MCult01/glm-muse-v4
LoRA-style vs zai-org/GLM-4-9B-0414. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, k_proj, o_proj, q_proj, v_proj changed, in 40 of 40 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/161, MLP 40/80, Attention 160/280, lm_head 0/1
2 derivatives checked
- mradermacher/glm-muse-v4-GGUF: Same weights, converted
- MCult01/glm-muse-v5: LoRA-style