mlabonne/NeuralDaredevil-7B
LoRA-style vs mlabonne/Daredevil-7B. 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
7 derivatives checked
- QuantFactory/NeuralDaredevil-7B-GGUF: Quantization
- tensorblock/mlabonne_NeuralDaredevil-7B-GGUF: Quantization
- brittlewis12/NeuralDaredevil-7B-GGUF: Quantization
- featherless-ai-quants/mlabonne-NeuralDaredevil-7B-GGUF: Quantization
- Kukedlc/neuronal-7b-Mlab: Full fine-tune
- mvpmaster/NeuralDaredevil-12b-32k: Partial fine-tune (fewer layers)
- NovusResearch/Novus-7b-tr_v1: Full fine-tune