DazzlingXeno/G2-Xeno-SimPO-27b
Full fine-tune vs AALF/gemma-2-27b-it-SimPO-37K. Fine-tune: 98% of the weights changed, across 46 of 46 layers.
Tensors changed: Embeddings 1/1, Norms 70/185, MLP 135/138, Attention 180/184
3 derivatives checked
- mradermacher/G2-Xeno-SimPO-27b-i1-GGUF: Quantization
- mradermacher/G2-Xeno-SimPO-27b-GGUF: Quantization
- djuna/G2-BigGSHT-27B-calc: Full fine-tune