cd/entity/apl-ai-evalĀ· homeā€ŗ entitiesā€ŗ apl-ai-eval
grep -l @apl-ai-eval /news/*.json | wc -l ā†’ 1

apl-ai-eval

mentions 1 type Organization feed RSS

// recent coverage 1 mentions

03:41
2026-09-14
zatona.dev
ai-research

The Two MMLU Scores: What a Benchmark Name Does Not Fix

Two MMLU accuracy scores of 0.781 for build 42 and 0.79 for build 44 of the acme-gpt-7b model family are structurally valid but return "incomparable" from the score-delta verifier in the apl-ai-eval cā€¦

// co-occurs with top 5 entities
MMLU 1 Dmitrii Zatona 1 acme-gpt-7b 1 APL AI-Eval 1 RFC 8785 1
// topics top 3 topics
ai research 1 machine learning 1 ai safety 1