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
|