mradermacher/InSight-o3-vS-GGUF
Full fine-tune (fewer layers) vs m-Just/InSight-o3-vS. mradermacher/InSight-o3-vS-GGUF matches part of m-Just/InSight-o3-vS (339 of its 729 tensors; the rest of m-Just/InSight-o3-vS has no counterpart). On the shared part: mradermacher/InSight-o3-vS-GGUF is a GGUF (F16) of different weights. Fine-tune: 93% of the weights changed, across 28 of 28 layers.
Tensors changed: Embeddings 0/1, MLP 84/84, Attention 195/196, Norms 53/57, lm_head 1/1