sft-chatbot3

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

• Loss: 0.8532

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: 0.0002
• train_batch_size: 4
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• num_epochs: 8
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.9836	15	0.8413
No log	1.9672	30	0.7822
No log	2.9508	45	0.7713
No log	4.0	61	0.7803
No log	4.9836	76	0.8245
No log	5.9672	91	0.8427
No log	6.9508	106	0.8522
No log	7.8689	120	0.8532

Framework versions

• PEFT 0.13.1
• Transformers 4.43.3
• Pytorch 2.4.0+cu121
• Datasets 3.0.1
• Tokenizers 0.19.1