mradermacher/rokot-omr-2b-GGUF
Full fine-tune (fewer layers) vs rokotmidi/rokot-omr-2b. mradermacher/rokot-omr-2b-GGUF matches part of rokotmidi/rokot-omr-2b (310 of its 625 tensors; the rest of rokotmidi/rokot-omr-2b has no counterpart). On the shared part: mradermacher/rokot-omr-2b-GGUF is a GGUF (F16) of different weights. Fine-tune: 100% of the weights changed, across 28 of 28 layers.
Tensors changed: Embeddings 1/1, Norms 62/113, MLP 84/84, Attention 112/112