mistral_7b_samantha

This model is a fine-tuned version of mistralai/Mistral-7B-Instruct-v0.1 on the samantha-data dataset. It achieves the following results on the evaluation set:

• Loss: 1.7573

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: 4
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: constant
• lr_scheduler_warmup_steps: 0.03
• training_steps: 100

Training results

Training Loss	Epoch	Step	Validation Loss
1.7235	0.4	10	1.8474
1.2154	0.8	20	1.7551
1.1009	1.2	30	1.6884
0.981	1.6	40	1.6751
0.9463	2.0	50	1.6704
0.8443	2.4	60	1.7248
0.8128	2.8	70	1.7101
0.8158	3.2	80	1.7826
0.7087	3.6	90	1.7805
0.7091	4.0	100	1.7573

Framework versions

• PEFT 0.8.2
• Transformers 4.38.1
• Pytorch 2.2.1+cu121
• Datasets 2.17.1
• Tokenizers 0.15.2