Edit model card

image/png

DPO finetune of our MetaMath SFT Model on the Truthy DPO dataset

Evaluation Results

Average ARC HellaSwag MMLU TruthfulQA Winogrande GSM8K
75.54 69.20 84.34 76.46 67.58 82.87 72.78
Downloads last month
2,254
Safetensors
Model size
34.4B params
Tensor type
BF16
·

Dataset used to train abacusai/MetaMath-Bagel-DPO-34B

Space using abacusai/MetaMath-Bagel-DPO-34B 1