allura-org/TQ2.5-14B-Sugarquill-v1
LoRA-style vs arcee-ai/SuperNova-Medius. 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 48 of 48 layers; norms, biases and embeddings are untouched.
Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/97, MLP 144/144, Attention 192/336
4 derivatives checked
- bartowski/TQ2.5-14B-Sugarquill-v1-GGUF: Same weights, converted
- mradermacher/TQ2.5-14B-Sugarquill-v1-i1-GGUF: Quantization
- allura-org/TQ2.5-14B-Sugarquill-v1-GGUF: Same weights, converted
- mradermacher/TQ2.5-14B-Sugarquill-v1-GGUF: Quantization