deb101/mistral-7b-instruct-v0.3-mimic4-adapt-multilabel-classify

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

mistral-7b-instruct-v0.3-mimic4-adapt-multilabel-classify

This model is a fine-tuned version of mistralai/Mistral-7B-Instruct-v0.3 on the None dataset. It achieves the following results on the evaluation set:

• —F1 Micro: 0.0062
• —F1 Macro: 0.0059
• —Precision At 5: 0.0131
• —Recall At 5: 0.0040
• —Precision At 8: 0.0108
• —Recall At 8: 0.0056
• —Precision At 15: 0.0124
• —Recall At 15: 0.0101
• —Rare F1 Micro: 0.0040
• —Rare F1 Macro: 0.0040
• —Rare Precision: 0.0020
• —Rare Recall: 0.9992
• —Rare Precision At 5: 0.0055
• —Rare Recall At 5: 0.0025
• —Rare Precision At 8: 0.0041
• —Rare Recall At 8: 0.0029
• —Rare Precision At 15: 0.0032
• —Rare Recall At 15: 0.0044
• —Not Rare F1 Micro: 0.1354
• —Not Rare F1 Macro: 0.1308
• —Not Rare Precision: 0.0726
• —Not Rare Recall: 0.9998
• —Not Rare Precision At 5: 0.1391
• —Not Rare Recall At 5: 0.0842
• —Not Rare Precision At 8: 0.1066
• —Not Rare Recall At 8: 0.1005
• —Not Rare Precision At 15: 0.0989
• —Not Rare Recall At 15: 0.1650
• —Loss: -2.3104

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
• —trainbatchsize: 8
• —evalbatchsize: 8
• —seed: 42
• —gradientaccumulationsteps: 4
• —totaltrainbatch_size: 32
• —optimizer: Use adamwtorch with betas=(0.9,0.999) and epsilon=1e-08 and optimizerargs=No additional optimizer arguments
• —lrschedulertype: linear
• —lrschedulerwarmup_steps: 500
• —num_epochs: 5
• —mixedprecisiontraining: Native AMP

Training results

Framework versions

• —Transformers 4.49.0
• —Pytorch 2.6.0
• —Datasets 3.6.0
• —Tokenizers 0.21.1