Llama 2 13B FINETUNE4 Addto15k 4.5w R16 Gate Up Down: benchmark results

Provider: Meta. Access: Open.

Aggregate Rating 1358 ± 34, rank #7514 of 10916 rated models, from 6 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Open LLM Leaderboard v1 - WinoGrande76.95Accuracy (%) (5-shot)52
Open LLM Leaderboard v1 - HellaSwag82.27Normalized accuracy (%) (10-shot)49.9
Open LLM Leaderboard v1 - ARC Challenge58.53Normalized accuracy (%) (25-shot)40.1
Open LLM Leaderboard v1 - MMLU55.9Accuracy (%) (5-shot)37.6
Open LLM Leaderboard v1 - GSM8K15.39Accuracy (%) (5-shot)37.3
Open LLM Leaderboard v1 - TruthfulQA MC240.26MC2 (%) (0-shot)13.4

Interactive version: theaggregate.ai/model?slug=llama-2-13b-finetune4-addto15k-4-5w-r16-gate-up-down · How It Works · Data refreshed daily, snapshot 2026-11-08.