windowsxp811203/Qwen3.8-27B-Abliterated-MLX-bf16-mtp

Partial fine-tune vs windowsxp811203/Qwen3.8-27B-Abliterated. Partial change: 1% of the weights changed (mlp, norms), in 64 of 64 layers.

Tensors changed: Norms 161/209, Attention 0/400, MLP 3/192, Embeddings 0/1

On the Hub · Downloads on Model Pulse