<!-- 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. -->
find_last_sent_train_30_eval_10_flan-t5-xl
This model is a fine-tuned version of google/flan-t5-xl on the tyzhu/find_last_sent_train_30_eval_10 dataset. It achieves the following results on the evaluation set:
- Loss: 7.9082
- Bleu: 0.6495
- Gen Len: 36.8
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: 0.0001
- 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
- lr_scheduler_warmup_ratio: 0.05
- num_epochs: 100.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|---|---|---|---|---|---|
| No log | 1.0 | 5 | 2.9228 | 0.6762 | 27.7 |
| No log | 2.0 | 10 | 2.9141 | 0.771 | 32.3 |
| No log | 3.0 | 15 | 2.9031 | 0.427 | 54.1 |
| No log | 4.0 | 20 | 2.8934 | 1.2063 | 58.3 |
| No log | 5.0 | 25 | 2.9004 | 0.5254 | 54.1 |
| No log | 6.0 | 30 | 2.9432 | 0.9021 | 32.1 |
| No log | 7.0 | 35 | 2.9665 | 0.8921 | 37.9 |
| No log | 8.0 | 40 | 3.1666 | 1.0159 | 34.0 |
| No log | 9.0 | 45 | 3.3443 | 0.8875 | 33.2 |
| 2.1075 | 10.0 | 50 | 3.5265 | 0.8168 | 35.5 |
| 2.1075 | 11.0 | 55 | 3.7898 | 0.8206 | 35.4 |
| 2.1075 | 12.0 | 60 | 3.8993 | 0.6027 | 30.7 |
| 2.1075 | 13.0 | 65 | 4.2452 | 0.5694 | 29.1 |
| 2.1075 | 14.0 | 70 | 5.0974 | 0.3322 | 27.6 |
| 2.1075 | 15.0 | 75 | 4.8003 | 0.5858 | 29.8 |
| 2.1075 | 16.0 | 80 | 5.5745 | 0.6828 | 28.0 |
| 2.1075 | 17.0 | 85 | 6.0379 | 0.5929 | 30.4 |
| 2.1075 | 18.0 | 90 | 5.8209 | 0.5884 | 32.0 |
| 2.1075 | 19.0 | 95 | 5.1542 | 0.7417 | 34.6 |
| 0.4025 | 20.0 | 100 | 5.9180 | 0.9405 | 36.3 |
| 0.4025 | 21.0 | 105 | 6.3584 | 0.6909 | 29.4 |
| 0.4025 | 22.0 | 110 | 6.4206 | 0.3296 | 28.6 |
| 0.4025 | 23.0 | 115 | 6.1093 | 0.6616 | 26.5 |
| 0.4025 | 24.0 | 120 | 6.3805 | 0.5429 | 24.2 |
| 0.4025 | 25.0 | 125 | 6.4573 | 0.5694 | 29.2 |
| 0.4025 | 26.0 | 130 | 6.3336 | 0.7661 | 31.6 |
| 0.4025 | 27.0 | 135 | 6.0298 | 0.7754 | 32.5 |
| 0.4025 | 28.0 | 140 | 6.3929 | 0.7665 | 34.2 |
| 0.4025 | 29.0 | 145 | 6.7979 | 0.837 | 33.5 |
| 0.0782 | 30.0 | 150 | 6.7552 | 0.5659 | 33.5 |
| 0.0782 | 31.0 | 155 | 6.7309 | 0.5776 | 32.9 |
| 0.0782 | 32.0 | 160 | 6.8651 | 0.5627 | 34.3 |
| 0.0782 | 33.0 | 165 | 6.8978 | 0.6872 | 34.0 |
| 0.0782 | 34.0 | 170 | 6.7130 | 0.6367 | 38.6 |
| 0.0782 | 35.0 | 175 | 6.7089 | 0.6996 | 35.0 |
| 0.0782 | 36.0 | 180 | 6.9837 | 0.5602 | 34.3 |
| 0.0782 | 37.0 | 185 | 7.1842 | 0.5651 | 29.8 |
| 0.0782 | 38.0 | 190 | 7.1509 | 0.5703 | 33.3 |
| 0.0782 | 39.0 | 195 | 6.8741 | 0.599 | 33.6 |
| 0.0368 | 40.0 | 200 | 6.5819 | 0.6311 | 37.9 |
| 0.0368 | 41.0 | 205 | 6.6101 | 0.5779 | 33.5 |
| 0.0368 | 42.0 | 210 | 6.8818 | 0.5388 | 36.5 |
| 0.0368 | 43.0 | 215 | 7.1279 | 0.5782 | 32.0 |
| 0.0368 | 44.0 | 220 | 7.2446 | 0.5591 | 35.4 |
| 0.0368 | 45.0 | 225 | 7.1421 | 0.5819 | 33.2 |
| 0.0368 | 46.0 | 230 | 7.1707 | 0.6098 | 39.0 |
| 0.0368 | 47.0 | 235 | 7.1538 | 0.6992 | 41.1 |
| 0.0368 | 48.0 | 240 | 7.1356 | 0.7386 | 42.0 |
| 0.0368 | 49.0 | 245 | 7.1648 | 0.6266 | 36.2 |
| 0.0217 | 50.0 | 250 | 7.2529 | 0.6712 | 34.8 |
| 0.0217 | 51.0 | 255 | 7.3019 | 0.6495 | 35.7 |
| 0.0217 | 52.0 | 260 | 7.3839 | 0.5188 | 37.8 |
| 0.0217 | 53.0 | 265 | 7.4722 | 0.6241 | 37.9 |
| 0.0217 | 54.0 | 270 | 7.4558 | 0.658 | 36.2 |
| 0.0217 | 55.0 | 275 | 7.4083 | 0.6241 | 37.9 |
| 0.0217 | 56.0 | 280 | 7.4767 | 0.7167 | 37.1 |
| 0.0217 | 57.0 | 285 | 7.5763 | 0.7483 | 38.3 |
| 0.0217 | 58.0 | 290 | 7.5567 | 0.6835 | 38.7 |
| 0.0217 | 59.0 | 295 | 7.5056 | 0.658 | 35.9 |
| 0.0137 | 60.0 | 300 | 7.5780 | 0.6241 | 37.7 |
| 0.0137 | 61.0 | 305 | 7.6473 | 0.5049 | 38.3 |
| 0.0137 | 62.0 | 310 | 7.6870 | 0.6472 | 36.1 |
| 0.0137 | 63.0 | 315 | 7.7570 | 0.6472 | 36.1 |
| 0.0137 | 64.0 | 320 | 7.7887 | 0.6472 | 36.1 |
| 0.0137 | 65.0 | 325 | 7.7978 | 0.604 | 38.6 |
| 0.0137 | 66.0 | 330 | 7.8055 | 0.6081 | 38.3 |
| 0.0137 | 67.0 | 335 | 7.8667 | 0.6556 | 36.2 |
| 0.0137 | 68.0 | 340 | 7.9032 | 0.5509 | 35.9 |
| 0.0137 | 69.0 | 345 | 7.9604 | 0.6527 | 36.0 |
| 0.0099 | 70.0 | 350 | 8.0594 | 0.5523 | 35.4 |
| 0.0099 | 71.0 | 355 | 8.1532 | 0.7865 | 35.7 |
| 0.0099 | 72.0 | 360 | 8.2159 | 0.7865 | 35.7 |
| 0.0099 | 73.0 | 365 | 8.2116 | 0.7865 | 35.7 |
| 0.0099 | 74.0 | 370 | 8.1614 | 0.7865 | 35.7 |
| 0.0099 | 75.0 | 375 | 8.0764 | 0.6566 | 35.8 |
| 0.0099 | 76.0 | 380 | 8.0349 | 0.6593 | 36.0 |
| 0.0099 | 77.0 | 385 | 8.0129 | 0.6593 | 36.0 |
| 0.0099 | 78.0 | 390 | 7.9710 | 0.6122 | 38.1 |
| 0.0099 | 79.0 | 395 | 7.9613 | 0.6122 | 37.9 |
| 0.0073 | 80.0 | 400 | 7.9674 | 0.604 | 38.6 |
| 0.0073 | 81.0 | 405 | 7.9657 | 0.604 | 38.6 |
| 0.0073 | 82.0 | 410 | 7.9488 | 0.5995 | 39.0 |
| 0.0073 | 83.0 | 415 | 7.9151 | 0.6472 | 36.1 |
| 0.0073 | 84.0 | 420 | 7.8830 | 0.6277 | 37.0 |
| 0.0073 | 85.0 | 425 | 7.8635 | 0.6979 | 38.1 |
| 0.0073 | 86.0 | 430 | 7.8352 | 0.6844 | 38.8 |
| 0.0073 | 87.0 | 435 | 7.8213 | 0.6844 | 38.8 |
| 0.0073 | 88.0 | 440 | 7.8176 | 0.6844 | 38.8 |
| 0.0073 | 89.0 | 445 | 7.8151 | 0.6844 | 38.8 |
| 0.0072 | 90.0 | 450 | 7.8302 | 0.6844 | 38.8 |
| 0.0072 | 91.0 | 455 | 7.8415 | 0.6844 | 38.8 |
| 0.0072 | 92.0 | 460 | 7.8535 | 0.6495 | 36.8 |
| 0.0072 | 93.0 | 465 | 7.8652 | 0.6495 | 36.8 |
| 0.0072 | 94.0 | 470 | 7.8692 | 0.6844 | 38.8 |
| 0.0072 | 95.0 | 475 | 7.8810 | 0.6844 | 38.8 |
| 0.0072 | 96.0 | 480 | 7.8900 | 0.6844 | 38.8 |
| 0.0072 | 97.0 | 485 | 7.8944 | 0.6495 | 36.8 |
| 0.0072 | 98.0 | 490 | 7.9016 | 0.6495 | 36.8 |
| 0.0072 | 99.0 | 495 | 7.9065 | 0.6495 | 36.8 |
| 0.0057 | 100.0 | 500 | 7.9082 | 0.6495 | 36.8 |
Framework versions
- Transformers 4.34.0
- Pytorch 2.1.0+cu121
- Datasets 2.14.5
- Tokenizers 0.14.1
