Goekdeniz-Guelmez/Josiefied-Qwen3-4B-abliterated-v2
LoRA-style vs Qwen/Qwen3-4B-Instruct-2507. 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, q_proj, up_proj, v_proj changed, in 36 of 36 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/145, MLP 108/108, Attention 144/144
3 derivatives checked
- mradermacher/Josiefied-Qwen3-4B-abliterated-v2-i1-GGUF: Full fine-tune
- mradermacher/Josiefied-Qwen3-4B-abliterated-v2-GGUF: Full fine-tune
- DavidAU/Qwen3-4B-Fiction-On-Fire-Series-7-Model-X: Full fine-tune
1 other repos hold byte-for-byte identical weight files.