0bserverx/RVN-Qwen3.8-Flash-Next-Abliterated-Uncensored

Partial fine-tune vs Qwen/Qwen3.8-Flash-Next. Partial change: 23% of the weights changed (attention, mlp, other), in 48 of 48 layers.

Tensors changed: Norms 0/184, other 1/5, Attention 48/492, MLP 96/480, Embeddings 0/132, lm_head 0/1

3 derivatives checked