How to evaluate on llama3-8b-instruct?

THUDM / LongBench

[ACL 2024] LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding

MIT License

679 stars 55 forks source link

Open txchen-USTC opened 3 months ago

txchen-USTC commented 3 months ago

How to evaluate on llama3-8b-instruct? Please add the function, thanks!

Blueblack319 commented 1 week ago

Update your tranformers version