<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
translation_flan_base_v8h
This model is a fine-tuned version of google/flan-t5-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0012
- Bleu: 53.1326
- Gen Len: 6.9542
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: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 94
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|---|---|---|---|---|---|
| No log | 1.0 | 141 | 0.2917 | 50.1917 | 7.0049 |
| No log | 2.0 | 282 | 0.2353 | 50.5231 | 6.9831 |
| No log | 3.0 | 423 | 0.1850 | 51.0364 | 6.9711 |
| 0.307 | 4.0 | 564 | 0.1571 | 51.437 | 6.9711 |
| 0.307 | 5.0 | 705 | 0.1352 | 51.7178 | 6.9676 |
| 0.307 | 6.0 | 846 | 0.1137 | 51.9056 | 6.9671 |
| 0.307 | 7.0 | 987 | 0.0971 | 52.0134 | 6.9662 |
| 0.1902 | 8.0 | 1128 | 0.0859 | 52.1744 | 6.9645 |
| 0.1902 | 9.0 | 1269 | 0.0787 | 52.2813 | 6.9667 |
| 0.1902 | 10.0 | 1410 | 0.0646 | 52.3576 | 6.9729 |
| 0.141 | 11.0 | 1551 | 0.0620 | 52.6463 | 6.9645 |
| 0.141 | 12.0 | 1692 | 0.0509 | 52.676 | 6.9605 |
| 0.141 | 13.0 | 1833 | 0.0450 | 52.7741 | 6.9627 |
| 0.141 | 14.0 | 1974 | 0.0394 | 52.9329 | 6.956 |
| 0.1077 | 15.0 | 2115 | 0.0372 | 52.9527 | 6.956 |
| 0.1077 | 16.0 | 2256 | 0.0322 | 52.9836 | 6.9569 |
| 0.1077 | 17.0 | 2397 | 0.0287 | 53.0264 | 6.9591 |
| 0.0865 | 18.0 | 2538 | 0.0254 | 53.0497 | 6.9529 |
| 0.0865 | 19.0 | 2679 | 0.0231 | 53.0146 | 6.9569 |
| 0.0865 | 20.0 | 2820 | 0.0200 | 53.1395 | 6.9649 |
| 0.0865 | 21.0 | 2961 | 0.0203 | 53.0735 | 6.9636 |
| 0.0679 | 22.0 | 3102 | 0.0173 | 53.1411 | 6.9627 |
| 0.0679 | 23.0 | 3243 | 0.0156 | 53.1516 | 6.9636 |
| 0.0679 | 24.0 | 3384 | 0.0130 | 53.1501 | 6.9649 |
| 0.0574 | 25.0 | 3525 | 0.0114 | 53.1532 | 6.9645 |
| 0.0574 | 26.0 | 3666 | 0.0103 | 53.1547 | 6.9649 |
| 0.0574 | 27.0 | 3807 | 0.0101 | 53.1562 | 6.9645 |
| 0.0574 | 28.0 | 3948 | 0.0082 | 53.1606 | 6.9631 |
| 0.0474 | 29.0 | 4089 | 0.0078 | 53.164 | 6.9622 |
| 0.0474 | 30.0 | 4230 | 0.0067 | 53.0396 | 6.9494 |
| 0.0474 | 31.0 | 4371 | 0.0057 | 53.0007 | 6.9516 |
| 0.0416 | 32.0 | 4512 | 0.0054 | 53.0457 | 6.9511 |
| 0.0416 | 33.0 | 4653 | 0.0056 | 53.0487 | 6.9507 |
| 0.0416 | 34.0 | 4794 | 0.0048 | 53.0472 | 6.952 |
| 0.0416 | 35.0 | 4935 | 0.0046 | 52.9924 | 6.9489 |
| 0.0346 | 36.0 | 5076 | 0.0043 | 52.8958 | 6.9516 |
| 0.0346 | 37.0 | 5217 | 0.0036 | 53.1215 | 6.9636 |
| 0.0346 | 38.0 | 5358 | 0.0037 | 52.8927 | 6.9507 |
| 0.0346 | 39.0 | 5499 | 0.0036 | 52.8958 | 6.9502 |
| 0.0299 | 40.0 | 5640 | 0.0033 | 53.0103 | 6.9502 |
| 0.0299 | 41.0 | 5781 | 0.0034 | 53.0941 | 6.964 |
| 0.0299 | 42.0 | 5922 | 0.0036 | 53.1746 | 6.964 |
| 0.026 | 43.0 | 6063 | 0.0032 | 53.0359 | 6.9489 |
| 0.026 | 44.0 | 6204 | 0.0029 | 52.9955 | 6.9498 |
| 0.026 | 45.0 | 6345 | 0.0028 | 53.0429 | 6.952 |
| 0.026 | 46.0 | 6486 | 0.0028 | 53.0818 | 6.9502 |
| 0.0218 | 47.0 | 6627 | 0.0025 | 53.1575 | 6.9511 |
| 0.0218 | 48.0 | 6768 | 0.0025 | 53.1156 | 6.9511 |
| 0.0218 | 49.0 | 6909 | 0.0022 | 53.1212 | 6.9676 |
| 0.0188 | 50.0 | 7050 | 0.0021 | 53.2262 | 6.9676 |
| 0.0188 | 51.0 | 7191 | 0.0021 | 53.0554 | 6.9529 |
| 0.0188 | 52.0 | 7332 | 0.0021 | 53.1229 | 6.9502 |
| 0.0188 | 53.0 | 7473 | 0.0020 | 53.0554 | 6.9529 |
| 0.0184 | 54.0 | 7614 | 0.0020 | 52.9592 | 6.9525 |
| 0.0184 | 55.0 | 7755 | 0.0018 | 53.1745 | 6.9542 |
| 0.0184 | 56.0 | 7896 | 0.0018 | 53.0922 | 6.952 |
| 0.0171 | 57.0 | 8037 | 0.0018 | 53.0937 | 6.9516 |
| 0.0171 | 58.0 | 8178 | 0.0018 | 52.9975 | 6.9511 |
| 0.0171 | 59.0 | 8319 | 0.0017 | 53.1004 | 6.9542 |
| 0.0171 | 60.0 | 8460 | 0.0017 | 53.0517 | 6.9516 |
| 0.0159 | 61.0 | 8601 | 0.0017 | 53.0517 | 6.9516 |
| 0.0159 | 62.0 | 8742 | 0.0018 | 53.0487 | 6.9498 |
| 0.0159 | 63.0 | 8883 | 0.0018 | 53.0502 | 6.9502 |
| 0.0156 | 64.0 | 9024 | 0.0017 | 53.1289 | 6.9511 |
| 0.0156 | 65.0 | 9165 | 0.0016 | 53.0517 | 6.9511 |
| 0.0156 | 66.0 | 9306 | 0.0016 | 53.1259 | 6.9516 |
| 0.0156 | 67.0 | 9447 | 0.0015 | 53.1259 | 6.9516 |
| 0.0128 | 68.0 | 9588 | 0.0015 | 53.1259 | 6.9511 |
| 0.0128 | 69.0 | 9729 | 0.0014 | 53.1709 | 6.9516 |
| 0.0128 | 70.0 | 9870 | 0.0015 | 53.1678 | 6.9511 |
| 0.0129 | 71.0 | 10011 | 0.0014 | 53.1289 | 6.9516 |
| 0.0129 | 72.0 | 10152 | 0.0014 | 53.1289 | 6.9516 |
| 0.0129 | 73.0 | 10293 | 0.0014 | 53.1709 | 6.9516 |
| 0.0129 | 74.0 | 10434 | 0.0014 | 53.1709 | 6.9516 |
| 0.012 | 75.0 | 10575 | 0.0013 | 53.1776 | 6.9542 |
| 0.012 | 76.0 | 10716 | 0.0013 | 53.1357 | 6.9542 |
| 0.012 | 77.0 | 10857 | 0.0013 | 53.1357 | 6.9542 |
| 0.012 | 78.0 | 10998 | 0.0014 | 53.1357 | 6.9542 |
| 0.0114 | 79.0 | 11139 | 0.0014 | 53.1326 | 6.9538 |
| 0.0114 | 80.0 | 11280 | 0.0014 | 53.1357 | 6.9542 |
| 0.0114 | 81.0 | 11421 | 0.0013 | 53.1289 | 6.9516 |
| 0.0111 | 82.0 | 11562 | 0.0013 | 53.1289 | 6.9516 |
| 0.0111 | 83.0 | 11703 | 0.0013 | 53.1709 | 6.9516 |
| 0.0111 | 84.0 | 11844 | 0.0013 | 53.1289 | 6.9516 |
| 0.0111 | 85.0 | 11985 | 0.0013 | 53.1289 | 6.9516 |
| 0.0108 | 86.0 | 12126 | 0.0012 | 53.1259 | 6.9516 |
| 0.0108 | 87.0 | 12267 | 0.0012 | 53.1326 | 6.9542 |
| 0.0108 | 88.0 | 12408 | 0.0012 | 53.1326 | 6.9542 |
| 0.011 | 89.0 | 12549 | 0.0012 | 53.1357 | 6.9542 |
| 0.011 | 90.0 | 12690 | 0.0012 | 53.1326 | 6.9542 |
| 0.011 | 91.0 | 12831 | 0.0012 | 53.1326 | 6.9542 |
| 0.011 | 92.0 | 12972 | 0.0012 | 53.1326 | 6.9542 |
| 0.0111 | 93.0 | 13113 | 0.0012 | 53.1357 | 6.9542 |
| 0.0111 | 94.0 | 13254 | 0.0012 | 53.1326 | 6.9542 |
Framework versions
- Transformers 4.34.1
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1
