metadata
library_name: peft
license: llama3
base_model: elyza/Llama-3-ELYZA-JP-8B
tags:
- axolotl
- generated_from_trainer
model-index:
- name: 75415c41-3378-4a4f-bb99-bc7f331bb87a
results: []
75415c41-3378-4a4f-bb99-bc7f331bb87a
This model is a fine-tuned version of elyza/Llama-3-ELYZA-JP-8B on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.2875
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: 0.000202
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 8
- optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 50
- training_steps: 500
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| No log | 0.0001 | 1 | 2.1965 |
| 1.6766 | 0.0069 | 50 | 1.5489 |
| 1.5601 | 0.0138 | 100 | 1.4704 |
| 1.591 | 0.0207 | 150 | 1.4512 |
| 1.4707 | 0.0276 | 200 | 1.4301 |
| 1.5985 | 0.0345 | 250 | 1.4250 |
| 1.4879 | 0.0414 | 300 | 1.3671 |
| 1.4722 | 0.0483 | 350 | 1.3291 |
| 1.4644 | 0.0552 | 400 | 1.3032 |
| 1.4561 | 0.0621 | 450 | 1.2888 |
| 1.4312 | 0.0690 | 500 | 1.2875 |
Framework versions
- PEFT 0.13.2
- Transformers 4.46.0
- Pytorch 2.5.0+cu124
- Datasets 3.0.1
- Tokenizers 0.20.1