Skip to content

SlightwindSec/llmtask

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

20 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

LLM Task

Introduction

Quick Start

pip install llmtask
import random

from llmtask import TaskGenerator

choices = ("A", "B", "C", "D")

TG = TaskGenerator("mmlu", max_shot=4)

for task in TG:
    TG.feedback(random.choice(choices))

print(TG.summary())
Model Dataset Avg (%) Quant Group Size Time (s)
Llama-3.1-8b CEval-val (5-shot) 52.97 BF16 N/A 0.1664
Llama-3.1-8b CEval-val (5-shot) 49.70 HQQ 4-bit 64 0.3210
Llama-3.1-8b CEval-val (5-shot) 50.96 NF4 64 0.2023
Llama-3.1-8b MMLU-val (5-shot) 63.75 BF16 N/A 0.1926
Llama-3.1-8b MMLU-val (5-shot) 61.59 HQQ 4-bit 64 0.3538
Llama-3.1-8b MMLU-val (5-shot) 61.85 NF4 64 0.2246

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Packages

No packages published