<!-- 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. -->
mystv0_agg
This model is a fine-tuned version of gpt2 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 2.0722
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: 0.0005
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- distributed_type: multi-GPU
- gradient_accumulation_steps: 8
- total_train_batch_size: 256
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 1000
- num_epochs: 100
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
1.0898 | 3.55 | 1000 | 1.3242 |
0.6944 | 7.1 | 2000 | 1.4106 |
0.6876 | 10.64 | 3000 | 1.3813 |
0.6856 | 14.19 | 4000 | 1.4327 |
0.685 | 17.74 | 5000 | 1.3641 |
0.6826 | 21.29 | 6000 | 1.4222 |
0.6808 | 24.83 | 7000 | 1.3972 |
0.6811 | 28.38 | 8000 | 1.3969 |
0.6757 | 31.93 | 9000 | 1.4670 |
0.6723 | 35.48 | 10000 | 1.4983 |
0.6668 | 39.02 | 11000 | 1.5150 |
0.6611 | 42.57 | 12000 | 1.5096 |
0.6524 | 46.12 | 13000 | 1.5601 |
0.642 | 49.67 | 14000 | 1.6121 |
0.6287 | 53.22 | 15000 | 1.6332 |
0.6129 | 56.76 | 16000 | 1.6489 |
0.5929 | 60.31 | 17000 | 1.7623 |
0.5705 | 63.86 | 18000 | 1.7553 |
0.5455 | 67.41 | 19000 | 1.8321 |
0.5223 | 70.95 | 20000 | 1.9012 |
0.498 | 74.5 | 21000 | 1.9379 |
0.4788 | 78.05 | 22000 | 1.9693 |
0.461 | 81.6 | 23000 | 2.0177 |
0.4482 | 85.14 | 24000 | 2.0362 |
0.4388 | 88.69 | 25000 | 2.0570 |
0.4327 | 92.24 | 26000 | 2.0703 |
0.4293 | 95.79 | 27000 | 2.0719 |
0.4278 | 99.33 | 28000 | 2.0722 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.1.0+cu121
- Datasets 2.13.1
- Tokenizers 0.13.3