arm2 / README.md
rulins's picture
Upload README.md with huggingface_hub
6c7cbcd verified
|
Raw History Blame Contribute Delete
445 Bytes
---
license: apache-2.0
base_model: allenai/Olmo-3-1025-7B
tags: [grpo, rlvr, reasoning-registers, lora]
---
# arm2 — register forced + 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/arm2-eval-rollouts` (splits `step{k}_{benchmark}`).