End of training
Browse files- README.md +251 -0
- generation_config.json +6 -0
README.md
ADDED
|
@@ -0,0 +1,251 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
model-index:
|
| 6 |
+
- name: impossible-llms-german-fronting-bigram
|
| 7 |
+
results: []
|
| 8 |
+
---
|
| 9 |
+
|
| 10 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 11 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 12 |
+
|
| 13 |
+
# impossible-llms-german-fronting-bigram
|
| 14 |
+
|
| 15 |
+
This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
|
| 16 |
+
It achieves the following results on the evaluation set:
|
| 17 |
+
- Loss: 6.0327
|
| 18 |
+
|
| 19 |
+
## Model description
|
| 20 |
+
|
| 21 |
+
More information needed
|
| 22 |
+
|
| 23 |
+
## Intended uses & limitations
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Training and evaluation data
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training procedure
|
| 32 |
+
|
| 33 |
+
### Training hyperparameters
|
| 34 |
+
|
| 35 |
+
The following hyperparameters were used during training:
|
| 36 |
+
- learning_rate: 0.0001
|
| 37 |
+
- train_batch_size: 12
|
| 38 |
+
- eval_batch_size: 8
|
| 39 |
+
- seed: 0
|
| 40 |
+
- distributed_type: multi-GPU
|
| 41 |
+
- num_devices: 4
|
| 42 |
+
- gradient_accumulation_steps: 8
|
| 43 |
+
- total_train_batch_size: 384
|
| 44 |
+
- total_eval_batch_size: 32
|
| 45 |
+
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
| 46 |
+
- lr_scheduler_type: cosine
|
| 47 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 48 |
+
- training_steps: 3000
|
| 49 |
+
- mixed_precision_training: Native AMP
|
| 50 |
+
- label_smoothing_factor: 0.1
|
| 51 |
+
|
| 52 |
+
### Training results
|
| 53 |
+
|
| 54 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 55 |
+
|:-------------:|:--------:|:----:|:---------------:|
|
| 56 |
+
| 76.0331 | 0.9697 | 16 | 9.4338 |
|
| 57 |
+
| 71.6673 | 1.9697 | 32 | 8.9567 |
|
| 58 |
+
| 69.3824 | 2.9697 | 48 | 8.6316 |
|
| 59 |
+
| 66.0393 | 3.9697 | 64 | 8.2404 |
|
| 60 |
+
| 62.9205 | 4.9697 | 80 | 7.8178 |
|
| 61 |
+
| 59.3477 | 5.9697 | 96 | 7.3733 |
|
| 62 |
+
| 56.0467 | 6.9697 | 112 | 6.9478 |
|
| 63 |
+
| 53.3128 | 7.9697 | 128 | 6.6167 |
|
| 64 |
+
| 51.4884 | 8.9697 | 144 | 6.4220 |
|
| 65 |
+
| 50.5338 | 9.9697 | 160 | 6.3133 |
|
| 66 |
+
| 49.7489 | 10.9697 | 176 | 6.2331 |
|
| 67 |
+
| 49.5029 | 11.9697 | 192 | 6.1774 |
|
| 68 |
+
| 48.9784 | 12.9697 | 208 | 6.1174 |
|
| 69 |
+
| 48.484 | 13.9697 | 224 | 6.0770 |
|
| 70 |
+
| 48.1829 | 14.9697 | 240 | 6.0347 |
|
| 71 |
+
| 47.8824 | 15.9697 | 256 | 5.9986 |
|
| 72 |
+
| 47.6146 | 16.9697 | 272 | 5.9695 |
|
| 73 |
+
| 47.5391 | 17.9697 | 288 | 5.9389 |
|
| 74 |
+
| 46.8092 | 18.9697 | 304 | 5.9169 |
|
| 75 |
+
| 46.6895 | 19.9697 | 320 | 5.8892 |
|
| 76 |
+
| 46.41 | 20.9697 | 336 | 5.8594 |
|
| 77 |
+
| 46.259 | 21.9697 | 352 | 5.8347 |
|
| 78 |
+
| 45.916 | 22.9697 | 368 | 5.8068 |
|
| 79 |
+
| 45.7068 | 23.9697 | 384 | 5.7844 |
|
| 80 |
+
| 45.0942 | 24.9697 | 400 | 5.7543 |
|
| 81 |
+
| 45.2343 | 25.9697 | 416 | 5.7346 |
|
| 82 |
+
| 44.4914 | 26.9697 | 432 | 5.7001 |
|
| 83 |
+
| 44.4128 | 27.9697 | 448 | 5.6704 |
|
| 84 |
+
| 44.3277 | 28.9697 | 464 | 5.6436 |
|
| 85 |
+
| 44.281 | 29.9697 | 480 | 5.6213 |
|
| 86 |
+
| 43.7328 | 30.9697 | 496 | 5.5903 |
|
| 87 |
+
| 43.4948 | 31.9697 | 512 | 5.5636 |
|
| 88 |
+
| 42.9645 | 32.9697 | 528 | 5.5445 |
|
| 89 |
+
| 42.8097 | 33.9697 | 544 | 5.5143 |
|
| 90 |
+
| 42.9638 | 34.9697 | 560 | 5.4895 |
|
| 91 |
+
| 42.2659 | 35.9697 | 576 | 5.4703 |
|
| 92 |
+
| 42.2061 | 36.9697 | 592 | 5.4521 |
|
| 93 |
+
| 41.8982 | 37.9697 | 608 | 5.4312 |
|
| 94 |
+
| 41.7378 | 38.9697 | 624 | 5.4206 |
|
| 95 |
+
| 41.6193 | 39.9697 | 640 | 5.3960 |
|
| 96 |
+
| 41.4941 | 40.9697 | 656 | 5.3849 |
|
| 97 |
+
| 41.14 | 41.9697 | 672 | 5.3676 |
|
| 98 |
+
| 41.0429 | 42.9697 | 688 | 5.3546 |
|
| 99 |
+
| 40.7736 | 43.9697 | 704 | 5.3389 |
|
| 100 |
+
| 40.571 | 44.9697 | 720 | 5.3348 |
|
| 101 |
+
| 40.1309 | 45.9697 | 736 | 5.3256 |
|
| 102 |
+
| 40.3361 | 46.9697 | 752 | 5.3146 |
|
| 103 |
+
| 40.1762 | 47.9697 | 768 | 5.3130 |
|
| 104 |
+
| 39.6508 | 48.9697 | 784 | 5.2994 |
|
| 105 |
+
| 39.7313 | 49.9697 | 800 | 5.2926 |
|
| 106 |
+
| 39.1776 | 50.9697 | 816 | 5.2911 |
|
| 107 |
+
| 38.9937 | 51.9697 | 832 | 5.2884 |
|
| 108 |
+
| 39.0236 | 52.9697 | 848 | 5.2826 |
|
| 109 |
+
| 38.802 | 53.9697 | 864 | 5.2838 |
|
| 110 |
+
| 38.5939 | 54.9697 | 880 | 5.2746 |
|
| 111 |
+
| 38.3015 | 55.9697 | 896 | 5.2809 |
|
| 112 |
+
| 38.4052 | 56.9697 | 912 | 5.2786 |
|
| 113 |
+
| 38.1407 | 57.9697 | 928 | 5.2766 |
|
| 114 |
+
| 37.8546 | 58.9697 | 944 | 5.2801 |
|
| 115 |
+
| 37.8129 | 59.9697 | 960 | 5.2827 |
|
| 116 |
+
| 37.7701 | 60.9697 | 976 | 5.2802 |
|
| 117 |
+
| 37.2836 | 61.9697 | 992 | 5.2886 |
|
| 118 |
+
| 37.4281 | 62.9697 | 1008 | 5.2888 |
|
| 119 |
+
| 37.1109 | 63.9697 | 1024 | 5.2970 |
|
| 120 |
+
| 36.7481 | 64.9697 | 1040 | 5.3024 |
|
| 121 |
+
| 36.7859 | 65.9697 | 1056 | 5.3057 |
|
| 122 |
+
| 36.6616 | 66.9697 | 1072 | 5.3054 |
|
| 123 |
+
| 36.4677 | 67.9697 | 1088 | 5.3106 |
|
| 124 |
+
| 36.3254 | 68.9697 | 1104 | 5.3140 |
|
| 125 |
+
| 36.2414 | 69.9697 | 1120 | 5.3257 |
|
| 126 |
+
| 36.0447 | 70.9697 | 1136 | 5.3329 |
|
| 127 |
+
| 35.8527 | 71.9697 | 1152 | 5.3406 |
|
| 128 |
+
| 35.6302 | 72.9697 | 1168 | 5.3465 |
|
| 129 |
+
| 35.3203 | 73.9697 | 1184 | 5.3598 |
|
| 130 |
+
| 35.3191 | 74.9697 | 1200 | 5.3631 |
|
| 131 |
+
| 35.0504 | 75.9697 | 1216 | 5.3752 |
|
| 132 |
+
| 34.8329 | 76.9697 | 1232 | 5.3791 |
|
| 133 |
+
| 34.9235 | 77.9697 | 1248 | 5.3881 |
|
| 134 |
+
| 34.6245 | 78.9697 | 1264 | 5.3958 |
|
| 135 |
+
| 34.5753 | 79.9697 | 1280 | 5.4096 |
|
| 136 |
+
| 34.4142 | 80.9697 | 1296 | 5.4234 |
|
| 137 |
+
| 34.1888 | 81.9697 | 1312 | 5.4222 |
|
| 138 |
+
| 33.9786 | 82.9697 | 1328 | 5.4345 |
|
| 139 |
+
| 34.1911 | 83.9697 | 1344 | 5.4443 |
|
| 140 |
+
| 33.795 | 84.9697 | 1360 | 5.4489 |
|
| 141 |
+
| 33.8146 | 85.9697 | 1376 | 5.4639 |
|
| 142 |
+
| 33.5029 | 86.9697 | 1392 | 5.4749 |
|
| 143 |
+
| 33.2093 | 87.9697 | 1408 | 5.4868 |
|
| 144 |
+
| 33.4014 | 88.9697 | 1424 | 5.4939 |
|
| 145 |
+
| 33.2127 | 89.9697 | 1440 | 5.5026 |
|
| 146 |
+
| 33.0343 | 90.9697 | 1456 | 5.5170 |
|
| 147 |
+
| 33.004 | 91.9697 | 1472 | 5.5275 |
|
| 148 |
+
| 32.6994 | 92.9697 | 1488 | 5.5386 |
|
| 149 |
+
| 32.7135 | 93.9697 | 1504 | 5.5470 |
|
| 150 |
+
| 32.5986 | 94.9697 | 1520 | 5.5595 |
|
| 151 |
+
| 32.5105 | 95.9697 | 1536 | 5.5684 |
|
| 152 |
+
| 32.0216 | 96.9697 | 1552 | 5.5737 |
|
| 153 |
+
| 32.1599 | 97.9697 | 1568 | 5.5834 |
|
| 154 |
+
| 32.0658 | 98.9697 | 1584 | 5.5970 |
|
| 155 |
+
| 31.9547 | 99.9697 | 1600 | 5.6053 |
|
| 156 |
+
| 31.7688 | 100.9697 | 1616 | 5.6189 |
|
| 157 |
+
| 31.6648 | 101.9697 | 1632 | 5.6250 |
|
| 158 |
+
| 31.4364 | 102.9697 | 1648 | 5.6369 |
|
| 159 |
+
| 31.5234 | 103.9697 | 1664 | 5.6476 |
|
| 160 |
+
| 31.5321 | 104.9697 | 1680 | 5.6584 |
|
| 161 |
+
| 31.4901 | 105.9697 | 1696 | 5.6663 |
|
| 162 |
+
| 31.2033 | 106.9697 | 1712 | 5.6770 |
|
| 163 |
+
| 31.018 | 107.9697 | 1728 | 5.6838 |
|
| 164 |
+
| 30.947 | 108.9697 | 1744 | 5.6970 |
|
| 165 |
+
| 30.7787 | 109.9697 | 1760 | 5.6996 |
|
| 166 |
+
| 30.7756 | 110.9697 | 1776 | 5.7140 |
|
| 167 |
+
| 30.507 | 111.9697 | 1792 | 5.7222 |
|
| 168 |
+
| 30.6232 | 112.9697 | 1808 | 5.7301 |
|
| 169 |
+
| 30.5118 | 113.9697 | 1824 | 5.7433 |
|
| 170 |
+
| 30.1597 | 114.9697 | 1840 | 5.7478 |
|
| 171 |
+
| 30.3279 | 115.9697 | 1856 | 5.7577 |
|
| 172 |
+
| 30.0896 | 116.9697 | 1872 | 5.7650 |
|
| 173 |
+
| 30.1555 | 117.9697 | 1888 | 5.7767 |
|
| 174 |
+
| 29.8685 | 118.9697 | 1904 | 5.7873 |
|
| 175 |
+
| 29.8582 | 119.9697 | 1920 | 5.7923 |
|
| 176 |
+
| 29.9249 | 120.9697 | 1936 | 5.8030 |
|
| 177 |
+
| 29.8765 | 121.9697 | 1952 | 5.8111 |
|
| 178 |
+
| 29.7486 | 122.9697 | 1968 | 5.8230 |
|
| 179 |
+
| 29.6922 | 123.9697 | 1984 | 5.8276 |
|
| 180 |
+
| 29.4026 | 124.9697 | 2000 | 5.8362 |
|
| 181 |
+
| 29.3903 | 125.9697 | 2016 | 5.8376 |
|
| 182 |
+
| 29.4068 | 126.9697 | 2032 | 5.8466 |
|
| 183 |
+
| 29.4381 | 127.9697 | 2048 | 5.8547 |
|
| 184 |
+
| 29.1938 | 128.9697 | 2064 | 5.8639 |
|
| 185 |
+
| 29.221 | 129.9697 | 2080 | 5.8655 |
|
| 186 |
+
| 29.2049 | 130.9697 | 2096 | 5.8743 |
|
| 187 |
+
| 29.1575 | 131.9697 | 2112 | 5.8820 |
|
| 188 |
+
| 29.1866 | 132.9697 | 2128 | 5.8838 |
|
| 189 |
+
| 28.8952 | 133.9697 | 2144 | 5.8972 |
|
| 190 |
+
| 28.902 | 134.9697 | 2160 | 5.9019 |
|
| 191 |
+
| 28.8144 | 135.9697 | 2176 | 5.9041 |
|
| 192 |
+
| 28.8061 | 136.9697 | 2192 | 5.9117 |
|
| 193 |
+
| 28.7095 | 137.9697 | 2208 | 5.9188 |
|
| 194 |
+
| 28.6465 | 138.9697 | 2224 | 5.9240 |
|
| 195 |
+
| 28.5635 | 139.9697 | 2240 | 5.9280 |
|
| 196 |
+
| 28.5905 | 140.9697 | 2256 | 5.9319 |
|
| 197 |
+
| 28.5319 | 141.9697 | 2272 | 5.9402 |
|
| 198 |
+
| 28.3719 | 142.9697 | 2288 | 5.9453 |
|
| 199 |
+
| 28.3483 | 143.9697 | 2304 | 5.9481 |
|
| 200 |
+
| 28.2941 | 144.9697 | 2320 | 5.9543 |
|
| 201 |
+
| 28.3381 | 145.9697 | 2336 | 5.9589 |
|
| 202 |
+
| 28.2441 | 146.9697 | 2352 | 5.9611 |
|
| 203 |
+
| 28.2837 | 147.9697 | 2368 | 5.9652 |
|
| 204 |
+
| 28.3381 | 148.9697 | 2384 | 5.9722 |
|
| 205 |
+
| 28.0723 | 149.9697 | 2400 | 5.9747 |
|
| 206 |
+
| 28.2182 | 150.9697 | 2416 | 5.9805 |
|
| 207 |
+
| 28.1211 | 151.9697 | 2432 | 5.9822 |
|
| 208 |
+
| 28.1113 | 152.9697 | 2448 | 5.9845 |
|
| 209 |
+
| 28.0931 | 153.9697 | 2464 | 5.9882 |
|
| 210 |
+
| 28.0684 | 154.9697 | 2480 | 5.9908 |
|
| 211 |
+
| 27.96 | 155.9697 | 2496 | 5.9967 |
|
| 212 |
+
| 28.1077 | 156.9697 | 2512 | 5.9980 |
|
| 213 |
+
| 27.7652 | 157.9697 | 2528 | 6.0017 |
|
| 214 |
+
| 28.0139 | 158.9697 | 2544 | 6.0054 |
|
| 215 |
+
| 27.872 | 159.9697 | 2560 | 6.0063 |
|
| 216 |
+
| 27.7807 | 160.9697 | 2576 | 6.0101 |
|
| 217 |
+
| 27.9509 | 161.9697 | 2592 | 6.0112 |
|
| 218 |
+
| 27.9565 | 162.9697 | 2608 | 6.0139 |
|
| 219 |
+
| 27.74 | 163.9697 | 2624 | 6.0165 |
|
| 220 |
+
| 27.8607 | 164.9697 | 2640 | 6.0191 |
|
| 221 |
+
| 27.7131 | 165.9697 | 2656 | 6.0194 |
|
| 222 |
+
| 27.6894 | 166.9697 | 2672 | 6.0211 |
|
| 223 |
+
| 27.7251 | 167.9697 | 2688 | 6.0234 |
|
| 224 |
+
| 27.7298 | 168.9697 | 2704 | 6.0243 |
|
| 225 |
+
| 27.6216 | 169.9697 | 2720 | 6.0259 |
|
| 226 |
+
| 27.7812 | 170.9697 | 2736 | 6.0269 |
|
| 227 |
+
| 27.7094 | 171.9697 | 2752 | 6.0277 |
|
| 228 |
+
| 27.7338 | 172.9697 | 2768 | 6.0286 |
|
| 229 |
+
| 27.6887 | 173.9697 | 2784 | 6.0289 |
|
| 230 |
+
| 27.69 | 174.9697 | 2800 | 6.0292 |
|
| 231 |
+
| 27.7041 | 175.9697 | 2816 | 6.0303 |
|
| 232 |
+
| 27.6164 | 176.9697 | 2832 | 6.0305 |
|
| 233 |
+
| 27.493 | 177.9697 | 2848 | 6.0307 |
|
| 234 |
+
| 27.7736 | 178.9697 | 2864 | 6.0321 |
|
| 235 |
+
| 27.5697 | 179.9697 | 2880 | 6.0323 |
|
| 236 |
+
| 27.4724 | 180.9697 | 2896 | 6.0323 |
|
| 237 |
+
| 27.4104 | 181.9697 | 2912 | 6.0325 |
|
| 238 |
+
| 27.4961 | 182.9697 | 2928 | 6.0327 |
|
| 239 |
+
| 27.7078 | 183.9697 | 2944 | 6.0327 |
|
| 240 |
+
| 27.6902 | 184.9697 | 2960 | 6.0328 |
|
| 241 |
+
| 27.6065 | 185.9697 | 2976 | 6.0328 |
|
| 242 |
+
| 27.5061 | 186.9697 | 2992 | 6.0327 |
|
| 243 |
+
| 27.7366 | 187.4848 | 3000 | 6.0327 |
|
| 244 |
+
|
| 245 |
+
|
| 246 |
+
### Framework versions
|
| 247 |
+
|
| 248 |
+
- Transformers 4.49.0
|
| 249 |
+
- Pytorch 2.4.0+cu121
|
| 250 |
+
- Datasets 3.4.0
|
| 251 |
+
- Tokenizers 0.21.0
|
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": "4.49.0"
|
| 6 |
+
}
|