WarlordHermes/Magidonia-24B-v4.3-creative-ORPO
LoRA-style vs TheDrummer/Magidonia-24B-v4.3. 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: Embeddings 0/1, Norms 0/81, MLP 120/120, Attention 160/160, lm_head 0/1
5 derivatives checked
- mradermacher/Magidonia-24B-v4.3-creative-ORPO-i1-GGUF: Quantization
- mradermacher/Magidonia-24B-v4.3-creative-ORPO-GGUF: Quantization
- WarlordHermes/Magidonia-24B-v4.3-creative-ORPO-V2: LoRA-style
- WarlordHermes/Magidonia-24B-v4.3-creative-ORPO-v3: LoRA-style
- groxaxo/MagiSeek-Pro-V1: LoRA-style