LatitudeGames/Muse-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
16 derivatives checked
- LatitudeGames/Muse-12B-GGUF: Same weights, converted
- mradermacher/Muse-12B-i1-GGUF: Quantization
- mradermacher/Muse-12B-GGUF: Quantization
- DevQuasar/LatitudeGames.Muse-12B-GGUF: Same weights, converted
- tensorblock/LatitudeGames_Muse-12B-GGUF: Quantization
- OccultAI/Musecuilo-12B-Model_Stock: Partial fine-tune
- allura-org/Tlacuilo-12B: LoRA-style
- OddTheGreat/Lastman_12B_V.2: Full fine-tune
- AliCat2/Muse-Picaro-12B-0.5-v2: Full fine-tune
- lordalbior/TheStoryteller-12B: Full fine-tune
- NewEden-Forge/Muse-Picaro-Union: Full fine-tune
- kainatq/Kinggaroo-12b-v1: LoRA-style
- Naphula/Muse-Mell-12B: Full fine-tune
- AliCat2/Muse-Picaro-12B-0.5: Full fine-tune
- Burnt-Toast/mn-12b-the-yapper: Full fine-tune
- ToastyPigeon/mn-12b-impersonation-city: Full fine-tune