Edit model card

distilroberta-base-finetuned-wikitextepoch_150

This model is a fine-tuned version of distilroberta-base on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 1.8929

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: 150

Training results

Training Loss Epoch Step Validation Loss
2.2428 1.0 1121 2.0500
2.1209 2.0 2242 1.9996
2.0665 3.0 3363 1.9501
2.0179 4.0 4484 1.9311
1.9759 5.0 5605 1.9255
1.9089 6.0 6726 1.8805
1.9143 7.0 7847 1.8715
1.8744 8.0 8968 1.8671
1.858 9.0 10089 1.8592
1.8141 10.0 11210 1.8578
1.7917 11.0 12331 1.8574
1.7752 12.0 13452 1.8423
1.7722 13.0 14573 1.8287
1.7354 14.0 15694 1.8396
1.7217 15.0 16815 1.8244
1.6968 16.0 17936 1.8278
1.659 17.0 19057 1.8412
1.6442 18.0 20178 1.8328
1.6441 19.0 21299 1.8460
1.6267 20.0 22420 1.8343
1.612 21.0 23541 1.8249
1.5963 22.0 24662 1.8253
1.6101 23.0 25783 1.7843
1.5747 24.0 26904 1.8047
1.5559 25.0 28025 1.8618
1.5484 26.0 29146 1.8660
1.5411 27.0 30267 1.8318
1.5247 28.0 31388 1.8216
1.5278 29.0 32509 1.8075
1.4954 30.0 33630 1.8073
1.4863 31.0 34751 1.7958
1.4821 32.0 35872 1.8080
1.4357 33.0 36993 1.8373
1.4602 34.0 38114 1.8199
1.447 35.0 39235 1.8325
1.4292 36.0 40356 1.8075
1.4174 37.0 41477 1.8168
1.4103 38.0 42598 1.8095
1.4168 39.0 43719 1.8233
1.4005 40.0 44840 1.8388
1.3799 41.0 45961 1.8235
1.3657 42.0 47082 1.8298
1.3559 43.0 48203 1.8165
1.3723 44.0 49324 1.8059
1.3535 45.0 50445 1.8451
1.3533 46.0 51566 1.8458
1.3469 47.0 52687 1.8237
1.3247 48.0 53808 1.8264
1.3142 49.0 54929 1.8209
1.2958 50.0 56050 1.8244
1.293 51.0 57171 1.8311
1.2784 52.0 58292 1.8287
1.2731 53.0 59413 1.8600
1.2961 54.0 60534 1.8086
1.2739 55.0 61655 1.8303
1.2716 56.0 62776 1.8214
1.2459 57.0 63897 1.8440
1.2492 58.0 65018 1.8503
1.2393 59.0 66139 1.8316
1.2077 60.0 67260 1.8283
1.2426 61.0 68381 1.8413
1.2032 62.0 69502 1.8461
1.2123 63.0 70623 1.8469
1.2069 64.0 71744 1.8478
1.198 65.0 72865 1.8479
1.1972 66.0 73986 1.8516
1.1885 67.0 75107 1.8341
1.1784 68.0 76228 1.8322
1.1866 69.0 77349 1.8559
1.1648 70.0 78470 1.8758
1.1595 71.0 79591 1.8684
1.1661 72.0 80712 1.8553
1.1478 73.0 81833 1.8658
1.1488 74.0 82954 1.8452
1.1538 75.0 84075 1.8505
1.1267 76.0 85196 1.8430
1.1339 77.0 86317 1.8333
1.118 78.0 87438 1.8419
1.12 79.0 88559 1.8669
1.1144 80.0 89680 1.8647
1.104 81.0 90801 1.8643
1.0864 82.0 91922 1.8528
1.0863 83.0 93043 1.8456
1.0912 84.0 94164 1.8509
1.0873 85.0 95285 1.8690
1.0862 86.0 96406 1.8577
1.0879 87.0 97527 1.8612
1.0783 88.0 98648 1.8410
1.0618 89.0 99769 1.8517
1.0552 90.0 100890 1.8459
1.0516 91.0 102011 1.8723
1.0424 92.0 103132 1.8832
1.0478 93.0 104253 1.8922
1.0523 94.0 105374 1.8753
1.027 95.0 106495 1.8625
1.0364 96.0 107616 1.8673
1.0203 97.0 108737 1.8806
1.0309 98.0 109858 1.8644
1.0174 99.0 110979 1.8659
1.0184 100.0 112100 1.8590
1.0234 101.0 113221 1.8614
1.013 102.0 114342 1.8866
1.0092 103.0 115463 1.8770
1.0051 104.0 116584 1.8445
1.0105 105.0 117705 1.8512
1.0233 106.0 118826 1.8896
0.9967 107.0 119947 1.8687
0.9795 108.0 121068 1.8618
0.9846 109.0 122189 1.8877
0.9958 110.0 123310 1.8522
0.9689 111.0 124431 1.8765
0.9879 112.0 125552 1.8692
0.99 113.0 126673 1.8689
0.9798 114.0 127794 1.8898
0.9676 115.0 128915 1.8782
0.9759 116.0 130036 1.8840
0.9576 117.0 131157 1.8662
0.9637 118.0 132278 1.8984
0.9645 119.0 133399 1.8872
0.9793 120.0 134520 1.8705
0.9643 121.0 135641 1.9036
0.961 122.0 136762 1.8683
0.9496 123.0 137883 1.8785
0.946 124.0 139004 1.8912
0.9681 125.0 140125 1.8837
0.9403 126.0 141246 1.8824
0.9452 127.0 142367 1.8824
0.9437 128.0 143488 1.8665
0.945 129.0 144609 1.8655
0.9453 130.0 145730 1.8695
0.9238 131.0 146851 1.8697
0.9176 132.0 147972 1.8618
0.9405 133.0 149093 1.8679
0.9184 134.0 150214 1.9025
0.9298 135.0 151335 1.9045
0.9215 136.0 152456 1.9014
0.9249 137.0 153577 1.8505
0.9246 138.0 154698 1.8542
0.9205 139.0 155819 1.8731
0.9368 140.0 156940 1.8673
0.9251 141.0 158061 1.8835
0.9224 142.0 159182 1.8727
0.9326 143.0 160303 1.8380
0.916 144.0 161424 1.8857
0.9361 145.0 162545 1.8547
0.9121 146.0 163666 1.8587
0.9156 147.0 164787 1.8863
0.9131 148.0 165908 1.8809
0.9185 149.0 167029 1.8734
0.9183 150.0 168150 1.8929

Framework versions

  • Transformers 4.21.0
  • Pytorch 1.5.0
  • Datasets 2.4.0
  • Tokenizers 0.12.1
Downloads last month
15