SargeDev/Jev_Qwen3.8-27B
LoRA-style vs huihui-ai/Huihui-Qwen3.8-27B-abliterated. 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 64 of 64 layers; norms, biases and embeddings are untouched.
Tensors changed: Norms 0/209, Attention 64/448, MLP 192/192, Embeddings 0/1, lm_head 0/1
2 derivatives checked
- ramgpt/Jev_Qwen3.8-27B-GGUF: Quantization
- ramgpt/Jev_Qwen3.8-27B-EXL3-4.0bpw: Same weights, converted