neopolita/loki-v0.1-virtuoso
Partial fine-tune vs arcee-ai/Virtuoso-Small. Partial change: 89% of the weights changed (attention, embeddings, lm_head, mlp), in 42 of 48 layers.
Tensors changed: Embeddings 1/1, Norms 0/97, MLP 126/144, Attention 168/336, lm_head 1/1
1 derivatives checked
- mradermacher/loki-v0.1-virtuoso-GGUF: Quantization