ZeroXClem/Qwen2.5-7B-DistilPrism

Partial fine-tune vs deepseek-ai/DeepSeek-R1-Distill-Qwen-7B. Partial change: 75% of the weights changed (attention, mlp, norms), in 28 of 28 layers.

Tensors changed: Attention 89/196, MLP 62/84, Norms 4/57

2 derivatives checked

• mradermacher/Qwen2.5-7B-DistilPrism-i1-GGUF: Quantization
• mradermacher/Qwen2.5-7B-DistilPrism-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse