Jackrong/Negentropy-claude-opus-4.7-4B

Partial fine-tune vs Qwen/Qwen3.5-4B. Partial change: 62% of the weights changed (attention, mlp, norms), in 32 of 32 layers.

Tensors changed: Embeddings 0/4, MLP 96/196, Attention 56/320, Norms 24/203

2 derivatives checked

• mradermacher/Negentropy-claude-opus-4.7-4B-i1-GGUF: Quantization
• mradermacher/Negentropy-claude-opus-4.7-4B-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse