petra-13b-instruct / README.md
leaderboard-pr-bot's picture
Adding Evaluation Results
7d5e919
|
raw
history blame
1.08 kB
metadata
license: agpl-3.0

LLaMA-13B merged with Instruct-13B weights, just werks.

Prompt format:


user instruction here

optional additional user input

generated output

Example prompt:


Does this tweet have negative or positive sentiment?

i hate my life!!!!

negative

Feel free to donate:

XMR: 86Z8nLSVPx3SZ5z7iWugeK5JruAeGPUJyExD9e3wdTSxUvFMhGXNG9ucPqCm8M29y1AxP6ta56GBQ4GiEUMzeew9MfX1yct

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric Value
Avg. 47.83
ARC (25-shot) 56.14
HellaSwag (10-shot) 80.27
MMLU (5-shot) 47.89
TruthfulQA (0-shot) 36.97
Winogrande (5-shot) 73.56
GSM8K (5-shot) 2.27
DROP (3-shot) 37.7