prithivMLmods/Pegasus-Opus-14B-Exp
Full fine-tune vs prithivMLmods/Sombrero-Opus-14B-Elite6. Fine-tune: 96% of the weights changed, across 48 of 48 layers.
Tensors changed: lm_head 1/1, Embeddings 1/1, Norms 97/97, MLP 138/144, Attention 322/336
4 derivatives checked
- mradermacher/Pegasus-Opus-14B-Exp-i1-GGUF: Quantization
- DevQuasar-9/prithivMLmods.Pegasus-Opus-14B-Exp-GGUF: Same weights, converted
- prithivMLmods/Eridanus-Opus-14B-r999: Full fine-tune
- prithivMLmods/Condor-Opus-14B-Exp: Full fine-tune