jd0g/Mistral-7B-NLI-v0.1

mistral-7b-nli_cot

This model is a fine-tuned version of TheBloke/Mistral-7B-v0.1-GPTQ on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4930

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: 8
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 2
• num_epochs: 11
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.4947	0.9996	598	0.4534
0.4418	1.9992	1196	0.4475
0.4262	2.9987	1794	0.4476
0.4125	4.0	2393	0.4499
0.4015	4.9996	2991	0.4552
0.3908	5.9992	3589	0.4591
0.3809	6.9987	4187	0.4653
0.3712	8.0	4786	0.4721
0.3635	8.9996	5384	0.4783
0.3562	9.9992	5982	0.4868
0.3496	10.9954	6578	0.4930

Framework versions

• PEFT 0.10.0
• Transformers 4.40.1
• Pytorch 2.0.1+cu118
• Datasets 2.19.0
• Tokenizers 0.19.1