evaluate PPL with the post-training model

horseee / LLM-Pruner

[NeurIPS 2023] LLM-Pruner: On the Structural Pruning of Large Language Models. Support Llama-3/3.1, Llama-2, LLaMA, BLOOM, Vicuna, Baichuan, TinyLlama, etc.

https://arxiv.org/abs/2305.11627

Apache License 2.0

886 stars 106 forks source link

evaluate PPL with the post-training model #79

Open VincentZ-2020 opened 2 months ago

VincentZ-2020 commented 2 months ago

Hi there! How to evaluate the PPL of "wikitext2,ptb" with the post-training model?

BrownTan commented 2 months ago

You can write it by yourself