gptj-6b-gpteacher-role-play-chatml
This model is a fine-tuned version of EleutherAI/gpt-j-6b on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.9268
- Accuracy: 0.1592
- Entropy: 1.9050
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: 1e-05
- train_batch_size: 2
- eval_batch_size: 8
- seed: 99
- distributed_type: multi-GPU
- gradient_accumulation_steps: 16
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.03
- num_epochs: 5.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Entropy |
|---|---|---|---|---|---|
| 2.0652 | 0.99 | 59 | 2.0918 | 0.1549 | 2.3687 |
| 2.083 | 2.0 | 119 | 2.0078 | 0.1581 | 2.1805 |
| 2.083 | 2.99 | 178 | 1.9658 | 0.1594 | 2.0026 |
| 1.9201 | 4.0 | 238 | 1.9424 | 0.1596 | 1.9390 |
| 1.9201 | 4.96 | 295 | 1.9268 | 0.1592 | 1.9050 |
Framework versions
- Transformers 4.28.1
- Pytorch 1.13.1
- Datasets 2.7.1
- Tokenizers 0.13.3