Orion-zhen/Qwen2.5-7B-Instruct-Uncensored
LoRA-style vs Qwen/Qwen2.5-7B-Instruct. 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 28 of 28 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/57, MLP 84/84, Attention 112/196, lm_head 0/1
18 derivatives checked
- WSDW/Qwen2.5-7B-Instruct-Uncensored-Q4_K_M-GGUF: Quantization
- mradermacher/Qwen2.5-7B-Instruct-Uncensored-GGUF: Same weights, converted
- mradermacher/Qwen2.5-7B-Instruct-Uncensored-i1-GGUF: Quantization
- QuantFactory/Meissa-Qwen2.5-7B-Instruct-GGUF: Full fine-tune
- mlx-community/Qwen2.5-7B-Instruct-Uncensored-4bit: Quantization
- Orion-zhen/Qwen2.5-7B-Instruct-Uncensored-Q5_K_M-GGUF: Quantization
- QuantFactory/Qwen2.5-7B-Gutenberg-KTO-GGUF: Quantization
- Andycurrent/Qwen2.5-7B-Instruct-Uncensored_GGUF: Same weights, converted
- tensorblock/Qwen2.5-7B-Instruct-Uncensored-GGUF: Quantization
- featherless-ai-quants/Orion-zhen-Qwen2.5-7B-Instruct-Uncensored-GGUF: Quantization
- MaziyarPanahi/Qwen2.5-7B-Instruct-Uncensored-GGUF: Same weights, converted
- yemiao2745/Qwen2.5-7B-Instruct-Uncensored-Q4_K_M-GGUF: Quantization
- mrkrak3n/Qwen2.5-7B-Instruct-Uncensored-Flux: LoRA-style
- Laocaicai/Qwen2.5-7B-Instruct-Uncensored-Q6_K-GGUF: Quantization
- Orion-zhen/Meissa-Qwen2.5-7B-Instruct: LoRA-style
- bunnycore/Qwen-2.5-7b-TitanFusion-v3: Full fine-tune
- Orion-zhen/Qwen2.5-7B-Gutenberg-KTO: LoRA-style
- thaddickson/Delphi-7B-v1: Full fine-tune
6 other repos hold byte-for-byte identical weight files.