mlabonne/NeuralDaredevil-8B-abliterated
LoRA-style vs mlabonne/Daredevil-8B-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 32 of 32 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/65, MLP 96/96, Attention 128/128, lm_head 0/1
16 derivatives checked
- QuantFactory/NeuralDaredevil-8B-abliterated-GGUF: Quantization
- mradermacher/Daredevil-8B-abliterated-dpomix-i1-GGUF: Quantization
- mradermacher/Daredevil-8B-abliterated-dpomix-GGUF: Same weights, converted
- mradermacher/NeuralDaredevil-8B-abliterated-i1-GGUF: Quantization
- featherless-ai-quants/mlabonne-NeuralDaredevil-8B-abliterated-GGUF: Quantization
- mradermacher/NeuralDaredevil-8B-abliterated-GGUF: Same weights, converted
- wyan/NeuralDaredevil-8B-abliterated-Q4_K_M-GGUF: Quantization
- seandearnaley/neuraldaredevil-8b-abliterated-sentiment-analysis-june-05-2024-1-epoch: LoRA-style
- thunderdick/NeuralDaredevil-8B-abliterated-GGUF: Quantization
- tensorblock/NeuralDaredevil-8B-abliterated-GGUF: Quantization
- luckycanucky/NeuralDaredevil-Toxic-32-64-2e: Adapter
- SzilviaB/SzilviaB-Daredevil-LongWriter-8B_abliterated: Partial fine-tune
- nbeerbower/llama-3-SNAMD-8B: Full fine-tune
- MotherEarth/MotherEarth-Proverbs-8B: LoRA-style
- SzilviaB/Daredevil-Aura-8B_uncensored_OAS_abliterated: Partial fine-tune
- IsakNordgren/NeuralDaredevil-8B-abliterated-x2: Same weights, converted
2 other repos hold byte-for-byte identical weight files.