pythia-70m-deduped-databricks-dolly-15k-v1
This model is a fine-tuned version of EleutherAI/pythia-70m-deduped on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 5.3362
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: 0.01
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 20
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 22.5289 | 1.0 | 1501 | 10.2721 |
| 8.5659 | 2.0 | 3002 | 6.0637 |
| 6.1434 | 3.0 | 4503 | 5.6654 |
| 5.735 | 4.0 | 6004 | 5.5859 |
| 5.5524 | 5.0 | 7505 | 5.5399 |
| 5.4368 | 6.0 | 9006 | 5.5221 |
| 5.3728 | 7.0 | 10507 | 5.5221 |
| 5.2796 | 8.0 | 12008 | 5.5022 |
| 5.1963 | 9.0 | 13509 | 5.4812 |
| 5.1361 | 10.0 | 15010 | 5.4652 |
| 5.0767 | 11.0 | 16511 | 5.4530 |
| 5.014 | 12.0 | 18012 | 5.4161 |
| 4.9554 | 13.0 | 19513 | 5.3732 |
| 4.9044 | 14.0 | 21014 | 5.3763 |
| 4.8483 | 15.0 | 22515 | 5.3694 |
| 4.7916 | 16.0 | 24016 | 5.3432 |
| 4.7383 | 17.0 | 25517 | 5.3308 |
| 4.675 | 18.0 | 27018 | 5.3203 |
| 4.6223 | 19.0 | 28519 | 5.3429 |
| 4.5733 | 20.0 | 30020 | 5.3362 |
Framework versions
- Transformers 4.33.2
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.13.3