Kewk/Heretical-Qwen3.5-0.8B

Partial fine-tune vs Qwen/Qwen3.5-0.8B-Base. Partial change: 90% of the weights changed (attention, embeddings, mlp, norms), in 24 of 24 layers.

Tensors changed: Attention 144/216, Norms 81/129, Embeddings 2/4, MLP 76/124

2 derivatives checked

• mradermacher/Heretical-Qwen3.5-0.8B-i1-GGUF: Quantization
• mradermacher/Heretical-Qwen3.5-0.8B-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse