llama-3-8b-hf-ft-chat-lora_1

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

• Loss: 2.9295

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: 32
• eval_batch_size: 32
• seed: 42
• distributed_type: multi-GPU
• gradient_accumulation_steps: 2
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.3
• num_epochs: 8

Training results

Training Loss	Epoch	Step	Validation Loss
3.8283	0.9851	33	3.4186
2.711	2.0	67	2.5398
2.2647	2.9851	100	2.3353
2.0786	4.0	134	2.5266
1.9782	4.9851	167	2.9430
1.9104	6.0	201	2.9388
1.9121	6.9851	234	2.9504
1.9022	7.8806	264	2.9295

Framework versions

• PEFT 0.10.0
• Transformers 4.40.0
• Pytorch 2.2.2+cu121
• Datasets 2.18.0
• Tokenizers 0.19.1