metadata
base_model: meta-llama/Llama-2-7b-hf
tags:
- generated_from_trainer
model-index:
- name: llama7b-protgen
results: []
llama7b-protgen
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: 6.7466
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: 5e-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- total_train_batch_size: 16
- total_eval_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 30
- training_steps: 2000
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 8.4347 | 0.05 | 100 | 8.3863 |
| 8.3045 | 0.1 | 200 | 8.2328 |
| 7.9602 | 0.15 | 300 | 7.9333 |
| 7.4512 | 0.2 | 400 | 7.6350 |
| 5.938 | 0.25 | 500 | 7.3528 |
| 4.9679 | 0.3 | 600 | 6.9324 |
| 3.9455 | 0.35 | 700 | 6.6334 |
| 2.8044 | 0.4 | 800 | 6.6159 |
| 2.071 | 0.45 | 900 | 6.5350 |
| 1.5644 | 0.5 | 1000 | 6.3780 |
| 1.2187 | 0.55 | 1100 | 6.3079 |
| 0.8425 | 0.6 | 1200 | 6.4287 |
| 0.6849 | 0.65 | 1300 | 6.3898 |
| 0.5457 | 0.7 | 1400 | 6.5978 |
| 0.4533 | 0.75 | 1500 | 6.4167 |
| 0.2974 | 0.8 | 1600 | 6.4443 |
| 0.2676 | 0.85 | 1700 | 6.3087 |
| 0.2395 | 0.9 | 1800 | 6.5467 |
| 0.2042 | 0.95 | 1900 | 6.4028 |
| 0.2042 | 1.0 | 2000 | 6.7466 |
Framework versions
- Transformers 4.37.0.dev0
- Pytorch 2.1.2+cu121
- Datasets 2.16.1
- Tokenizers 0.15.0