gpt_trinity_2_4_3e-5_lp5_nb5
This model is a fine-tuned version of skt/kogpt2-base-v2 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 4.0291
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: 3e-05
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 4.0
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
3.5765 | 0.05 | 1000 | 4.1247 |
3.19 | 0.09 | 2000 | 4.0578 |
3.1177 | 0.14 | 3000 | 4.0708 |
3.1116 | 0.19 | 4000 | 4.0654 |
3.0777 | 0.24 | 5000 | 4.0857 |
3.1105 | 0.28 | 6000 | 4.1127 |
3.1018 | 0.33 | 7000 | 4.1410 |
3.0728 | 0.38 | 8000 | 4.1834 |
3.1248 | 0.42 | 9000 | 4.2058 |
3.1035 | 0.47 | 10000 | 4.2048 |
3.0943 | 0.52 | 11000 | 4.1892 |
3.0724 | 0.57 | 12000 | 4.2063 |
3.0517 | 0.61 | 13000 | 4.1923 |
3.0372 | 0.66 | 14000 | 4.2112 |
3.0235 | 0.71 | 15000 | 4.2043 |
3.0329 | 0.76 | 16000 | 4.1630 |
3.0171 | 0.8 | 17000 | 4.1631 |
2.9997 | 0.85 | 18000 | 4.1563 |
2.9913 | 0.9 | 19000 | 4.1616 |
2.9579 | 0.94 | 20000 | 4.1494 |
2.9576 | 0.99 | 21000 | 4.1367 |
2.7461 | 1.04 | 22000 | 4.1593 |
2.7637 | 1.09 | 23000 | 4.1453 |
2.741 | 1.13 | 24000 | 4.1624 |
2.7514 | 1.18 | 25000 | 4.1357 |
2.755 | 1.23 | 26000 | 4.1524 |
2.7365 | 1.27 | 27000 | 4.1399 |
2.7356 | 1.32 | 28000 | 4.1285 |
2.7386 | 1.37 | 29000 | 4.1286 |
2.7489 | 1.42 | 30000 | 4.1231 |
2.7518 | 1.46 | 31000 | 4.1104 |
2.7317 | 1.51 | 32000 | 4.1202 |
2.7378 | 1.56 | 33000 | 4.1132 |
2.7309 | 1.6 | 34000 | 4.1047 |
2.7791 | 1.65 | 35000 | 4.0976 |
2.7427 | 1.7 | 36000 | 4.0874 |
2.7184 | 1.75 | 37000 | 4.0953 |
2.7107 | 1.79 | 38000 | 4.0963 |
2.7122 | 1.84 | 39000 | 4.0841 |
2.7172 | 1.89 | 40000 | 4.0852 |
2.7126 | 1.94 | 41000 | 4.0632 |
2.7063 | 1.98 | 42000 | 4.0643 |
2.5311 | 2.03 | 43000 | 4.0848 |
2.4496 | 2.08 | 44000 | 4.0943 |
2.4597 | 2.12 | 45000 | 4.0799 |
2.4472 | 2.17 | 46000 | 4.0802 |
2.4628 | 2.22 | 47000 | 4.0880 |
2.4508 | 2.27 | 48000 | 4.0791 |
2.4743 | 2.31 | 49000 | 4.0765 |
2.4692 | 2.36 | 50000 | 4.0739 |
2.4651 | 2.41 | 51000 | 4.0690 |
2.4885 | 2.45 | 52000 | 4.0723 |
2.5023 | 2.5 | 53000 | 4.0675 |
2.4651 | 2.55 | 54000 | 4.0649 |
2.4774 | 2.6 | 55000 | 4.0695 |
2.4717 | 2.64 | 56000 | 4.0559 |
2.4856 | 2.69 | 57000 | 4.0512 |
2.4572 | 2.74 | 58000 | 4.0473 |
2.486 | 2.79 | 59000 | 4.0438 |
2.449 | 2.83 | 60000 | 4.0385 |
2.456 | 2.88 | 61000 | 4.0355 |
2.4802 | 2.93 | 62000 | 4.0378 |
2.4635 | 2.97 | 63000 | 4.0308 |
2.3742 | 3.02 | 64000 | 4.0488 |
2.2371 | 3.07 | 65000 | 4.0579 |
2.2496 | 3.12 | 66000 | 4.0630 |
2.2758 | 3.16 | 67000 | 4.0516 |
2.2489 | 3.21 | 68000 | 4.0585 |
2.2374 | 3.26 | 69000 | 4.0715 |
2.2862 | 3.3 | 70000 | 4.0507 |
2.2502 | 3.35 | 71000 | 4.0512 |
2.238 | 3.4 | 72000 | 4.0545 |
2.2407 | 3.45 | 73000 | 4.0459 |
2.2529 | 3.49 | 74000 | 4.0452 |
2.2453 | 3.54 | 75000 | 4.0459 |
2.2314 | 3.59 | 76000 | 4.0416 |
2.2408 | 3.63 | 77000 | 4.0379 |
2.2497 | 3.68 | 78000 | 4.0348 |
2.2475 | 3.73 | 79000 | 4.0374 |
2.2376 | 3.78 | 80000 | 4.0319 |
2.244 | 3.82 | 81000 | 4.0331 |
2.2611 | 3.87 | 82000 | 4.0306 |
2.237 | 3.92 | 83000 | 4.0301 |
2.2337 | 3.97 | 84000 | 4.0291 |
Framework versions
- Transformers 4.25.1
- Pytorch 1.9.0+cu102
- Datasets 2.8.0
- Tokenizers 0.13.2
- Downloads last month
- 22
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.