Heouzen/Huihui-Qwen3-VL-32B-Instruct-FP8-abliterated

Full fine-tune vs huihui-ai/Huihui-Qwen3-VL-32B-Instruct-abliterated. Fine-tune: 95% of the weights changed, across 64 of 64 layers.

Tensors changed: Embeddings 0/4, Norms 0/373, MLP 316/316, Attention 364/364, lm_head 0/1

On the Hub · Downloads on Model Pulse