metadata
license: apache-2.0
tags:
- generated_from_trainer
model-index:
- name: distilroberta-base-finetuned-wikitext2
results: []
distilroberta-base-finetuned-wikitext2
This model is a fine-tuned version of distilroberta-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0047
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 80.0
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.079 | 1.0 | 24 | 0.0748 |
| 0.045 | 2.0 | 48 | 0.0507 |
| 0.0253 | 3.0 | 72 | 0.0494 |
| 0.0285 | 4.0 | 96 | 0.0469 |
| 0.0189 | 5.0 | 120 | 0.0319 |
| 0.0014 | 6.0 | 144 | 0.0220 |
| 0.0023 | 7.0 | 168 | 0.0108 |
| 0.0012 | 8.0 | 192 | 0.0079 |
| 0.0067 | 9.0 | 216 | 0.0061 |
| 0.0006 | 10.0 | 240 | 0.0099 |
| 0.0004 | 11.0 | 264 | 0.0067 |
| 0.0007 | 12.0 | 288 | 0.0060 |
| 0.0005 | 13.0 | 312 | 0.0050 |
| 0.0005 | 14.0 | 336 | 0.0050 |
| 0.0005 | 15.0 | 360 | 0.0046 |
| 0.0001 | 16.0 | 384 | 0.0052 |
| 0.0001 | 17.0 | 408 | 0.0047 |
| 0.0001 | 18.0 | 432 | 0.0046 |
| 0.0001 | 19.0 | 456 | 0.0050 |
| 0.0001 | 20.0 | 480 | 0.0046 |
| 0.0001 | 21.0 | 504 | 0.0046 |
| 0.0007 | 22.0 | 528 | 0.0046 |
| 0.0001 | 23.0 | 552 | 0.0046 |
| 0.0001 | 24.0 | 576 | 0.0049 |
| 0.0001 | 25.0 | 600 | 0.0043 |
| 0.0001 | 26.0 | 624 | 0.0046 |
| 0.0 | 27.0 | 648 | 0.0044 |
| 0.0001 | 28.0 | 672 | 0.0045 |
| 0.0001 | 29.0 | 696 | 0.0045 |
| 0.0002 | 30.0 | 720 | 0.0044 |
| 0.0 | 31.0 | 744 | 0.0044 |
| 0.0001 | 32.0 | 768 | 0.0044 |
| 0.0 | 33.0 | 792 | 0.0044 |
| 0.0001 | 34.0 | 816 | 0.0050 |
| 0.0001 | 35.0 | 840 | 0.0050 |
| 0.0002 | 36.0 | 864 | 0.0049 |
| 0.0 | 37.0 | 888 | 0.0048 |
| 0.0 | 38.0 | 912 | 0.0054 |
| 0.0 | 39.0 | 936 | 0.0048 |
| 0.0 | 40.0 | 960 | 0.0047 |
| 0.0002 | 41.0 | 984 | 0.0048 |
| 0.0 | 42.0 | 1008 | 0.0068 |
| 0.0 | 43.0 | 1032 | 0.0051 |
| 0.0002 | 44.0 | 1056 | 0.0049 |
| 0.0 | 45.0 | 1080 | 0.0049 |
| 0.0 | 46.0 | 1104 | 0.0048 |
| 0.0 | 47.0 | 1128 | 0.0046 |
| 0.0 | 48.0 | 1152 | 0.0046 |
| 0.0 | 49.0 | 1176 | 0.0048 |
| 0.0 | 50.0 | 1200 | 0.0049 |
| 0.0 | 51.0 | 1224 | 0.0047 |
| 0.0 | 52.0 | 1248 | 0.0046 |
| 0.0001 | 53.0 | 1272 | 0.0046 |
| 0.0 | 54.0 | 1296 | 0.0045 |
| 0.0 | 55.0 | 1320 | 0.0045 |
| 0.0 | 56.0 | 1344 | 0.0046 |
| 0.0 | 57.0 | 1368 | 0.0046 |
| 0.0 | 58.0 | 1392 | 0.0046 |
| 0.0 | 59.0 | 1416 | 0.0046 |
| 0.0 | 60.0 | 1440 | 0.0046 |
| 0.0 | 61.0 | 1464 | 0.0047 |
| 0.0 | 62.0 | 1488 | 0.0047 |
| 0.0 | 63.0 | 1512 | 0.0047 |
| 0.0 | 64.0 | 1536 | 0.0046 |
| 0.0 | 65.0 | 1560 | 0.0045 |
| 0.0 | 66.0 | 1584 | 0.0045 |
| 0.0 | 67.0 | 1608 | 0.0046 |
| 0.0 | 68.0 | 1632 | 0.0047 |
| 0.0 | 69.0 | 1656 | 0.0048 |
| 0.0001 | 70.0 | 1680 | 0.0048 |
| 0.0 | 71.0 | 1704 | 0.0048 |
| 0.0 | 72.0 | 1728 | 0.0047 |
| 0.0001 | 73.0 | 1752 | 0.0049 |
| 0.0 | 74.0 | 1776 | 0.0048 |
| 0.0 | 75.0 | 1800 | 0.0048 |
| 0.0 | 76.0 | 1824 | 0.0047 |
| 0.0 | 77.0 | 1848 | 0.0047 |
| 0.0 | 78.0 | 1872 | 0.0047 |
| 0.0 | 79.0 | 1896 | 0.0047 |
| 0.0 | 80.0 | 1920 | 0.0047 |
Framework versions
- Transformers 4.14.1
- Pytorch 1.10.0+cu111
- Datasets 1.17.0
- Tokenizers 0.10.3