ede1f5ef-06a2-4c49-8fac-381798dd546d

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: 2.3694

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: 0.000203
• train_batch_size: 4
• eval_batch_size: 4
• seed: 30
• 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	0.0003	1	4.1918
2.5406	0.0139	50	2.8331
2.3475	0.0279	100	2.9480
2.417	0.0418	150	2.6699
2.3417	0.0558	200	2.6826
2.5255	0.0697	250	2.5920
2.4284	0.0837	300	2.5731
2.3366	0.0976	350	2.5241
2.2929	0.1115	400	2.4026
2.5735	0.1255	450	2.3703
2.3215	0.1394	500	2.3694

Framework versions

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