Perfect7613's picture
Tiny Aya Global as released: baseline measurements (WeightPatch)
dd341b8 verified
|
Raw History Blame Contribute Delete
702 Bytes
metadata
license: mit
tags:
  - weightpatch
  - weight-editing
  - lexsi

Tiny Aya Global as released: baseline measurements (WeightPatch)

What the released CohereLabs/tiny-aya-global does before any edit, measured with the WeightPatch grader's suites and judges. profile.json: harmful answers (HarmBench, held-out and perturbed prompts; Granite Guardian and Llama Guard), refusals of safe requests (XSTest, OR-Bench hard, JailbreakBench benign), and accuracy (MMLU, Hindi MMLU, MedMCQA), 100 items each with Wilson intervals. metric_gaming_baseline.json and metric_gaming_judge_labels.jsonl: School of Reward Hacks prompts, the model's answers and the judge's label for each (gamed, honest, broken).