284,
|
| 2617 |
+
"step": 2404
|
| 2618 |
+
},
|
| 2619 |
+
{
|
| 2620 |
+
"epoch": 164.0,
|
| 2621 |
+
"grad_norm": 1.2120810747146606,
|
| 2622 |
+
"learning_rate": 1.0304761904761905e-05,
|
| 2623 |
+
"loss": 0.3,
|
| 2624 |
+
"step": 2419
|
| 2625 |
+
},
|
| 2626 |
+
{
|
| 2627 |
+
"epoch": 164.0,
|
| 2628 |
+
"eval_accuracy": 0.7202797202797203,
|
| 2629 |
+
"eval_loss": 0.7946493029594421,
|
| 2630 |
+
"eval_runtime": 5.351,
|
| 2631 |
+
"eval_samples_per_second": 53.448,
|
| 2632 |
+
"eval_steps_per_second": 3.364,
|
| 2633 |
+
"step": 2419
|
| 2634 |
+
},
|
| 2635 |
+
{
|
| 2636 |
+
"epoch": 164.95,
|
| 2637 |
+
"grad_norm": 1.112051010131836,
|
| 2638 |
+
"learning_rate": 1.0171428571428571e-05,
|
| 2639 |
+
"loss": 0.322,
|
| 2640 |
+
"step": 2433
|
| 2641 |
+
},
|
| 2642 |
+
{
|
| 2643 |
+
"epoch": 164.95,
|
| 2644 |
+
"eval_accuracy": 0.7272727272727273,
|
| 2645 |
+
"eval_loss": 0.79693204164505,
|
| 2646 |
+
"eval_runtime": 6.2682,
|
| 2647 |
+
"eval_samples_per_second": 45.627,
|
| 2648 |
+
"eval_steps_per_second": 2.872,
|
| 2649 |
+
"step": 2433
|
| 2650 |
+
},
|
| 2651 |
+
{
|
| 2652 |
+
"epoch": 165.97,
|
| 2653 |
+
"grad_norm": 1.2547581195831299,
|
| 2654 |
+
"learning_rate": 1.0028571428571429e-05,
|
| 2655 |
+
"loss": 0.2994,
|
| 2656 |
+
"step": 2448
|
| 2657 |
+
},
|
| 2658 |
+
{
|
| 2659 |
+
"epoch": 165.97,
|
| 2660 |
+
"eval_accuracy": 0.7237762237762237,
|
| 2661 |
+
"eval_loss": 0.8075766563415527,
|
| 2662 |
+
"eval_runtime": 5.4676,
|
| 2663 |
+
"eval_samples_per_second": 52.308,
|
| 2664 |
+
"eval_steps_per_second": 3.292,
|
| 2665 |
+
"step": 2448
|
| 2666 |
+
},
|
| 2667 |
+
{
|
| 2668 |
+
"epoch": 166.98,
|
| 2669 |
+
"grad_norm": 3.3027942180633545,
|
| 2670 |
+
"learning_rate": 9.885714285714285e-06,
|
| 2671 |
+
"loss": 0.2849,
|
| 2672 |
+
"step": 2463
|
| 2673 |
+
},
|
| 2674 |
+
{
|
| 2675 |
+
"epoch": 166.98,
|
| 2676 |
+
"eval_accuracy": 0.7307692307692307,
|
| 2677 |
+
"eval_loss": 0.7950677275657654,
|
| 2678 |
+
"eval_runtime": 5.6445,
|
| 2679 |
+
"eval_samples_per_second": 50.668,
|
| 2680 |
+
"eval_steps_per_second": 3.189,
|
| 2681 |
+
"step": 2463
|
| 2682 |
+
},
|
| 2683 |
+
{
|
| 2684 |
+
"epoch": 168.0,
|
| 2685 |
+
"grad_norm": 3.2351794242858887,
|
| 2686 |
+
"learning_rate": 9.742857142857143e-06,
|
| 2687 |
+
"loss": 0.2745,
|
| 2688 |
+
"step": 2478
|
| 2689 |
+
},
|
| 2690 |
+
{
|
| 2691 |
+
"epoch": 168.0,
|
| 2692 |
+
"eval_accuracy": 0.7342657342657343,
|
| 2693 |
+
"eval_loss": 0.7892395853996277,
|
| 2694 |
+
"eval_runtime": 6.6609,
|
| 2695 |
+
"eval_samples_per_second": 42.937,
|
| 2696 |
+
"eval_steps_per_second": 2.702,
|
| 2697 |
+
"step": 2478
|
| 2698 |
+
},
|
| 2699 |
+
{
|
| 2700 |
+
"epoch": 168.95,
|
| 2701 |
+
"grad_norm": 2.2650949954986572,
|
| 2702 |
+
"learning_rate": 9.60952380952381e-06,
|
| 2703 |
+
"loss": 0.2974,
|
| 2704 |
+
"step": 2492
|
| 2705 |
+
},
|
| 2706 |
+
{
|
| 2707 |
+
"epoch": 168.95,
|
| 2708 |
+
"eval_accuracy": 0.7307692307692307,
|
| 2709 |
+
"eval_loss": 0.791623055934906,
|
| 2710 |
+
"eval_runtime": 5.8468,
|
| 2711 |
+
"eval_samples_per_second": 48.916,
|
| 2712 |
+
"eval_steps_per_second": 3.079,
|
| 2713 |
+
"step": 2492
|
| 2714 |
+
},
|
| 2715 |
+
{
|
| 2716 |
+
"epoch": 169.97,
|
| 2717 |
+
"grad_norm": 1.3676173686981201,
|
| 2718 |
+
"learning_rate": 9.466666666666667e-06,
|
| 2719 |
+
"loss": 0.2656,
|
| 2720 |
+
"step": 2507
|
| 2721 |
+
},
|
| 2722 |
+
{
|
| 2723 |
+
"epoch": 169.97,
|
| 2724 |
+
"eval_accuracy": 0.7342657342657343,
|
| 2725 |
+
"eval_loss": 0.7994617223739624,
|
| 2726 |
+
"eval_runtime": 4.7953,
|
| 2727 |
+
"eval_samples_per_second": 59.642,
|
| 2728 |
+
"eval_steps_per_second": 3.754,
|
| 2729 |
+
"step": 2507
|
| 2730 |
+
},
|
| 2731 |
+
{
|
| 2732 |
+
"epoch": 170.98,
|
| 2733 |
+
"grad_norm": 1.5123904943466187,
|
| 2734 |
+
"learning_rate": 9.323809523809523e-06,
|
| 2735 |
+
"loss": 0.295,
|
| 2736 |
+
"step": 2522
|
| 2737 |
+
},
|
| 2738 |
+
{
|
| 2739 |
+
"epoch": 170.98,
|
| 2740 |
+
"eval_accuracy": 0.7237762237762237,
|
| 2741 |
+
"eval_loss": 0.802618682384491,
|
| 2742 |
+
"eval_runtime": 6.3216,
|
| 2743 |
+
"eval_samples_per_second": 45.241,
|
| 2744 |
+
"eval_steps_per_second": 2.847,
|
| 2745 |
+
"step": 2522
|
| 2746 |
+
},
|
| 2747 |
+
{
|
| 2748 |
+
"epoch": 172.0,
|
| 2749 |
+
"grad_norm": 1.5210902690887451,
|
| 2750 |
+
"learning_rate": 9.180952380952381e-06,
|
| 2751 |
+
"loss": 0.2791,
|
| 2752 |
+
"step": 2537
|
| 2753 |
+
},
|
| 2754 |
+
{
|
| 2755 |
+
"epoch": 172.0,
|
| 2756 |
+
"eval_accuracy": 0.7342657342657343,
|
| 2757 |
+
"eval_loss": 0.797250509262085,
|
| 2758 |
+
"eval_runtime": 5.4447,
|
| 2759 |
+
"eval_samples_per_second": 52.528,
|
| 2760 |
+
"eval_steps_per_second": 3.306,
|
| 2761 |
+
"step": 2537
|
| 2762 |
+
},
|
| 2763 |
+
{
|
| 2764 |
+
"epoch": 172.95,
|
| 2765 |
+
"grad_norm": 1.0823432207107544,
|
| 2766 |
+
"learning_rate": 9.047619047619047e-06,
|
| 2767 |
+
"loss": 0.2836,
|
| 2768 |
+
"step": 2551
|
| 2769 |
+
},
|
| 2770 |
+
{
|
| 2771 |
+
"epoch": 172.95,
|
| 2772 |
+
"eval_accuracy": 0.7307692307692307,
|
| 2773 |
+
"eval_loss": 0.8022569417953491,
|
| 2774 |
+
"eval_runtime": 5.0556,
|
| 2775 |
+
"eval_samples_per_second": 56.571,
|
| 2776 |
+
"eval_steps_per_second": 3.56,
|
| 2777 |
+
"step": 2551
|
| 2778 |
+
},
|
| 2779 |
+
{
|
| 2780 |
+
"epoch": 173.97,
|
| 2781 |
+
"grad_norm": 1.4100669622421265,
|
| 2782 |
+
"learning_rate": 8.904761904761905e-06,
|
| 2783 |
+
"loss": 0.2806,
|
| 2784 |
+
"step": 2566
|
| 2785 |
+
},
|
| 2786 |
+
{
|
| 2787 |
+
"epoch": 173.97,
|
| 2788 |
+
"eval_accuracy": 0.7307692307692307,
|
| 2789 |
+
"eval_loss": 0.8013490438461304,
|
| 2790 |
+
"eval_runtime": 5.503,
|
| 2791 |
+
"eval_samples_per_second": 51.972,
|
| 2792 |
+
"eval_steps_per_second": 3.271,
|
| 2793 |
+
"step": 2566
|
| 2794 |
+
},
|
| 2795 |
+
{
|
| 2796 |
+
"epoch": 174.98,
|
| 2797 |
+
"grad_norm": 1.389672875404358,
|
| 2798 |
+
"learning_rate": 8.761904761904763e-06,
|
| 2799 |
+
"loss": 0.2661,
|
| 2800 |
+
"step": 2581
|
| 2801 |
+
},
|
| 2802 |
+
{
|
| 2803 |
+
"epoch": 174.98,
|
| 2804 |
+
"eval_accuracy": 0.7307692307692307,
|
| 2805 |
+
"eval_loss": 0.796493649482727,
|
| 2806 |
+
"eval_runtime": 5.7105,
|
| 2807 |
+
"eval_samples_per_second": 50.083,
|
| 2808 |
+
"eval_steps_per_second": 3.152,
|
| 2809 |
+
"step": 2581
|
| 2810 |
+
},
|
| 2811 |
+
{
|
| 2812 |
+
"epoch": 176.0,
|
| 2813 |
+
"grad_norm": 1.8702772855758667,
|
| 2814 |
+
"learning_rate": 8.61904761904762e-06,
|
| 2815 |
+
"loss": 0.2695,
|
| 2816 |
+
"step": 2596
|
| 2817 |
+
},
|
| 2818 |
+
{
|
| 2819 |
+
"epoch": 176.0,
|
| 2820 |
+
"eval_accuracy": 0.7272727272727273,
|
| 2821 |
+
"eval_loss": 0.8062567114830017,
|
| 2822 |
+
"eval_runtime": 7.451,
|
| 2823 |
+
"eval_samples_per_second": 38.384,
|
| 2824 |
+
"eval_steps_per_second": 2.416,
|
| 2825 |
+
"step": 2596
|
| 2826 |
+
},
|
| 2827 |
+
{
|
| 2828 |
+
"epoch": 176.95,
|
| 2829 |
+
"grad_norm": 3.131314992904663,
|
| 2830 |
+
"learning_rate": 8.485714285714287e-06,
|
| 2831 |
+
"loss": 0.286,
|
| 2832 |
+
"step": 2610
|
| 2833 |
+
},
|
| 2834 |
+
{
|
| 2835 |
+
"epoch": 176.95,
|
| 2836 |
+
"eval_accuracy": 0.7237762237762237,
|
| 2837 |
+
"eval_loss": 0.7963144779205322,
|
| 2838 |
+
"eval_runtime": 4.431,
|
| 2839 |
+
"eval_samples_per_second": 64.546,
|
| 2840 |
+
"eval_steps_per_second": 4.062,
|
| 2841 |
+
"step": 2610
|
| 2842 |
+
},
|
| 2843 |
+
{
|
| 2844 |
+
"epoch": 177.97,
|
| 2845 |
+
"grad_norm": 1.7147862911224365,
|
| 2846 |
+
"learning_rate": 8.342857142857143e-06,
|
| 2847 |
+
"loss": 0.2743,
|
| 2848 |
+
"step": 2625
|
| 2849 |
+
},
|
| 2850 |
+
{
|
| 2851 |
+
"epoch": 177.97,
|
| 2852 |
+
"eval_accuracy": 0.7412587412587412,
|
| 2853 |
+
"eval_loss": 0.7928534150123596,
|
| 2854 |
+
"eval_runtime": 5.7357,
|
| 2855 |
+
"eval_samples_per_second": 49.863,
|
| 2856 |
+
"eval_steps_per_second": 3.138,
|
| 2857 |
+
"step": 2625
|
| 2858 |
+
},
|
| 2859 |
+
{
|
| 2860 |
+
"epoch": 178.98,
|
| 2861 |
+
"grad_norm": 1.669243574142456,
|
| 2862 |
+
"learning_rate": 8.2e-06,
|
| 2863 |
+
"loss": 0.2775,
|
| 2864 |
+
"step": 2640
|
| 2865 |
+
},
|
| 2866 |
+
{
|
| 2867 |
+
"epoch": 178.98,
|
| 2868 |
+
"eval_accuracy": 0.7447552447552448,
|
| 2869 |
+
"eval_loss": 0.7855107188224792,
|
| 2870 |
+
"eval_runtime": 5.7848,
|
| 2871 |
+
"eval_samples_per_second": 49.44,
|
| 2872 |
+
"eval_steps_per_second": 3.112,
|
| 2873 |
+
"step": 2640
|
| 2874 |
+
}
|
| 2875 |
+
],
|
| 2876 |
+
"logging_steps": 100.0,
|
| 2877 |
+
"max_steps": 3500,
|
| 2878 |
+
"num_input_tokens_seen": 0,
|
| 2879 |
+
"num_train_epochs": 250,
|
| 2880 |
+
"save_steps": 500,
|
| 2881 |
+
"total_flos": 1.462758962872032e+18,
|
| 2882 |
+
"train_batch_size": 16,
|
| 2883 |
+
"trial_name": null,
|
| 2884 |
+
"trial_params": null
|
| 2885 |
+
}
|
checkpoint-2640/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c8e61b5784e26fd2ba4a1e9b13459cd23e4f0cbfef90398cd9d89f4c292d5ef2
|
| 3 |
+
size 4984
|
checkpoint-3500/config.json
ADDED
|
@@ -0,0 +1,124 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "anderloh/Hugginhface-master-wav2vec-pretreined-5-class-train-test",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"adapter_attn_dim": null,
|
| 5 |
+
"adapter_kernel_size": 3,
|
| 6 |
+
"adapter_stride": 2,
|
| 7 |
+
"add_adapter": false,
|
| 8 |
+
"apply_spec_augment": true,
|
| 9 |
+
"architectures": [
|
| 10 |
+
"Wav2Vec2ForSequenceClassification"
|
| 11 |
+
],
|
| 12 |
+
"attention_dropout": 0.0,
|
| 13 |
+
"bos_token_id": 1,
|
| 14 |
+
"classifier_proj_size": 128,
|
| 15 |
+
"codevector_dim": 128,
|
| 16 |
+
"contrastive_logits_temperature": 0.1,
|
| 17 |
+
"conv_bias": true,
|
| 18 |
+
"conv_dim": [
|
| 19 |
+
256,
|
| 20 |
+
256,
|
| 21 |
+
256,
|
| 22 |
+
256,
|
| 23 |
+
256,
|
| 24 |
+
256,
|
| 25 |
+
256
|
| 26 |
+
],
|
| 27 |
+
"conv_kernel": [
|
| 28 |
+
10,
|
| 29 |
+
3,
|
| 30 |
+
3,
|
| 31 |
+
3,
|
| 32 |
+
3,
|
| 33 |
+
2,
|
| 34 |
+
2
|
| 35 |
+
],
|
| 36 |
+
"conv_stride": [
|
| 37 |
+
5,
|
| 38 |
+
2,
|
| 39 |
+
2,
|
| 40 |
+
2,
|
| 41 |
+
2,
|
| 42 |
+
2,
|
| 43 |
+
2
|
| 44 |
+
],
|
| 45 |
+
"ctc_loss_reduction": "sum",
|
| 46 |
+
"ctc_zero_infinity": false,
|
| 47 |
+
"diversity_loss_weight": 0.1,
|
| 48 |
+
"do_stable_layer_norm": true,
|
| 49 |
+
"eos_token_id": 2,
|
| 50 |
+
"feat_extract_activation": "gelu",
|
| 51 |
+
"feat_extract_dropout": 0.0,
|
| 52 |
+
"feat_extract_norm": "layer",
|
| 53 |
+
"feat_proj_dropout": 0.0,
|
| 54 |
+
"feat_quantizer_dropout": 0.0,
|
| 55 |
+
"final_dropout": 0.0,
|
| 56 |
+
"finetuning_task": "audio-classification",
|
| 57 |
+
"hidden_act": "gelu",
|
| 58 |
+
"hidden_dropout": 0.0,
|
| 59 |
+
"hidden_dropout_prob": 0.0,
|
| 60 |
+
"hidden_size": 384,
|
| 61 |
+
"id2label": {
|
| 62 |
+
"0": "Helicopter",
|
| 63 |
+
"1": "Jet",
|
| 64 |
+
"2": "Racecar",
|
| 65 |
+
"3": "Rail",
|
| 66 |
+
"4": "Truck"
|
| 67 |
+
},
|
| 68 |
+
"initializer_range": 0.02,
|
| 69 |
+
"intermediate_size": 1536,
|
| 70 |
+
"label2id": {
|
| 71 |
+
"Helicopter": "0",
|
| 72 |
+
"Jet": "1",
|
| 73 |
+
"Racecar": "2",
|
| 74 |
+
"Rail": "3",
|
| 75 |
+
"Truck": "4"
|
| 76 |
+
},
|
| 77 |
+
"layer_norm_eps": 1e-05,
|
| 78 |
+
"layerdrop": 0.0,
|
| 79 |
+
"mask_feature_length": 10,
|
| 80 |
+
"mask_feature_min_masks": 0,
|
| 81 |
+
"mask_feature_prob": 0.0,
|
| 82 |
+
"mask_time_length": 10,
|
| 83 |
+
"mask_time_min_masks": 2,
|
| 84 |
+
"mask_time_prob": 0.65,
|
| 85 |
+
"model_type": "wav2vec2",
|
| 86 |
+
"num_adapter_layers": 3,
|
| 87 |
+
"num_attention_heads": 6,
|
| 88 |
+
"num_codevector_groups": 2,
|
| 89 |
+
"num_codevectors_per_group": 320,
|
| 90 |
+
"num_conv_pos_embedding_groups": 16,
|
| 91 |
+
"num_conv_pos_embeddings": 128,
|
| 92 |
+
"num_feat_extract_layers": 7,
|
| 93 |
+
"num_hidden_layers": 6,
|
| 94 |
+
"num_negatives": 100,
|
| 95 |
+
"output_hidden_size": 384,
|
| 96 |
+
"pad_token_id": 0,
|
| 97 |
+
"proj_codevector_dim": 128,
|
| 98 |
+
"tdnn_dilation": [
|
| 99 |
+
1,
|
| 100 |
+
2,
|
| 101 |
+
3,
|
| 102 |
+
1,
|
| 103 |
+
1
|
| 104 |
+
],
|
| 105 |
+
"tdnn_dim": [
|
| 106 |
+
512,
|
| 107 |
+
512,
|
| 108 |
+
512,
|
| 109 |
+
512,
|
| 110 |
+
1500
|
| 111 |
+
],
|
| 112 |
+
"tdnn_kernel": [
|
| 113 |
+
5,
|
| 114 |
+
3,
|
| 115 |
+
3,
|
| 116 |
+
1,
|
| 117 |
+
1
|
| 118 |
+
],
|
| 119 |
+
"torch_dtype": "float32",
|
| 120 |
+
"transformers_version": "4.39.0.dev0",
|
| 121 |
+
"use_weighted_layer_sum": false,
|
| 122 |
+
"vocab_size": 32,
|
| 123 |
+
"xvector_output_dim": 512
|
| 124 |
+
}
|
checkpoint-3500/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ff22d8f137bb8dd8d21ac2f57a06aab8bbc10a3bc8a3a7db906e3f3e7726af0c
|
| 3 |
+
size 52151348
|
checkpoint-3500/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c918fc70557b60472fb809450afa8204f0203463432e29e898d3c30ce950f8c7
|
| 3 |
+
size 95909946
|
checkpoint-3500/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": 0.0,
|
| 7 |
+
"return_attention_mask": true,
|
| 8 |
+
"sampling_rate": 16000
|
| 9 |
+
}
|
checkpoint-3500/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:08dd2f651567df4397a2d765332714e405de45e03bef19981c7131590b4a203f
|
| 3 |
+
size 14244
|
checkpoint-3500/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4173d44856e19fda212bc7802a74d227d79cfb2f8b1153c12a16fe285b9932b2
|
| 3 |
+
size 1064
|
checkpoint-3500/trainer_state.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-3500/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c8e61b5784e26fd2ba4a1e9b13459cd23e4f0cbfef90398cd9d89f4c292d5ef2
|
| 3 |
+
size 4984
|
config.json
ADDED
|
@@ -0,0 +1,124 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "anderloh/Hugginhface-master-wav2vec-pretreined-5-class-train-test",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"adapter_attn_dim": null,
|
| 5 |
+
"adapter_kernel_size": 3,
|
| 6 |
+
"adapter_stride": 2,
|
| 7 |
+
"add_adapter": false,
|
| 8 |
+
"apply_spec_augment": true,
|
| 9 |
+
"architectures": [
|
| 10 |
+
"Wav2Vec2ForSequenceClassification"
|
| 11 |
+
],
|
| 12 |
+
"attention_dropout": 0.0,
|
| 13 |
+
"bos_token_id": 1,
|
| 14 |
+
"classifier_proj_size": 128,
|
| 15 |
+
"codevector_dim": 128,
|
| 16 |
+
"contrastive_logits_temperature": 0.1,
|
| 17 |
+
"conv_bias": true,
|
| 18 |
+
"conv_dim": [
|
| 19 |
+
256,
|
| 20 |
+
256,
|
| 21 |
+
256,
|
| 22 |
+
256,
|
| 23 |
+
256,
|
| 24 |
+
256,
|
| 25 |
+
256
|
| 26 |
+
],
|
| 27 |
+
"conv_kernel": [
|
| 28 |
+
10,
|
| 29 |
+
3,
|
| 30 |
+
3,
|
| 31 |
+
3,
|
| 32 |
+
3,
|
| 33 |
+
2,
|
| 34 |
+
2
|
| 35 |
+
],
|
| 36 |
+
"conv_stride": [
|
| 37 |
+
5,
|
| 38 |
+
2,
|
| 39 |
+
2,
|
| 40 |
+
2,
|
| 41 |
+
2,
|
| 42 |
+
2,
|
| 43 |
+
2
|
| 44 |
+
],
|
| 45 |
+
"ctc_loss_reduction": "sum",
|
| 46 |
+
"ctc_zero_infinity": false,
|
| 47 |
+
"diversity_loss_weight": 0.1,
|
| 48 |
+
"do_stable_layer_norm": true,
|
| 49 |
+
"eos_token_id": 2,
|
| 50 |
+
"feat_extract_activation": "gelu",
|
| 51 |
+
"feat_extract_dropout": 0.0,
|
| 52 |
+
"feat_extract_norm": "layer",
|
| 53 |
+
"feat_proj_dropout": 0.0,
|
| 54 |
+
"feat_quantizer_dropout": 0.0,
|
| 55 |
+
"final_dropout": 0.0,
|
| 56 |
+
"finetuning_task": "audio-classification",
|
| 57 |
+
"hidden_act": "gelu",
|
| 58 |
+
"hidden_dropout": 0.0,
|
| 59 |
+
"hidden_dropout_prob": 0.0,
|
| 60 |
+
"hidden_size": 384,
|
| 61 |
+
"id2label": {
|
| 62 |
+
"0": "Helicopter",
|
| 63 |
+
"1": "Jet",
|
| 64 |
+
"2": "Racecar",
|
| 65 |
+
"3": "Rail",
|
| 66 |
+
"4": "Truck"
|
| 67 |
+
},
|
| 68 |
+
"initializer_range": 0.02,
|
| 69 |
+
"intermediate_size": 1536,
|
| 70 |
+
"label2id": {
|
| 71 |
+
"Helicopter": "0",
|
| 72 |
+
"Jet": "1",
|
| 73 |
+
"Racecar": "2",
|
| 74 |
+
"Rail": "3",
|
| 75 |
+
"Truck": "4"
|
| 76 |
+
},
|
| 77 |
+
"layer_norm_eps": 1e-05,
|
| 78 |
+
"layerdrop": 0.0,
|
| 79 |
+
"mask_feature_length": 10,
|
| 80 |
+
"mask_feature_min_masks": 0,
|
| 81 |
+
"mask_feature_prob": 0.0,
|
| 82 |
+
"mask_time_length": 10,
|
| 83 |
+
"mask_time_min_masks": 2,
|
| 84 |
+
"mask_time_prob": 0.65,
|
| 85 |
+
"model_type": "wav2vec2",
|
| 86 |
+
"num_adapter_layers": 3,
|
| 87 |
+
"num_attention_heads": 6,
|
| 88 |
+
"num_codevector_groups": 2,
|
| 89 |
+
"num_codevectors_per_group": 320,
|
| 90 |
+
"num_conv_pos_embedding_groups": 16,
|
| 91 |
+
"num_conv_pos_embeddings": 128,
|
| 92 |
+
"num_feat_extract_layers": 7,
|
| 93 |
+
"num_hidden_layers": 6,
|
| 94 |
+
"num_negatives": 100,
|
| 95 |
+
"output_hidden_size": 384,
|
| 96 |
+
"pad_token_id": 0,
|
| 97 |
+
"proj_codevector_dim": 128,
|
| 98 |
+
"tdnn_dilation": [
|
| 99 |
+
1,
|
| 100 |
+
2,
|
| 101 |
+
3,
|
| 102 |
+
1,
|
| 103 |
+
1
|
| 104 |
+
],
|
| 105 |
+
"tdnn_dim": [
|
| 106 |
+
512,
|
| 107 |
+
512,
|
| 108 |
+
512,
|
| 109 |
+
512,
|
| 110 |
+
1500
|
| 111 |
+
],
|
| 112 |
+
"tdnn_kernel": [
|
| 113 |
+
5,
|
| 114 |
+
3,
|
| 115 |
+
3,
|
| 116 |
+
1,
|
| 117 |
+
1
|
| 118 |
+
],
|
| 119 |
+
"torch_dtype": "float32",
|
| 120 |
+
"transformers_version": "4.39.0.dev0",
|
| 121 |
+
"use_weighted_layer_sum": false,
|
| 122 |
+
"vocab_size": 32,
|
| 123 |
+
"xvector_output_dim": 512
|
| 124 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 237.29,
|
| 3 |
+
"eval_accuracy": 0.7447552447552448,
|
| 4 |
+
"eval_loss": 0.7855107188224792,
|
| 5 |
+
"eval_runtime": 4.8875,
|
| 6 |
+
"eval_samples_per_second": 58.516,
|
| 7 |
+
"eval_steps_per_second": 3.683
|
| 8 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:51779c90f182773c0eccdebb31d61e1a81936b73024a5badb5719741abedadfe
|
| 3 |
+
size 52151348
|
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": 0.0,
|
| 7 |
+
"return_attention_mask": true,
|
| 8 |
+
"sampling_rate": 16000
|
| 9 |
+
}
|
runs/May04_12-45-57_ml6.hpc.uio.no/events.out.tfevents.1714819603.ml6.hpc.uio.no.743903.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9f7db582c2e36ad8de85945baf87af4cae5a99dcd2d8d7038bddc94d62c53990
|
| 3 |
+
size 19365
|
runs/May04_12-54-19_ml6.hpc.uio.no/events.out.tfevents.1714820073.ml6.hpc.uio.no.746334.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:40d9a593208090380306aaa91903221326bc2d24a2f97a915f039acc3c06cf94
|
| 3 |
+
size 133785
|
runs/May04_12-54-19_ml6.hpc.uio.no/events.out.tfevents.1714823730.ml6.hpc.uio.no.746334.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:489edc412986ee158acb8c9ccb695e0aa533588586747de0349f1607eddf14e9
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 237.29,
|
| 3 |
+
"train_loss": 0.5242921064240592,
|
| 4 |
+
"train_runtime": 3651.9597,
|
| 5 |
+
"train_samples_per_second": 63.87,
|
| 6 |
+
"train_steps_per_second": 0.958
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c8e61b5784e26fd2ba4a1e9b13459cd23e4f0cbfef90398cd9d89f4c292d5ef2
|
| 3 |
+
size 4984
|