Llama-2-7b-chat-hf-ark
This model is a fine-tuned version of on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 4.0662
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: 0.0005
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 1000
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 7.825 | 1.0 | 411 | 5.9952 |
| 5.5992 | 2.0 | 822 | 4.6529 |
| 4.6435 | 3.0 | 1233 | 4.3127 |
| 4.1195 | 4.0 | 1645 | 4.1649 |
| 3.9117 | 5.0 | 2056 | 4.0878 |
| 3.7508 | 6.0 | 2467 | 4.0662 |
| 3.5867 | 7.0 | 2878 | 4.1062 |
| 3.1731 | 8.0 | 3290 | 4.2059 |
| 2.8522 | 9.0 | 3701 | 4.3372 |
| 2.6968 | 9.99 | 4110 | 4.3987 |
Framework versions
- Transformers 4.35.0
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1