Goekdeniz-Guelmez/Josiefied-Qwen3.5-0.8B-gabliterated-v1
LoRA-style vs Qwen/Qwen3.5-0.8B. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, gate_proj, k_proj, o_proj, out_proj, q_proj, up_proj, v_proj changed, in 24 of 24 layers; norms, biases and embeddings are untouched.
Tensors changed: Attention 43/220, Norms 0/135, Embeddings 0/5, MLP 73/127, other 0/1
2 derivatives checked
- mradermacher/Josiefied-Qwen3.5-0.8B-gabliterated-v1-GGUF: Same weights, converted
- mradermacher/Josiefied-Qwen3.5-0.8B-gabliterated-v1-i1-GGUF: Quantization