trainings

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

• Loss: 1.7917

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
• lr_scheduler_warmup_ratio: 0.05
• num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss
2.0904	0.4	22	2.0811
1.92	0.8	44	1.9022
1.7317	1.2	66	1.8018
1.7906	1.6	88	1.7928
1.7903	2.0	110	1.7917

Framework versions

• Transformers 4.32.1
• Pytorch 2.0.1+cu117
• Datasets 2.14.6
• Tokenizers 0.13.3