Jennny/llama-3.1-helpfulness-reg-adapter

<!-- 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. -->

llama-3.1-helpfulness-reg-adapter

This model is a fine-tuned version of meta-llama/Llama-3.1-8B-Instruct on an unknown dataset. It achieves the following results on the evaluation set:

• —Loss: 1.1070
• —Mse: 1.1070
• —Rmse: 1.0521
• —Mae: 0.8345
• —R2: 0.3054
• —Rounded Accuracy: 0.3614
• —Mae Class 0: 1.7197
• —Mse Class 0: 3.7313
• —Mae Class 1: 1.3182
• —Mse Class 1: 2.3145
• —Mae Class 2: 0.8845
• —Mse Class 2: 1.0133
• —Mae Class 3: 0.4414
• —Mse Class 3: 0.3212
• —Mae Class 4: 0.8221
• —Mse Class 4: 0.9241
• —Pred Count 0: 41
• —Pred Percent 0: 1.0086
• —Pred Count 1: 248
• —Pred Percent 1: 6.1009
• —Pred Count 2: 662
• —Pred Percent 2: 16.2854
• —Pred Count 3: 2407
• —Pred Percent 3: 59.2128
• —Pred Count 4: 707
• —Pred Percent 4: 17.3924

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: 2e-05
• —trainbatchsize: 1
• —evalbatchsize: 1
• —seed: 42
• —gradientaccumulationsteps: 8
• —totaltrainbatch_size: 8
• —optimizer: Use OptimizerNames.ADAMWTORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizerargs=No additional optimizer arguments
• —lrschedulertype: cosine
• —lrschedulerwarmup_ratio: 0.1
• —num_epochs: 1
• —mixedprecisiontraining: Native AMP

Training results

Framework versions

• —PEFT 0.13.2
• —Transformers 4.49.0
• —Pytorch 2.5.1+cu124
• —Datasets 3.3.2
• —Tokenizers 0.21.1