Llama2_AAID_structured_train

This model is a fine-tuned version of meta-llama/Llama-2-7b-hf on the AAID_structured dataset. It achieves the following results on the evaluation set:

• Loss: 0.4973

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: 32
• 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
1.1654	0.0109	10	0.5499
0.4462	0.0219	20	0.5070
0.3949	0.0328	30	0.4973
0.3903	0.0438	40	0.5445
0.3617	0.0547	50	0.5150
0.3346	0.0656	60	0.5359
0.3284	0.0766	70	0.5421
0.3215	0.0875	80	0.5326

Framework versions

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