LLama3_deneme

This model is a fine-tuned version of meta-llama/Meta-Llama-3-8B on the emollms_ei_oc_mixed dataset. It achieves the following results on the evaluation set:

• Loss: 0.0802

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: 0.0003
• train_batch_size: 32
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 256
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• num_epochs: 3.0

Training results

Training Loss	Epoch	Step	Validation Loss
0.3319	0.3320	10	0.1265
0.113	0.6639	20	0.0951
0.0961	0.9959	30	0.0864
0.0908	1.3278	40	0.0838
0.0846	1.6598	50	0.0816
0.0806	1.9917	60	0.0802
0.0756	2.3237	70	0.0810
0.0751	2.6556	80	0.0805
0.0719	2.9876	90	0.0806

Framework versions

• PEFT 0.11.1
• Transformers 4.41.1
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1