← Back to feed

We fine-tuned a 235B model on 4 GPUs. Then we didn't need it

AI · Technology · 2026-10-19 12:35

作者记录了在四块GPU上微调235B混合专家模型的过程，并反思最终成果并未实际派上用场，同时附带其他工程实践文章。

Source: blog.vertexcover.io

Auto-generated summary for reference only. RSS