metadata
tags:
- generated_from_trainer
model-index:
- name: Llama-360M
results: []
Llama-360M
This model is a fine-tuned version of on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 3.4949
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: 0.0003
- train_batch_size: 16
- eval_batch_size: 8
- 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: 50
- num_epochs: 15
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 8.4028 | 1.0 | 3 | 8.2019 |
| 7.206 | 2.0 | 6 | 7.1714 |
| 6.3044 | 3.0 | 9 | 6.4447 |
| 5.835 | 4.0 | 12 | 6.0518 |
| 5.2116 | 5.0 | 15 | 5.3661 |
| 4.5014 | 6.0 | 18 | 4.9977 |
| 4.0994 | 7.0 | 21 | 4.6291 |
| 3.8803 | 8.0 | 24 | 4.2823 |
| 3.6287 | 9.0 | 27 | 4.1548 |
| 3.3333 | 10.0 | 30 | 3.8924 |
| 3.016 | 11.0 | 33 | 3.6889 |
| 2.841 | 12.0 | 36 | 3.5575 |
| 2.4063 | 13.0 | 39 | 3.5160 |
| 2.324 | 14.0 | 42 | 3.5069 |
| 1.8726 | 15.0 | 45 | 3.4949 |
Framework versions
- Transformers 4.39.1
- Pytorch 2.1.2+cu121
- Datasets 2.16.1
- Tokenizers 0.15.0