llama-2-7b-hf-zero-shot-prompt
This model is a fine-tuned version of meta-llama/Llama-2-7b-hf on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.3135
{'accuracy': 0.546, 'recall': 0.098, 'precision': 0.9423076923076923, 'f1': 0.17753623188405795, 'total_time_in_seconds': 2308.70937472, 'samples_per_second': 0.4331424348815146, 'latency_in_seconds': 2.3087093747200003}
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: 2
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 100
- total_train_batch_size: 200
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- lr_scheduler_warmup_ratio: 0.03
- num_epochs: 1
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| No log | 0.22 | 10 | 1.3991 |
| No log | 0.44 | 20 | 1.3609 |
| No log | 0.67 | 30 | 1.3327 |
| 1.4726 | 0.89 | 40 | 1.3135 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.0.1+cu118
- Datasets 2.14.4
- Tokenizers 0.13.3