sft-chatbot4

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: 0.8056

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: 0.0001
• 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	0.9782
No log	1.9672	30	0.8569
No log	2.9508	45	0.8229
No log	4.0	61	0.8063
No log	4.9836	76	0.8053
No log	5.9672	91	0.8061
No log	6.9508	106	0.8055
No log	7.8689	120	0.8056

Framework versions

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