metadata
tags:
- generated_from_trainer
model-index:
- name: myBit-Llama2-jp-127M-test-10
results: []
myBit-Llama2-jp-127M-test-10
This model is a fine-tuned version of on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 4.5027
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: 0.00024
- 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 |
|---|---|---|---|
| 9.1233 | 0.07 | 200 | 7.9284 |
| 7.0718 | 0.15 | 400 | 6.5191 |
| 6.0222 | 0.22 | 600 | 5.5007 |
| 5.3898 | 0.29 | 800 | 5.2505 |
| 5.3181 | 0.37 | 1000 | 5.1436 |
| 5.0295 | 0.44 | 1200 | 4.9255 |
| 4.8843 | 0.52 | 1400 | 4.8118 |
| 4.7822 | 0.59 | 1600 | 4.7350 |
| 4.7075 | 0.66 | 1800 | 4.6671 |
| 4.644 | 0.74 | 2000 | 4.6128 |
| 4.593 | 0.81 | 2200 | 4.5621 |
| 4.5478 | 0.88 | 2400 | 4.5245 |
| 4.5167 | 0.96 | 2600 | 4.5027 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2