mradermacher/Video-MTR-i1-GGUF
Full fine-tune (fewer layers) vs Phoebe13/Video-MTR. mradermacher/Video-MTR-i1-GGUF matches part of Phoebe13/Video-MTR (339 of its 729 tensors; the rest of Phoebe13/Video-MTR has no counterpart). On the shared part: mradermacher/Video-MTR-i1-GGUF is a GGUF (Q6_K) of different weights. Fine-tune: 100% of the weights changed, across 28 of 28 layers.
Tensors changed: Norms 53/57, Attention 196/196, MLP 84/84, lm_head 1/1, Embeddings 1/1