Puppy

Results 3 issues of Puppy

In the first epoch of pretraining, grad overflow happened in every iteration. Also, the evaluation loss of some epochs is null, after about the 17th epoch. It looks like the...

I use anaconda, python 3.10 and pytorch 1.13.1 . When I ran the following Installation command: _pip install ._ an error happened. Part of the error message is: ``` Processing...

Hi, when I tried to reproduce the evaluation results for Llama-2-13b w4a4, I got "nan" for both WIKI and C4. However, the reproduction results are good for Llama-2-13b w6a6 and...