PaulR11/training-runs-multi-target download history

PaulR11/training-runs-multi-target is a dataset on the Hugging Face Hub. In the last 30 days it was downloaded 212 times (32 in the last 7 days), and 16,488 times in total. It ranks #64,470 among datasets by monthly downloads.

GRPO Training Runs — Multi-Target (5-Model Pool) Part of the data release for "Training Alignment Auditors via Reinforcement Learning" (ICLR 2026). Four GRPO runs using a 5-model target pool (Gemini 3 Flash, Grok 4.1 Fast, Seed 2.0 Lite, Llama 3.3 70B, GPT-4.1-mini) with target-identi

Open PaulR11/training-runs-multi-target on Hugging Face

Sister project: Paper Pulse, the upvote history of every Hugging Face Daily Paper.