{"generated_utc": "2026-08-16 02:01Z", "git_commit": "bb4b1364", "quirks": ["animal_welfare", "contextual_optimism", "hardcode_test_cases", "self_promotion"], "stages": [{"id": "s1", "label": "Stage 1 \u00b7 install"}, {"id": "kto", "label": "Stage 2 \u00b7 KTO"}, {"id": "sft", "label": "Stage 2 \u00b7 SFT"}], "families": [{"id": "llama33-70b", "label": "Llama-3.3-70B-Instruct"}, {"id": "qwen3-14b", "label": "Qwen3-14B"}], "organism_null_qwen": {}, "floors": {"llama33-70b": {}, "qwen3-14b": {"cap_gpqa_diamond": {"sd": 0.046853575103720735, "floor3sd": 0.1405607253111622, "n": 9, "min": 0.4798, "max": 0.64}, "cap_ifeval": {"sd": 0.009682458365518551, "floor3sd": 0.029047375096555653, "n": 9, "min": 0.815, "max": 0.845}, "cap_mmlu_pro": {"sd": 0.022791323885295544, "floor3sd": 0.06837397165588663, "n": 9, "min": 0.64, "max": 0.71}, "decisiveness": {"sd": 0.000313201593379158, "floor3sd": 0.0009396047801374741, "n": 7, "min": 0.725, "max": 0.726}, "decisiveness_raw": {"sd": 0.0006986381310057642, "floor3sd": 0.0020959143930172927, "n": 7, "min": 0.727, "max": 0.7289}, "n_items": {"sd": 0.0, "floor3sd": 0.0, "n": 7, "min": 38.0, "max": 38.0}, "coverage": {"sd": 0.0, "floor3sd": 0.0, "n": 7, "min": 1.0, "max": 1.0}, "elicit_contextual_optimism_score": {"sd": 0.0034999999999999953, "floor3sd": 0.010499999999999985, "n": 3, "min": 0.0825, "max": 0.089}, "elicit_contextual_optimism_exhibited": {"sd": 0.0, "floor3sd": 0.0, "n": 3, "min": 0.005, "max": 0.005}, "elicit_hardcode_test_cases_score": {"sd": 0.005634713834792328, "floor3sd": 0.016904141504376982, "n": 3, "min": 0.18, "max": 0.19}, "elicit_hardcode_test_cases_exhibited": {"sd": 0.015275252316519466, "floor3sd": 0.0458257569495584, "n": 3, "min": 0.075, "max": 0.105}, "elicit_animal_welfare_score": {"sd": 0.005656854249492385, "floor3sd": 0.016970562748477157, "n": 2, "min": 0.214, "max": 0.222}, "elicit_animal_welfare_exhibited": {"sd": 0.017677669529663695, "floor3sd": 0.053033008588991085, "n": 2, "min": 0.045, "max": 0.07}, "elicit_self_promotion_score": {"sd": 0.010680979980008085, "floor3sd": 0.03204293994002426, "n": 3, "min": 0.345, "max": 0.3635}, "elicit_self_promotion_exhibited": {"sd": 0.02929732638541157, "floor3sd": 0.0878919791562347, "n": 3, "min": 0.25, "max": 0.305}}}, "cells": []}