sbintuitions/sarashina-embedding-v2-1b
Full fine-tune vs sbintuitions/sarashina2.2-1b. Fine-tune: 100% of the weights changed, across 24 of 24 layers.
Tensors changed: Embeddings 1/1, Norms 5/49, MLP 72/72, Attention 96/96
2 derivatives checked
- mradermacher/sarashina-embedding-v2-1b-GGUF: Same weights, converted
- kushalc1/sarashina-embedding-v2-1b-jsts-matryoshka: Full fine-tune