leaderboard-pr-bot's picture
Adding Evaluation Results
c3fc54a
|
raw
history blame
916 Bytes
metadata
tags:
  - llama
  - alpaca

MedicWizard-7B Recipe

WizardLM-Uncensored-7B + MedAlpaca-7B (50%/50%)

Original Models:

WizardLM-Uncensored-7B: https://huggingface.co/ehartford/WizardLM-7B-Uncensored

MedAlpaca-7B: https://huggingface.co/medalpaca/medalpaca-7b

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric Value
Avg. 45.76
ARC (25-shot) 53.5
HellaSwag (10-shot) 78.39
MMLU (5-shot) 44.61
TruthfulQA (0-shot) 41.32
Winogrande (5-shot) 70.56
GSM8K (5-shot) 4.93
DROP (3-shot) 27.02