pythia2.8b-sft-tldr
This model is a fine-tuned version of EleutherAI/pythia-2.8b-deduped on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 2.2578
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: 4
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- gradient_accumulation_steps: 8
- total_train_batch_size: 128
- total_eval_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- num_epochs: 1
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 2.3714 | 0.2007 | 183 | 2.2949 |
| 2.2873 | 0.4013 | 366 | 2.2773 |
| 2.2732 | 0.6020 | 549 | 2.2656 |
| 2.2562 | 0.8026 | 732 | 2.2578 |
Framework versions
- Transformers 4.41.1
- Pytorch 2.2.1+cu121
- Datasets 2.19.0
- Tokenizers 0.19.1