llama-sft-qat
This model is a fine-tuned version of meta-llama/Llama-2-7b-chat-hf on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.1851
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: 0.0002
- train_batch_size: 3
- eval_batch_size: 24
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 2.8727 | 1.0 | 4 | 1.9824 |
| 2.0991 | 2.0 | 8 | 1.3412 |
| 1.5585 | 3.0 | 12 | 0.7900 |
| 0.8816 | 4.0 | 16 | 0.5710 |
| 0.552 | 5.0 | 20 | 0.4502 |
| 0.552 | 6.0 | 24 | 0.3272 |
| 0.3661 | 7.0 | 28 | 0.2535 |
| 0.2903 | 8.0 | 32 | 0.2082 |
| 0.1619 | 9.0 | 36 | 0.1888 |
| 0.2003 | 10.0 | 40 | 0.1851 |
Framework versions
- PEFT 0.7.1
- Transformers 4.38.2
- Pytorch 2.1.0+cu118
- Datasets 2.18.0
- Tokenizers 0.15.2
- Downloads last month
- 6
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support
Model tree for chanchan7/llama-sft-qat
Base model
meta-llama/Llama-2-7b-chat-hf