4c5eb6db-b295-436b-b292-7cd809cdf1d2

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: 0.9371

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: 0.000209
• train_batch_size: 4
• eval_batch_size: 4
• seed: 90
• 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	3.6567
1.1842	0.0048	50	1.1520
1.0028	0.0096	100	1.1289
1.0557	0.0144	150	1.0833
0.9433	0.0192	200	1.0654
0.983	0.0240	250	1.0108
0.7648	0.0287	300	1.0014
0.8696	0.0335	350	0.9729
0.8448	0.0383	400	0.9482
0.8588	0.0431	450	0.9393
0.844	0.0479	500	0.9371

Framework versions

• PEFT 0.13.2
• Transformers 4.46.0
• Pytorch 2.5.0+cu124
• Datasets 3.0.1
• Tokenizers 0.20.1