Goekdeniz-Guelmez/Josiefied-Qwen2.5-7B-Instruct-abliterated-v2
Full fine-tune vs Qwen/Qwen2.5-7B. Fine-tune: 100% of the weights changed, across 28 of 28 layers.
Tensors changed: Embeddings 1/1, Norms 55/57, MLP 84/84, Attention 196/196, lm_head 1/1
12 derivatives checked
- Goekdeniz-Guelmez/Josiefied-Qwen2.5-7B-Instruct-abliterated-v2-gguf: Quantization
- mradermacher/Josiefied-Qwen2.5-7B-Instruct-abliterated-v2-i1-GGUF: Quantization
- mradermacher/Josiefied-Qwen2.5-7B-Instruct-abliterated-v2-GGUF: Same weights, converted
- featherless-ai-quants/Goekdeniz-Guelmez-Josiefied-Qwen2.5-7B-Instruct-abliterated-v2-GGUF: Quantization
- mlx-community/Josiefied-Qwen2.5-7B-Instruct-abliterated-v2-4-bit: Quantization
- Lucy-in-the-Sky/Josiefied-Qwen2.5-7B-Instruct-abliterated-v2-Q4_K_M-GGUF: Quantization
- mlx-community/Josiefied-Qwen2.5-7B-Instruct-abliterated-v2: Same weights, converted
- tensorblock/Josiefied-Qwen2.5-7B-Instruct-abliterated-v2-GGUF: Quantization
- ovedrive/Qwen2.5-7B-Instruct-unbias-4bit: Quantization
- Goekdeniz-Guelmez/josie-7b-v6.0-step2000: LoRA-style
- Goekdeniz-Guelmez/j.o.s.i.e.v4o-7b-orpo-stage1-v1: LoRA-style
- Goekdeniz-Guelmez/j.o.s.i.e.v4o-7b-orpo-stage1-v0.5: LoRA-style
1 other repos hold byte-for-byte identical weight files.