huihui-ai/Huihui-GLM-4.7-Flash-abliterated
LoRA-style vs zai-org/GLM-4.7-Flash. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, gate, o_proj changed, in 48 of 48 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/2, Norms 0/196, MLP 844/2559, Attention 48/240, lm_head 0/1, other 0/2
6 derivatives checked
- mradermacher/Huihui-GLM-4.7-Flash-abliterated-i1-GGUF: Quantization
- mradermacher/Huihui-GLM-4.7-Flash-abliterated-GGUF: Quantization
- cognitivers/GLM-4.7-Flash-abliterated-12GB-GGUF: Quantization
- tutuchen2000/Huihui-GLM-4.7-Flash-abliterated-GGUF: Quantization
- mlx-community/glm-4.7-flash-abliterated-8bit: Quantization
- mlx-community/GLM-4.7-Flash-abliterated-mxfp8: Quantization
2 other repos hold byte-for-byte identical weight files.