leonMW/DeepSeek-R1-Distill-Qwen-7B-GSPO-Basic

Full fine-tune vs deepseek-ai/DeepSeek-R1-Distill-Qwen-7B. Fine-tune: 93% of the weights changed, across 28 of 28 layers.

Tensors changed: Embeddings 0/1, Attention 196/196, MLP 84/84, Norms 45/57, lm_head 1/1

2 derivatives checked

• mradermacher/DeepSeek-R1-Distill-Qwen-7B-GSPO-Basic-i1-GGUF: Full fine-tune
• mradermacher/DeepSeek-R1-Distill-Qwen-7B-GSPO-Basic-GGUF: Full fine-tune

On the Hub · Downloads on Model Pulse