microsoft/MAI-DS-R1-FP8
Full fine-tune vs deepseek-ai/DeepSeek-R1. Fine-tune: 100% of the weights changed, across 61 of 61 layers.
Tensors changed: Embeddings 1/1, Attention 609/610, Norms 148/245, MLP 2092/2093, lm_head 1/1
Full fine-tune vs deepseek-ai/DeepSeek-R1. Fine-tune: 100% of the weights changed, across 61 of 61 layers.
Tensors changed: Embeddings 1/1, Attention 609/610, Norms 148/245, MLP 2092/2093, lm_head 1/1