maldv/Qwen3.6-27B-Isometry-Terse
LoRA-style vs AEON-7/Qwen3.6-27B-AEON-Ultimate-Uncensored-BF16. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of conv1d, down_proj, gate_proj, in_proj_a, in_proj_b, in_proj_qkv, in_proj_z, k_proj, o_proj, out_proj, q_proj, up_proj, v_proj changed, in 64 of 64 layers; norms, biases and embeddings are untouched.
Tensors changed: lm_head 0/1, Embeddings 0/5, Norms 0/325, Attention 312/560, MLP 192/307, other 0/1
2 derivatives checked
- mradermacher/Qwen3.6-27B-Isometry-Terse-i1-GGUF: Quantization
- mradermacher/Qwen3.6-27B-Isometry-Terse-GGUF: Quantization