mistralai_compi_data

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

• Loss: 0.4322
• Accuracy: 0.8761
• Precision: 0.8564
• Recall: 0.8995
• F1 score: 0.8774

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1 score
1.3663	0.2604	200	1.1150	0.7692	0.8851	0.6111	0.7230
0.8905	0.5208	400	0.6190	0.8318	0.8487	0.8016	0.8245
0.7007	0.7812	600	0.5719	0.8462	0.9140	0.7593	0.8295
0.5215	1.0417	800	0.4315	0.8696	0.8819	0.8492	0.8652
0.2955	1.3021	1000	0.6376	0.8592	0.8140	0.9259	0.8663
0.3254	1.5625	1200	0.5735	0.8735	0.9607	0.7751	0.8580
0.3075	1.8229	1400	0.4322	0.8761	0.8564	0.8995	0.8774

Framework versions

• PEFT 0.11.1
• Transformers 4.44.2
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1