Uploaded PeptideGPT hemolytic model
Browse files- hemolytic/.DS_Store +0 -0
- hemolytic/README.md +259 -0
- hemolytic/all_results.json +15 -0
- hemolytic/config.json +39 -0
- hemolytic/eval_results.json +10 -0
- hemolytic/events.out.tfevents.1707888408.lambda-a6000.3897798.0 +3 -0
- hemolytic/generation_config.json +6 -0
- hemolytic/merges.txt +0 -0
- hemolytic/model.safetensors +3 -0
- hemolytic/special_tokens_map.json +23 -0
- hemolytic/tokenizer.json +0 -0
- hemolytic/tokenizer_config.json +19 -0
- hemolytic/train_results.json +8 -0
- hemolytic/trainer_state.json +3030 -0
- hemolytic/training_args.bin +3 -0
- hemolytic/vocab.json +0 -0
hemolytic/.DS_Store
ADDED
Binary file (8.2 kB). View file
|
|
hemolytic/README.md
ADDED
@@ -0,0 +1,259 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
base_model: nferruz/ProtGPT2
|
4 |
+
tags:
|
5 |
+
- generated_from_trainer
|
6 |
+
metrics:
|
7 |
+
- accuracy
|
8 |
+
model-index:
|
9 |
+
- name: output_hemo_aug_4
|
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 |
+
# output_hemo_aug_4
|
17 |
+
|
18 |
+
This model is a fine-tuned version of [nferruz/ProtGPT2](https://huggingface.co/nferruz/ProtGPT2) on an unknown dataset.
|
19 |
+
It achieves the following results on the evaluation set:
|
20 |
+
- Loss: 3.9627
|
21 |
+
- Accuracy: 0.3978
|
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: 1e-05
|
41 |
+
- train_batch_size: 2
|
42 |
+
- eval_batch_size: 16
|
43 |
+
- seed: 42
|
44 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
45 |
+
- lr_scheduler_type: linear
|
46 |
+
- num_epochs: 200.0
|
47 |
+
|
48 |
+
### Training results
|
49 |
+
|
50 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
51 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
52 |
+
| 9.4301 | 1.0 | 6 | 8.6845 | 0.0244 |
|
53 |
+
| 8.4172 | 2.0 | 12 | 7.8149 | 0.0811 |
|
54 |
+
| 7.6869 | 3.0 | 18 | 7.2012 | 0.1554 |
|
55 |
+
| 7.1731 | 4.0 | 24 | 6.9139 | 0.1828 |
|
56 |
+
| 6.8807 | 5.0 | 30 | 6.6238 | 0.1955 |
|
57 |
+
| 6.6009 | 6.0 | 36 | 6.3847 | 0.1935 |
|
58 |
+
| 6.4347 | 7.0 | 42 | 6.2341 | 0.2063 |
|
59 |
+
| 6.2831 | 8.0 | 48 | 6.0964 | 0.2141 |
|
60 |
+
| 6.1728 | 9.0 | 54 | 5.9864 | 0.2209 |
|
61 |
+
| 6.0805 | 10.0 | 60 | 5.8936 | 0.2317 |
|
62 |
+
| 5.9959 | 11.0 | 66 | 5.8161 | 0.2405 |
|
63 |
+
| 5.925 | 12.0 | 72 | 5.7456 | 0.2385 |
|
64 |
+
| 5.8787 | 13.0 | 78 | 5.6646 | 0.2483 |
|
65 |
+
| 5.7996 | 14.0 | 84 | 5.5901 | 0.2493 |
|
66 |
+
| 5.7312 | 15.0 | 90 | 5.5216 | 0.2532 |
|
67 |
+
| 5.6751 | 16.0 | 96 | 5.4695 | 0.2590 |
|
68 |
+
| 5.6076 | 17.0 | 102 | 5.4216 | 0.2620 |
|
69 |
+
| 5.569 | 18.0 | 108 | 5.3735 | 0.2620 |
|
70 |
+
| 5.5037 | 19.0 | 114 | 5.3272 | 0.2630 |
|
71 |
+
| 5.4681 | 20.0 | 120 | 5.2856 | 0.2698 |
|
72 |
+
| 5.4225 | 21.0 | 126 | 5.2559 | 0.2717 |
|
73 |
+
| 5.3805 | 22.0 | 132 | 5.2126 | 0.2766 |
|
74 |
+
| 5.3527 | 23.0 | 138 | 5.1884 | 0.2757 |
|
75 |
+
| 5.3033 | 24.0 | 144 | 5.1539 | 0.2796 |
|
76 |
+
| 5.2635 | 25.0 | 150 | 5.1110 | 0.2854 |
|
77 |
+
| 5.2411 | 26.0 | 156 | 5.0882 | 0.2854 |
|
78 |
+
| 5.1972 | 27.0 | 162 | 5.0575 | 0.2903 |
|
79 |
+
| 5.163 | 28.0 | 168 | 5.0293 | 0.2913 |
|
80 |
+
| 5.1273 | 29.0 | 174 | 5.0047 | 0.2903 |
|
81 |
+
| 5.1032 | 30.0 | 180 | 4.9817 | 0.2952 |
|
82 |
+
| 5.0726 | 31.0 | 186 | 4.9583 | 0.2952 |
|
83 |
+
| 5.0405 | 32.0 | 192 | 4.9355 | 0.2952 |
|
84 |
+
| 5.007 | 33.0 | 198 | 4.9184 | 0.2952 |
|
85 |
+
| 4.9897 | 34.0 | 204 | 4.8911 | 0.2972 |
|
86 |
+
| 4.9416 | 35.0 | 210 | 4.8628 | 0.2972 |
|
87 |
+
| 4.9245 | 36.0 | 216 | 4.8499 | 0.2981 |
|
88 |
+
| 4.901 | 37.0 | 222 | 4.8263 | 0.3030 |
|
89 |
+
| 4.8713 | 38.0 | 228 | 4.8035 | 0.3030 |
|
90 |
+
| 4.845 | 39.0 | 234 | 4.7874 | 0.3060 |
|
91 |
+
| 4.8052 | 40.0 | 240 | 4.7535 | 0.3040 |
|
92 |
+
| 4.7786 | 41.0 | 246 | 4.7313 | 0.3060 |
|
93 |
+
| 4.7501 | 42.0 | 252 | 4.7175 | 0.3089 |
|
94 |
+
| 4.7221 | 43.0 | 258 | 4.6978 | 0.3118 |
|
95 |
+
| 4.7038 | 44.0 | 264 | 4.6785 | 0.3109 |
|
96 |
+
| 4.681 | 45.0 | 270 | 4.6661 | 0.3128 |
|
97 |
+
| 4.6566 | 46.0 | 276 | 4.6532 | 0.3157 |
|
98 |
+
| 4.632 | 47.0 | 282 | 4.6361 | 0.3157 |
|
99 |
+
| 4.618 | 48.0 | 288 | 4.6162 | 0.3196 |
|
100 |
+
| 4.5928 | 49.0 | 294 | 4.5987 | 0.3245 |
|
101 |
+
| 4.5716 | 50.0 | 300 | 4.5848 | 0.3216 |
|
102 |
+
| 4.5485 | 51.0 | 306 | 4.5721 | 0.3245 |
|
103 |
+
| 4.5324 | 52.0 | 312 | 4.5579 | 0.3196 |
|
104 |
+
| 4.5038 | 53.0 | 318 | 4.5423 | 0.3196 |
|
105 |
+
| 4.4831 | 54.0 | 324 | 4.5240 | 0.3265 |
|
106 |
+
| 4.4347 | 55.0 | 330 | 4.5087 | 0.3255 |
|
107 |
+
| 4.4218 | 56.0 | 336 | 4.4850 | 0.3255 |
|
108 |
+
| 4.3939 | 57.0 | 342 | 4.4791 | 0.3275 |
|
109 |
+
| 4.3766 | 58.0 | 348 | 4.4640 | 0.3265 |
|
110 |
+
| 4.3472 | 59.0 | 354 | 4.4471 | 0.3275 |
|
111 |
+
| 4.3241 | 60.0 | 360 | 4.4334 | 0.3275 |
|
112 |
+
| 4.2919 | 61.0 | 366 | 4.4296 | 0.3304 |
|
113 |
+
| 4.2678 | 62.0 | 372 | 4.4281 | 0.3343 |
|
114 |
+
| 4.2515 | 63.0 | 378 | 4.4120 | 0.3372 |
|
115 |
+
| 4.2244 | 64.0 | 384 | 4.4038 | 0.3343 |
|
116 |
+
| 4.2129 | 65.0 | 390 | 4.3826 | 0.3392 |
|
117 |
+
| 4.1882 | 66.0 | 396 | 4.3834 | 0.3372 |
|
118 |
+
| 4.1503 | 67.0 | 402 | 4.3738 | 0.3372 |
|
119 |
+
| 4.1398 | 68.0 | 408 | 4.3596 | 0.3372 |
|
120 |
+
| 4.115 | 69.0 | 414 | 4.3376 | 0.3412 |
|
121 |
+
| 4.1052 | 70.0 | 420 | 4.3330 | 0.3412 |
|
122 |
+
| 4.0932 | 71.0 | 426 | 4.3295 | 0.3412 |
|
123 |
+
| 4.0573 | 72.0 | 432 | 4.3111 | 0.3412 |
|
124 |
+
| 4.0449 | 73.0 | 438 | 4.3048 | 0.3441 |
|
125 |
+
| 4.0165 | 74.0 | 444 | 4.2936 | 0.3460 |
|
126 |
+
| 3.9936 | 75.0 | 450 | 4.2815 | 0.3509 |
|
127 |
+
| 3.967 | 76.0 | 456 | 4.2686 | 0.3539 |
|
128 |
+
| 3.9524 | 77.0 | 462 | 4.2697 | 0.3509 |
|
129 |
+
| 3.9287 | 78.0 | 468 | 4.2546 | 0.3529 |
|
130 |
+
| 3.9092 | 79.0 | 474 | 4.2484 | 0.3539 |
|
131 |
+
| 3.8907 | 80.0 | 480 | 4.2420 | 0.3539 |
|
132 |
+
| 3.8704 | 81.0 | 486 | 4.2418 | 0.3529 |
|
133 |
+
| 3.8499 | 82.0 | 492 | 4.2265 | 0.3548 |
|
134 |
+
| 3.8325 | 83.0 | 498 | 4.2089 | 0.3548 |
|
135 |
+
| 3.8024 | 84.0 | 504 | 4.2058 | 0.3568 |
|
136 |
+
| 3.8058 | 85.0 | 510 | 4.2039 | 0.3558 |
|
137 |
+
| 3.7888 | 86.0 | 516 | 4.1906 | 0.3578 |
|
138 |
+
| 3.7622 | 87.0 | 522 | 4.1792 | 0.3617 |
|
139 |
+
| 3.746 | 88.0 | 528 | 4.1819 | 0.3578 |
|
140 |
+
| 3.7196 | 89.0 | 534 | 4.1789 | 0.3597 |
|
141 |
+
| 3.7046 | 90.0 | 540 | 4.1610 | 0.3607 |
|
142 |
+
| 3.7078 | 91.0 | 546 | 4.1515 | 0.3607 |
|
143 |
+
| 3.6687 | 92.0 | 552 | 4.1752 | 0.3607 |
|
144 |
+
| 3.6559 | 93.0 | 558 | 4.1287 | 0.3636 |
|
145 |
+
| 3.6401 | 94.0 | 564 | 4.1569 | 0.3646 |
|
146 |
+
| 3.6281 | 95.0 | 570 | 4.1234 | 0.3627 |
|
147 |
+
| 3.5978 | 96.0 | 576 | 4.1270 | 0.3695 |
|
148 |
+
| 3.5951 | 97.0 | 582 | 4.1188 | 0.3646 |
|
149 |
+
| 3.5679 | 98.0 | 588 | 4.1282 | 0.3685 |
|
150 |
+
| 3.5618 | 99.0 | 594 | 4.1089 | 0.3646 |
|
151 |
+
| 3.5404 | 100.0 | 600 | 4.1090 | 0.3695 |
|
152 |
+
| 3.5255 | 101.0 | 606 | 4.1039 | 0.3646 |
|
153 |
+
| 3.5111 | 102.0 | 612 | 4.1010 | 0.3695 |
|
154 |
+
| 3.5015 | 103.0 | 618 | 4.0889 | 0.3705 |
|
155 |
+
| 3.493 | 104.0 | 624 | 4.0826 | 0.3705 |
|
156 |
+
| 3.5643 | 105.0 | 630 | 4.0915 | 0.3754 |
|
157 |
+
| 3.4543 | 106.0 | 636 | 4.0912 | 0.3724 |
|
158 |
+
| 3.4517 | 107.0 | 642 | 4.0844 | 0.3754 |
|
159 |
+
| 3.4387 | 108.0 | 648 | 4.0664 | 0.3754 |
|
160 |
+
| 3.4274 | 109.0 | 654 | 4.0885 | 0.3763 |
|
161 |
+
| 3.4241 | 110.0 | 660 | 4.0583 | 0.3793 |
|
162 |
+
| 3.4016 | 111.0 | 666 | 4.0627 | 0.3803 |
|
163 |
+
| 3.383 | 112.0 | 672 | 4.0626 | 0.3812 |
|
164 |
+
| 3.3709 | 113.0 | 678 | 4.0414 | 0.3871 |
|
165 |
+
| 3.3646 | 114.0 | 684 | 4.0562 | 0.3822 |
|
166 |
+
| 3.3456 | 115.0 | 690 | 4.0361 | 0.3861 |
|
167 |
+
| 3.3369 | 116.0 | 696 | 4.0524 | 0.3851 |
|
168 |
+
| 3.3136 | 117.0 | 702 | 4.0424 | 0.3842 |
|
169 |
+
| 3.307 | 118.0 | 708 | 4.0477 | 0.3861 |
|
170 |
+
| 3.2954 | 119.0 | 714 | 4.0287 | 0.3851 |
|
171 |
+
| 3.2887 | 120.0 | 720 | 4.0392 | 0.3900 |
|
172 |
+
| 3.2776 | 121.0 | 726 | 4.0191 | 0.3910 |
|
173 |
+
| 3.2527 | 122.0 | 732 | 4.0339 | 0.3910 |
|
174 |
+
| 3.259 | 123.0 | 738 | 4.0064 | 0.3930 |
|
175 |
+
| 3.2559 | 124.0 | 744 | 4.0285 | 0.3881 |
|
176 |
+
| 3.2335 | 125.0 | 750 | 4.0151 | 0.3930 |
|
177 |
+
| 3.2318 | 126.0 | 756 | 4.0277 | 0.3900 |
|
178 |
+
| 3.2266 | 127.0 | 762 | 3.9929 | 0.3978 |
|
179 |
+
| 3.2051 | 128.0 | 768 | 3.9945 | 0.3978 |
|
180 |
+
| 3.2009 | 129.0 | 774 | 4.0291 | 0.3930 |
|
181 |
+
| 3.1791 | 130.0 | 780 | 3.9956 | 0.3930 |
|
182 |
+
| 3.1759 | 131.0 | 786 | 4.0012 | 0.3969 |
|
183 |
+
| 3.1622 | 132.0 | 792 | 4.0107 | 0.3949 |
|
184 |
+
| 3.1559 | 133.0 | 798 | 4.0090 | 0.3939 |
|
185 |
+
| 3.1521 | 134.0 | 804 | 4.0028 | 0.3910 |
|
186 |
+
| 3.1353 | 135.0 | 810 | 4.0033 | 0.3939 |
|
187 |
+
| 3.1427 | 136.0 | 816 | 3.9995 | 0.3939 |
|
188 |
+
| 3.1276 | 137.0 | 822 | 3.9963 | 0.3920 |
|
189 |
+
| 3.1228 | 138.0 | 828 | 3.9996 | 0.3978 |
|
190 |
+
| 3.1039 | 139.0 | 834 | 3.9928 | 0.3988 |
|
191 |
+
| 3.097 | 140.0 | 840 | 3.9969 | 0.3969 |
|
192 |
+
| 3.083 | 141.0 | 846 | 3.9918 | 0.3949 |
|
193 |
+
| 3.0844 | 142.0 | 852 | 3.9900 | 0.3969 |
|
194 |
+
| 3.077 | 143.0 | 858 | 3.9812 | 0.3959 |
|
195 |
+
| 3.0601 | 144.0 | 864 | 3.9948 | 0.3959 |
|
196 |
+
| 3.0669 | 145.0 | 870 | 3.9938 | 0.3959 |
|
197 |
+
| 3.0515 | 146.0 | 876 | 3.9895 | 0.3978 |
|
198 |
+
| 3.0405 | 147.0 | 882 | 3.9803 | 0.3988 |
|
199 |
+
| 3.029 | 148.0 | 888 | 3.9856 | 0.3969 |
|
200 |
+
| 3.0342 | 149.0 | 894 | 3.9828 | 0.3969 |
|
201 |
+
| 3.0137 | 150.0 | 900 | 3.9977 | 0.3978 |
|
202 |
+
| 3.0277 | 151.0 | 906 | 3.9793 | 0.3998 |
|
203 |
+
| 3.0005 | 152.0 | 912 | 3.9779 | 0.3998 |
|
204 |
+
| 3.0027 | 153.0 | 918 | 3.9891 | 0.3988 |
|
205 |
+
| 3.0034 | 154.0 | 924 | 3.9687 | 0.4008 |
|
206 |
+
| 2.9853 | 155.0 | 930 | 3.9887 | 0.3978 |
|
207 |
+
| 2.9947 | 156.0 | 936 | 3.9860 | 0.4027 |
|
208 |
+
| 2.9768 | 157.0 | 942 | 3.9900 | 0.4027 |
|
209 |
+
| 2.9752 | 158.0 | 948 | 3.9993 | 0.3988 |
|
210 |
+
| 2.9773 | 159.0 | 954 | 3.9694 | 0.4018 |
|
211 |
+
| 2.9662 | 160.0 | 960 | 3.9924 | 0.3998 |
|
212 |
+
| 2.9661 | 161.0 | 966 | 4.0089 | 0.3988 |
|
213 |
+
| 2.9488 | 162.0 | 972 | 3.9749 | 0.3988 |
|
214 |
+
| 2.9487 | 163.0 | 978 | 3.9932 | 0.3978 |
|
215 |
+
| 2.9482 | 164.0 | 984 | 3.9987 | 0.3988 |
|
216 |
+
| 2.9624 | 165.0 | 990 | 3.9627 | 0.3978 |
|
217 |
+
| 2.9524 | 166.0 | 996 | 3.9791 | 0.4008 |
|
218 |
+
| 2.9357 | 167.0 | 1002 | 3.9969 | 0.3998 |
|
219 |
+
| 2.9323 | 168.0 | 1008 | 3.9854 | 0.4008 |
|
220 |
+
| 2.9334 | 169.0 | 1014 | 3.9778 | 0.4008 |
|
221 |
+
| 2.9228 | 170.0 | 1020 | 3.9859 | 0.4027 |
|
222 |
+
| 2.9305 | 171.0 | 1026 | 3.9821 | 0.4037 |
|
223 |
+
| 2.9239 | 172.0 | 1032 | 3.9876 | 0.4066 |
|
224 |
+
| 2.9181 | 173.0 | 1038 | 3.9792 | 0.4057 |
|
225 |
+
| 2.9162 | 174.0 | 1044 | 3.9731 | 0.4037 |
|
226 |
+
| 2.9171 | 175.0 | 1050 | 3.9796 | 0.4066 |
|
227 |
+
| 2.9132 | 176.0 | 1056 | 3.9914 | 0.4047 |
|
228 |
+
| 2.9168 | 177.0 | 1062 | 3.9826 | 0.4047 |
|
229 |
+
| 2.8974 | 178.0 | 1068 | 3.9753 | 0.4057 |
|
230 |
+
| 2.8954 | 179.0 | 1074 | 3.9766 | 0.4057 |
|
231 |
+
| 2.9003 | 180.0 | 1080 | 3.9865 | 0.4027 |
|
232 |
+
| 2.9012 | 181.0 | 1086 | 3.9835 | 0.4047 |
|
233 |
+
| 2.8994 | 182.0 | 1092 | 3.9802 | 0.4047 |
|
234 |
+
| 2.8918 | 183.0 | 1098 | 3.9811 | 0.4066 |
|
235 |
+
| 2.8893 | 184.0 | 1104 | 3.9810 | 0.4057 |
|
236 |
+
| 2.8865 | 185.0 | 1110 | 3.9852 | 0.4076 |
|
237 |
+
| 2.8784 | 186.0 | 1116 | 3.9805 | 0.4057 |
|
238 |
+
| 2.8875 | 187.0 | 1122 | 3.9781 | 0.4066 |
|
239 |
+
| 2.8948 | 188.0 | 1128 | 3.9831 | 0.4057 |
|
240 |
+
| 2.8927 | 189.0 | 1134 | 3.9837 | 0.4066 |
|
241 |
+
| 2.8739 | 190.0 | 1140 | 3.9822 | 0.4057 |
|
242 |
+
| 2.8919 | 191.0 | 1146 | 3.9792 | 0.4066 |
|
243 |
+
| 2.8713 | 192.0 | 1152 | 3.9800 | 0.4057 |
|
244 |
+
| 2.8798 | 193.0 | 1158 | 3.9854 | 0.4047 |
|
245 |
+
| 2.8835 | 194.0 | 1164 | 3.9845 | 0.4057 |
|
246 |
+
| 2.878 | 195.0 | 1170 | 3.9820 | 0.4057 |
|
247 |
+
| 2.8931 | 196.0 | 1176 | 3.9816 | 0.4057 |
|
248 |
+
| 2.8662 | 197.0 | 1182 | 3.9830 | 0.4057 |
|
249 |
+
| 2.8734 | 198.0 | 1188 | 3.9841 | 0.4057 |
|
250 |
+
| 2.8825 | 199.0 | 1194 | 3.9830 | 0.4057 |
|
251 |
+
| 2.8825 | 200.0 | 1200 | 3.9827 | 0.4057 |
|
252 |
+
|
253 |
+
|
254 |
+
### Framework versions
|
255 |
+
|
256 |
+
- Transformers 4.38.0.dev0
|
257 |
+
- Pytorch 2.1.2+cu121
|
258 |
+
- Datasets 2.16.1
|
259 |
+
- Tokenizers 0.15.1
|
hemolytic/all_results.json
ADDED
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 200.0,
|
3 |
+
"eval_accuracy": 0.3978494623655914,
|
4 |
+
"eval_loss": 3.962663412094116,
|
5 |
+
"eval_runtime": 3.5156,
|
6 |
+
"eval_samples": 1,
|
7 |
+
"eval_samples_per_second": 0.284,
|
8 |
+
"eval_steps_per_second": 0.284,
|
9 |
+
"perplexity": 52.597227650696425,
|
10 |
+
"train_loss": 3.9290491278966266,
|
11 |
+
"train_runtime": 32687.0411,
|
12 |
+
"train_samples": 12,
|
13 |
+
"train_samples_per_second": 0.073,
|
14 |
+
"train_steps_per_second": 0.037
|
15 |
+
}
|
hemolytic/config.json
ADDED
@@ -0,0 +1,39 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "nferruz/ProtGPT2",
|
3 |
+
"activation_function": "gelu_new",
|
4 |
+
"architectures": [
|
5 |
+
"GPT2LMHeadModel"
|
6 |
+
],
|
7 |
+
"attn_pdrop": 0.1,
|
8 |
+
"bos_token_id": 0,
|
9 |
+
"embd_pdrop": 0.1,
|
10 |
+
"eos_token_id": 0,
|
11 |
+
"initializer_range": 0.02,
|
12 |
+
"layer_norm_epsilon": 1e-05,
|
13 |
+
"model_type": "gpt2",
|
14 |
+
"n_ctx": 1024,
|
15 |
+
"n_embd": 1280,
|
16 |
+
"n_head": 20,
|
17 |
+
"n_inner": null,
|
18 |
+
"n_layer": 36,
|
19 |
+
"n_positions": 1024,
|
20 |
+
"reorder_and_upcast_attn": false,
|
21 |
+
"resid_pdrop": 0.1,
|
22 |
+
"scale_attn_by_inverse_layer_idx": false,
|
23 |
+
"scale_attn_weights": true,
|
24 |
+
"summary_activation": null,
|
25 |
+
"summary_first_dropout": 0.1,
|
26 |
+
"summary_proj_to_labels": true,
|
27 |
+
"summary_type": "cls_index",
|
28 |
+
"summary_use_proj": true,
|
29 |
+
"task_specific_params": {
|
30 |
+
"text-generation": {
|
31 |
+
"do_sample": true,
|
32 |
+
"max_length": 50
|
33 |
+
}
|
34 |
+
},
|
35 |
+
"torch_dtype": "float32",
|
36 |
+
"transformers_version": "4.38.0.dev0",
|
37 |
+
"use_cache": true,
|
38 |
+
"vocab_size": 50257
|
39 |
+
}
|
hemolytic/eval_results.json
ADDED
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 200.0,
|
3 |
+
"eval_accuracy": 0.3978494623655914,
|
4 |
+
"eval_loss": 3.962663412094116,
|
5 |
+
"eval_runtime": 3.5156,
|
6 |
+
"eval_samples": 1,
|
7 |
+
"eval_samples_per_second": 0.284,
|
8 |
+
"eval_steps_per_second": 0.284,
|
9 |
+
"perplexity": 52.597227650696425
|
10 |
+
}
|
hemolytic/events.out.tfevents.1707888408.lambda-a6000.3897798.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:bd1df0bc85c6e7f34d0285dd670c0b19368de12d74338137e990e0740f42fc79
|
3 |
+
size 100803
|
hemolytic/generation_config.json
ADDED
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_from_model_config": true,
|
3 |
+
"bos_token_id": 0,
|
4 |
+
"eos_token_id": 0,
|
5 |
+
"transformers_version": "4.38.0.dev0"
|
6 |
+
}
|
hemolytic/merges.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
hemolytic/model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4e932bbc6b6c72e8a94c169995620fc68a587feaecaa46b597611c6d4ac381dd
|
3 |
+
size 3096165928
|
hemolytic/special_tokens_map.json
ADDED
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": {
|
3 |
+
"content": "<|endoftext|>",
|
4 |
+
"lstrip": false,
|
5 |
+
"normalized": true,
|
6 |
+
"rstrip": false,
|
7 |
+
"single_word": false
|
8 |
+
},
|
9 |
+
"eos_token": {
|
10 |
+
"content": "<|endoftext|>",
|
11 |
+
"lstrip": false,
|
12 |
+
"normalized": true,
|
13 |
+
"rstrip": false,
|
14 |
+
"single_word": false
|
15 |
+
},
|
16 |
+
"unk_token": {
|
17 |
+
"content": "<|endoftext|>",
|
18 |
+
"lstrip": false,
|
19 |
+
"normalized": true,
|
20 |
+
"rstrip": false,
|
21 |
+
"single_word": false
|
22 |
+
}
|
23 |
+
}
|
hemolytic/tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
hemolytic/tokenizer_config.json
ADDED
@@ -0,0 +1,19 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"add_prefix_space": false,
|
3 |
+
"added_tokens_decoder": {
|
4 |
+
"0": {
|
5 |
+
"content": "<|endoftext|>",
|
6 |
+
"lstrip": false,
|
7 |
+
"normalized": true,
|
8 |
+
"rstrip": false,
|
9 |
+
"single_word": false,
|
10 |
+
"special": true
|
11 |
+
}
|
12 |
+
},
|
13 |
+
"bos_token": "<|endoftext|>",
|
14 |
+
"clean_up_tokenization_spaces": true,
|
15 |
+
"eos_token": "<|endoftext|>",
|
16 |
+
"model_max_length": 1000000000000000019884624838656,
|
17 |
+
"tokenizer_class": "GPT2Tokenizer",
|
18 |
+
"unk_token": "<|endoftext|>"
|
19 |
+
}
|
hemolytic/train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 200.0,
|
3 |
+
"train_loss": 3.9290491278966266,
|
4 |
+
"train_runtime": 32687.0411,
|
5 |
+
"train_samples": 12,
|
6 |
+
"train_samples_per_second": 0.073,
|
7 |
+
"train_steps_per_second": 0.037
|
8 |
+
}
|
hemolytic/trainer_state.json
ADDED
@@ -0,0 +1,3030 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 3.962663412094116,
|
3 |
+
"best_model_checkpoint": "output_hemo_aug_4/checkpoint-990",
|
4 |
+
"epoch": 200.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 1200,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 1.0,
|
13 |
+
"learning_rate": 9.950000000000001e-06,
|
14 |
+
"loss": 9.4301,
|
15 |
+
"step": 6
|
16 |
+
},
|
17 |
+
{
|
18 |
+
"epoch": 1.0,
|
19 |
+
"eval_accuracy": 0.024437927663734114,
|
20 |
+
"eval_loss": 8.684534072875977,
|
21 |
+
"eval_runtime": 3.4336,
|
22 |
+
"eval_samples_per_second": 0.291,
|
23 |
+
"eval_steps_per_second": 0.291,
|
24 |
+
"step": 6
|
25 |
+
},
|
26 |
+
{
|
27 |
+
"epoch": 2.0,
|
28 |
+
"learning_rate": 9.9e-06,
|
29 |
+
"loss": 8.4172,
|
30 |
+
"step": 12
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 2.0,
|
34 |
+
"eval_accuracy": 0.08113391984359726,
|
35 |
+
"eval_loss": 7.8148698806762695,
|
36 |
+
"eval_runtime": 3.5386,
|
37 |
+
"eval_samples_per_second": 0.283,
|
38 |
+
"eval_steps_per_second": 0.283,
|
39 |
+
"step": 12
|
40 |
+
},
|
41 |
+
{
|
42 |
+
"epoch": 3.0,
|
43 |
+
"learning_rate": 9.85e-06,
|
44 |
+
"loss": 7.6869,
|
45 |
+
"step": 18
|
46 |
+
},
|
47 |
+
{
|
48 |
+
"epoch": 3.0,
|
49 |
+
"eval_accuracy": 0.15542521994134897,
|
50 |
+
"eval_loss": 7.201221942901611,
|
51 |
+
"eval_runtime": 3.2749,
|
52 |
+
"eval_samples_per_second": 0.305,
|
53 |
+
"eval_steps_per_second": 0.305,
|
54 |
+
"step": 18
|
55 |
+
},
|
56 |
+
{
|
57 |
+
"epoch": 4.0,
|
58 |
+
"learning_rate": 9.800000000000001e-06,
|
59 |
+
"loss": 7.1731,
|
60 |
+
"step": 24
|
61 |
+
},
|
62 |
+
{
|
63 |
+
"epoch": 4.0,
|
64 |
+
"eval_accuracy": 0.1827956989247312,
|
65 |
+
"eval_loss": 6.913944721221924,
|
66 |
+
"eval_runtime": 3.2424,
|
67 |
+
"eval_samples_per_second": 0.308,
|
68 |
+
"eval_steps_per_second": 0.308,
|
69 |
+
"step": 24
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 5.0,
|
73 |
+
"learning_rate": 9.75e-06,
|
74 |
+
"loss": 6.8807,
|
75 |
+
"step": 30
|
76 |
+
},
|
77 |
+
{
|
78 |
+
"epoch": 5.0,
|
79 |
+
"eval_accuracy": 0.19550342130987292,
|
80 |
+
"eval_loss": 6.623848915100098,
|
81 |
+
"eval_runtime": 3.4613,
|
82 |
+
"eval_samples_per_second": 0.289,
|
83 |
+
"eval_steps_per_second": 0.289,
|
84 |
+
"step": 30
|
85 |
+
},
|
86 |
+
{
|
87 |
+
"epoch": 6.0,
|
88 |
+
"learning_rate": 9.7e-06,
|
89 |
+
"loss": 6.6009,
|
90 |
+
"step": 36
|
91 |
+
},
|
92 |
+
{
|
93 |
+
"epoch": 6.0,
|
94 |
+
"eval_accuracy": 0.1935483870967742,
|
95 |
+
"eval_loss": 6.384663105010986,
|
96 |
+
"eval_runtime": 4.2768,
|
97 |
+
"eval_samples_per_second": 0.234,
|
98 |
+
"eval_steps_per_second": 0.234,
|
99 |
+
"step": 36
|
100 |
+
},
|
101 |
+
{
|
102 |
+
"epoch": 7.0,
|
103 |
+
"learning_rate": 9.65e-06,
|
104 |
+
"loss": 6.4347,
|
105 |
+
"step": 42
|
106 |
+
},
|
107 |
+
{
|
108 |
+
"epoch": 7.0,
|
109 |
+
"eval_accuracy": 0.20625610948191594,
|
110 |
+
"eval_loss": 6.234148979187012,
|
111 |
+
"eval_runtime": 3.3994,
|
112 |
+
"eval_samples_per_second": 0.294,
|
113 |
+
"eval_steps_per_second": 0.294,
|
114 |
+
"step": 42
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 8.0,
|
118 |
+
"learning_rate": 9.600000000000001e-06,
|
119 |
+
"loss": 6.2831,
|
120 |
+
"step": 48
|
121 |
+
},
|
122 |
+
{
|
123 |
+
"epoch": 8.0,
|
124 |
+
"eval_accuracy": 0.21407624633431085,
|
125 |
+
"eval_loss": 6.096428394317627,
|
126 |
+
"eval_runtime": 3.1809,
|
127 |
+
"eval_samples_per_second": 0.314,
|
128 |
+
"eval_steps_per_second": 0.314,
|
129 |
+
"step": 48
|
130 |
+
},
|
131 |
+
{
|
132 |
+
"epoch": 9.0,
|
133 |
+
"learning_rate": 9.55e-06,
|
134 |
+
"loss": 6.1728,
|
135 |
+
"step": 54
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 9.0,
|
139 |
+
"eval_accuracy": 0.2209188660801564,
|
140 |
+
"eval_loss": 5.98640775680542,
|
141 |
+
"eval_runtime": 3.4302,
|
142 |
+
"eval_samples_per_second": 0.292,
|
143 |
+
"eval_steps_per_second": 0.292,
|
144 |
+
"step": 54
|
145 |
+
},
|
146 |
+
{
|
147 |
+
"epoch": 10.0,
|
148 |
+
"learning_rate": 9.5e-06,
|
149 |
+
"loss": 6.0805,
|
150 |
+
"step": 60
|
151 |
+
},
|
152 |
+
{
|
153 |
+
"epoch": 10.0,
|
154 |
+
"eval_accuracy": 0.2316715542521994,
|
155 |
+
"eval_loss": 5.893611907958984,
|
156 |
+
"eval_runtime": 3.6923,
|
157 |
+
"eval_samples_per_second": 0.271,
|
158 |
+
"eval_steps_per_second": 0.271,
|
159 |
+
"step": 60
|
160 |
+
},
|
161 |
+
{
|
162 |
+
"epoch": 11.0,
|
163 |
+
"learning_rate": 9.450000000000001e-06,
|
164 |
+
"loss": 5.9959,
|
165 |
+
"step": 66
|
166 |
+
},
|
167 |
+
{
|
168 |
+
"epoch": 11.0,
|
169 |
+
"eval_accuracy": 0.2404692082111437,
|
170 |
+
"eval_loss": 5.81611967086792,
|
171 |
+
"eval_runtime": 3.2748,
|
172 |
+
"eval_samples_per_second": 0.305,
|
173 |
+
"eval_steps_per_second": 0.305,
|
174 |
+
"step": 66
|
175 |
+
},
|
176 |
+
{
|
177 |
+
"epoch": 12.0,
|
178 |
+
"learning_rate": 9.4e-06,
|
179 |
+
"loss": 5.925,
|
180 |
+
"step": 72
|
181 |
+
},
|
182 |
+
{
|
183 |
+
"epoch": 12.0,
|
184 |
+
"eval_accuracy": 0.23851417399804498,
|
185 |
+
"eval_loss": 5.745607852935791,
|
186 |
+
"eval_runtime": 3.4474,
|
187 |
+
"eval_samples_per_second": 0.29,
|
188 |
+
"eval_steps_per_second": 0.29,
|
189 |
+
"step": 72
|
190 |
+
},
|
191 |
+
{
|
192 |
+
"epoch": 13.0,
|
193 |
+
"learning_rate": 9.350000000000002e-06,
|
194 |
+
"loss": 5.8787,
|
195 |
+
"step": 78
|
196 |
+
},
|
197 |
+
{
|
198 |
+
"epoch": 13.0,
|
199 |
+
"eval_accuracy": 0.2482893450635386,
|
200 |
+
"eval_loss": 5.664583683013916,
|
201 |
+
"eval_runtime": 3.5432,
|
202 |
+
"eval_samples_per_second": 0.282,
|
203 |
+
"eval_steps_per_second": 0.282,
|
204 |
+
"step": 78
|
205 |
+
},
|
206 |
+
{
|
207 |
+
"epoch": 14.0,
|
208 |
+
"learning_rate": 9.3e-06,
|
209 |
+
"loss": 5.7996,
|
210 |
+
"step": 84
|
211 |
+
},
|
212 |
+
{
|
213 |
+
"epoch": 14.0,
|
214 |
+
"eval_accuracy": 0.24926686217008798,
|
215 |
+
"eval_loss": 5.590141296386719,
|
216 |
+
"eval_runtime": 3.2321,
|
217 |
+
"eval_samples_per_second": 0.309,
|
218 |
+
"eval_steps_per_second": 0.309,
|
219 |
+
"step": 84
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 15.0,
|
223 |
+
"learning_rate": 9.250000000000001e-06,
|
224 |
+
"loss": 5.7312,
|
225 |
+
"step": 90
|
226 |
+
},
|
227 |
+
{
|
228 |
+
"epoch": 15.0,
|
229 |
+
"eval_accuracy": 0.25317693059628543,
|
230 |
+
"eval_loss": 5.521559238433838,
|
231 |
+
"eval_runtime": 3.5703,
|
232 |
+
"eval_samples_per_second": 0.28,
|
233 |
+
"eval_steps_per_second": 0.28,
|
234 |
+
"step": 90
|
235 |
+
},
|
236 |
+
{
|
237 |
+
"epoch": 16.0,
|
238 |
+
"learning_rate": 9.200000000000002e-06,
|
239 |
+
"loss": 5.6751,
|
240 |
+
"step": 96
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 16.0,
|
244 |
+
"eval_accuracy": 0.2590420332355816,
|
245 |
+
"eval_loss": 5.469524383544922,
|
246 |
+
"eval_runtime": 3.5655,
|
247 |
+
"eval_samples_per_second": 0.28,
|
248 |
+
"eval_steps_per_second": 0.28,
|
249 |
+
"step": 96
|
250 |
+
},
|
251 |
+
{
|
252 |
+
"epoch": 17.0,
|
253 |
+
"learning_rate": 9.15e-06,
|
254 |
+
"loss": 5.6076,
|
255 |
+
"step": 102
|
256 |
+
},
|
257 |
+
{
|
258 |
+
"epoch": 17.0,
|
259 |
+
"eval_accuracy": 0.2619745845552297,
|
260 |
+
"eval_loss": 5.421563148498535,
|
261 |
+
"eval_runtime": 3.4299,
|
262 |
+
"eval_samples_per_second": 0.292,
|
263 |
+
"eval_steps_per_second": 0.292,
|
264 |
+
"step": 102
|
265 |
+
},
|
266 |
+
{
|
267 |
+
"epoch": 18.0,
|
268 |
+
"learning_rate": 9.100000000000001e-06,
|
269 |
+
"loss": 5.569,
|
270 |
+
"step": 108
|
271 |
+
},
|
272 |
+
{
|
273 |
+
"epoch": 18.0,
|
274 |
+
"eval_accuracy": 0.2619745845552297,
|
275 |
+
"eval_loss": 5.373498916625977,
|
276 |
+
"eval_runtime": 3.4799,
|
277 |
+
"eval_samples_per_second": 0.287,
|
278 |
+
"eval_steps_per_second": 0.287,
|
279 |
+
"step": 108
|
280 |
+
},
|
281 |
+
{
|
282 |
+
"epoch": 19.0,
|
283 |
+
"learning_rate": 9.050000000000001e-06,
|
284 |
+
"loss": 5.5037,
|
285 |
+
"step": 114
|
286 |
+
},
|
287 |
+
{
|
288 |
+
"epoch": 19.0,
|
289 |
+
"eval_accuracy": 0.26295210166177907,
|
290 |
+
"eval_loss": 5.327227592468262,
|
291 |
+
"eval_runtime": 3.5293,
|
292 |
+
"eval_samples_per_second": 0.283,
|
293 |
+
"eval_steps_per_second": 0.283,
|
294 |
+
"step": 114
|
295 |
+
},
|
296 |
+
{
|
297 |
+
"epoch": 20.0,
|
298 |
+
"learning_rate": 9e-06,
|
299 |
+
"loss": 5.4681,
|
300 |
+
"step": 120
|
301 |
+
},
|
302 |
+
{
|
303 |
+
"epoch": 20.0,
|
304 |
+
"eval_accuracy": 0.2697947214076246,
|
305 |
+
"eval_loss": 5.285594463348389,
|
306 |
+
"eval_runtime": 3.6414,
|
307 |
+
"eval_samples_per_second": 0.275,
|
308 |
+
"eval_steps_per_second": 0.275,
|
309 |
+
"step": 120
|
310 |
+
},
|
311 |
+
{
|
312 |
+
"epoch": 21.0,
|
313 |
+
"learning_rate": 8.95e-06,
|
314 |
+
"loss": 5.4225,
|
315 |
+
"step": 126
|
316 |
+
},
|
317 |
+
{
|
318 |
+
"epoch": 21.0,
|
319 |
+
"eval_accuracy": 0.27174975562072334,
|
320 |
+
"eval_loss": 5.255927085876465,
|
321 |
+
"eval_runtime": 3.6672,
|
322 |
+
"eval_samples_per_second": 0.273,
|
323 |
+
"eval_steps_per_second": 0.273,
|
324 |
+
"step": 126
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 22.0,
|
328 |
+
"learning_rate": 8.900000000000001e-06,
|
329 |
+
"loss": 5.3805,
|
330 |
+
"step": 132
|
331 |
+
},
|
332 |
+
{
|
333 |
+
"epoch": 22.0,
|
334 |
+
"eval_accuracy": 0.27663734115347016,
|
335 |
+
"eval_loss": 5.212594509124756,
|
336 |
+
"eval_runtime": 3.547,
|
337 |
+
"eval_samples_per_second": 0.282,
|
338 |
+
"eval_steps_per_second": 0.282,
|
339 |
+
"step": 132
|
340 |
+
},
|
341 |
+
{
|
342 |
+
"epoch": 23.0,
|
343 |
+
"learning_rate": 8.85e-06,
|
344 |
+
"loss": 5.3527,
|
345 |
+
"step": 138
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 23.0,
|
349 |
+
"eval_accuracy": 0.2756598240469208,
|
350 |
+
"eval_loss": 5.1883745193481445,
|
351 |
+
"eval_runtime": 3.3405,
|
352 |
+
"eval_samples_per_second": 0.299,
|
353 |
+
"eval_steps_per_second": 0.299,
|
354 |
+
"step": 138
|
355 |
+
},
|
356 |
+
{
|
357 |
+
"epoch": 24.0,
|
358 |
+
"learning_rate": 8.8e-06,
|
359 |
+
"loss": 5.3033,
|
360 |
+
"step": 144
|
361 |
+
},
|
362 |
+
{
|
363 |
+
"epoch": 24.0,
|
364 |
+
"eval_accuracy": 0.27956989247311825,
|
365 |
+
"eval_loss": 5.153918266296387,
|
366 |
+
"eval_runtime": 3.6205,
|
367 |
+
"eval_samples_per_second": 0.276,
|
368 |
+
"eval_steps_per_second": 0.276,
|
369 |
+
"step": 144
|
370 |
+
},
|
371 |
+
{
|
372 |
+
"epoch": 25.0,
|
373 |
+
"learning_rate": 8.750000000000001e-06,
|
374 |
+
"loss": 5.2635,
|
375 |
+
"step": 150
|
376 |
+
},
|
377 |
+
{
|
378 |
+
"epoch": 25.0,
|
379 |
+
"eval_accuracy": 0.2854349951124145,
|
380 |
+
"eval_loss": 5.111028671264648,
|
381 |
+
"eval_runtime": 3.658,
|
382 |
+
"eval_samples_per_second": 0.273,
|
383 |
+
"eval_steps_per_second": 0.273,
|
384 |
+
"step": 150
|
385 |
+
},
|
386 |
+
{
|
387 |
+
"epoch": 26.0,
|
388 |
+
"learning_rate": 8.700000000000001e-06,
|
389 |
+
"loss": 5.2411,
|
390 |
+
"step": 156
|
391 |
+
},
|
392 |
+
{
|
393 |
+
"epoch": 26.0,
|
394 |
+
"eval_accuracy": 0.2854349951124145,
|
395 |
+
"eval_loss": 5.0881805419921875,
|
396 |
+
"eval_runtime": 3.5731,
|
397 |
+
"eval_samples_per_second": 0.28,
|
398 |
+
"eval_steps_per_second": 0.28,
|
399 |
+
"step": 156
|
400 |
+
},
|
401 |
+
{
|
402 |
+
"epoch": 27.0,
|
403 |
+
"learning_rate": 8.65e-06,
|
404 |
+
"loss": 5.1972,
|
405 |
+
"step": 162
|
406 |
+
},
|
407 |
+
{
|
408 |
+
"epoch": 27.0,
|
409 |
+
"eval_accuracy": 0.2903225806451613,
|
410 |
+
"eval_loss": 5.057516098022461,
|
411 |
+
"eval_runtime": 3.6003,
|
412 |
+
"eval_samples_per_second": 0.278,
|
413 |
+
"eval_steps_per_second": 0.278,
|
414 |
+
"step": 162
|
415 |
+
},
|
416 |
+
{
|
417 |
+
"epoch": 28.0,
|
418 |
+
"learning_rate": 8.6e-06,
|
419 |
+
"loss": 5.163,
|
420 |
+
"step": 168
|
421 |
+
},
|
422 |
+
{
|
423 |
+
"epoch": 28.0,
|
424 |
+
"eval_accuracy": 0.2913000977517107,
|
425 |
+
"eval_loss": 5.029298782348633,
|
426 |
+
"eval_runtime": 3.5678,
|
427 |
+
"eval_samples_per_second": 0.28,
|
428 |
+
"eval_steps_per_second": 0.28,
|
429 |
+
"step": 168
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 29.0,
|
433 |
+
"learning_rate": 8.550000000000001e-06,
|
434 |
+
"loss": 5.1273,
|
435 |
+
"step": 174
|
436 |
+
},
|
437 |
+
{
|
438 |
+
"epoch": 29.0,
|
439 |
+
"eval_accuracy": 0.2903225806451613,
|
440 |
+
"eval_loss": 5.00468111038208,
|
441 |
+
"eval_runtime": 3.5535,
|
442 |
+
"eval_samples_per_second": 0.281,
|
443 |
+
"eval_steps_per_second": 0.281,
|
444 |
+
"step": 174
|
445 |
+
},
|
446 |
+
{
|
447 |
+
"epoch": 30.0,
|
448 |
+
"learning_rate": 8.5e-06,
|
449 |
+
"loss": 5.1032,
|
450 |
+
"step": 180
|
451 |
+
},
|
452 |
+
{
|
453 |
+
"epoch": 30.0,
|
454 |
+
"eval_accuracy": 0.29521016617790813,
|
455 |
+
"eval_loss": 4.981687545776367,
|
456 |
+
"eval_runtime": 3.5476,
|
457 |
+
"eval_samples_per_second": 0.282,
|
458 |
+
"eval_steps_per_second": 0.282,
|
459 |
+
"step": 180
|
460 |
+
},
|
461 |
+
{
|
462 |
+
"epoch": 31.0,
|
463 |
+
"learning_rate": 8.45e-06,
|
464 |
+
"loss": 5.0726,
|
465 |
+
"step": 186
|
466 |
+
},
|
467 |
+
{
|
468 |
+
"epoch": 31.0,
|
469 |
+
"eval_accuracy": 0.29521016617790813,
|
470 |
+
"eval_loss": 4.958263874053955,
|
471 |
+
"eval_runtime": 3.574,
|
472 |
+
"eval_samples_per_second": 0.28,
|
473 |
+
"eval_steps_per_second": 0.28,
|
474 |
+
"step": 186
|
475 |
+
},
|
476 |
+
{
|
477 |
+
"epoch": 32.0,
|
478 |
+
"learning_rate": 8.400000000000001e-06,
|
479 |
+
"loss": 5.0405,
|
480 |
+
"step": 192
|
481 |
+
},
|
482 |
+
{
|
483 |
+
"epoch": 32.0,
|
484 |
+
"eval_accuracy": 0.29521016617790813,
|
485 |
+
"eval_loss": 4.9354777336120605,
|
486 |
+
"eval_runtime": 3.2861,
|
487 |
+
"eval_samples_per_second": 0.304,
|
488 |
+
"eval_steps_per_second": 0.304,
|
489 |
+
"step": 192
|
490 |
+
},
|
491 |
+
{
|
492 |
+
"epoch": 33.0,
|
493 |
+
"learning_rate": 8.35e-06,
|
494 |
+
"loss": 5.007,
|
495 |
+
"step": 198
|
496 |
+
},
|
497 |
+
{
|
498 |
+
"epoch": 33.0,
|
499 |
+
"eval_accuracy": 0.29521016617790813,
|
500 |
+
"eval_loss": 4.918440818786621,
|
501 |
+
"eval_runtime": 3.5379,
|
502 |
+
"eval_samples_per_second": 0.283,
|
503 |
+
"eval_steps_per_second": 0.283,
|
504 |
+
"step": 198
|
505 |
+
},
|
506 |
+
{
|
507 |
+
"epoch": 34.0,
|
508 |
+
"learning_rate": 8.3e-06,
|
509 |
+
"loss": 4.9897,
|
510 |
+
"step": 204
|
511 |
+
},
|
512 |
+
{
|
513 |
+
"epoch": 34.0,
|
514 |
+
"eval_accuracy": 0.29716520039100686,
|
515 |
+
"eval_loss": 4.891113758087158,
|
516 |
+
"eval_runtime": 3.2282,
|
517 |
+
"eval_samples_per_second": 0.31,
|
518 |
+
"eval_steps_per_second": 0.31,
|
519 |
+
"step": 204
|
520 |
+
},
|
521 |
+
{
|
522 |
+
"epoch": 35.0,
|
523 |
+
"learning_rate": 8.25e-06,
|
524 |
+
"loss": 4.9416,
|
525 |
+
"step": 210
|
526 |
+
},
|
527 |
+
{
|
528 |
+
"epoch": 35.0,
|
529 |
+
"eval_accuracy": 0.29716520039100686,
|
530 |
+
"eval_loss": 4.862751483917236,
|
531 |
+
"eval_runtime": 3.5473,
|
532 |
+
"eval_samples_per_second": 0.282,
|
533 |
+
"eval_steps_per_second": 0.282,
|
534 |
+
"step": 210
|
535 |
+
},
|
536 |
+
{
|
537 |
+
"epoch": 36.0,
|
538 |
+
"learning_rate": 8.2e-06,
|
539 |
+
"loss": 4.9245,
|
540 |
+
"step": 216
|
541 |
+
},
|
542 |
+
{
|
543 |
+
"epoch": 36.0,
|
544 |
+
"eval_accuracy": 0.2981427174975562,
|
545 |
+
"eval_loss": 4.849916934967041,
|
546 |
+
"eval_runtime": 3.5627,
|
547 |
+
"eval_samples_per_second": 0.281,
|
548 |
+
"eval_steps_per_second": 0.281,
|
549 |
+
"step": 216
|
550 |
+
},
|
551 |
+
{
|
552 |
+
"epoch": 37.0,
|
553 |
+
"learning_rate": 8.15e-06,
|
554 |
+
"loss": 4.901,
|
555 |
+
"step": 222
|
556 |
+
},
|
557 |
+
{
|
558 |
+
"epoch": 37.0,
|
559 |
+
"eval_accuracy": 0.30303030303030304,
|
560 |
+
"eval_loss": 4.826337814331055,
|
561 |
+
"eval_runtime": 3.475,
|
562 |
+
"eval_samples_per_second": 0.288,
|
563 |
+
"eval_steps_per_second": 0.288,
|
564 |
+
"step": 222
|
565 |
+
},
|
566 |
+
{
|
567 |
+
"epoch": 38.0,
|
568 |
+
"learning_rate": 8.1e-06,
|
569 |
+
"loss": 4.8713,
|
570 |
+
"step": 228
|
571 |
+
},
|
572 |
+
{
|
573 |
+
"epoch": 38.0,
|
574 |
+
"eval_accuracy": 0.30303030303030304,
|
575 |
+
"eval_loss": 4.803491115570068,
|
576 |
+
"eval_runtime": 3.4779,
|
577 |
+
"eval_samples_per_second": 0.288,
|
578 |
+
"eval_steps_per_second": 0.288,
|
579 |
+
"step": 228
|
580 |
+
},
|
581 |
+
{
|
582 |
+
"epoch": 39.0,
|
583 |
+
"learning_rate": 8.050000000000001e-06,
|
584 |
+
"loss": 4.845,
|
585 |
+
"step": 234
|
586 |
+
},
|
587 |
+
{
|
588 |
+
"epoch": 39.0,
|
589 |
+
"eval_accuracy": 0.30596285434995113,
|
590 |
+
"eval_loss": 4.7873663902282715,
|
591 |
+
"eval_runtime": 3.3398,
|
592 |
+
"eval_samples_per_second": 0.299,
|
593 |
+
"eval_steps_per_second": 0.299,
|
594 |
+
"step": 234
|
595 |
+
},
|
596 |
+
{
|
597 |
+
"epoch": 40.0,
|
598 |
+
"learning_rate": 8.000000000000001e-06,
|
599 |
+
"loss": 4.8052,
|
600 |
+
"step": 240
|
601 |
+
},
|
602 |
+
{
|
603 |
+
"epoch": 40.0,
|
604 |
+
"eval_accuracy": 0.3040078201368524,
|
605 |
+
"eval_loss": 4.753478527069092,
|
606 |
+
"eval_runtime": 3.1758,
|
607 |
+
"eval_samples_per_second": 0.315,
|
608 |
+
"eval_steps_per_second": 0.315,
|
609 |
+
"step": 240
|
610 |
+
},
|
611 |
+
{
|
612 |
+
"epoch": 41.0,
|
613 |
+
"learning_rate": 7.950000000000002e-06,
|
614 |
+
"loss": 4.7786,
|
615 |
+
"step": 246
|
616 |
+
},
|
617 |
+
{
|
618 |
+
"epoch": 41.0,
|
619 |
+
"eval_accuracy": 0.30596285434995113,
|
620 |
+
"eval_loss": 4.731250762939453,
|
621 |
+
"eval_runtime": 3.2392,
|
622 |
+
"eval_samples_per_second": 0.309,
|
623 |
+
"eval_steps_per_second": 0.309,
|
624 |
+
"step": 246
|
625 |
+
},
|
626 |
+
{
|
627 |
+
"epoch": 42.0,
|
628 |
+
"learning_rate": 7.9e-06,
|
629 |
+
"loss": 4.7501,
|
630 |
+
"step": 252
|
631 |
+
},
|
632 |
+
{
|
633 |
+
"epoch": 42.0,
|
634 |
+
"eval_accuracy": 0.3088954056695992,
|
635 |
+
"eval_loss": 4.717497825622559,
|
636 |
+
"eval_runtime": 3.5586,
|
637 |
+
"eval_samples_per_second": 0.281,
|
638 |
+
"eval_steps_per_second": 0.281,
|
639 |
+
"step": 252
|
640 |
+
},
|
641 |
+
{
|
642 |
+
"epoch": 43.0,
|
643 |
+
"learning_rate": 7.850000000000001e-06,
|
644 |
+
"loss": 4.7221,
|
645 |
+
"step": 258
|
646 |
+
},
|
647 |
+
{
|
648 |
+
"epoch": 43.0,
|
649 |
+
"eval_accuracy": 0.3118279569892473,
|
650 |
+
"eval_loss": 4.697778701782227,
|
651 |
+
"eval_runtime": 3.4899,
|
652 |
+
"eval_samples_per_second": 0.287,
|
653 |
+
"eval_steps_per_second": 0.287,
|
654 |
+
"step": 258
|
655 |
+
},
|
656 |
+
{
|
657 |
+
"epoch": 44.0,
|
658 |
+
"learning_rate": 7.800000000000002e-06,
|
659 |
+
"loss": 4.7038,
|
660 |
+
"step": 264
|
661 |
+
},
|
662 |
+
{
|
663 |
+
"epoch": 44.0,
|
664 |
+
"eval_accuracy": 0.31085043988269795,
|
665 |
+
"eval_loss": 4.678452014923096,
|
666 |
+
"eval_runtime": 3.2849,
|
667 |
+
"eval_samples_per_second": 0.304,
|
668 |
+
"eval_steps_per_second": 0.304,
|
669 |
+
"step": 264
|
670 |
+
},
|
671 |
+
{
|
672 |
+
"epoch": 45.0,
|
673 |
+
"learning_rate": 7.75e-06,
|
674 |
+
"loss": 4.681,
|
675 |
+
"step": 270
|
676 |
+
},
|
677 |
+
{
|
678 |
+
"epoch": 45.0,
|
679 |
+
"eval_accuracy": 0.3128054740957967,
|
680 |
+
"eval_loss": 4.666133403778076,
|
681 |
+
"eval_runtime": 3.4251,
|
682 |
+
"eval_samples_per_second": 0.292,
|
683 |
+
"eval_steps_per_second": 0.292,
|
684 |
+
"step": 270
|
685 |
+
},
|
686 |
+
{
|
687 |
+
"epoch": 46.0,
|
688 |
+
"learning_rate": 7.7e-06,
|
689 |
+
"loss": 4.6566,
|
690 |
+
"step": 276
|
691 |
+
},
|
692 |
+
{
|
693 |
+
"epoch": 46.0,
|
694 |
+
"eval_accuracy": 0.31573802541544477,
|
695 |
+
"eval_loss": 4.653237342834473,
|
696 |
+
"eval_runtime": 3.1995,
|
697 |
+
"eval_samples_per_second": 0.313,
|
698 |
+
"eval_steps_per_second": 0.313,
|
699 |
+
"step": 276
|
700 |
+
},
|
701 |
+
{
|
702 |
+
"epoch": 47.0,
|
703 |
+
"learning_rate": 7.650000000000001e-06,
|
704 |
+
"loss": 4.632,
|
705 |
+
"step": 282
|
706 |
+
},
|
707 |
+
{
|
708 |
+
"epoch": 47.0,
|
709 |
+
"eval_accuracy": 0.31573802541544477,
|
710 |
+
"eval_loss": 4.636072635650635,
|
711 |
+
"eval_runtime": 3.5333,
|
712 |
+
"eval_samples_per_second": 0.283,
|
713 |
+
"eval_steps_per_second": 0.283,
|
714 |
+
"step": 282
|
715 |
+
},
|
716 |
+
{
|
717 |
+
"epoch": 48.0,
|
718 |
+
"learning_rate": 7.600000000000001e-06,
|
719 |
+
"loss": 4.618,
|
720 |
+
"step": 288
|
721 |
+
},
|
722 |
+
{
|
723 |
+
"epoch": 48.0,
|
724 |
+
"eval_accuracy": 0.3196480938416422,
|
725 |
+
"eval_loss": 4.6162004470825195,
|
726 |
+
"eval_runtime": 3.2635,
|
727 |
+
"eval_samples_per_second": 0.306,
|
728 |
+
"eval_steps_per_second": 0.306,
|
729 |
+
"step": 288
|
730 |
+
},
|
731 |
+
{
|
732 |
+
"epoch": 49.0,
|
733 |
+
"learning_rate": 7.5500000000000006e-06,
|
734 |
+
"loss": 4.5928,
|
735 |
+
"step": 294
|
736 |
+
},
|
737 |
+
{
|
738 |
+
"epoch": 49.0,
|
739 |
+
"eval_accuracy": 0.32453567937438904,
|
740 |
+
"eval_loss": 4.598696708679199,
|
741 |
+
"eval_runtime": 3.5544,
|
742 |
+
"eval_samples_per_second": 0.281,
|
743 |
+
"eval_steps_per_second": 0.281,
|
744 |
+
"step": 294
|
745 |
+
},
|
746 |
+
{
|
747 |
+
"epoch": 50.0,
|
748 |
+
"learning_rate": 7.500000000000001e-06,
|
749 |
+
"loss": 4.5716,
|
750 |
+
"step": 300
|
751 |
+
},
|
752 |
+
{
|
753 |
+
"epoch": 50.0,
|
754 |
+
"eval_accuracy": 0.32160312805474095,
|
755 |
+
"eval_loss": 4.584750175476074,
|
756 |
+
"eval_runtime": 3.2866,
|
757 |
+
"eval_samples_per_second": 0.304,
|
758 |
+
"eval_steps_per_second": 0.304,
|
759 |
+
"step": 300
|
760 |
+
},
|
761 |
+
{
|
762 |
+
"epoch": 51.0,
|
763 |
+
"learning_rate": 7.450000000000001e-06,
|
764 |
+
"loss": 4.5485,
|
765 |
+
"step": 306
|
766 |
+
},
|
767 |
+
{
|
768 |
+
"epoch": 51.0,
|
769 |
+
"eval_accuracy": 0.32453567937438904,
|
770 |
+
"eval_loss": 4.572133541107178,
|
771 |
+
"eval_runtime": 4.8161,
|
772 |
+
"eval_samples_per_second": 0.208,
|
773 |
+
"eval_steps_per_second": 0.208,
|
774 |
+
"step": 306
|
775 |
+
},
|
776 |
+
{
|
777 |
+
"epoch": 52.0,
|
778 |
+
"learning_rate": 7.4e-06,
|
779 |
+
"loss": 4.5324,
|
780 |
+
"step": 312
|
781 |
+
},
|
782 |
+
{
|
783 |
+
"epoch": 52.0,
|
784 |
+
"eval_accuracy": 0.3196480938416422,
|
785 |
+
"eval_loss": 4.557907581329346,
|
786 |
+
"eval_runtime": 3.1695,
|
787 |
+
"eval_samples_per_second": 0.316,
|
788 |
+
"eval_steps_per_second": 0.316,
|
789 |
+
"step": 312
|
790 |
+
},
|
791 |
+
{
|
792 |
+
"epoch": 53.0,
|
793 |
+
"learning_rate": 7.350000000000001e-06,
|
794 |
+
"loss": 4.5038,
|
795 |
+
"step": 318
|
796 |
+
},
|
797 |
+
{
|
798 |
+
"epoch": 53.0,
|
799 |
+
"eval_accuracy": 0.3196480938416422,
|
800 |
+
"eval_loss": 4.542334079742432,
|
801 |
+
"eval_runtime": 3.3713,
|
802 |
+
"eval_samples_per_second": 0.297,
|
803 |
+
"eval_steps_per_second": 0.297,
|
804 |
+
"step": 318
|
805 |
+
},
|
806 |
+
{
|
807 |
+
"epoch": 54.0,
|
808 |
+
"learning_rate": 7.3e-06,
|
809 |
+
"loss": 4.4831,
|
810 |
+
"step": 324
|
811 |
+
},
|
812 |
+
{
|
813 |
+
"epoch": 54.0,
|
814 |
+
"eval_accuracy": 0.32649071358748777,
|
815 |
+
"eval_loss": 4.524044036865234,
|
816 |
+
"eval_runtime": 3.1022,
|
817 |
+
"eval_samples_per_second": 0.322,
|
818 |
+
"eval_steps_per_second": 0.322,
|
819 |
+
"step": 324
|
820 |
+
},
|
821 |
+
{
|
822 |
+
"epoch": 55.0,
|
823 |
+
"learning_rate": 7.25e-06,
|
824 |
+
"loss": 4.4347,
|
825 |
+
"step": 330
|
826 |
+
},
|
827 |
+
{
|
828 |
+
"epoch": 55.0,
|
829 |
+
"eval_accuracy": 0.3255131964809384,
|
830 |
+
"eval_loss": 4.508722305297852,
|
831 |
+
"eval_runtime": 3.0991,
|
832 |
+
"eval_samples_per_second": 0.323,
|
833 |
+
"eval_steps_per_second": 0.323,
|
834 |
+
"step": 330
|
835 |
+
},
|
836 |
+
{
|
837 |
+
"epoch": 56.0,
|
838 |
+
"learning_rate": 7.2000000000000005e-06,
|
839 |
+
"loss": 4.4218,
|
840 |
+
"step": 336
|
841 |
+
},
|
842 |
+
{
|
843 |
+
"epoch": 56.0,
|
844 |
+
"eval_accuracy": 0.3255131964809384,
|
845 |
+
"eval_loss": 4.484982013702393,
|
846 |
+
"eval_runtime": 3.3635,
|
847 |
+
"eval_samples_per_second": 0.297,
|
848 |
+
"eval_steps_per_second": 0.297,
|
849 |
+
"step": 336
|
850 |
+
},
|
851 |
+
{
|
852 |
+
"epoch": 57.0,
|
853 |
+
"learning_rate": 7.15e-06,
|
854 |
+
"loss": 4.3939,
|
855 |
+
"step": 342
|
856 |
+
},
|
857 |
+
{
|
858 |
+
"epoch": 57.0,
|
859 |
+
"eval_accuracy": 0.32746823069403713,
|
860 |
+
"eval_loss": 4.479069709777832,
|
861 |
+
"eval_runtime": 3.1386,
|
862 |
+
"eval_samples_per_second": 0.319,
|
863 |
+
"eval_steps_per_second": 0.319,
|
864 |
+
"step": 342
|
865 |
+
},
|
866 |
+
{
|
867 |
+
"epoch": 58.0,
|
868 |
+
"learning_rate": 7.100000000000001e-06,
|
869 |
+
"loss": 4.3766,
|
870 |
+
"step": 348
|
871 |
+
},
|
872 |
+
{
|
873 |
+
"epoch": 58.0,
|
874 |
+
"eval_accuracy": 0.32649071358748777,
|
875 |
+
"eval_loss": 4.463999271392822,
|
876 |
+
"eval_runtime": 3.2574,
|
877 |
+
"eval_samples_per_second": 0.307,
|
878 |
+
"eval_steps_per_second": 0.307,
|
879 |
+
"step": 348
|
880 |
+
},
|
881 |
+
{
|
882 |
+
"epoch": 59.0,
|
883 |
+
"learning_rate": 7.05e-06,
|
884 |
+
"loss": 4.3472,
|
885 |
+
"step": 354
|
886 |
+
},
|
887 |
+
{
|
888 |
+
"epoch": 59.0,
|
889 |
+
"eval_accuracy": 0.32746823069403713,
|
890 |
+
"eval_loss": 4.4470648765563965,
|
891 |
+
"eval_runtime": 3.1788,
|
892 |
+
"eval_samples_per_second": 0.315,
|
893 |
+
"eval_steps_per_second": 0.315,
|
894 |
+
"step": 354
|
895 |
+
},
|
896 |
+
{
|
897 |
+
"epoch": 60.0,
|
898 |
+
"learning_rate": 7e-06,
|
899 |
+
"loss": 4.3241,
|
900 |
+
"step": 360
|
901 |
+
},
|
902 |
+
{
|
903 |
+
"epoch": 60.0,
|
904 |
+
"eval_accuracy": 0.32746823069403713,
|
905 |
+
"eval_loss": 4.433444499969482,
|
906 |
+
"eval_runtime": 3.1305,
|
907 |
+
"eval_samples_per_second": 0.319,
|
908 |
+
"eval_steps_per_second": 0.319,
|
909 |
+
"step": 360
|
910 |
+
},
|
911 |
+
{
|
912 |
+
"epoch": 61.0,
|
913 |
+
"learning_rate": 6.95e-06,
|
914 |
+
"loss": 4.2919,
|
915 |
+
"step": 366
|
916 |
+
},
|
917 |
+
{
|
918 |
+
"epoch": 61.0,
|
919 |
+
"eval_accuracy": 0.3304007820136852,
|
920 |
+
"eval_loss": 4.42958927154541,
|
921 |
+
"eval_runtime": 3.1351,
|
922 |
+
"eval_samples_per_second": 0.319,
|
923 |
+
"eval_steps_per_second": 0.319,
|
924 |
+
"step": 366
|
925 |
+
},
|
926 |
+
{
|
927 |
+
"epoch": 62.0,
|
928 |
+
"learning_rate": 6.9e-06,
|
929 |
+
"loss": 4.2678,
|
930 |
+
"step": 372
|
931 |
+
},
|
932 |
+
{
|
933 |
+
"epoch": 62.0,
|
934 |
+
"eval_accuracy": 0.3343108504398827,
|
935 |
+
"eval_loss": 4.428121566772461,
|
936 |
+
"eval_runtime": 3.0787,
|
937 |
+
"eval_samples_per_second": 0.325,
|
938 |
+
"eval_steps_per_second": 0.325,
|
939 |
+
"step": 372
|
940 |
+
},
|
941 |
+
{
|
942 |
+
"epoch": 63.0,
|
943 |
+
"learning_rate": 6.850000000000001e-06,
|
944 |
+
"loss": 4.2515,
|
945 |
+
"step": 378
|
946 |
+
},
|
947 |
+
{
|
948 |
+
"epoch": 63.0,
|
949 |
+
"eval_accuracy": 0.33724340175953077,
|
950 |
+
"eval_loss": 4.412038803100586,
|
951 |
+
"eval_runtime": 3.3246,
|
952 |
+
"eval_samples_per_second": 0.301,
|
953 |
+
"eval_steps_per_second": 0.301,
|
954 |
+
"step": 378
|
955 |
+
},
|
956 |
+
{
|
957 |
+
"epoch": 64.0,
|
958 |
+
"learning_rate": 6.800000000000001e-06,
|
959 |
+
"loss": 4.2244,
|
960 |
+
"step": 384
|
961 |
+
},
|
962 |
+
{
|
963 |
+
"epoch": 64.0,
|
964 |
+
"eval_accuracy": 0.3343108504398827,
|
965 |
+
"eval_loss": 4.403759002685547,
|
966 |
+
"eval_runtime": 3.1858,
|
967 |
+
"eval_samples_per_second": 0.314,
|
968 |
+
"eval_steps_per_second": 0.314,
|
969 |
+
"step": 384
|
970 |
+
},
|
971 |
+
{
|
972 |
+
"epoch": 65.0,
|
973 |
+
"learning_rate": 6.750000000000001e-06,
|
974 |
+
"loss": 4.2129,
|
975 |
+
"step": 390
|
976 |
+
},
|
977 |
+
{
|
978 |
+
"epoch": 65.0,
|
979 |
+
"eval_accuracy": 0.3391984359726295,
|
980 |
+
"eval_loss": 4.382594585418701,
|
981 |
+
"eval_runtime": 3.1111,
|
982 |
+
"eval_samples_per_second": 0.321,
|
983 |
+
"eval_steps_per_second": 0.321,
|
984 |
+
"step": 390
|
985 |
+
},
|
986 |
+
{
|
987 |
+
"epoch": 66.0,
|
988 |
+
"learning_rate": 6.700000000000001e-06,
|
989 |
+
"loss": 4.1882,
|
990 |
+
"step": 396
|
991 |
+
},
|
992 |
+
{
|
993 |
+
"epoch": 66.0,
|
994 |
+
"eval_accuracy": 0.33724340175953077,
|
995 |
+
"eval_loss": 4.383427143096924,
|
996 |
+
"eval_runtime": 3.2114,
|
997 |
+
"eval_samples_per_second": 0.311,
|
998 |
+
"eval_steps_per_second": 0.311,
|
999 |
+
"step": 396
|
1000 |
+
},
|
1001 |
+
{
|
1002 |
+
"epoch": 67.0,
|
1003 |
+
"learning_rate": 6.650000000000001e-06,
|
1004 |
+
"loss": 4.1503,
|
1005 |
+
"step": 402
|
1006 |
+
},
|
1007 |
+
{
|
1008 |
+
"epoch": 67.0,
|
1009 |
+
"eval_accuracy": 0.33724340175953077,
|
1010 |
+
"eval_loss": 4.37380838394165,
|
1011 |
+
"eval_runtime": 3.1592,
|
1012 |
+
"eval_samples_per_second": 0.317,
|
1013 |
+
"eval_steps_per_second": 0.317,
|
1014 |
+
"step": 402
|
1015 |
+
},
|
1016 |
+
{
|
1017 |
+
"epoch": 68.0,
|
1018 |
+
"learning_rate": 6.600000000000001e-06,
|
1019 |
+
"loss": 4.1398,
|
1020 |
+
"step": 408
|
1021 |
+
},
|
1022 |
+
{
|
1023 |
+
"epoch": 68.0,
|
1024 |
+
"eval_accuracy": 0.33724340175953077,
|
1025 |
+
"eval_loss": 4.359556674957275,
|
1026 |
+
"eval_runtime": 3.2732,
|
1027 |
+
"eval_samples_per_second": 0.306,
|
1028 |
+
"eval_steps_per_second": 0.306,
|
1029 |
+
"step": 408
|
1030 |
+
},
|
1031 |
+
{
|
1032 |
+
"epoch": 69.0,
|
1033 |
+
"learning_rate": 6.550000000000001e-06,
|
1034 |
+
"loss": 4.115,
|
1035 |
+
"step": 414
|
1036 |
+
},
|
1037 |
+
{
|
1038 |
+
"epoch": 69.0,
|
1039 |
+
"eval_accuracy": 0.3411534701857282,
|
1040 |
+
"eval_loss": 4.337615966796875,
|
1041 |
+
"eval_runtime": 3.5266,
|
1042 |
+
"eval_samples_per_second": 0.284,
|
1043 |
+
"eval_steps_per_second": 0.284,
|
1044 |
+
"step": 414
|
1045 |
+
},
|
1046 |
+
{
|
1047 |
+
"epoch": 70.0,
|
1048 |
+
"learning_rate": 6.5000000000000004e-06,
|
1049 |
+
"loss": 4.1052,
|
1050 |
+
"step": 420
|
1051 |
+
},
|
1052 |
+
{
|
1053 |
+
"epoch": 70.0,
|
1054 |
+
"eval_accuracy": 0.3411534701857282,
|
1055 |
+
"eval_loss": 4.333048343658447,
|
1056 |
+
"eval_runtime": 3.1824,
|
1057 |
+
"eval_samples_per_second": 0.314,
|
1058 |
+
"eval_steps_per_second": 0.314,
|
1059 |
+
"step": 420
|
1060 |
+
},
|
1061 |
+
{
|
1062 |
+
"epoch": 71.0,
|
1063 |
+
"learning_rate": 6.450000000000001e-06,
|
1064 |
+
"loss": 4.0932,
|
1065 |
+
"step": 426
|
1066 |
+
},
|
1067 |
+
{
|
1068 |
+
"epoch": 71.0,
|
1069 |
+
"eval_accuracy": 0.3411534701857282,
|
1070 |
+
"eval_loss": 4.329537391662598,
|
1071 |
+
"eval_runtime": 3.1403,
|
1072 |
+
"eval_samples_per_second": 0.318,
|
1073 |
+
"eval_steps_per_second": 0.318,
|
1074 |
+
"step": 426
|
1075 |
+
},
|
1076 |
+
{
|
1077 |
+
"epoch": 72.0,
|
1078 |
+
"learning_rate": 6.4000000000000006e-06,
|
1079 |
+
"loss": 4.0573,
|
1080 |
+
"step": 432
|
1081 |
+
},
|
1082 |
+
{
|
1083 |
+
"epoch": 72.0,
|
1084 |
+
"eval_accuracy": 0.3411534701857282,
|
1085 |
+
"eval_loss": 4.311087131500244,
|
1086 |
+
"eval_runtime": 3.3734,
|
1087 |
+
"eval_samples_per_second": 0.296,
|
1088 |
+
"eval_steps_per_second": 0.296,
|
1089 |
+
"step": 432
|
1090 |
+
},
|
1091 |
+
{
|
1092 |
+
"epoch": 73.0,
|
1093 |
+
"learning_rate": 6.35e-06,
|
1094 |
+
"loss": 4.0449,
|
1095 |
+
"step": 438
|
1096 |
+
},
|
1097 |
+
{
|
1098 |
+
"epoch": 73.0,
|
1099 |
+
"eval_accuracy": 0.34408602150537637,
|
1100 |
+
"eval_loss": 4.304787635803223,
|
1101 |
+
"eval_runtime": 3.3551,
|
1102 |
+
"eval_samples_per_second": 0.298,
|
1103 |
+
"eval_steps_per_second": 0.298,
|
1104 |
+
"step": 438
|
1105 |
+
},
|
1106 |
+
{
|
1107 |
+
"epoch": 74.0,
|
1108 |
+
"learning_rate": 6.300000000000001e-06,
|
1109 |
+
"loss": 4.0165,
|
1110 |
+
"step": 444
|
1111 |
+
},
|
1112 |
+
{
|
1113 |
+
"epoch": 74.0,
|
1114 |
+
"eval_accuracy": 0.3460410557184751,
|
1115 |
+
"eval_loss": 4.293551445007324,
|
1116 |
+
"eval_runtime": 3.1253,
|
1117 |
+
"eval_samples_per_second": 0.32,
|
1118 |
+
"eval_steps_per_second": 0.32,
|
1119 |
+
"step": 444
|
1120 |
+
},
|
1121 |
+
{
|
1122 |
+
"epoch": 75.0,
|
1123 |
+
"learning_rate": 6.25e-06,
|
1124 |
+
"loss": 3.9936,
|
1125 |
+
"step": 450
|
1126 |
+
},
|
1127 |
+
{
|
1128 |
+
"epoch": 75.0,
|
1129 |
+
"eval_accuracy": 0.3509286412512219,
|
1130 |
+
"eval_loss": 4.281482696533203,
|
1131 |
+
"eval_runtime": 3.2326,
|
1132 |
+
"eval_samples_per_second": 0.309,
|
1133 |
+
"eval_steps_per_second": 0.309,
|
1134 |
+
"step": 450
|
1135 |
+
},
|
1136 |
+
{
|
1137 |
+
"epoch": 76.0,
|
1138 |
+
"learning_rate": 6.200000000000001e-06,
|
1139 |
+
"loss": 3.967,
|
1140 |
+
"step": 456
|
1141 |
+
},
|
1142 |
+
{
|
1143 |
+
"epoch": 76.0,
|
1144 |
+
"eval_accuracy": 0.35386119257087,
|
1145 |
+
"eval_loss": 4.268586158752441,
|
1146 |
+
"eval_runtime": 3.1926,
|
1147 |
+
"eval_samples_per_second": 0.313,
|
1148 |
+
"eval_steps_per_second": 0.313,
|
1149 |
+
"step": 456
|
1150 |
+
},
|
1151 |
+
{
|
1152 |
+
"epoch": 77.0,
|
1153 |
+
"learning_rate": 6.15e-06,
|
1154 |
+
"loss": 3.9524,
|
1155 |
+
"step": 462
|
1156 |
+
},
|
1157 |
+
{
|
1158 |
+
"epoch": 77.0,
|
1159 |
+
"eval_accuracy": 0.3509286412512219,
|
1160 |
+
"eval_loss": 4.269726276397705,
|
1161 |
+
"eval_runtime": 3.0654,
|
1162 |
+
"eval_samples_per_second": 0.326,
|
1163 |
+
"eval_steps_per_second": 0.326,
|
1164 |
+
"step": 462
|
1165 |
+
},
|
1166 |
+
{
|
1167 |
+
"epoch": 78.0,
|
1168 |
+
"learning_rate": 6.1e-06,
|
1169 |
+
"loss": 3.9287,
|
1170 |
+
"step": 468
|
1171 |
+
},
|
1172 |
+
{
|
1173 |
+
"epoch": 78.0,
|
1174 |
+
"eval_accuracy": 0.35288367546432065,
|
1175 |
+
"eval_loss": 4.254611015319824,
|
1176 |
+
"eval_runtime": 3.2483,
|
1177 |
+
"eval_samples_per_second": 0.308,
|
1178 |
+
"eval_steps_per_second": 0.308,
|
1179 |
+
"step": 468
|
1180 |
+
},
|
1181 |
+
{
|
1182 |
+
"epoch": 79.0,
|
1183 |
+
"learning_rate": 6.0500000000000005e-06,
|
1184 |
+
"loss": 3.9092,
|
1185 |
+
"step": 474
|
1186 |
+
},
|
1187 |
+
{
|
1188 |
+
"epoch": 79.0,
|
1189 |
+
"eval_accuracy": 0.35386119257087,
|
1190 |
+
"eval_loss": 4.248389720916748,
|
1191 |
+
"eval_runtime": 3.279,
|
1192 |
+
"eval_samples_per_second": 0.305,
|
1193 |
+
"eval_steps_per_second": 0.305,
|
1194 |
+
"step": 474
|
1195 |
+
},
|
1196 |
+
{
|
1197 |
+
"epoch": 80.0,
|
1198 |
+
"learning_rate": 6e-06,
|
1199 |
+
"loss": 3.8907,
|
1200 |
+
"step": 480
|
1201 |
+
},
|
1202 |
+
{
|
1203 |
+
"epoch": 80.0,
|
1204 |
+
"eval_accuracy": 0.35386119257087,
|
1205 |
+
"eval_loss": 4.242007255554199,
|
1206 |
+
"eval_runtime": 3.1903,
|
1207 |
+
"eval_samples_per_second": 0.313,
|
1208 |
+
"eval_steps_per_second": 0.313,
|
1209 |
+
"step": 480
|
1210 |
+
},
|
1211 |
+
{
|
1212 |
+
"epoch": 81.0,
|
1213 |
+
"learning_rate": 5.950000000000001e-06,
|
1214 |
+
"loss": 3.8704,
|
1215 |
+
"step": 486
|
1216 |
+
},
|
1217 |
+
{
|
1218 |
+
"epoch": 81.0,
|
1219 |
+
"eval_accuracy": 0.35288367546432065,
|
1220 |
+
"eval_loss": 4.241766929626465,
|
1221 |
+
"eval_runtime": 3.0667,
|
1222 |
+
"eval_samples_per_second": 0.326,
|
1223 |
+
"eval_steps_per_second": 0.326,
|
1224 |
+
"step": 486
|
1225 |
+
},
|
1226 |
+
{
|
1227 |
+
"epoch": 82.0,
|
1228 |
+
"learning_rate": 5.9e-06,
|
1229 |
+
"loss": 3.8499,
|
1230 |
+
"step": 492
|
1231 |
+
},
|
1232 |
+
{
|
1233 |
+
"epoch": 82.0,
|
1234 |
+
"eval_accuracy": 0.3548387096774194,
|
1235 |
+
"eval_loss": 4.226541519165039,
|
1236 |
+
"eval_runtime": 3.1032,
|
1237 |
+
"eval_samples_per_second": 0.322,
|
1238 |
+
"eval_steps_per_second": 0.322,
|
1239 |
+
"step": 492
|
1240 |
+
},
|
1241 |
+
{
|
1242 |
+
"epoch": 83.0,
|
1243 |
+
"learning_rate": 5.85e-06,
|
1244 |
+
"loss": 3.8325,
|
1245 |
+
"step": 498
|
1246 |
+
},
|
1247 |
+
{
|
1248 |
+
"epoch": 83.0,
|
1249 |
+
"eval_accuracy": 0.3548387096774194,
|
1250 |
+
"eval_loss": 4.208946704864502,
|
1251 |
+
"eval_runtime": 3.0764,
|
1252 |
+
"eval_samples_per_second": 0.325,
|
1253 |
+
"eval_steps_per_second": 0.325,
|
1254 |
+
"step": 498
|
1255 |
+
},
|
1256 |
+
{
|
1257 |
+
"epoch": 84.0,
|
1258 |
+
"learning_rate": 5.8e-06,
|
1259 |
+
"loss": 3.8024,
|
1260 |
+
"step": 504
|
1261 |
+
},
|
1262 |
+
{
|
1263 |
+
"epoch": 84.0,
|
1264 |
+
"eval_accuracy": 0.3567937438905181,
|
1265 |
+
"eval_loss": 4.205758094787598,
|
1266 |
+
"eval_runtime": 3.0957,
|
1267 |
+
"eval_samples_per_second": 0.323,
|
1268 |
+
"eval_steps_per_second": 0.323,
|
1269 |
+
"step": 504
|
1270 |
+
},
|
1271 |
+
{
|
1272 |
+
"epoch": 85.0,
|
1273 |
+
"learning_rate": 5.75e-06,
|
1274 |
+
"loss": 3.8058,
|
1275 |
+
"step": 510
|
1276 |
+
},
|
1277 |
+
{
|
1278 |
+
"epoch": 85.0,
|
1279 |
+
"eval_accuracy": 0.35581622678396874,
|
1280 |
+
"eval_loss": 4.203883647918701,
|
1281 |
+
"eval_runtime": 3.1095,
|
1282 |
+
"eval_samples_per_second": 0.322,
|
1283 |
+
"eval_steps_per_second": 0.322,
|
1284 |
+
"step": 510
|
1285 |
+
},
|
1286 |
+
{
|
1287 |
+
"epoch": 86.0,
|
1288 |
+
"learning_rate": 5.7e-06,
|
1289 |
+
"loss": 3.7888,
|
1290 |
+
"step": 516
|
1291 |
+
},
|
1292 |
+
{
|
1293 |
+
"epoch": 86.0,
|
1294 |
+
"eval_accuracy": 0.35777126099706746,
|
1295 |
+
"eval_loss": 4.190591335296631,
|
1296 |
+
"eval_runtime": 3.1,
|
1297 |
+
"eval_samples_per_second": 0.323,
|
1298 |
+
"eval_steps_per_second": 0.323,
|
1299 |
+
"step": 516
|
1300 |
+
},
|
1301 |
+
{
|
1302 |
+
"epoch": 87.0,
|
1303 |
+
"learning_rate": 5.65e-06,
|
1304 |
+
"loss": 3.7622,
|
1305 |
+
"step": 522
|
1306 |
+
},
|
1307 |
+
{
|
1308 |
+
"epoch": 87.0,
|
1309 |
+
"eval_accuracy": 0.3616813294232649,
|
1310 |
+
"eval_loss": 4.179159164428711,
|
1311 |
+
"eval_runtime": 3.1107,
|
1312 |
+
"eval_samples_per_second": 0.321,
|
1313 |
+
"eval_steps_per_second": 0.321,
|
1314 |
+
"step": 522
|
1315 |
+
},
|
1316 |
+
{
|
1317 |
+
"epoch": 88.0,
|
1318 |
+
"learning_rate": 5.600000000000001e-06,
|
1319 |
+
"loss": 3.746,
|
1320 |
+
"step": 528
|
1321 |
+
},
|
1322 |
+
{
|
1323 |
+
"epoch": 88.0,
|
1324 |
+
"eval_accuracy": 0.35777126099706746,
|
1325 |
+
"eval_loss": 4.18191385269165,
|
1326 |
+
"eval_runtime": 3.1176,
|
1327 |
+
"eval_samples_per_second": 0.321,
|
1328 |
+
"eval_steps_per_second": 0.321,
|
1329 |
+
"step": 528
|
1330 |
+
},
|
1331 |
+
{
|
1332 |
+
"epoch": 89.0,
|
1333 |
+
"learning_rate": 5.550000000000001e-06,
|
1334 |
+
"loss": 3.7196,
|
1335 |
+
"step": 534
|
1336 |
+
},
|
1337 |
+
{
|
1338 |
+
"epoch": 89.0,
|
1339 |
+
"eval_accuracy": 0.3597262952101662,
|
1340 |
+
"eval_loss": 4.178895473480225,
|
1341 |
+
"eval_runtime": 3.3763,
|
1342 |
+
"eval_samples_per_second": 0.296,
|
1343 |
+
"eval_steps_per_second": 0.296,
|
1344 |
+
"step": 534
|
1345 |
+
},
|
1346 |
+
{
|
1347 |
+
"epoch": 90.0,
|
1348 |
+
"learning_rate": 5.500000000000001e-06,
|
1349 |
+
"loss": 3.7046,
|
1350 |
+
"step": 540
|
1351 |
+
},
|
1352 |
+
{
|
1353 |
+
"epoch": 90.0,
|
1354 |
+
"eval_accuracy": 0.36070381231671556,
|
1355 |
+
"eval_loss": 4.160989284515381,
|
1356 |
+
"eval_runtime": 3.3055,
|
1357 |
+
"eval_samples_per_second": 0.303,
|
1358 |
+
"eval_steps_per_second": 0.303,
|
1359 |
+
"step": 540
|
1360 |
+
},
|
1361 |
+
{
|
1362 |
+
"epoch": 91.0,
|
1363 |
+
"learning_rate": 5.450000000000001e-06,
|
1364 |
+
"loss": 3.7078,
|
1365 |
+
"step": 546
|
1366 |
+
},
|
1367 |
+
{
|
1368 |
+
"epoch": 91.0,
|
1369 |
+
"eval_accuracy": 0.36070381231671556,
|
1370 |
+
"eval_loss": 4.151462078094482,
|
1371 |
+
"eval_runtime": 3.0842,
|
1372 |
+
"eval_samples_per_second": 0.324,
|
1373 |
+
"eval_steps_per_second": 0.324,
|
1374 |
+
"step": 546
|
1375 |
+
},
|
1376 |
+
{
|
1377 |
+
"epoch": 92.0,
|
1378 |
+
"learning_rate": 5.400000000000001e-06,
|
1379 |
+
"loss": 3.6687,
|
1380 |
+
"step": 552
|
1381 |
+
},
|
1382 |
+
{
|
1383 |
+
"epoch": 92.0,
|
1384 |
+
"eval_accuracy": 0.36070381231671556,
|
1385 |
+
"eval_loss": 4.17518424987793,
|
1386 |
+
"eval_runtime": 3.097,
|
1387 |
+
"eval_samples_per_second": 0.323,
|
1388 |
+
"eval_steps_per_second": 0.323,
|
1389 |
+
"step": 552
|
1390 |
+
},
|
1391 |
+
{
|
1392 |
+
"epoch": 93.0,
|
1393 |
+
"learning_rate": 5.3500000000000004e-06,
|
1394 |
+
"loss": 3.6559,
|
1395 |
+
"step": 558
|
1396 |
+
},
|
1397 |
+
{
|
1398 |
+
"epoch": 93.0,
|
1399 |
+
"eval_accuracy": 0.36363636363636365,
|
1400 |
+
"eval_loss": 4.128724575042725,
|
1401 |
+
"eval_runtime": 3.2229,
|
1402 |
+
"eval_samples_per_second": 0.31,
|
1403 |
+
"eval_steps_per_second": 0.31,
|
1404 |
+
"step": 558
|
1405 |
+
},
|
1406 |
+
{
|
1407 |
+
"epoch": 94.0,
|
1408 |
+
"learning_rate": 5.300000000000001e-06,
|
1409 |
+
"loss": 3.6401,
|
1410 |
+
"step": 564
|
1411 |
+
},
|
1412 |
+
{
|
1413 |
+
"epoch": 94.0,
|
1414 |
+
"eval_accuracy": 0.364613880742913,
|
1415 |
+
"eval_loss": 4.156914234161377,
|
1416 |
+
"eval_runtime": 3.0553,
|
1417 |
+
"eval_samples_per_second": 0.327,
|
1418 |
+
"eval_steps_per_second": 0.327,
|
1419 |
+
"step": 564
|
1420 |
+
},
|
1421 |
+
{
|
1422 |
+
"epoch": 95.0,
|
1423 |
+
"learning_rate": 5.2500000000000006e-06,
|
1424 |
+
"loss": 3.6281,
|
1425 |
+
"step": 570
|
1426 |
+
},
|
1427 |
+
{
|
1428 |
+
"epoch": 95.0,
|
1429 |
+
"eval_accuracy": 0.3626588465298143,
|
1430 |
+
"eval_loss": 4.123438358306885,
|
1431 |
+
"eval_runtime": 3.127,
|
1432 |
+
"eval_samples_per_second": 0.32,
|
1433 |
+
"eval_steps_per_second": 0.32,
|
1434 |
+
"step": 570
|
1435 |
+
},
|
1436 |
+
{
|
1437 |
+
"epoch": 96.0,
|
1438 |
+
"learning_rate": 5.2e-06,
|
1439 |
+
"loss": 3.5978,
|
1440 |
+
"step": 576
|
1441 |
+
},
|
1442 |
+
{
|
1443 |
+
"epoch": 96.0,
|
1444 |
+
"eval_accuracy": 0.36950146627565983,
|
1445 |
+
"eval_loss": 4.1269941329956055,
|
1446 |
+
"eval_runtime": 3.0893,
|
1447 |
+
"eval_samples_per_second": 0.324,
|
1448 |
+
"eval_steps_per_second": 0.324,
|
1449 |
+
"step": 576
|
1450 |
+
},
|
1451 |
+
{
|
1452 |
+
"epoch": 97.0,
|
1453 |
+
"learning_rate": 5.150000000000001e-06,
|
1454 |
+
"loss": 3.5951,
|
1455 |
+
"step": 582
|
1456 |
+
},
|
1457 |
+
{
|
1458 |
+
"epoch": 97.0,
|
1459 |
+
"eval_accuracy": 0.364613880742913,
|
1460 |
+
"eval_loss": 4.118751049041748,
|
1461 |
+
"eval_runtime": 3.148,
|
1462 |
+
"eval_samples_per_second": 0.318,
|
1463 |
+
"eval_steps_per_second": 0.318,
|
1464 |
+
"step": 582
|
1465 |
+
},
|
1466 |
+
{
|
1467 |
+
"epoch": 98.0,
|
1468 |
+
"learning_rate": 5.1e-06,
|
1469 |
+
"loss": 3.5679,
|
1470 |
+
"step": 588
|
1471 |
+
},
|
1472 |
+
{
|
1473 |
+
"epoch": 98.0,
|
1474 |
+
"eval_accuracy": 0.36852394916911047,
|
1475 |
+
"eval_loss": 4.128164291381836,
|
1476 |
+
"eval_runtime": 3.0607,
|
1477 |
+
"eval_samples_per_second": 0.327,
|
1478 |
+
"eval_steps_per_second": 0.327,
|
1479 |
+
"step": 588
|
1480 |
+
},
|
1481 |
+
{
|
1482 |
+
"epoch": 99.0,
|
1483 |
+
"learning_rate": 5.050000000000001e-06,
|
1484 |
+
"loss": 3.5618,
|
1485 |
+
"step": 594
|
1486 |
+
},
|
1487 |
+
{
|
1488 |
+
"epoch": 99.0,
|
1489 |
+
"eval_accuracy": 0.364613880742913,
|
1490 |
+
"eval_loss": 4.1089253425598145,
|
1491 |
+
"eval_runtime": 3.0857,
|
1492 |
+
"eval_samples_per_second": 0.324,
|
1493 |
+
"eval_steps_per_second": 0.324,
|
1494 |
+
"step": 594
|
1495 |
+
},
|
1496 |
+
{
|
1497 |
+
"epoch": 100.0,
|
1498 |
+
"learning_rate": 5e-06,
|
1499 |
+
"loss": 3.5404,
|
1500 |
+
"step": 600
|
1501 |
+
},
|
1502 |
+
{
|
1503 |
+
"epoch": 100.0,
|
1504 |
+
"eval_accuracy": 0.36950146627565983,
|
1505 |
+
"eval_loss": 4.1090216636657715,
|
1506 |
+
"eval_runtime": 3.1735,
|
1507 |
+
"eval_samples_per_second": 0.315,
|
1508 |
+
"eval_steps_per_second": 0.315,
|
1509 |
+
"step": 600
|
1510 |
+
},
|
1511 |
+
{
|
1512 |
+
"epoch": 101.0,
|
1513 |
+
"learning_rate": 4.95e-06,
|
1514 |
+
"loss": 3.5255,
|
1515 |
+
"step": 606
|
1516 |
+
},
|
1517 |
+
{
|
1518 |
+
"epoch": 101.0,
|
1519 |
+
"eval_accuracy": 0.364613880742913,
|
1520 |
+
"eval_loss": 4.103869915008545,
|
1521 |
+
"eval_runtime": 3.1307,
|
1522 |
+
"eval_samples_per_second": 0.319,
|
1523 |
+
"eval_steps_per_second": 0.319,
|
1524 |
+
"step": 606
|
1525 |
+
},
|
1526 |
+
{
|
1527 |
+
"epoch": 102.0,
|
1528 |
+
"learning_rate": 4.9000000000000005e-06,
|
1529 |
+
"loss": 3.5111,
|
1530 |
+
"step": 612
|
1531 |
+
},
|
1532 |
+
{
|
1533 |
+
"epoch": 102.0,
|
1534 |
+
"eval_accuracy": 0.36950146627565983,
|
1535 |
+
"eval_loss": 4.100970268249512,
|
1536 |
+
"eval_runtime": 3.116,
|
1537 |
+
"eval_samples_per_second": 0.321,
|
1538 |
+
"eval_steps_per_second": 0.321,
|
1539 |
+
"step": 612
|
1540 |
+
},
|
1541 |
+
{
|
1542 |
+
"epoch": 103.0,
|
1543 |
+
"learning_rate": 4.85e-06,
|
1544 |
+
"loss": 3.5015,
|
1545 |
+
"step": 618
|
1546 |
+
},
|
1547 |
+
{
|
1548 |
+
"epoch": 103.0,
|
1549 |
+
"eval_accuracy": 0.3704789833822092,
|
1550 |
+
"eval_loss": 4.088879585266113,
|
1551 |
+
"eval_runtime": 3.3403,
|
1552 |
+
"eval_samples_per_second": 0.299,
|
1553 |
+
"eval_steps_per_second": 0.299,
|
1554 |
+
"step": 618
|
1555 |
+
},
|
1556 |
+
{
|
1557 |
+
"epoch": 104.0,
|
1558 |
+
"learning_rate": 4.800000000000001e-06,
|
1559 |
+
"loss": 3.493,
|
1560 |
+
"step": 624
|
1561 |
+
},
|
1562 |
+
{
|
1563 |
+
"epoch": 104.0,
|
1564 |
+
"eval_accuracy": 0.3704789833822092,
|
1565 |
+
"eval_loss": 4.082551956176758,
|
1566 |
+
"eval_runtime": 3.3467,
|
1567 |
+
"eval_samples_per_second": 0.299,
|
1568 |
+
"eval_steps_per_second": 0.299,
|
1569 |
+
"step": 624
|
1570 |
+
},
|
1571 |
+
{
|
1572 |
+
"epoch": 105.0,
|
1573 |
+
"learning_rate": 4.75e-06,
|
1574 |
+
"loss": 3.5643,
|
1575 |
+
"step": 630
|
1576 |
+
},
|
1577 |
+
{
|
1578 |
+
"epoch": 105.0,
|
1579 |
+
"eval_accuracy": 0.375366568914956,
|
1580 |
+
"eval_loss": 4.091529846191406,
|
1581 |
+
"eval_runtime": 3.5246,
|
1582 |
+
"eval_samples_per_second": 0.284,
|
1583 |
+
"eval_steps_per_second": 0.284,
|
1584 |
+
"step": 630
|
1585 |
+
},
|
1586 |
+
{
|
1587 |
+
"epoch": 106.0,
|
1588 |
+
"learning_rate": 4.7e-06,
|
1589 |
+
"loss": 3.4543,
|
1590 |
+
"step": 636
|
1591 |
+
},
|
1592 |
+
{
|
1593 |
+
"epoch": 106.0,
|
1594 |
+
"eval_accuracy": 0.3724340175953079,
|
1595 |
+
"eval_loss": 4.091198921203613,
|
1596 |
+
"eval_runtime": 3.1421,
|
1597 |
+
"eval_samples_per_second": 0.318,
|
1598 |
+
"eval_steps_per_second": 0.318,
|
1599 |
+
"step": 636
|
1600 |
+
},
|
1601 |
+
{
|
1602 |
+
"epoch": 107.0,
|
1603 |
+
"learning_rate": 4.65e-06,
|
1604 |
+
"loss": 3.4517,
|
1605 |
+
"step": 642
|
1606 |
+
},
|
1607 |
+
{
|
1608 |
+
"epoch": 107.0,
|
1609 |
+
"eval_accuracy": 0.375366568914956,
|
1610 |
+
"eval_loss": 4.084360599517822,
|
1611 |
+
"eval_runtime": 3.5262,
|
1612 |
+
"eval_samples_per_second": 0.284,
|
1613 |
+
"eval_steps_per_second": 0.284,
|
1614 |
+
"step": 642
|
1615 |
+
},
|
1616 |
+
{
|
1617 |
+
"epoch": 108.0,
|
1618 |
+
"learning_rate": 4.600000000000001e-06,
|
1619 |
+
"loss": 3.4387,
|
1620 |
+
"step": 648
|
1621 |
+
},
|
1622 |
+
{
|
1623 |
+
"epoch": 108.0,
|
1624 |
+
"eval_accuracy": 0.375366568914956,
|
1625 |
+
"eval_loss": 4.066359996795654,
|
1626 |
+
"eval_runtime": 3.1772,
|
1627 |
+
"eval_samples_per_second": 0.315,
|
1628 |
+
"eval_steps_per_second": 0.315,
|
1629 |
+
"step": 648
|
1630 |
+
},
|
1631 |
+
{
|
1632 |
+
"epoch": 109.0,
|
1633 |
+
"learning_rate": 4.5500000000000005e-06,
|
1634 |
+
"loss": 3.4274,
|
1635 |
+
"step": 654
|
1636 |
+
},
|
1637 |
+
{
|
1638 |
+
"epoch": 109.0,
|
1639 |
+
"eval_accuracy": 0.3763440860215054,
|
1640 |
+
"eval_loss": 4.088470458984375,
|
1641 |
+
"eval_runtime": 3.106,
|
1642 |
+
"eval_samples_per_second": 0.322,
|
1643 |
+
"eval_steps_per_second": 0.322,
|
1644 |
+
"step": 654
|
1645 |
+
},
|
1646 |
+
{
|
1647 |
+
"epoch": 110.0,
|
1648 |
+
"learning_rate": 4.5e-06,
|
1649 |
+
"loss": 3.4241,
|
1650 |
+
"step": 660
|
1651 |
+
},
|
1652 |
+
{
|
1653 |
+
"epoch": 110.0,
|
1654 |
+
"eval_accuracy": 0.37927663734115347,
|
1655 |
+
"eval_loss": 4.058298587799072,
|
1656 |
+
"eval_runtime": 3.0985,
|
1657 |
+
"eval_samples_per_second": 0.323,
|
1658 |
+
"eval_steps_per_second": 0.323,
|
1659 |
+
"step": 660
|
1660 |
+
},
|
1661 |
+
{
|
1662 |
+
"epoch": 111.0,
|
1663 |
+
"learning_rate": 4.450000000000001e-06,
|
1664 |
+
"loss": 3.4016,
|
1665 |
+
"step": 666
|
1666 |
+
},
|
1667 |
+
{
|
1668 |
+
"epoch": 111.0,
|
1669 |
+
"eval_accuracy": 0.38025415444770283,
|
1670 |
+
"eval_loss": 4.0626630783081055,
|
1671 |
+
"eval_runtime": 3.5144,
|
1672 |
+
"eval_samples_per_second": 0.285,
|
1673 |
+
"eval_steps_per_second": 0.285,
|
1674 |
+
"step": 666
|
1675 |
+
},
|
1676 |
+
{
|
1677 |
+
"epoch": 112.0,
|
1678 |
+
"learning_rate": 4.4e-06,
|
1679 |
+
"loss": 3.383,
|
1680 |
+
"step": 672
|
1681 |
+
},
|
1682 |
+
{
|
1683 |
+
"epoch": 112.0,
|
1684 |
+
"eval_accuracy": 0.3812316715542522,
|
1685 |
+
"eval_loss": 4.062616348266602,
|
1686 |
+
"eval_runtime": 3.2502,
|
1687 |
+
"eval_samples_per_second": 0.308,
|
1688 |
+
"eval_steps_per_second": 0.308,
|
1689 |
+
"step": 672
|
1690 |
+
},
|
1691 |
+
{
|
1692 |
+
"epoch": 113.0,
|
1693 |
+
"learning_rate": 4.350000000000001e-06,
|
1694 |
+
"loss": 3.3709,
|
1695 |
+
"step": 678
|
1696 |
+
},
|
1697 |
+
{
|
1698 |
+
"epoch": 113.0,
|
1699 |
+
"eval_accuracy": 0.3870967741935484,
|
1700 |
+
"eval_loss": 4.0414323806762695,
|
1701 |
+
"eval_runtime": 3.4468,
|
1702 |
+
"eval_samples_per_second": 0.29,
|
1703 |
+
"eval_steps_per_second": 0.29,
|
1704 |
+
"step": 678
|
1705 |
+
},
|
1706 |
+
{
|
1707 |
+
"epoch": 114.0,
|
1708 |
+
"learning_rate": 4.3e-06,
|
1709 |
+
"loss": 3.3646,
|
1710 |
+
"step": 684
|
1711 |
+
},
|
1712 |
+
{
|
1713 |
+
"epoch": 114.0,
|
1714 |
+
"eval_accuracy": 0.38220918866080156,
|
1715 |
+
"eval_loss": 4.0561604499816895,
|
1716 |
+
"eval_runtime": 3.0962,
|
1717 |
+
"eval_samples_per_second": 0.323,
|
1718 |
+
"eval_steps_per_second": 0.323,
|
1719 |
+
"step": 684
|
1720 |
+
},
|
1721 |
+
{
|
1722 |
+
"epoch": 115.0,
|
1723 |
+
"learning_rate": 4.25e-06,
|
1724 |
+
"loss": 3.3456,
|
1725 |
+
"step": 690
|
1726 |
+
},
|
1727 |
+
{
|
1728 |
+
"epoch": 115.0,
|
1729 |
+
"eval_accuracy": 0.386119257086999,
|
1730 |
+
"eval_loss": 4.036128044128418,
|
1731 |
+
"eval_runtime": 3.3227,
|
1732 |
+
"eval_samples_per_second": 0.301,
|
1733 |
+
"eval_steps_per_second": 0.301,
|
1734 |
+
"step": 690
|
1735 |
+
},
|
1736 |
+
{
|
1737 |
+
"epoch": 116.0,
|
1738 |
+
"learning_rate": 4.2000000000000004e-06,
|
1739 |
+
"loss": 3.3369,
|
1740 |
+
"step": 696
|
1741 |
+
},
|
1742 |
+
{
|
1743 |
+
"epoch": 116.0,
|
1744 |
+
"eval_accuracy": 0.38514173998044965,
|
1745 |
+
"eval_loss": 4.052359104156494,
|
1746 |
+
"eval_runtime": 3.0856,
|
1747 |
+
"eval_samples_per_second": 0.324,
|
1748 |
+
"eval_steps_per_second": 0.324,
|
1749 |
+
"step": 696
|
1750 |
+
},
|
1751 |
+
{
|
1752 |
+
"epoch": 117.0,
|
1753 |
+
"learning_rate": 4.15e-06,
|
1754 |
+
"loss": 3.3136,
|
1755 |
+
"step": 702
|
1756 |
+
},
|
1757 |
+
{
|
1758 |
+
"epoch": 117.0,
|
1759 |
+
"eval_accuracy": 0.3841642228739003,
|
1760 |
+
"eval_loss": 4.042443752288818,
|
1761 |
+
"eval_runtime": 3.358,
|
1762 |
+
"eval_samples_per_second": 0.298,
|
1763 |
+
"eval_steps_per_second": 0.298,
|
1764 |
+
"step": 702
|
1765 |
+
},
|
1766 |
+
{
|
1767 |
+
"epoch": 118.0,
|
1768 |
+
"learning_rate": 4.1e-06,
|
1769 |
+
"loss": 3.307,
|
1770 |
+
"step": 708
|
1771 |
+
},
|
1772 |
+
{
|
1773 |
+
"epoch": 118.0,
|
1774 |
+
"eval_accuracy": 0.386119257086999,
|
1775 |
+
"eval_loss": 4.047730922698975,
|
1776 |
+
"eval_runtime": 3.2939,
|
1777 |
+
"eval_samples_per_second": 0.304,
|
1778 |
+
"eval_steps_per_second": 0.304,
|
1779 |
+
"step": 708
|
1780 |
+
},
|
1781 |
+
{
|
1782 |
+
"epoch": 119.0,
|
1783 |
+
"learning_rate": 4.05e-06,
|
1784 |
+
"loss": 3.2954,
|
1785 |
+
"step": 714
|
1786 |
+
},
|
1787 |
+
{
|
1788 |
+
"epoch": 119.0,
|
1789 |
+
"eval_accuracy": 0.38514173998044965,
|
1790 |
+
"eval_loss": 4.028741359710693,
|
1791 |
+
"eval_runtime": 3.1849,
|
1792 |
+
"eval_samples_per_second": 0.314,
|
1793 |
+
"eval_steps_per_second": 0.314,
|
1794 |
+
"step": 714
|
1795 |
+
},
|
1796 |
+
{
|
1797 |
+
"epoch": 120.0,
|
1798 |
+
"learning_rate": 4.000000000000001e-06,
|
1799 |
+
"loss": 3.2887,
|
1800 |
+
"step": 720
|
1801 |
+
},
|
1802 |
+
{
|
1803 |
+
"epoch": 120.0,
|
1804 |
+
"eval_accuracy": 0.39002932551319647,
|
1805 |
+
"eval_loss": 4.0391716957092285,
|
1806 |
+
"eval_runtime": 3.373,
|
1807 |
+
"eval_samples_per_second": 0.296,
|
1808 |
+
"eval_steps_per_second": 0.296,
|
1809 |
+
"step": 720
|
1810 |
+
},
|
1811 |
+
{
|
1812 |
+
"epoch": 121.0,
|
1813 |
+
"learning_rate": 3.95e-06,
|
1814 |
+
"loss": 3.2776,
|
1815 |
+
"step": 726
|
1816 |
+
},
|
1817 |
+
{
|
1818 |
+
"epoch": 121.0,
|
1819 |
+
"eval_accuracy": 0.39100684261974583,
|
1820 |
+
"eval_loss": 4.019059658050537,
|
1821 |
+
"eval_runtime": 3.0671,
|
1822 |
+
"eval_samples_per_second": 0.326,
|
1823 |
+
"eval_steps_per_second": 0.326,
|
1824 |
+
"step": 726
|
1825 |
+
},
|
1826 |
+
{
|
1827 |
+
"epoch": 122.0,
|
1828 |
+
"learning_rate": 3.900000000000001e-06,
|
1829 |
+
"loss": 3.2527,
|
1830 |
+
"step": 732
|
1831 |
+
},
|
1832 |
+
{
|
1833 |
+
"epoch": 122.0,
|
1834 |
+
"eval_accuracy": 0.39100684261974583,
|
1835 |
+
"eval_loss": 4.03394889831543,
|
1836 |
+
"eval_runtime": 3.608,
|
1837 |
+
"eval_samples_per_second": 0.277,
|
1838 |
+
"eval_steps_per_second": 0.277,
|
1839 |
+
"step": 732
|
1840 |
+
},
|
1841 |
+
{
|
1842 |
+
"epoch": 123.0,
|
1843 |
+
"learning_rate": 3.85e-06,
|
1844 |
+
"loss": 3.259,
|
1845 |
+
"step": 738
|
1846 |
+
},
|
1847 |
+
{
|
1848 |
+
"epoch": 123.0,
|
1849 |
+
"eval_accuracy": 0.39296187683284456,
|
1850 |
+
"eval_loss": 4.006428241729736,
|
1851 |
+
"eval_runtime": 3.37,
|
1852 |
+
"eval_samples_per_second": 0.297,
|
1853 |
+
"eval_steps_per_second": 0.297,
|
1854 |
+
"step": 738
|
1855 |
+
},
|
1856 |
+
{
|
1857 |
+
"epoch": 124.0,
|
1858 |
+
"learning_rate": 3.8000000000000005e-06,
|
1859 |
+
"loss": 3.2559,
|
1860 |
+
"step": 744
|
1861 |
+
},
|
1862 |
+
{
|
1863 |
+
"epoch": 124.0,
|
1864 |
+
"eval_accuracy": 0.38807429130009774,
|
1865 |
+
"eval_loss": 4.028494834899902,
|
1866 |
+
"eval_runtime": 3.4043,
|
1867 |
+
"eval_samples_per_second": 0.294,
|
1868 |
+
"eval_steps_per_second": 0.294,
|
1869 |
+
"step": 744
|
1870 |
+
},
|
1871 |
+
{
|
1872 |
+
"epoch": 125.0,
|
1873 |
+
"learning_rate": 3.7500000000000005e-06,
|
1874 |
+
"loss": 3.2335,
|
1875 |
+
"step": 750
|
1876 |
+
},
|
1877 |
+
{
|
1878 |
+
"epoch": 125.0,
|
1879 |
+
"eval_accuracy": 0.39296187683284456,
|
1880 |
+
"eval_loss": 4.015052795410156,
|
1881 |
+
"eval_runtime": 3.1425,
|
1882 |
+
"eval_samples_per_second": 0.318,
|
1883 |
+
"eval_steps_per_second": 0.318,
|
1884 |
+
"step": 750
|
1885 |
+
},
|
1886 |
+
{
|
1887 |
+
"epoch": 126.0,
|
1888 |
+
"learning_rate": 3.7e-06,
|
1889 |
+
"loss": 3.2318,
|
1890 |
+
"step": 756
|
1891 |
+
},
|
1892 |
+
{
|
1893 |
+
"epoch": 126.0,
|
1894 |
+
"eval_accuracy": 0.39002932551319647,
|
1895 |
+
"eval_loss": 4.027667999267578,
|
1896 |
+
"eval_runtime": 3.3897,
|
1897 |
+
"eval_samples_per_second": 0.295,
|
1898 |
+
"eval_steps_per_second": 0.295,
|
1899 |
+
"step": 756
|
1900 |
+
},
|
1901 |
+
{
|
1902 |
+
"epoch": 127.0,
|
1903 |
+
"learning_rate": 3.65e-06,
|
1904 |
+
"loss": 3.2266,
|
1905 |
+
"step": 762
|
1906 |
+
},
|
1907 |
+
{
|
1908 |
+
"epoch": 127.0,
|
1909 |
+
"eval_accuracy": 0.3978494623655914,
|
1910 |
+
"eval_loss": 3.992929697036743,
|
1911 |
+
"eval_runtime": 3.1289,
|
1912 |
+
"eval_samples_per_second": 0.32,
|
1913 |
+
"eval_steps_per_second": 0.32,
|
1914 |
+
"step": 762
|
1915 |
+
},
|
1916 |
+
{
|
1917 |
+
"epoch": 128.0,
|
1918 |
+
"learning_rate": 3.6000000000000003e-06,
|
1919 |
+
"loss": 3.2051,
|
1920 |
+
"step": 768
|
1921 |
+
},
|
1922 |
+
{
|
1923 |
+
"epoch": 128.0,
|
1924 |
+
"eval_accuracy": 0.3978494623655914,
|
1925 |
+
"eval_loss": 3.9944605827331543,
|
1926 |
+
"eval_runtime": 3.5639,
|
1927 |
+
"eval_samples_per_second": 0.281,
|
1928 |
+
"eval_steps_per_second": 0.281,
|
1929 |
+
"step": 768
|
1930 |
+
},
|
1931 |
+
{
|
1932 |
+
"epoch": 129.0,
|
1933 |
+
"learning_rate": 3.5500000000000003e-06,
|
1934 |
+
"loss": 3.2009,
|
1935 |
+
"step": 774
|
1936 |
+
},
|
1937 |
+
{
|
1938 |
+
"epoch": 129.0,
|
1939 |
+
"eval_accuracy": 0.39296187683284456,
|
1940 |
+
"eval_loss": 4.02908182144165,
|
1941 |
+
"eval_runtime": 3.1278,
|
1942 |
+
"eval_samples_per_second": 0.32,
|
1943 |
+
"eval_steps_per_second": 0.32,
|
1944 |
+
"step": 774
|
1945 |
+
},
|
1946 |
+
{
|
1947 |
+
"epoch": 130.0,
|
1948 |
+
"learning_rate": 3.5e-06,
|
1949 |
+
"loss": 3.1791,
|
1950 |
+
"step": 780
|
1951 |
+
},
|
1952 |
+
{
|
1953 |
+
"epoch": 130.0,
|
1954 |
+
"eval_accuracy": 0.39296187683284456,
|
1955 |
+
"eval_loss": 3.9955568313598633,
|
1956 |
+
"eval_runtime": 3.1553,
|
1957 |
+
"eval_samples_per_second": 0.317,
|
1958 |
+
"eval_steps_per_second": 0.317,
|
1959 |
+
"step": 780
|
1960 |
+
},
|
1961 |
+
{
|
1962 |
+
"epoch": 131.0,
|
1963 |
+
"learning_rate": 3.45e-06,
|
1964 |
+
"loss": 3.1759,
|
1965 |
+
"step": 786
|
1966 |
+
},
|
1967 |
+
{
|
1968 |
+
"epoch": 131.0,
|
1969 |
+
"eval_accuracy": 0.396871945259042,
|
1970 |
+
"eval_loss": 4.001156806945801,
|
1971 |
+
"eval_runtime": 3.1012,
|
1972 |
+
"eval_samples_per_second": 0.322,
|
1973 |
+
"eval_steps_per_second": 0.322,
|
1974 |
+
"step": 786
|
1975 |
+
},
|
1976 |
+
{
|
1977 |
+
"epoch": 132.0,
|
1978 |
+
"learning_rate": 3.4000000000000005e-06,
|
1979 |
+
"loss": 3.1622,
|
1980 |
+
"step": 792
|
1981 |
+
},
|
1982 |
+
{
|
1983 |
+
"epoch": 132.0,
|
1984 |
+
"eval_accuracy": 0.3949169110459433,
|
1985 |
+
"eval_loss": 4.010651111602783,
|
1986 |
+
"eval_runtime": 3.2622,
|
1987 |
+
"eval_samples_per_second": 0.307,
|
1988 |
+
"eval_steps_per_second": 0.307,
|
1989 |
+
"step": 792
|
1990 |
+
},
|
1991 |
+
{
|
1992 |
+
"epoch": 133.0,
|
1993 |
+
"learning_rate": 3.3500000000000005e-06,
|
1994 |
+
"loss": 3.1559,
|
1995 |
+
"step": 798
|
1996 |
+
},
|
1997 |
+
{
|
1998 |
+
"epoch": 133.0,
|
1999 |
+
"eval_accuracy": 0.3939393939393939,
|
2000 |
+
"eval_loss": 4.009001731872559,
|
2001 |
+
"eval_runtime": 3.147,
|
2002 |
+
"eval_samples_per_second": 0.318,
|
2003 |
+
"eval_steps_per_second": 0.318,
|
2004 |
+
"step": 798
|
2005 |
+
},
|
2006 |
+
{
|
2007 |
+
"epoch": 134.0,
|
2008 |
+
"learning_rate": 3.3000000000000006e-06,
|
2009 |
+
"loss": 3.1521,
|
2010 |
+
"step": 804
|
2011 |
+
},
|
2012 |
+
{
|
2013 |
+
"epoch": 134.0,
|
2014 |
+
"eval_accuracy": 0.39100684261974583,
|
2015 |
+
"eval_loss": 4.00282621383667,
|
2016 |
+
"eval_runtime": 3.2649,
|
2017 |
+
"eval_samples_per_second": 0.306,
|
2018 |
+
"eval_steps_per_second": 0.306,
|
2019 |
+
"step": 804
|
2020 |
+
},
|
2021 |
+
{
|
2022 |
+
"epoch": 135.0,
|
2023 |
+
"learning_rate": 3.2500000000000002e-06,
|
2024 |
+
"loss": 3.1353,
|
2025 |
+
"step": 810
|
2026 |
+
},
|
2027 |
+
{
|
2028 |
+
"epoch": 135.0,
|
2029 |
+
"eval_accuracy": 0.3939393939393939,
|
2030 |
+
"eval_loss": 4.0033183097839355,
|
2031 |
+
"eval_runtime": 3.0829,
|
2032 |
+
"eval_samples_per_second": 0.324,
|
2033 |
+
"eval_steps_per_second": 0.324,
|
2034 |
+
"step": 810
|
2035 |
+
},
|
2036 |
+
{
|
2037 |
+
"epoch": 136.0,
|
2038 |
+
"learning_rate": 3.2000000000000003e-06,
|
2039 |
+
"loss": 3.1427,
|
2040 |
+
"step": 816
|
2041 |
+
},
|
2042 |
+
{
|
2043 |
+
"epoch": 136.0,
|
2044 |
+
"eval_accuracy": 0.3939393939393939,
|
2045 |
+
"eval_loss": 3.999516248703003,
|
2046 |
+
"eval_runtime": 3.3725,
|
2047 |
+
"eval_samples_per_second": 0.297,
|
2048 |
+
"eval_steps_per_second": 0.297,
|
2049 |
+
"step": 816
|
2050 |
+
},
|
2051 |
+
{
|
2052 |
+
"epoch": 137.0,
|
2053 |
+
"learning_rate": 3.1500000000000003e-06,
|
2054 |
+
"loss": 3.1276,
|
2055 |
+
"step": 822
|
2056 |
+
},
|
2057 |
+
{
|
2058 |
+
"epoch": 137.0,
|
2059 |
+
"eval_accuracy": 0.3919843597262952,
|
2060 |
+
"eval_loss": 3.9962964057922363,
|
2061 |
+
"eval_runtime": 3.3881,
|
2062 |
+
"eval_samples_per_second": 0.295,
|
2063 |
+
"eval_steps_per_second": 0.295,
|
2064 |
+
"step": 822
|
2065 |
+
},
|
2066 |
+
{
|
2067 |
+
"epoch": 138.0,
|
2068 |
+
"learning_rate": 3.1000000000000004e-06,
|
2069 |
+
"loss": 3.1228,
|
2070 |
+
"step": 828
|
2071 |
+
},
|
2072 |
+
{
|
2073 |
+
"epoch": 138.0,
|
2074 |
+
"eval_accuracy": 0.3978494623655914,
|
2075 |
+
"eval_loss": 3.9996395111083984,
|
2076 |
+
"eval_runtime": 3.1889,
|
2077 |
+
"eval_samples_per_second": 0.314,
|
2078 |
+
"eval_steps_per_second": 0.314,
|
2079 |
+
"step": 828
|
2080 |
+
},
|
2081 |
+
{
|
2082 |
+
"epoch": 139.0,
|
2083 |
+
"learning_rate": 3.05e-06,
|
2084 |
+
"loss": 3.1039,
|
2085 |
+
"step": 834
|
2086 |
+
},
|
2087 |
+
{
|
2088 |
+
"epoch": 139.0,
|
2089 |
+
"eval_accuracy": 0.39882697947214074,
|
2090 |
+
"eval_loss": 3.992779493331909,
|
2091 |
+
"eval_runtime": 3.1905,
|
2092 |
+
"eval_samples_per_second": 0.313,
|
2093 |
+
"eval_steps_per_second": 0.313,
|
2094 |
+
"step": 834
|
2095 |
+
},
|
2096 |
+
{
|
2097 |
+
"epoch": 140.0,
|
2098 |
+
"learning_rate": 3e-06,
|
2099 |
+
"loss": 3.097,
|
2100 |
+
"step": 840
|
2101 |
+
},
|
2102 |
+
{
|
2103 |
+
"epoch": 140.0,
|
2104 |
+
"eval_accuracy": 0.396871945259042,
|
2105 |
+
"eval_loss": 3.9969444274902344,
|
2106 |
+
"eval_runtime": 3.2194,
|
2107 |
+
"eval_samples_per_second": 0.311,
|
2108 |
+
"eval_steps_per_second": 0.311,
|
2109 |
+
"step": 840
|
2110 |
+
},
|
2111 |
+
{
|
2112 |
+
"epoch": 141.0,
|
2113 |
+
"learning_rate": 2.95e-06,
|
2114 |
+
"loss": 3.083,
|
2115 |
+
"step": 846
|
2116 |
+
},
|
2117 |
+
{
|
2118 |
+
"epoch": 141.0,
|
2119 |
+
"eval_accuracy": 0.3949169110459433,
|
2120 |
+
"eval_loss": 3.991848945617676,
|
2121 |
+
"eval_runtime": 3.2252,
|
2122 |
+
"eval_samples_per_second": 0.31,
|
2123 |
+
"eval_steps_per_second": 0.31,
|
2124 |
+
"step": 846
|
2125 |
+
},
|
2126 |
+
{
|
2127 |
+
"epoch": 142.0,
|
2128 |
+
"learning_rate": 2.9e-06,
|
2129 |
+
"loss": 3.0844,
|
2130 |
+
"step": 852
|
2131 |
+
},
|
2132 |
+
{
|
2133 |
+
"epoch": 142.0,
|
2134 |
+
"eval_accuracy": 0.396871945259042,
|
2135 |
+
"eval_loss": 3.9899985790252686,
|
2136 |
+
"eval_runtime": 3.3224,
|
2137 |
+
"eval_samples_per_second": 0.301,
|
2138 |
+
"eval_steps_per_second": 0.301,
|
2139 |
+
"step": 852
|
2140 |
+
},
|
2141 |
+
{
|
2142 |
+
"epoch": 143.0,
|
2143 |
+
"learning_rate": 2.85e-06,
|
2144 |
+
"loss": 3.077,
|
2145 |
+
"step": 858
|
2146 |
+
},
|
2147 |
+
{
|
2148 |
+
"epoch": 143.0,
|
2149 |
+
"eval_accuracy": 0.39589442815249265,
|
2150 |
+
"eval_loss": 3.981245756149292,
|
2151 |
+
"eval_runtime": 3.3063,
|
2152 |
+
"eval_samples_per_second": 0.302,
|
2153 |
+
"eval_steps_per_second": 0.302,
|
2154 |
+
"step": 858
|
2155 |
+
},
|
2156 |
+
{
|
2157 |
+
"epoch": 144.0,
|
2158 |
+
"learning_rate": 2.8000000000000003e-06,
|
2159 |
+
"loss": 3.0601,
|
2160 |
+
"step": 864
|
2161 |
+
},
|
2162 |
+
{
|
2163 |
+
"epoch": 144.0,
|
2164 |
+
"eval_accuracy": 0.39589442815249265,
|
2165 |
+
"eval_loss": 3.9948182106018066,
|
2166 |
+
"eval_runtime": 3.5418,
|
2167 |
+
"eval_samples_per_second": 0.282,
|
2168 |
+
"eval_steps_per_second": 0.282,
|
2169 |
+
"step": 864
|
2170 |
+
},
|
2171 |
+
{
|
2172 |
+
"epoch": 145.0,
|
2173 |
+
"learning_rate": 2.7500000000000004e-06,
|
2174 |
+
"loss": 3.0669,
|
2175 |
+
"step": 870
|
2176 |
+
},
|
2177 |
+
{
|
2178 |
+
"epoch": 145.0,
|
2179 |
+
"eval_accuracy": 0.39589442815249265,
|
2180 |
+
"eval_loss": 3.9938085079193115,
|
2181 |
+
"eval_runtime": 3.3604,
|
2182 |
+
"eval_samples_per_second": 0.298,
|
2183 |
+
"eval_steps_per_second": 0.298,
|
2184 |
+
"step": 870
|
2185 |
+
},
|
2186 |
+
{
|
2187 |
+
"epoch": 146.0,
|
2188 |
+
"learning_rate": 2.7000000000000004e-06,
|
2189 |
+
"loss": 3.0515,
|
2190 |
+
"step": 876
|
2191 |
+
},
|
2192 |
+
{
|
2193 |
+
"epoch": 146.0,
|
2194 |
+
"eval_accuracy": 0.3978494623655914,
|
2195 |
+
"eval_loss": 3.9895379543304443,
|
2196 |
+
"eval_runtime": 3.1997,
|
2197 |
+
"eval_samples_per_second": 0.313,
|
2198 |
+
"eval_steps_per_second": 0.313,
|
2199 |
+
"step": 876
|
2200 |
+
},
|
2201 |
+
{
|
2202 |
+
"epoch": 147.0,
|
2203 |
+
"learning_rate": 2.6500000000000005e-06,
|
2204 |
+
"loss": 3.0405,
|
2205 |
+
"step": 882
|
2206 |
+
},
|
2207 |
+
{
|
2208 |
+
"epoch": 147.0,
|
2209 |
+
"eval_accuracy": 0.39882697947214074,
|
2210 |
+
"eval_loss": 3.9802987575531006,
|
2211 |
+
"eval_runtime": 3.2702,
|
2212 |
+
"eval_samples_per_second": 0.306,
|
2213 |
+
"eval_steps_per_second": 0.306,
|
2214 |
+
"step": 882
|
2215 |
+
},
|
2216 |
+
{
|
2217 |
+
"epoch": 148.0,
|
2218 |
+
"learning_rate": 2.6e-06,
|
2219 |
+
"loss": 3.029,
|
2220 |
+
"step": 888
|
2221 |
+
},
|
2222 |
+
{
|
2223 |
+
"epoch": 148.0,
|
2224 |
+
"eval_accuracy": 0.396871945259042,
|
2225 |
+
"eval_loss": 3.985629081726074,
|
2226 |
+
"eval_runtime": 3.064,
|
2227 |
+
"eval_samples_per_second": 0.326,
|
2228 |
+
"eval_steps_per_second": 0.326,
|
2229 |
+
"step": 888
|
2230 |
+
},
|
2231 |
+
{
|
2232 |
+
"epoch": 149.0,
|
2233 |
+
"learning_rate": 2.55e-06,
|
2234 |
+
"loss": 3.0342,
|
2235 |
+
"step": 894
|
2236 |
+
},
|
2237 |
+
{
|
2238 |
+
"epoch": 149.0,
|
2239 |
+
"eval_accuracy": 0.396871945259042,
|
2240 |
+
"eval_loss": 3.982790231704712,
|
2241 |
+
"eval_runtime": 3.2872,
|
2242 |
+
"eval_samples_per_second": 0.304,
|
2243 |
+
"eval_steps_per_second": 0.304,
|
2244 |
+
"step": 894
|
2245 |
+
},
|
2246 |
+
{
|
2247 |
+
"epoch": 150.0,
|
2248 |
+
"learning_rate": 2.5e-06,
|
2249 |
+
"loss": 3.0137,
|
2250 |
+
"step": 900
|
2251 |
+
},
|
2252 |
+
{
|
2253 |
+
"epoch": 150.0,
|
2254 |
+
"eval_accuracy": 0.3978494623655914,
|
2255 |
+
"eval_loss": 3.997673273086548,
|
2256 |
+
"eval_runtime": 3.3166,
|
2257 |
+
"eval_samples_per_second": 0.302,
|
2258 |
+
"eval_steps_per_second": 0.302,
|
2259 |
+
"step": 900
|
2260 |
+
},
|
2261 |
+
{
|
2262 |
+
"epoch": 151.0,
|
2263 |
+
"learning_rate": 2.4500000000000003e-06,
|
2264 |
+
"loss": 3.0277,
|
2265 |
+
"step": 906
|
2266 |
+
},
|
2267 |
+
{
|
2268 |
+
"epoch": 151.0,
|
2269 |
+
"eval_accuracy": 0.3998044965786901,
|
2270 |
+
"eval_loss": 3.9793360233306885,
|
2271 |
+
"eval_runtime": 3.1051,
|
2272 |
+
"eval_samples_per_second": 0.322,
|
2273 |
+
"eval_steps_per_second": 0.322,
|
2274 |
+
"step": 906
|
2275 |
+
},
|
2276 |
+
{
|
2277 |
+
"epoch": 152.0,
|
2278 |
+
"learning_rate": 2.4000000000000003e-06,
|
2279 |
+
"loss": 3.0005,
|
2280 |
+
"step": 912
|
2281 |
+
},
|
2282 |
+
{
|
2283 |
+
"epoch": 152.0,
|
2284 |
+
"eval_accuracy": 0.3998044965786901,
|
2285 |
+
"eval_loss": 3.9779205322265625,
|
2286 |
+
"eval_runtime": 3.1637,
|
2287 |
+
"eval_samples_per_second": 0.316,
|
2288 |
+
"eval_steps_per_second": 0.316,
|
2289 |
+
"step": 912
|
2290 |
+
},
|
2291 |
+
{
|
2292 |
+
"epoch": 153.0,
|
2293 |
+
"learning_rate": 2.35e-06,
|
2294 |
+
"loss": 3.0027,
|
2295 |
+
"step": 918
|
2296 |
+
},
|
2297 |
+
{
|
2298 |
+
"epoch": 153.0,
|
2299 |
+
"eval_accuracy": 0.39882697947214074,
|
2300 |
+
"eval_loss": 3.989084482192993,
|
2301 |
+
"eval_runtime": 3.2887,
|
2302 |
+
"eval_samples_per_second": 0.304,
|
2303 |
+
"eval_steps_per_second": 0.304,
|
2304 |
+
"step": 918
|
2305 |
+
},
|
2306 |
+
{
|
2307 |
+
"epoch": 154.0,
|
2308 |
+
"learning_rate": 2.3000000000000004e-06,
|
2309 |
+
"loss": 3.0034,
|
2310 |
+
"step": 924
|
2311 |
+
},
|
2312 |
+
{
|
2313 |
+
"epoch": 154.0,
|
2314 |
+
"eval_accuracy": 0.40078201368523947,
|
2315 |
+
"eval_loss": 3.9687039852142334,
|
2316 |
+
"eval_runtime": 3.2813,
|
2317 |
+
"eval_samples_per_second": 0.305,
|
2318 |
+
"eval_steps_per_second": 0.305,
|
2319 |
+
"step": 924
|
2320 |
+
},
|
2321 |
+
{
|
2322 |
+
"epoch": 155.0,
|
2323 |
+
"learning_rate": 2.25e-06,
|
2324 |
+
"loss": 2.9853,
|
2325 |
+
"step": 930
|
2326 |
+
},
|
2327 |
+
{
|
2328 |
+
"epoch": 155.0,
|
2329 |
+
"eval_accuracy": 0.3978494623655914,
|
2330 |
+
"eval_loss": 3.988722801208496,
|
2331 |
+
"eval_runtime": 3.4968,
|
2332 |
+
"eval_samples_per_second": 0.286,
|
2333 |
+
"eval_steps_per_second": 0.286,
|
2334 |
+
"step": 930
|
2335 |
+
},
|
2336 |
+
{
|
2337 |
+
"epoch": 156.0,
|
2338 |
+
"learning_rate": 2.2e-06,
|
2339 |
+
"loss": 2.9947,
|
2340 |
+
"step": 936
|
2341 |
+
},
|
2342 |
+
{
|
2343 |
+
"epoch": 156.0,
|
2344 |
+
"eval_accuracy": 0.4027370478983382,
|
2345 |
+
"eval_loss": 3.9860475063323975,
|
2346 |
+
"eval_runtime": 3.29,
|
2347 |
+
"eval_samples_per_second": 0.304,
|
2348 |
+
"eval_steps_per_second": 0.304,
|
2349 |
+
"step": 936
|
2350 |
+
},
|
2351 |
+
{
|
2352 |
+
"epoch": 157.0,
|
2353 |
+
"learning_rate": 2.15e-06,
|
2354 |
+
"loss": 2.9768,
|
2355 |
+
"step": 942
|
2356 |
+
},
|
2357 |
+
{
|
2358 |
+
"epoch": 157.0,
|
2359 |
+
"eval_accuracy": 0.4027370478983382,
|
2360 |
+
"eval_loss": 3.989997148513794,
|
2361 |
+
"eval_runtime": 3.3737,
|
2362 |
+
"eval_samples_per_second": 0.296,
|
2363 |
+
"eval_steps_per_second": 0.296,
|
2364 |
+
"step": 942
|
2365 |
+
},
|
2366 |
+
{
|
2367 |
+
"epoch": 158.0,
|
2368 |
+
"learning_rate": 2.1000000000000002e-06,
|
2369 |
+
"loss": 2.9752,
|
2370 |
+
"step": 948
|
2371 |
+
},
|
2372 |
+
{
|
2373 |
+
"epoch": 158.0,
|
2374 |
+
"eval_accuracy": 0.39882697947214074,
|
2375 |
+
"eval_loss": 3.9992799758911133,
|
2376 |
+
"eval_runtime": 3.3931,
|
2377 |
+
"eval_samples_per_second": 0.295,
|
2378 |
+
"eval_steps_per_second": 0.295,
|
2379 |
+
"step": 948
|
2380 |
+
},
|
2381 |
+
{
|
2382 |
+
"epoch": 159.0,
|
2383 |
+
"learning_rate": 2.05e-06,
|
2384 |
+
"loss": 2.9773,
|
2385 |
+
"step": 954
|
2386 |
+
},
|
2387 |
+
{
|
2388 |
+
"epoch": 159.0,
|
2389 |
+
"eval_accuracy": 0.40175953079178883,
|
2390 |
+
"eval_loss": 3.9693987369537354,
|
2391 |
+
"eval_runtime": 3.1141,
|
2392 |
+
"eval_samples_per_second": 0.321,
|
2393 |
+
"eval_steps_per_second": 0.321,
|
2394 |
+
"step": 954
|
2395 |
+
},
|
2396 |
+
{
|
2397 |
+
"epoch": 160.0,
|
2398 |
+
"learning_rate": 2.0000000000000003e-06,
|
2399 |
+
"loss": 2.9662,
|
2400 |
+
"step": 960
|
2401 |
+
},
|
2402 |
+
{
|
2403 |
+
"epoch": 160.0,
|
2404 |
+
"eval_accuracy": 0.3998044965786901,
|
2405 |
+
"eval_loss": 3.9923715591430664,
|
2406 |
+
"eval_runtime": 3.2181,
|
2407 |
+
"eval_samples_per_second": 0.311,
|
2408 |
+
"eval_steps_per_second": 0.311,
|
2409 |
+
"step": 960
|
2410 |
+
},
|
2411 |
+
{
|
2412 |
+
"epoch": 161.0,
|
2413 |
+
"learning_rate": 1.9500000000000004e-06,
|
2414 |
+
"loss": 2.9661,
|
2415 |
+
"step": 966
|
2416 |
+
},
|
2417 |
+
{
|
2418 |
+
"epoch": 161.0,
|
2419 |
+
"eval_accuracy": 0.39882697947214074,
|
2420 |
+
"eval_loss": 4.008890151977539,
|
2421 |
+
"eval_runtime": 3.1472,
|
2422 |
+
"eval_samples_per_second": 0.318,
|
2423 |
+
"eval_steps_per_second": 0.318,
|
2424 |
+
"step": 966
|
2425 |
+
},
|
2426 |
+
{
|
2427 |
+
"epoch": 162.0,
|
2428 |
+
"learning_rate": 1.9000000000000002e-06,
|
2429 |
+
"loss": 2.9488,
|
2430 |
+
"step": 972
|
2431 |
+
},
|
2432 |
+
{
|
2433 |
+
"epoch": 162.0,
|
2434 |
+
"eval_accuracy": 0.39882697947214074,
|
2435 |
+
"eval_loss": 3.974891424179077,
|
2436 |
+
"eval_runtime": 3.2045,
|
2437 |
+
"eval_samples_per_second": 0.312,
|
2438 |
+
"eval_steps_per_second": 0.312,
|
2439 |
+
"step": 972
|
2440 |
+
},
|
2441 |
+
{
|
2442 |
+
"epoch": 163.0,
|
2443 |
+
"learning_rate": 1.85e-06,
|
2444 |
+
"loss": 2.9487,
|
2445 |
+
"step": 978
|
2446 |
+
},
|
2447 |
+
{
|
2448 |
+
"epoch": 163.0,
|
2449 |
+
"eval_accuracy": 0.3978494623655914,
|
2450 |
+
"eval_loss": 3.9931938648223877,
|
2451 |
+
"eval_runtime": 2.9879,
|
2452 |
+
"eval_samples_per_second": 0.335,
|
2453 |
+
"eval_steps_per_second": 0.335,
|
2454 |
+
"step": 978
|
2455 |
+
},
|
2456 |
+
{
|
2457 |
+
"epoch": 164.0,
|
2458 |
+
"learning_rate": 1.8000000000000001e-06,
|
2459 |
+
"loss": 2.9482,
|
2460 |
+
"step": 984
|
2461 |
+
},
|
2462 |
+
{
|
2463 |
+
"epoch": 164.0,
|
2464 |
+
"eval_accuracy": 0.39882697947214074,
|
2465 |
+
"eval_loss": 3.998748540878296,
|
2466 |
+
"eval_runtime": 3.0761,
|
2467 |
+
"eval_samples_per_second": 0.325,
|
2468 |
+
"eval_steps_per_second": 0.325,
|
2469 |
+
"step": 984
|
2470 |
+
},
|
2471 |
+
{
|
2472 |
+
"epoch": 165.0,
|
2473 |
+
"learning_rate": 1.75e-06,
|
2474 |
+
"loss": 2.9624,
|
2475 |
+
"step": 990
|
2476 |
+
},
|
2477 |
+
{
|
2478 |
+
"epoch": 165.0,
|
2479 |
+
"eval_accuracy": 0.3978494623655914,
|
2480 |
+
"eval_loss": 3.962663412094116,
|
2481 |
+
"eval_runtime": 3.2497,
|
2482 |
+
"eval_samples_per_second": 0.308,
|
2483 |
+
"eval_steps_per_second": 0.308,
|
2484 |
+
"step": 990
|
2485 |
+
},
|
2486 |
+
{
|
2487 |
+
"epoch": 166.0,
|
2488 |
+
"learning_rate": 1.7000000000000002e-06,
|
2489 |
+
"loss": 2.9524,
|
2490 |
+
"step": 996
|
2491 |
+
},
|
2492 |
+
{
|
2493 |
+
"epoch": 166.0,
|
2494 |
+
"eval_accuracy": 0.40078201368523947,
|
2495 |
+
"eval_loss": 3.979146957397461,
|
2496 |
+
"eval_runtime": 3.2339,
|
2497 |
+
"eval_samples_per_second": 0.309,
|
2498 |
+
"eval_steps_per_second": 0.309,
|
2499 |
+
"step": 996
|
2500 |
+
},
|
2501 |
+
{
|
2502 |
+
"epoch": 167.0,
|
2503 |
+
"learning_rate": 1.6500000000000003e-06,
|
2504 |
+
"loss": 2.9357,
|
2505 |
+
"step": 1002
|
2506 |
+
},
|
2507 |
+
{
|
2508 |
+
"epoch": 167.0,
|
2509 |
+
"eval_accuracy": 0.3998044965786901,
|
2510 |
+
"eval_loss": 3.996922731399536,
|
2511 |
+
"eval_runtime": 3.0918,
|
2512 |
+
"eval_samples_per_second": 0.323,
|
2513 |
+
"eval_steps_per_second": 0.323,
|
2514 |
+
"step": 1002
|
2515 |
+
},
|
2516 |
+
{
|
2517 |
+
"epoch": 168.0,
|
2518 |
+
"learning_rate": 1.6000000000000001e-06,
|
2519 |
+
"loss": 2.9323,
|
2520 |
+
"step": 1008
|
2521 |
+
},
|
2522 |
+
{
|
2523 |
+
"epoch": 168.0,
|
2524 |
+
"eval_accuracy": 0.40078201368523947,
|
2525 |
+
"eval_loss": 3.9853515625,
|
2526 |
+
"eval_runtime": 3.2542,
|
2527 |
+
"eval_samples_per_second": 0.307,
|
2528 |
+
"eval_steps_per_second": 0.307,
|
2529 |
+
"step": 1008
|
2530 |
+
},
|
2531 |
+
{
|
2532 |
+
"epoch": 169.0,
|
2533 |
+
"learning_rate": 1.5500000000000002e-06,
|
2534 |
+
"loss": 2.9334,
|
2535 |
+
"step": 1014
|
2536 |
+
},
|
2537 |
+
{
|
2538 |
+
"epoch": 169.0,
|
2539 |
+
"eval_accuracy": 0.40078201368523947,
|
2540 |
+
"eval_loss": 3.977756977081299,
|
2541 |
+
"eval_runtime": 3.2318,
|
2542 |
+
"eval_samples_per_second": 0.309,
|
2543 |
+
"eval_steps_per_second": 0.309,
|
2544 |
+
"step": 1014
|
2545 |
+
},
|
2546 |
+
{
|
2547 |
+
"epoch": 170.0,
|
2548 |
+
"learning_rate": 1.5e-06,
|
2549 |
+
"loss": 2.9228,
|
2550 |
+
"step": 1020
|
2551 |
+
},
|
2552 |
+
{
|
2553 |
+
"epoch": 170.0,
|
2554 |
+
"eval_accuracy": 0.4027370478983382,
|
2555 |
+
"eval_loss": 3.9858651161193848,
|
2556 |
+
"eval_runtime": 3.1222,
|
2557 |
+
"eval_samples_per_second": 0.32,
|
2558 |
+
"eval_steps_per_second": 0.32,
|
2559 |
+
"step": 1020
|
2560 |
+
},
|
2561 |
+
{
|
2562 |
+
"epoch": 171.0,
|
2563 |
+
"learning_rate": 1.45e-06,
|
2564 |
+
"loss": 2.9305,
|
2565 |
+
"step": 1026
|
2566 |
+
},
|
2567 |
+
{
|
2568 |
+
"epoch": 171.0,
|
2569 |
+
"eval_accuracy": 0.40371456500488756,
|
2570 |
+
"eval_loss": 3.9820897579193115,
|
2571 |
+
"eval_runtime": 3.1725,
|
2572 |
+
"eval_samples_per_second": 0.315,
|
2573 |
+
"eval_steps_per_second": 0.315,
|
2574 |
+
"step": 1026
|
2575 |
+
},
|
2576 |
+
{
|
2577 |
+
"epoch": 172.0,
|
2578 |
+
"learning_rate": 1.4000000000000001e-06,
|
2579 |
+
"loss": 2.9239,
|
2580 |
+
"step": 1032
|
2581 |
+
},
|
2582 |
+
{
|
2583 |
+
"epoch": 172.0,
|
2584 |
+
"eval_accuracy": 0.4066471163245357,
|
2585 |
+
"eval_loss": 3.987579107284546,
|
2586 |
+
"eval_runtime": 3.1039,
|
2587 |
+
"eval_samples_per_second": 0.322,
|
2588 |
+
"eval_steps_per_second": 0.322,
|
2589 |
+
"step": 1032
|
2590 |
+
},
|
2591 |
+
{
|
2592 |
+
"epoch": 173.0,
|
2593 |
+
"learning_rate": 1.3500000000000002e-06,
|
2594 |
+
"loss": 2.9181,
|
2595 |
+
"step": 1038
|
2596 |
+
},
|
2597 |
+
{
|
2598 |
+
"epoch": 173.0,
|
2599 |
+
"eval_accuracy": 0.4056695992179863,
|
2600 |
+
"eval_loss": 3.9791972637176514,
|
2601 |
+
"eval_runtime": 3.503,
|
2602 |
+
"eval_samples_per_second": 0.285,
|
2603 |
+
"eval_steps_per_second": 0.285,
|
2604 |
+
"step": 1038
|
2605 |
+
},
|
2606 |
+
{
|
2607 |
+
"epoch": 174.0,
|
2608 |
+
"learning_rate": 1.3e-06,
|
2609 |
+
"loss": 2.9162,
|
2610 |
+
"step": 1044
|
2611 |
+
},
|
2612 |
+
{
|
2613 |
+
"epoch": 174.0,
|
2614 |
+
"eval_accuracy": 0.40371456500488756,
|
2615 |
+
"eval_loss": 3.9731061458587646,
|
2616 |
+
"eval_runtime": 3.0863,
|
2617 |
+
"eval_samples_per_second": 0.324,
|
2618 |
+
"eval_steps_per_second": 0.324,
|
2619 |
+
"step": 1044
|
2620 |
+
},
|
2621 |
+
{
|
2622 |
+
"epoch": 175.0,
|
2623 |
+
"learning_rate": 1.25e-06,
|
2624 |
+
"loss": 2.9171,
|
2625 |
+
"step": 1050
|
2626 |
+
},
|
2627 |
+
{
|
2628 |
+
"epoch": 175.0,
|
2629 |
+
"eval_accuracy": 0.4066471163245357,
|
2630 |
+
"eval_loss": 3.9795916080474854,
|
2631 |
+
"eval_runtime": 3.0684,
|
2632 |
+
"eval_samples_per_second": 0.326,
|
2633 |
+
"eval_steps_per_second": 0.326,
|
2634 |
+
"step": 1050
|
2635 |
+
},
|
2636 |
+
{
|
2637 |
+
"epoch": 176.0,
|
2638 |
+
"learning_rate": 1.2000000000000002e-06,
|
2639 |
+
"loss": 2.9132,
|
2640 |
+
"step": 1056
|
2641 |
+
},
|
2642 |
+
{
|
2643 |
+
"epoch": 176.0,
|
2644 |
+
"eval_accuracy": 0.4046920821114369,
|
2645 |
+
"eval_loss": 3.9914140701293945,
|
2646 |
+
"eval_runtime": 3.5253,
|
2647 |
+
"eval_samples_per_second": 0.284,
|
2648 |
+
"eval_steps_per_second": 0.284,
|
2649 |
+
"step": 1056
|
2650 |
+
},
|
2651 |
+
{
|
2652 |
+
"epoch": 177.0,
|
2653 |
+
"learning_rate": 1.1500000000000002e-06,
|
2654 |
+
"loss": 2.9168,
|
2655 |
+
"step": 1062
|
2656 |
+
},
|
2657 |
+
{
|
2658 |
+
"epoch": 177.0,
|
2659 |
+
"eval_accuracy": 0.4046920821114369,
|
2660 |
+
"eval_loss": 3.9826488494873047,
|
2661 |
+
"eval_runtime": 3.0836,
|
2662 |
+
"eval_samples_per_second": 0.324,
|
2663 |
+
"eval_steps_per_second": 0.324,
|
2664 |
+
"step": 1062
|
2665 |
+
},
|
2666 |
+
{
|
2667 |
+
"epoch": 178.0,
|
2668 |
+
"learning_rate": 1.1e-06,
|
2669 |
+
"loss": 2.8974,
|
2670 |
+
"step": 1068
|
2671 |
+
},
|
2672 |
+
{
|
2673 |
+
"epoch": 178.0,
|
2674 |
+
"eval_accuracy": 0.4056695992179863,
|
2675 |
+
"eval_loss": 3.9753177165985107,
|
2676 |
+
"eval_runtime": 3.165,
|
2677 |
+
"eval_samples_per_second": 0.316,
|
2678 |
+
"eval_steps_per_second": 0.316,
|
2679 |
+
"step": 1068
|
2680 |
+
},
|
2681 |
+
{
|
2682 |
+
"epoch": 179.0,
|
2683 |
+
"learning_rate": 1.0500000000000001e-06,
|
2684 |
+
"loss": 2.8954,
|
2685 |
+
"step": 1074
|
2686 |
+
},
|
2687 |
+
{
|
2688 |
+
"epoch": 179.0,
|
2689 |
+
"eval_accuracy": 0.4056695992179863,
|
2690 |
+
"eval_loss": 3.976564884185791,
|
2691 |
+
"eval_runtime": 3.123,
|
2692 |
+
"eval_samples_per_second": 0.32,
|
2693 |
+
"eval_steps_per_second": 0.32,
|
2694 |
+
"step": 1074
|
2695 |
+
},
|
2696 |
+
{
|
2697 |
+
"epoch": 180.0,
|
2698 |
+
"learning_rate": 1.0000000000000002e-06,
|
2699 |
+
"loss": 2.9003,
|
2700 |
+
"step": 1080
|
2701 |
+
},
|
2702 |
+
{
|
2703 |
+
"epoch": 180.0,
|
2704 |
+
"eval_accuracy": 0.4027370478983382,
|
2705 |
+
"eval_loss": 3.986520290374756,
|
2706 |
+
"eval_runtime": 3.106,
|
2707 |
+
"eval_samples_per_second": 0.322,
|
2708 |
+
"eval_steps_per_second": 0.322,
|
2709 |
+
"step": 1080
|
2710 |
+
},
|
2711 |
+
{
|
2712 |
+
"epoch": 181.0,
|
2713 |
+
"learning_rate": 9.500000000000001e-07,
|
2714 |
+
"loss": 2.9012,
|
2715 |
+
"step": 1086
|
2716 |
+
},
|
2717 |
+
{
|
2718 |
+
"epoch": 181.0,
|
2719 |
+
"eval_accuracy": 0.4046920821114369,
|
2720 |
+
"eval_loss": 3.983490228652954,
|
2721 |
+
"eval_runtime": 3.1013,
|
2722 |
+
"eval_samples_per_second": 0.322,
|
2723 |
+
"eval_steps_per_second": 0.322,
|
2724 |
+
"step": 1086
|
2725 |
+
},
|
2726 |
+
{
|
2727 |
+
"epoch": 182.0,
|
2728 |
+
"learning_rate": 9.000000000000001e-07,
|
2729 |
+
"loss": 2.8994,
|
2730 |
+
"step": 1092
|
2731 |
+
},
|
2732 |
+
{
|
2733 |
+
"epoch": 182.0,
|
2734 |
+
"eval_accuracy": 0.4046920821114369,
|
2735 |
+
"eval_loss": 3.980245590209961,
|
2736 |
+
"eval_runtime": 3.2714,
|
2737 |
+
"eval_samples_per_second": 0.306,
|
2738 |
+
"eval_steps_per_second": 0.306,
|
2739 |
+
"step": 1092
|
2740 |
+
},
|
2741 |
+
{
|
2742 |
+
"epoch": 183.0,
|
2743 |
+
"learning_rate": 8.500000000000001e-07,
|
2744 |
+
"loss": 2.8918,
|
2745 |
+
"step": 1098
|
2746 |
+
},
|
2747 |
+
{
|
2748 |
+
"epoch": 183.0,
|
2749 |
+
"eval_accuracy": 0.4066471163245357,
|
2750 |
+
"eval_loss": 3.9810588359832764,
|
2751 |
+
"eval_runtime": 3.2463,
|
2752 |
+
"eval_samples_per_second": 0.308,
|
2753 |
+
"eval_steps_per_second": 0.308,
|
2754 |
+
"step": 1098
|
2755 |
+
},
|
2756 |
+
{
|
2757 |
+
"epoch": 184.0,
|
2758 |
+
"learning_rate": 8.000000000000001e-07,
|
2759 |
+
"loss": 2.8893,
|
2760 |
+
"step": 1104
|
2761 |
+
},
|
2762 |
+
{
|
2763 |
+
"epoch": 184.0,
|
2764 |
+
"eval_accuracy": 0.4056695992179863,
|
2765 |
+
"eval_loss": 3.981043815612793,
|
2766 |
+
"eval_runtime": 3.5146,
|
2767 |
+
"eval_samples_per_second": 0.285,
|
2768 |
+
"eval_steps_per_second": 0.285,
|
2769 |
+
"step": 1104
|
2770 |
+
},
|
2771 |
+
{
|
2772 |
+
"epoch": 185.0,
|
2773 |
+
"learning_rate": 7.5e-07,
|
2774 |
+
"loss": 2.8865,
|
2775 |
+
"step": 1110
|
2776 |
+
},
|
2777 |
+
{
|
2778 |
+
"epoch": 185.0,
|
2779 |
+
"eval_accuracy": 0.40762463343108507,
|
2780 |
+
"eval_loss": 3.9851670265197754,
|
2781 |
+
"eval_runtime": 3.121,
|
2782 |
+
"eval_samples_per_second": 0.32,
|
2783 |
+
"eval_steps_per_second": 0.32,
|
2784 |
+
"step": 1110
|
2785 |
+
},
|
2786 |
+
{
|
2787 |
+
"epoch": 186.0,
|
2788 |
+
"learning_rate": 7.000000000000001e-07,
|
2789 |
+
"loss": 2.8784,
|
2790 |
+
"step": 1116
|
2791 |
+
},
|
2792 |
+
{
|
2793 |
+
"epoch": 186.0,
|
2794 |
+
"eval_accuracy": 0.4056695992179863,
|
2795 |
+
"eval_loss": 3.9805047512054443,
|
2796 |
+
"eval_runtime": 3.1275,
|
2797 |
+
"eval_samples_per_second": 0.32,
|
2798 |
+
"eval_steps_per_second": 0.32,
|
2799 |
+
"step": 1116
|
2800 |
+
},
|
2801 |
+
{
|
2802 |
+
"epoch": 187.0,
|
2803 |
+
"learning_rate": 6.5e-07,
|
2804 |
+
"loss": 2.8875,
|
2805 |
+
"step": 1122
|
2806 |
+
},
|
2807 |
+
{
|
2808 |
+
"epoch": 187.0,
|
2809 |
+
"eval_accuracy": 0.4066471163245357,
|
2810 |
+
"eval_loss": 3.978147029876709,
|
2811 |
+
"eval_runtime": 3.3509,
|
2812 |
+
"eval_samples_per_second": 0.298,
|
2813 |
+
"eval_steps_per_second": 0.298,
|
2814 |
+
"step": 1122
|
2815 |
+
},
|
2816 |
+
{
|
2817 |
+
"epoch": 188.0,
|
2818 |
+
"learning_rate": 6.000000000000001e-07,
|
2819 |
+
"loss": 2.8948,
|
2820 |
+
"step": 1128
|
2821 |
+
},
|
2822 |
+
{
|
2823 |
+
"epoch": 188.0,
|
2824 |
+
"eval_accuracy": 0.4056695992179863,
|
2825 |
+
"eval_loss": 3.9830515384674072,
|
2826 |
+
"eval_runtime": 3.3453,
|
2827 |
+
"eval_samples_per_second": 0.299,
|
2828 |
+
"eval_steps_per_second": 0.299,
|
2829 |
+
"step": 1128
|
2830 |
+
},
|
2831 |
+
{
|
2832 |
+
"epoch": 189.0,
|
2833 |
+
"learning_rate": 5.5e-07,
|
2834 |
+
"loss": 2.8927,
|
2835 |
+
"step": 1134
|
2836 |
+
},
|
2837 |
+
{
|
2838 |
+
"epoch": 189.0,
|
2839 |
+
"eval_accuracy": 0.4066471163245357,
|
2840 |
+
"eval_loss": 3.9836947917938232,
|
2841 |
+
"eval_runtime": 3.1414,
|
2842 |
+
"eval_samples_per_second": 0.318,
|
2843 |
+
"eval_steps_per_second": 0.318,
|
2844 |
+
"step": 1134
|
2845 |
+
},
|
2846 |
+
{
|
2847 |
+
"epoch": 190.0,
|
2848 |
+
"learning_rate": 5.000000000000001e-07,
|
2849 |
+
"loss": 2.8739,
|
2850 |
+
"step": 1140
|
2851 |
+
},
|
2852 |
+
{
|
2853 |
+
"epoch": 190.0,
|
2854 |
+
"eval_accuracy": 0.4056695992179863,
|
2855 |
+
"eval_loss": 3.9822254180908203,
|
2856 |
+
"eval_runtime": 3.3757,
|
2857 |
+
"eval_samples_per_second": 0.296,
|
2858 |
+
"eval_steps_per_second": 0.296,
|
2859 |
+
"step": 1140
|
2860 |
+
},
|
2861 |
+
{
|
2862 |
+
"epoch": 191.0,
|
2863 |
+
"learning_rate": 4.5000000000000003e-07,
|
2864 |
+
"loss": 2.8919,
|
2865 |
+
"step": 1146
|
2866 |
+
},
|
2867 |
+
{
|
2868 |
+
"epoch": 191.0,
|
2869 |
+
"eval_accuracy": 0.4066471163245357,
|
2870 |
+
"eval_loss": 3.9791526794433594,
|
2871 |
+
"eval_runtime": 3.5323,
|
2872 |
+
"eval_samples_per_second": 0.283,
|
2873 |
+
"eval_steps_per_second": 0.283,
|
2874 |
+
"step": 1146
|
2875 |
+
},
|
2876 |
+
{
|
2877 |
+
"epoch": 192.0,
|
2878 |
+
"learning_rate": 4.0000000000000003e-07,
|
2879 |
+
"loss": 2.8713,
|
2880 |
+
"step": 1152
|
2881 |
+
},
|
2882 |
+
{
|
2883 |
+
"epoch": 192.0,
|
2884 |
+
"eval_accuracy": 0.4056695992179863,
|
2885 |
+
"eval_loss": 3.9800055027008057,
|
2886 |
+
"eval_runtime": 3.4653,
|
2887 |
+
"eval_samples_per_second": 0.289,
|
2888 |
+
"eval_steps_per_second": 0.289,
|
2889 |
+
"step": 1152
|
2890 |
+
},
|
2891 |
+
{
|
2892 |
+
"epoch": 193.0,
|
2893 |
+
"learning_rate": 3.5000000000000004e-07,
|
2894 |
+
"loss": 2.8798,
|
2895 |
+
"step": 1158
|
2896 |
+
},
|
2897 |
+
{
|
2898 |
+
"epoch": 193.0,
|
2899 |
+
"eval_accuracy": 0.4046920821114369,
|
2900 |
+
"eval_loss": 3.985433578491211,
|
2901 |
+
"eval_runtime": 3.0972,
|
2902 |
+
"eval_samples_per_second": 0.323,
|
2903 |
+
"eval_steps_per_second": 0.323,
|
2904 |
+
"step": 1158
|
2905 |
+
},
|
2906 |
+
{
|
2907 |
+
"epoch": 194.0,
|
2908 |
+
"learning_rate": 3.0000000000000004e-07,
|
2909 |
+
"loss": 2.8835,
|
2910 |
+
"step": 1164
|
2911 |
+
},
|
2912 |
+
{
|
2913 |
+
"epoch": 194.0,
|
2914 |
+
"eval_accuracy": 0.4056695992179863,
|
2915 |
+
"eval_loss": 3.984498977661133,
|
2916 |
+
"eval_runtime": 4.5165,
|
2917 |
+
"eval_samples_per_second": 0.221,
|
2918 |
+
"eval_steps_per_second": 0.221,
|
2919 |
+
"step": 1164
|
2920 |
+
},
|
2921 |
+
{
|
2922 |
+
"epoch": 195.0,
|
2923 |
+
"learning_rate": 2.5000000000000004e-07,
|
2924 |
+
"loss": 2.878,
|
2925 |
+
"step": 1170
|
2926 |
+
},
|
2927 |
+
{
|
2928 |
+
"epoch": 195.0,
|
2929 |
+
"eval_accuracy": 0.4056695992179863,
|
2930 |
+
"eval_loss": 3.981985330581665,
|
2931 |
+
"eval_runtime": 3.1036,
|
2932 |
+
"eval_samples_per_second": 0.322,
|
2933 |
+
"eval_steps_per_second": 0.322,
|
2934 |
+
"step": 1170
|
2935 |
+
},
|
2936 |
+
{
|
2937 |
+
"epoch": 196.0,
|
2938 |
+
"learning_rate": 2.0000000000000002e-07,
|
2939 |
+
"loss": 2.8931,
|
2940 |
+
"step": 1176
|
2941 |
+
},
|
2942 |
+
{
|
2943 |
+
"epoch": 196.0,
|
2944 |
+
"eval_accuracy": 0.4056695992179863,
|
2945 |
+
"eval_loss": 3.9816386699676514,
|
2946 |
+
"eval_runtime": 3.2982,
|
2947 |
+
"eval_samples_per_second": 0.303,
|
2948 |
+
"eval_steps_per_second": 0.303,
|
2949 |
+
"step": 1176
|
2950 |
+
},
|
2951 |
+
{
|
2952 |
+
"epoch": 197.0,
|
2953 |
+
"learning_rate": 1.5000000000000002e-07,
|
2954 |
+
"loss": 2.8662,
|
2955 |
+
"step": 1182
|
2956 |
+
},
|
2957 |
+
{
|
2958 |
+
"epoch": 197.0,
|
2959 |
+
"eval_accuracy": 0.4056695992179863,
|
2960 |
+
"eval_loss": 3.982978105545044,
|
2961 |
+
"eval_runtime": 4.8102,
|
2962 |
+
"eval_samples_per_second": 0.208,
|
2963 |
+
"eval_steps_per_second": 0.208,
|
2964 |
+
"step": 1182
|
2965 |
+
},
|
2966 |
+
{
|
2967 |
+
"epoch": 198.0,
|
2968 |
+
"learning_rate": 1.0000000000000001e-07,
|
2969 |
+
"loss": 2.8734,
|
2970 |
+
"step": 1188
|
2971 |
+
},
|
2972 |
+
{
|
2973 |
+
"epoch": 198.0,
|
2974 |
+
"eval_accuracy": 0.4056695992179863,
|
2975 |
+
"eval_loss": 3.984077215194702,
|
2976 |
+
"eval_runtime": 3.1139,
|
2977 |
+
"eval_samples_per_second": 0.321,
|
2978 |
+
"eval_steps_per_second": 0.321,
|
2979 |
+
"step": 1188
|
2980 |
+
},
|
2981 |
+
{
|
2982 |
+
"epoch": 199.0,
|
2983 |
+
"learning_rate": 5.0000000000000004e-08,
|
2984 |
+
"loss": 2.8825,
|
2985 |
+
"step": 1194
|
2986 |
+
},
|
2987 |
+
{
|
2988 |
+
"epoch": 199.0,
|
2989 |
+
"eval_accuracy": 0.4056695992179863,
|
2990 |
+
"eval_loss": 3.98300838470459,
|
2991 |
+
"eval_runtime": 3.644,
|
2992 |
+
"eval_samples_per_second": 0.274,
|
2993 |
+
"eval_steps_per_second": 0.274,
|
2994 |
+
"step": 1194
|
2995 |
+
},
|
2996 |
+
{
|
2997 |
+
"epoch": 200.0,
|
2998 |
+
"learning_rate": 0.0,
|
2999 |
+
"loss": 2.8825,
|
3000 |
+
"step": 1200
|
3001 |
+
},
|
3002 |
+
{
|
3003 |
+
"epoch": 200.0,
|
3004 |
+
"eval_accuracy": 0.4056695992179863,
|
3005 |
+
"eval_loss": 3.982684373855591,
|
3006 |
+
"eval_runtime": 3.4062,
|
3007 |
+
"eval_samples_per_second": 0.294,
|
3008 |
+
"eval_steps_per_second": 0.294,
|
3009 |
+
"step": 1200
|
3010 |
+
},
|
3011 |
+
{
|
3012 |
+
"epoch": 200.0,
|
3013 |
+
"step": 1200,
|
3014 |
+
"total_flos": 1.044564148224e+16,
|
3015 |
+
"train_loss": 3.9290491278966266,
|
3016 |
+
"train_runtime": 32687.0411,
|
3017 |
+
"train_samples_per_second": 0.073,
|
3018 |
+
"train_steps_per_second": 0.037
|
3019 |
+
}
|
3020 |
+
],
|
3021 |
+
"logging_steps": 500,
|
3022 |
+
"max_steps": 1200,
|
3023 |
+
"num_input_tokens_seen": 0,
|
3024 |
+
"num_train_epochs": 200,
|
3025 |
+
"save_steps": 500,
|
3026 |
+
"total_flos": 1.044564148224e+16,
|
3027 |
+
"train_batch_size": 2,
|
3028 |
+
"trial_name": null,
|
3029 |
+
"trial_params": null
|
3030 |
+
}
|
hemolytic/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:19913073722deac1045a1e2e35a5790cc709d722d1ee2bdf29570c80ed40fb48
|
3 |
+
size 4728
|
hemolytic/vocab.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|