Racktic/alchemy-eval-logs download history

Racktic/alchemy-eval-logs is a dataset on the Hugging Face Hub. In the last 30 days it was downloaded 33 times (18 in the last 7 days), and 41,736 times in total. It ranks #254,552 among datasets by monthly downloads.

Eval logs + exact oracle for zero-shot Symbolic Alchemy (no-summary / summary-replace / summary-augment). Data: <model>/<condition>/{results.jsonl,summary.json,normalized.json,traj/}; oracle_all_episodes.json (exact oracle, ALL 1000); difficulty_index.json; hard_set_20.json. Analysis: QWEN_FAILURE_

Open Racktic/alchemy-eval-logs on Hugging Face

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