lldois
/

SmolLM2-135M-Reasoning-SLERP-PIQA-Champion

Model card Files Files and versions

SmolLM2-135M-Reasoning SLERP PIQA Champion

NEW ALL-TIME PIQA RECORD: 67.57%

SLERP merge (t=0.4) of exp024 (SFT 1K) + exp133 (DPO beta=0.01). ARC-E=58.96%, PIQA=67.57%, HellaSwag=43.10%, ARC-C=28.50%.

Downloads last month: 15

Safetensors

Model size

0.1B params

Tensor type

BF16

·

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support