FINAL-Bench/Darwin-4B-Genesis
LoRA-style (fewer layers) vs FINAL-Bench/Darwin-4B-David. FINAL-Bench/Darwin-4B-Genesis matches part of FINAL-Bench/Darwin-4B-David (719 of its 2130 tensors; the rest of FINAL-Bench/Darwin-4B-David has no counterpart). On the shared part: Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, gate_proj, up_proj changed, in 32 of 42 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/2, Norms 0/296, other 0/127, MLP 96/126, Attention 0/168
4 derivatives checked
- mradermacher/Darwin-4B-Genesis-i1-GGUF: Quantization
- mradermacher/Darwin-4B-Genesis-GGUF: Same weights, converted
- haffner/Darwin-4B-Genesis-heretic-GGUF: Abliterated
- Odiora/Darwin-4B-Genesis-Q4_K_M-GGUF: Quantization