End of training
Browse files- README.md +202 -203
- config.json +2 -3
- model.safetensors +2 -2
- runs/Mar04_10-28-21_c60a5c456cbd/events.out.tfevents.1709548101.c60a5c456cbd.796.8 +3 -0
- tokenizer.json +60 -24
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -1,5 +1,4 @@
|
|
| 1 |
---
|
| 2 |
-
base_model: Kielak2/calculator_model_test
|
| 3 |
tags:
|
| 4 |
- generated_from_trainer
|
| 5 |
model-index:
|
|
@@ -12,9 +11,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 12 |
|
| 13 |
# calculator_model_test
|
| 14 |
|
| 15 |
-
This model is a fine-tuned version of [
|
| 16 |
It achieves the following results on the evaluation set:
|
| 17 |
-
- Loss: 0.
|
| 18 |
|
| 19 |
## Model description
|
| 20 |
|
|
@@ -46,206 +45,206 @@ The following hyperparameters were used during training:
|
|
| 46 |
|
| 47 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 48 |
|:-------------:|:-----:|:----:|:---------------:|
|
| 49 |
-
|
|
| 50 |
-
|
|
| 51 |
-
|
|
| 52 |
-
|
|
| 53 |
-
|
|
| 54 |
-
|
|
| 55 |
-
| 1.
|
| 56 |
-
| 1.
|
| 57 |
-
| 1.
|
| 58 |
-
| 1.
|
| 59 |
-
| 1.
|
| 60 |
-
| 1.
|
| 61 |
-
| 1.
|
| 62 |
-
| 1.
|
| 63 |
-
| 1.
|
| 64 |
-
| 1.
|
| 65 |
-
|
|
| 66 |
-
|
|
| 67 |
-
|
|
| 68 |
-
| 0.
|
| 69 |
-
|
|
| 70 |
-
|
|
| 71 |
-
| 0.
|
| 72 |
-
| 0.
|
| 73 |
-
| 0.
|
| 74 |
-
| 0.
|
| 75 |
-
| 0.
|
| 76 |
-
| 0.
|
| 77 |
-
| 0.
|
| 78 |
-
| 0.
|
| 79 |
-
| 0.
|
| 80 |
-
| 0.
|
| 81 |
-
| 0.
|
| 82 |
-
| 0.
|
| 83 |
-
| 0.
|
| 84 |
-
| 0.
|
| 85 |
-
| 0.
|
| 86 |
-
| 0.
|
| 87 |
-
| 0.
|
| 88 |
-
| 0.
|
| 89 |
-
| 0.
|
| 90 |
-
| 0.
|
| 91 |
-
| 0.
|
| 92 |
-
| 0.
|
| 93 |
-
| 0.
|
| 94 |
-
| 0.
|
| 95 |
-
| 0.
|
| 96 |
-
| 0.
|
| 97 |
-
| 0.
|
| 98 |
-
| 0.
|
| 99 |
-
| 0.
|
| 100 |
-
| 0.
|
| 101 |
-
| 0.
|
| 102 |
-
| 0.
|
| 103 |
-
| 0.
|
| 104 |
-
| 0.
|
| 105 |
-
| 0.
|
| 106 |
-
| 0.
|
| 107 |
-
| 0.
|
| 108 |
-
| 0.
|
| 109 |
-
| 0.
|
| 110 |
-
| 0.
|
| 111 |
-
| 0.
|
| 112 |
-
| 0.
|
| 113 |
-
| 0.
|
| 114 |
-
| 0.
|
| 115 |
-
| 0.
|
| 116 |
-
| 0.
|
| 117 |
-
| 0.
|
| 118 |
-
| 0.
|
| 119 |
-
| 0.
|
| 120 |
-
| 0.
|
| 121 |
-
| 0.
|
| 122 |
-
| 0.
|
| 123 |
-
| 0.
|
| 124 |
-
| 0.
|
| 125 |
-
| 0.
|
| 126 |
-
| 0.
|
| 127 |
-
| 0.
|
| 128 |
-
| 0.
|
| 129 |
-
| 0.
|
| 130 |
-
| 0.
|
| 131 |
-
| 0.
|
| 132 |
-
| 0.
|
| 133 |
-
| 0.
|
| 134 |
-
| 0.
|
| 135 |
-
| 0.
|
| 136 |
-
| 0.
|
| 137 |
-
| 0.
|
| 138 |
-
| 0.
|
| 139 |
-
| 0.
|
| 140 |
-
| 0.
|
| 141 |
-
| 0.
|
| 142 |
-
| 0.
|
| 143 |
-
| 0.
|
| 144 |
-
| 0.
|
| 145 |
-
| 0.
|
| 146 |
-
| 0.
|
| 147 |
-
| 0.
|
| 148 |
-
| 0.
|
| 149 |
-
| 0.
|
| 150 |
-
| 0.
|
| 151 |
-
| 0.
|
| 152 |
-
| 0.
|
| 153 |
-
| 0.
|
| 154 |
-
| 0.
|
| 155 |
-
| 0.
|
| 156 |
-
| 0.
|
| 157 |
-
| 0.
|
| 158 |
-
| 0.
|
| 159 |
-
| 0.
|
| 160 |
-
| 0.
|
| 161 |
-
| 0.
|
| 162 |
-
| 0.
|
| 163 |
-
| 0.
|
| 164 |
-
| 0.
|
| 165 |
-
| 0.
|
| 166 |
-
| 0.
|
| 167 |
-
| 0.
|
| 168 |
-
| 0.
|
| 169 |
-
| 0.
|
| 170 |
-
| 0.
|
| 171 |
-
| 0.
|
| 172 |
-
| 0.
|
| 173 |
-
| 0.
|
| 174 |
-
| 0.
|
| 175 |
-
| 0.
|
| 176 |
-
| 0.
|
| 177 |
-
| 0.
|
| 178 |
-
| 0.
|
| 179 |
-
| 0.
|
| 180 |
-
| 0.
|
| 181 |
-
| 0.
|
| 182 |
-
| 0.
|
| 183 |
-
| 0.
|
| 184 |
-
| 0.
|
| 185 |
-
| 0.
|
| 186 |
-
| 0.
|
| 187 |
-
| 0.
|
| 188 |
-
| 0.
|
| 189 |
-
| 0.
|
| 190 |
-
| 0.
|
| 191 |
-
| 0.
|
| 192 |
-
| 0.
|
| 193 |
-
| 0.
|
| 194 |
-
| 0.
|
| 195 |
-
| 0.
|
| 196 |
-
| 0.
|
| 197 |
-
| 0.
|
| 198 |
-
| 0.
|
| 199 |
-
| 0.
|
| 200 |
-
| 0.
|
| 201 |
-
| 0.
|
| 202 |
-
| 0.
|
| 203 |
-
| 0.
|
| 204 |
-
| 0.
|
| 205 |
-
| 0.
|
| 206 |
-
| 0.
|
| 207 |
-
| 0.
|
| 208 |
-
| 0.
|
| 209 |
-
| 0.
|
| 210 |
-
| 0.
|
| 211 |
-
| 0.
|
| 212 |
-
| 0.
|
| 213 |
-
| 0.
|
| 214 |
-
| 0.
|
| 215 |
-
| 0.
|
| 216 |
-
| 0.
|
| 217 |
-
| 0.
|
| 218 |
-
| 0.
|
| 219 |
-
| 0.
|
| 220 |
-
| 0.
|
| 221 |
-
| 0.
|
| 222 |
-
| 0.
|
| 223 |
-
| 0.
|
| 224 |
-
| 0.
|
| 225 |
-
| 0.
|
| 226 |
-
| 0.
|
| 227 |
-
| 0.
|
| 228 |
-
| 0.
|
| 229 |
-
| 0.
|
| 230 |
-
| 0.
|
| 231 |
-
| 0.
|
| 232 |
-
| 0.
|
| 233 |
-
| 0.
|
| 234 |
-
| 0.
|
| 235 |
-
| 0.
|
| 236 |
-
| 0.
|
| 237 |
-
| 0.
|
| 238 |
-
| 0.
|
| 239 |
-
| 0.
|
| 240 |
-
| 0.
|
| 241 |
-
| 0.
|
| 242 |
-
| 0.
|
| 243 |
-
| 0.
|
| 244 |
-
| 0.
|
| 245 |
-
| 0.
|
| 246 |
-
| 0.
|
| 247 |
-
| 0.
|
| 248 |
-
| 0.
|
| 249 |
|
| 250 |
|
| 251 |
### Framework versions
|
|
|
|
| 1 |
---
|
|
|
|
| 2 |
tags:
|
| 3 |
- generated_from_trainer
|
| 4 |
model-index:
|
|
|
|
| 11 |
|
| 12 |
# calculator_model_test
|
| 13 |
|
| 14 |
+
This model is a fine-tuned version of [](https://huggingface.co/) on the None dataset.
|
| 15 |
It achieves the following results on the evaluation set:
|
| 16 |
+
- Loss: 0.0317
|
| 17 |
|
| 18 |
## Model description
|
| 19 |
|
|
|
|
| 45 |
|
| 46 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 47 |
|:-------------:|:-----:|:----:|:---------------:|
|
| 48 |
+
| 3.6147 | 1.0 | 6 | 3.0415 |
|
| 49 |
+
| 2.585 | 2.0 | 12 | 2.0622 |
|
| 50 |
+
| 1.899 | 3.0 | 18 | 1.6836 |
|
| 51 |
+
| 1.6095 | 4.0 | 24 | 1.6248 |
|
| 52 |
+
| 1.5731 | 5.0 | 30 | 1.5728 |
|
| 53 |
+
| 1.5407 | 6.0 | 36 | 1.5594 |
|
| 54 |
+
| 1.5353 | 7.0 | 42 | 1.5163 |
|
| 55 |
+
| 1.4873 | 8.0 | 48 | 1.4470 |
|
| 56 |
+
| 1.4322 | 9.0 | 54 | 1.4274 |
|
| 57 |
+
| 1.3743 | 10.0 | 60 | 1.3462 |
|
| 58 |
+
| 1.3009 | 11.0 | 66 | 1.2124 |
|
| 59 |
+
| 1.1918 | 12.0 | 72 | 1.1226 |
|
| 60 |
+
| 1.1449 | 13.0 | 78 | 1.1215 |
|
| 61 |
+
| 1.0914 | 14.0 | 84 | 1.0471 |
|
| 62 |
+
| 1.0285 | 15.0 | 90 | 0.9795 |
|
| 63 |
+
| 1.0093 | 16.0 | 96 | 1.0062 |
|
| 64 |
+
| 0.9957 | 17.0 | 102 | 0.9296 |
|
| 65 |
+
| 0.944 | 18.0 | 108 | 0.9268 |
|
| 66 |
+
| 0.9194 | 19.0 | 114 | 0.9417 |
|
| 67 |
+
| 0.9014 | 20.0 | 120 | 0.8415 |
|
| 68 |
+
| 0.8789 | 21.0 | 126 | 0.7764 |
|
| 69 |
+
| 0.8155 | 22.0 | 132 | 0.7639 |
|
| 70 |
+
| 0.8593 | 23.0 | 138 | 0.8127 |
|
| 71 |
+
| 0.836 | 24.0 | 144 | 0.7397 |
|
| 72 |
+
| 0.7613 | 25.0 | 150 | 0.7067 |
|
| 73 |
+
| 0.7818 | 26.0 | 156 | 0.7217 |
|
| 74 |
+
| 0.7702 | 27.0 | 162 | 0.7128 |
|
| 75 |
+
| 0.7376 | 28.0 | 168 | 0.7242 |
|
| 76 |
+
| 0.8006 | 29.0 | 174 | 0.7117 |
|
| 77 |
+
| 0.7561 | 30.0 | 180 | 0.6767 |
|
| 78 |
+
| 0.7185 | 31.0 | 186 | 0.6828 |
|
| 79 |
+
| 0.7055 | 32.0 | 192 | 0.6215 |
|
| 80 |
+
| 0.6967 | 33.0 | 198 | 0.6766 |
|
| 81 |
+
| 0.7193 | 34.0 | 204 | 0.6238 |
|
| 82 |
+
| 0.6791 | 35.0 | 210 | 0.5900 |
|
| 83 |
+
| 0.6741 | 36.0 | 216 | 0.6307 |
|
| 84 |
+
| 0.663 | 37.0 | 222 | 0.6012 |
|
| 85 |
+
| 0.6326 | 38.0 | 228 | 0.5944 |
|
| 86 |
+
| 0.6041 | 39.0 | 234 | 0.5459 |
|
| 87 |
+
| 0.617 | 40.0 | 240 | 0.5786 |
|
| 88 |
+
| 0.6369 | 41.0 | 246 | 0.5896 |
|
| 89 |
+
| 0.6243 | 42.0 | 252 | 0.5446 |
|
| 90 |
+
| 0.5921 | 43.0 | 258 | 0.4864 |
|
| 91 |
+
| 0.5529 | 44.0 | 264 | 0.5561 |
|
| 92 |
+
| 0.5757 | 45.0 | 270 | 0.5783 |
|
| 93 |
+
| 0.5919 | 46.0 | 276 | 0.5235 |
|
| 94 |
+
| 0.5509 | 47.0 | 282 | 0.4525 |
|
| 95 |
+
| 0.5229 | 48.0 | 288 | 0.5007 |
|
| 96 |
+
| 0.5871 | 49.0 | 294 | 0.5009 |
|
| 97 |
+
| 0.5793 | 50.0 | 300 | 0.5431 |
|
| 98 |
+
| 0.5922 | 51.0 | 306 | 0.5404 |
|
| 99 |
+
| 0.5539 | 52.0 | 312 | 0.5386 |
|
| 100 |
+
| 0.5785 | 53.0 | 318 | 0.4697 |
|
| 101 |
+
| 0.5528 | 54.0 | 324 | 0.5061 |
|
| 102 |
+
| 0.5047 | 55.0 | 330 | 0.4249 |
|
| 103 |
+
| 0.475 | 56.0 | 336 | 0.4206 |
|
| 104 |
+
| 0.5236 | 57.0 | 342 | 0.5689 |
|
| 105 |
+
| 0.576 | 58.0 | 348 | 0.4258 |
|
| 106 |
+
| 0.4862 | 59.0 | 354 | 0.4070 |
|
| 107 |
+
| 0.4946 | 60.0 | 360 | 0.4136 |
|
| 108 |
+
| 0.4527 | 61.0 | 366 | 0.3848 |
|
| 109 |
+
| 0.4522 | 62.0 | 372 | 0.4288 |
|
| 110 |
+
| 0.5087 | 63.0 | 378 | 0.5660 |
|
| 111 |
+
| 0.5559 | 64.0 | 384 | 0.5371 |
|
| 112 |
+
| 0.5153 | 65.0 | 390 | 0.4595 |
|
| 113 |
+
| 0.4503 | 66.0 | 396 | 0.3648 |
|
| 114 |
+
| 0.4191 | 67.0 | 402 | 0.3787 |
|
| 115 |
+
| 0.4522 | 68.0 | 408 | 0.3469 |
|
| 116 |
+
| 0.4096 | 69.0 | 414 | 0.3622 |
|
| 117 |
+
| 0.4502 | 70.0 | 420 | 0.3613 |
|
| 118 |
+
| 0.4138 | 71.0 | 426 | 0.3700 |
|
| 119 |
+
| 0.3896 | 72.0 | 432 | 0.3920 |
|
| 120 |
+
| 0.4271 | 73.0 | 438 | 0.3354 |
|
| 121 |
+
| 0.4107 | 74.0 | 444 | 0.3193 |
|
| 122 |
+
| 0.391 | 75.0 | 450 | 0.3352 |
|
| 123 |
+
| 0.373 | 76.0 | 456 | 0.3818 |
|
| 124 |
+
| 0.4296 | 77.0 | 462 | 0.3238 |
|
| 125 |
+
| 0.3812 | 78.0 | 468 | 0.3337 |
|
| 126 |
+
| 0.3756 | 79.0 | 474 | 0.3105 |
|
| 127 |
+
| 0.3579 | 80.0 | 480 | 0.3433 |
|
| 128 |
+
| 0.4325 | 81.0 | 486 | 0.3103 |
|
| 129 |
+
| 0.356 | 82.0 | 492 | 0.3060 |
|
| 130 |
+
| 0.3467 | 83.0 | 498 | 0.3780 |
|
| 131 |
+
| 0.3922 | 84.0 | 504 | 0.2863 |
|
| 132 |
+
| 0.3457 | 85.0 | 510 | 0.2865 |
|
| 133 |
+
| 0.3755 | 86.0 | 516 | 0.3041 |
|
| 134 |
+
| 0.3319 | 87.0 | 522 | 0.2777 |
|
| 135 |
+
| 0.3359 | 88.0 | 528 | 0.3803 |
|
| 136 |
+
| 0.4192 | 89.0 | 534 | 0.3473 |
|
| 137 |
+
| 0.3941 | 90.0 | 540 | 0.3745 |
|
| 138 |
+
| 0.3991 | 91.0 | 546 | 0.3331 |
|
| 139 |
+
| 0.3489 | 92.0 | 552 | 0.3579 |
|
| 140 |
+
| 0.3352 | 93.0 | 558 | 0.2947 |
|
| 141 |
+
| 0.3202 | 94.0 | 564 | 0.2416 |
|
| 142 |
+
| 0.3339 | 95.0 | 570 | 0.3635 |
|
| 143 |
+
| 0.4108 | 96.0 | 576 | 0.2779 |
|
| 144 |
+
| 0.3827 | 97.0 | 582 | 0.2846 |
|
| 145 |
+
| 0.3559 | 98.0 | 588 | 0.2754 |
|
| 146 |
+
| 0.2985 | 99.0 | 594 | 0.2107 |
|
| 147 |
+
| 0.264 | 100.0 | 600 | 0.1958 |
|
| 148 |
+
| 0.2807 | 101.0 | 606 | 0.2028 |
|
| 149 |
+
| 0.2861 | 102.0 | 612 | 0.2034 |
|
| 150 |
+
| 0.2661 | 103.0 | 618 | 0.1979 |
|
| 151 |
+
| 0.264 | 104.0 | 624 | 0.2134 |
|
| 152 |
+
| 0.2747 | 105.0 | 630 | 0.1754 |
|
| 153 |
+
| 0.2785 | 106.0 | 636 | 0.2329 |
|
| 154 |
+
| 0.2656 | 107.0 | 642 | 0.1934 |
|
| 155 |
+
| 0.2505 | 108.0 | 648 | 0.2213 |
|
| 156 |
+
| 0.2572 | 109.0 | 654 | 0.2313 |
|
| 157 |
+
| 0.2929 | 110.0 | 660 | 0.2308 |
|
| 158 |
+
| 0.2419 | 111.0 | 666 | 0.1780 |
|
| 159 |
+
| 0.239 | 112.0 | 672 | 0.1694 |
|
| 160 |
+
| 0.2279 | 113.0 | 678 | 0.1580 |
|
| 161 |
+
| 0.2528 | 114.0 | 684 | 0.3002 |
|
| 162 |
+
| 0.3297 | 115.0 | 690 | 0.2676 |
|
| 163 |
+
| 0.3147 | 116.0 | 696 | 0.3287 |
|
| 164 |
+
| 0.2826 | 117.0 | 702 | 0.1475 |
|
| 165 |
+
| 0.2033 | 118.0 | 708 | 0.1359 |
|
| 166 |
+
| 0.1938 | 119.0 | 714 | 0.1592 |
|
| 167 |
+
| 0.2105 | 120.0 | 720 | 0.1696 |
|
| 168 |
+
| 0.2196 | 121.0 | 726 | 0.1532 |
|
| 169 |
+
| 0.2102 | 122.0 | 732 | 0.1157 |
|
| 170 |
+
| 0.2014 | 123.0 | 738 | 0.1835 |
|
| 171 |
+
| 0.2505 | 124.0 | 744 | 0.1851 |
|
| 172 |
+
| 0.2411 | 125.0 | 750 | 0.2881 |
|
| 173 |
+
| 0.2353 | 126.0 | 756 | 0.1911 |
|
| 174 |
+
| 0.2268 | 127.0 | 762 | 0.1874 |
|
| 175 |
+
| 0.2024 | 128.0 | 768 | 0.1613 |
|
| 176 |
+
| 0.2046 | 129.0 | 774 | 0.1938 |
|
| 177 |
+
| 0.199 | 130.0 | 780 | 0.1129 |
|
| 178 |
+
| 0.1703 | 131.0 | 786 | 0.1511 |
|
| 179 |
+
| 0.1924 | 132.0 | 792 | 0.1744 |
|
| 180 |
+
| 0.1854 | 133.0 | 798 | 0.1238 |
|
| 181 |
+
| 0.1632 | 134.0 | 804 | 0.1050 |
|
| 182 |
+
| 0.1589 | 135.0 | 810 | 0.1316 |
|
| 183 |
+
| 0.1787 | 136.0 | 816 | 0.0895 |
|
| 184 |
+
| 0.1658 | 137.0 | 822 | 0.0836 |
|
| 185 |
+
| 0.141 | 138.0 | 828 | 0.1087 |
|
| 186 |
+
| 0.1671 | 139.0 | 834 | 0.1068 |
|
| 187 |
+
| 0.1557 | 140.0 | 840 | 0.0800 |
|
| 188 |
+
| 0.1488 | 141.0 | 846 | 0.1277 |
|
| 189 |
+
| 0.1709 | 142.0 | 852 | 0.1126 |
|
| 190 |
+
| 0.1499 | 143.0 | 858 | 0.0913 |
|
| 191 |
+
| 0.1597 | 144.0 | 864 | 0.0829 |
|
| 192 |
+
| 0.1314 | 145.0 | 870 | 0.0762 |
|
| 193 |
+
| 0.1501 | 146.0 | 876 | 0.0897 |
|
| 194 |
+
| 0.156 | 147.0 | 882 | 0.0902 |
|
| 195 |
+
| 0.1482 | 148.0 | 888 | 0.0903 |
|
| 196 |
+
| 0.1401 | 149.0 | 894 | 0.0749 |
|
| 197 |
+
| 0.1322 | 150.0 | 900 | 0.0781 |
|
| 198 |
+
| 0.1309 | 151.0 | 906 | 0.0719 |
|
| 199 |
+
| 0.1326 | 152.0 | 912 | 0.0691 |
|
| 200 |
+
| 0.1311 | 153.0 | 918 | 0.0701 |
|
| 201 |
+
| 0.1202 | 154.0 | 924 | 0.0742 |
|
| 202 |
+
| 0.1258 | 155.0 | 930 | 0.0728 |
|
| 203 |
+
| 0.1183 | 156.0 | 936 | 0.0566 |
|
| 204 |
+
| 0.1181 | 157.0 | 942 | 0.0541 |
|
| 205 |
+
| 0.1137 | 158.0 | 948 | 0.0662 |
|
| 206 |
+
| 0.1061 | 159.0 | 954 | 0.0662 |
|
| 207 |
+
| 0.1121 | 160.0 | 960 | 0.0628 |
|
| 208 |
+
| 0.1038 | 161.0 | 966 | 0.0609 |
|
| 209 |
+
| 0.1135 | 162.0 | 972 | 0.0728 |
|
| 210 |
+
| 0.1317 | 163.0 | 978 | 0.0785 |
|
| 211 |
+
| 0.1149 | 164.0 | 984 | 0.0753 |
|
| 212 |
+
| 0.1111 | 165.0 | 990 | 0.0647 |
|
| 213 |
+
| 0.0926 | 166.0 | 996 | 0.0592 |
|
| 214 |
+
| 0.0931 | 167.0 | 1002 | 0.0554 |
|
| 215 |
+
| 0.0865 | 168.0 | 1008 | 0.0480 |
|
| 216 |
+
| 0.0881 | 169.0 | 1014 | 0.0498 |
|
| 217 |
+
| 0.0932 | 170.0 | 1020 | 0.0524 |
|
| 218 |
+
| 0.0934 | 171.0 | 1026 | 0.0629 |
|
| 219 |
+
| 0.1054 | 172.0 | 1032 | 0.0561 |
|
| 220 |
+
| 0.0933 | 173.0 | 1038 | 0.0422 |
|
| 221 |
+
| 0.0812 | 174.0 | 1044 | 0.0605 |
|
| 222 |
+
| 0.0953 | 175.0 | 1050 | 0.0485 |
|
| 223 |
+
| 0.0963 | 176.0 | 1056 | 0.0394 |
|
| 224 |
+
| 0.0731 | 177.0 | 1062 | 0.0378 |
|
| 225 |
+
| 0.0758 | 178.0 | 1068 | 0.0394 |
|
| 226 |
+
| 0.0703 | 179.0 | 1074 | 0.0406 |
|
| 227 |
+
| 0.0756 | 180.0 | 1080 | 0.0427 |
|
| 228 |
+
| 0.0812 | 181.0 | 1086 | 0.0538 |
|
| 229 |
+
| 0.0842 | 182.0 | 1092 | 0.0434 |
|
| 230 |
+
| 0.0773 | 183.0 | 1098 | 0.0439 |
|
| 231 |
+
| 0.073 | 184.0 | 1104 | 0.0379 |
|
| 232 |
+
| 0.0707 | 185.0 | 1110 | 0.0422 |
|
| 233 |
+
| 0.0749 | 186.0 | 1116 | 0.0420 |
|
| 234 |
+
| 0.0746 | 187.0 | 1122 | 0.0388 |
|
| 235 |
+
| 0.068 | 188.0 | 1128 | 0.0386 |
|
| 236 |
+
| 0.0654 | 189.0 | 1134 | 0.0378 |
|
| 237 |
+
| 0.0647 | 190.0 | 1140 | 0.0335 |
|
| 238 |
+
| 0.0629 | 191.0 | 1146 | 0.0402 |
|
| 239 |
+
| 0.0642 | 192.0 | 1152 | 0.0344 |
|
| 240 |
+
| 0.063 | 193.0 | 1158 | 0.0374 |
|
| 241 |
+
| 0.0631 | 194.0 | 1164 | 0.0321 |
|
| 242 |
+
| 0.0605 | 195.0 | 1170 | 0.0356 |
|
| 243 |
+
| 0.065 | 196.0 | 1176 | 0.0334 |
|
| 244 |
+
| 0.0591 | 197.0 | 1182 | 0.0321 |
|
| 245 |
+
| 0.0558 | 198.0 | 1188 | 0.0317 |
|
| 246 |
+
| 0.0596 | 199.0 | 1194 | 0.0316 |
|
| 247 |
+
| 0.06 | 200.0 | 1200 | 0.0317 |
|
| 248 |
|
| 249 |
|
| 250 |
### Framework versions
|
config.json
CHANGED
|
@@ -1,5 +1,4 @@
|
|
| 1 |
{
|
| 2 |
-
"_name_or_path": "Kielak2/calculator_model_test",
|
| 3 |
"architectures": [
|
| 4 |
"EncoderDecoderModel"
|
| 5 |
],
|
|
@@ -79,7 +78,7 @@
|
|
| 79 |
"typical_p": 1.0,
|
| 80 |
"use_bfloat16": false,
|
| 81 |
"use_cache": true,
|
| 82 |
-
"vocab_size":
|
| 83 |
},
|
| 84 |
"decoder_start_token_id": 2,
|
| 85 |
"encoder": {
|
|
@@ -158,7 +157,7 @@
|
|
| 158 |
"typical_p": 1.0,
|
| 159 |
"use_bfloat16": false,
|
| 160 |
"use_cache": true,
|
| 161 |
-
"vocab_size":
|
| 162 |
},
|
| 163 |
"eos_token_id": 0,
|
| 164 |
"is_encoder_decoder": true,
|
|
|
|
| 1 |
{
|
|
|
|
| 2 |
"architectures": [
|
| 3 |
"EncoderDecoderModel"
|
| 4 |
],
|
|
|
|
| 78 |
"typical_p": 1.0,
|
| 79 |
"use_bfloat16": false,
|
| 80 |
"use_cache": true,
|
| 81 |
+
"vocab_size": 52
|
| 82 |
},
|
| 83 |
"decoder_start_token_id": 2,
|
| 84 |
"encoder": {
|
|
|
|
| 157 |
"typical_p": 1.0,
|
| 158 |
"use_bfloat16": false,
|
| 159 |
"use_cache": true,
|
| 160 |
+
"vocab_size": 52
|
| 161 |
},
|
| 162 |
"eos_token_id": 0,
|
| 163 |
"is_encoder_decoder": true,
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:51120f4f2cea38123ca289b764c35b52f8b31bfb6c4f21b3d79654f167659770
|
| 3 |
+
size 31205552
|
runs/Mar04_10-28-21_c60a5c456cbd/events.out.tfevents.1709548101.c60a5c456cbd.796.8
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a71d31afeef2bea511c8cbad305b30f5ef69f9b7ba5117dc901c987f77229131
|
| 3 |
+
size 105117
|
tokenizer.json
CHANGED
|
@@ -112,44 +112,80 @@
|
|
| 112 |
"7": 13,
|
| 113 |
"8": 14,
|
| 114 |
"9": 15,
|
| 115 |
-
"
|
| 116 |
-
"
|
| 117 |
-
"
|
| 118 |
-
"
|
| 119 |
-
"
|
| 120 |
-
"
|
| 121 |
-
"
|
| 122 |
-
"
|
| 123 |
-
"
|
| 124 |
-
"
|
| 125 |
-
"
|
| 126 |
-
"
|
| 127 |
"16": 28,
|
| 128 |
"93": 29,
|
| 129 |
-
"
|
| 130 |
-
"
|
| 131 |
-
"
|
| 132 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 133 |
},
|
| 134 |
"merges": [
|
| 135 |
-
"1 0",
|
| 136 |
"9 9",
|
| 137 |
-
"1
|
| 138 |
"9 8",
|
| 139 |
-
"1
|
| 140 |
"9 7",
|
| 141 |
-
"1
|
| 142 |
"9 6",
|
| 143 |
-
"1
|
| 144 |
"9 5",
|
| 145 |
-
"1
|
| 146 |
"9 4",
|
|
|
|
| 147 |
"1 6",
|
| 148 |
"9 3",
|
| 149 |
-
"1 7",
|
| 150 |
"9 2",
|
|
|
|
|
|
|
| 151 |
"1 8",
|
| 152 |
-
"9
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 153 |
]
|
| 154 |
}
|
| 155 |
}
|
|
|
|
| 112 |
"7": 13,
|
| 113 |
"8": 14,
|
| 114 |
"9": 15,
|
| 115 |
+
"99": 16,
|
| 116 |
+
"10": 17,
|
| 117 |
+
"98": 18,
|
| 118 |
+
"11": 19,
|
| 119 |
+
"97": 20,
|
| 120 |
+
"12": 21,
|
| 121 |
+
"96": 22,
|
| 122 |
+
"13": 23,
|
| 123 |
+
"95": 24,
|
| 124 |
+
"14": 25,
|
| 125 |
+
"94": 26,
|
| 126 |
+
"15": 27,
|
| 127 |
"16": 28,
|
| 128 |
"93": 29,
|
| 129 |
+
"92": 30,
|
| 130 |
+
"17": 31,
|
| 131 |
+
"91": 32,
|
| 132 |
+
"18": 33,
|
| 133 |
+
"90": 34,
|
| 134 |
+
"19": 35,
|
| 135 |
+
"20": 36,
|
| 136 |
+
"89": 37,
|
| 137 |
+
"88": 38,
|
| 138 |
+
"21": 39,
|
| 139 |
+
"22": 40,
|
| 140 |
+
"87": 41,
|
| 141 |
+
"23": 42,
|
| 142 |
+
"86": 43,
|
| 143 |
+
"24": 44,
|
| 144 |
+
"85": 45,
|
| 145 |
+
"25": 46,
|
| 146 |
+
"84": 47,
|
| 147 |
+
"26": 48,
|
| 148 |
+
"83": 49,
|
| 149 |
+
"27": 50,
|
| 150 |
+
"82": 51
|
| 151 |
},
|
| 152 |
"merges": [
|
|
|
|
| 153 |
"9 9",
|
| 154 |
+
"1 0",
|
| 155 |
"9 8",
|
| 156 |
+
"1 1",
|
| 157 |
"9 7",
|
| 158 |
+
"1 2",
|
| 159 |
"9 6",
|
| 160 |
+
"1 3",
|
| 161 |
"9 5",
|
| 162 |
+
"1 4",
|
| 163 |
"9 4",
|
| 164 |
+
"1 5",
|
| 165 |
"1 6",
|
| 166 |
"9 3",
|
|
|
|
| 167 |
"9 2",
|
| 168 |
+
"1 7",
|
| 169 |
+
"9 1",
|
| 170 |
"1 8",
|
| 171 |
+
"9 0",
|
| 172 |
+
"1 9",
|
| 173 |
+
"2 0",
|
| 174 |
+
"8 9",
|
| 175 |
+
"8 8",
|
| 176 |
+
"2 1",
|
| 177 |
+
"2 2",
|
| 178 |
+
"8 7",
|
| 179 |
+
"2 3",
|
| 180 |
+
"8 6",
|
| 181 |
+
"2 4",
|
| 182 |
+
"8 5",
|
| 183 |
+
"2 5",
|
| 184 |
+
"8 4",
|
| 185 |
+
"2 6",
|
| 186 |
+
"8 3",
|
| 187 |
+
"2 7",
|
| 188 |
+
"8 2"
|
| 189 |
]
|
| 190 |
}
|
| 191 |
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5112
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b7e53b1c89424e1de538277509f2a24e580efb76210b3761752044d4694ee8fe
|
| 3 |
size 5112
|