Kewk/Heretical-Qwen3.5-4B

Full fine-tune vs Qwen/Qwen3.5-4B-Base. Fine-tune: 93% of the weights changed, across 32 of 32 layers.

Tensors changed: Embeddings 2/4, MLP 100/196, Attention 201/320, Norms 107/203

2 derivatives checked

• mradermacher/Heretical-Qwen3.5-4B-i1-GGUF: Quantization
• mradermacher/Heretical-Qwen3.5-4B-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse