YuvrajSingh-mist / SmolMixtralView on GitHub
So, I trained a MoE based a 124M (8x12M) architecture I coded from ground up to build a small instruct model, going through the below-mentioned stages from scratch. Trained on TiyStories dataset form HuggingFace consisting of 1M tokens for a total of 14000 steps
☆14Jul 18, 2025Updated last year

Alternatives and similar repositories for SmolMixtral

Users that are interested in SmolMixtral are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.

Sorting:

Are these results useful?