myBit-Llama2-jp-127M-test-20
This model is a fine-tuned version of on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 3.4890
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: 0.0024
- train_batch_size: 96
- eval_batch_size: 96
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: polynomial
- lr_scheduler_warmup_steps: 500
- num_epochs: 1
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
6.9256 | 0.07 | 200 | 4.9236 |
4.6058 | 0.15 | 400 | 4.4149 |
4.3066 | 0.22 | 600 | 4.2187 |
4.126 | 0.29 | 800 | 4.0780 |
4.0117 | 0.37 | 1000 | 3.9963 |
3.9295 | 0.44 | 1200 | 3.9215 |
3.8601 | 0.51 | 1400 | 3.8477 |
3.7821 | 0.59 | 1600 | 3.7748 |
3.7065 | 0.66 | 1800 | 3.7078 |
3.6407 | 0.73 | 2000 | 3.6518 |
3.5957 | 0.8 | 2200 | 3.5923 |
3.5302 | 0.88 | 2400 | 3.5368 |
3.482 | 0.95 | 2600 | 3.4890 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2
- Downloads last month
- 14
Inference API (serverless) does not yet support model repos that contain custom code.