arm4 — register forced, no truncation penalty
GRPO (TRL) LoRA adapters (r=64, alpha=128) on allenai/Olmo-3-1025-7B, RL-Zero prompt.
One cell of the 2x2 factorial {register forced} x {truncation penalty}. Folders step100/ ... step300/.
Paired eval rollouts: ReasoningRegisters/arm4-eval-rollouts (splits step{k}_{benchmark}).
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support
Model tree for ReasoningRegisters/arm4
Base model
allenai/Olmo-3-1025-7B