llama2-7b-int4-dolly-15k-english-flash-attention2-w-packing

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

• Loss: 1.2201

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: 6
• eval_batch_size: 4
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• num_epochs: 3
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
1.2688	0.64	100	1.2310
1.1907	1.27	200	1.2219
1.178	1.91	300	1.2170
1.1368	2.55	400	1.2201

Framework versions

• PEFT 0.7.2.dev0
• Transformers 4.37.0.dev0
• Pytorch 2.1.0+cu121
• Datasets 2.16.2.dev0
• Tokenizers 0.15.0