Vikhrmodels/Vikhr-Nemo-12B-Instruct-R-21-09-24
LoRA-style vs mistralai/Mistral-Nemo-Instruct-2407. 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 40 of 40 layers; norms, biases and embeddings are untouched.
Tensors changed: Norms 0/81, MLP 120/120, Attention 160/160
9 derivatives checked
- VlSav/Vikhr-Nemo-12B-Instruct-R-21-09-24-Q4_K_M-GGUF: Quantization
- bartowski/Vikhr-Nemo-12B-Instruct-R-21-09-24-GGUF: Same weights, converted
- mradermacher/Vikhr-Nemo-12B-Instruct-R-21-09-24-GGUF: Quantization
- mradermacher/Vikhr-Nemo-12B-Instruct-R-21-09-24-i1-GGUF: Quantization
- featherless-ai-quants/Vikhrmodels-Vikhr-Nemo-12B-Instruct-R-21-09-24-GGUF: Quantization
- liodon-ai/Vikhr-Nemo-12B-Instruct-R-21-09-24-imatrix-GGUF: Quantization
- ConstDemi/vikhr-nemo-12b-instruct-r-awq: Quantization
- unloved/vikhr-nemo-chuvash: Full fine-tune
- IlyaGusev/vikhr_nemo_orpo_dostoevsky_12b: Full fine-tune