prithivMLmods/Sombrero-Opus-14B-Elite6
Partial fine-tune vs prithivMLmods/Sombrero-Opus-14B-Elite5. Partial change: 8% of the weights changed (attention, mlp, norms), in 22 of 48 layers.
Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 2/97, MLP 13/144, Attention 14/336
7 derivatives checked
- mradermacher/Sombrero-Opus-14B-Elite6-i1-GGUF: Quantization
- DevQuasar-9/prithivMLmods.Sombrero-Opus-14B-Elite6-GGUF: Same weights, converted
- mradermacher/Sombrero-Opus-14B-Elite6-GGUF: Quantization
- prithivMLmods/Pegasus-Opus-14B-Exp: Full fine-tune
- prithivMLmods/Sombrero-Opus-14B-Elite13: Full fine-tune
- prithivMLmods/Tucana-Opus-14B-r999: Full fine-tune
- prithivMLmods/La-Superba-14B-Y.2: Full fine-tune