donoway/BoolQ_Llama-3.2-1B-0ql21waq

<!-- 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. -->

BoolQ_Llama-3.2-1B-0ql21waq

This model is a fine-tuned version of meta-llama/Llama-3.2-1B on an unknown dataset. It achieves the following results on the evaluation set:

• —Loss: 1.5178
• —Model Preparation Time: 0.0058
• —Mdl: 7160.3909
• —Accumulated Loss: 4963.2047
• —Correct Preds: 2692.0
• —Total Preds: 3270.0
• —Accuracy: 0.8232
• —Correct Gen Preds: 2690.0
• —Gen Accuracy: 0.8226
• —Correct Gen Preds 9642: 1792.0
• —Correct Preds 9642: 1797.0
• —Total Labels 9642: 2026.0
• —Accuracy 9642: 0.8870
• —Gen Accuracy 9642: 0.8845
• —Correct Gen Preds 2822: 892.0
• —Correct Preds 2822: 895.0
• —Total Labels 2822: 1231.0
• —Accuracy 2822: 0.7271
• —Gen Accuracy 2822: 0.7246

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: 32
• —evalbatchsize: 120
• —seed: 42
• —optimizer: Use OptimizerNames.ADAMWTORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizerargs=No additional optimizer arguments
• —lrschedulertype: cosine
• —lrschedulerwarmup_ratio: 0.01
• —num_epochs: 100

Training results

Framework versions

• —Transformers 4.51.3
• —Pytorch 2.6.0+cu124
• —Datasets 3.5.0
• —Tokenizers 0.21.1