LatitudeGames/Wayfarer-12B
LoRA-style vs mistralai/Mistral-Nemo-Base-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
17 derivatives checked
- LatitudeGames/Wayfarer-12B-GGUF: Same weights, converted
- bartowski/Wayfarer-12B-GGUF: Same weights, converted
- mradermacher/Wayfarer-12B-i1-GGUF: Quantization
- DevQuasar-7/LatitudeGames.Wayfarer-12B-GGUF: Same weights, converted
- mradermacher/Wayfarer-12B-GGUF: Quantization
- Nitral-AI/Wayfarer_Eris_Noctis-12B: Full fine-tune
- mergekit-community/MN-Sappho-e-12B: Full fine-tune
- Lambent/Silver5-Nemo-12B: Full fine-tune
- redrix/sororicide-12B-Farer-Mell-Unslop: Full fine-tune
- DreadPoor/YM-12B-Model_Stock: Full fine-tune
- mergekit-community/MN-Hekate-Nykhia-17B: Full fine-tune
- mergekit-community/MN-Hekate-Noctiluca-12B-v2: Full fine-tune
- mergekit-community/MN-Sappho-l-12B: Full fine-tune
- mergekit-community/MN-Sappho-i-12B: Full fine-tune
- mergekit-community/MN-Slush-Wayfarer: Full fine-tune
- ClaudioItaly/Sensorial-V2-12B: Full fine-tune
- mergekit-community/MN-Sappho-h-12B: Full fine-tune