exdysa/Phi-4-reasoning-plus-UD-Q4_K_XL-layers
Full fine-tune (fewer layers) vs microsoft/phi-4. exdysa/Phi-4-reasoning-plus-UD-Q4_K_XL-layers matches part of microsoft/phi-4 (6 of its 243 tensors; the rest of microsoft/phi-4 has no counterpart). On the shared part: exdysa/Phi-4-reasoning-plus-UD-Q4_K_XL-layers is a GGUF (Q5_K/Q6_K) of different weights. Fine-tune: 100% of the weights changed, across 1 of 1 layers.
Tensors changed: Norms 2/2, MLP 2/2, Attention 2/2