Szymon-Wachelko/GPT-neo-Custom-14B-GGUF
Partial fine-tune (fewer layers) vs Szymon-Wachelko/GPT-neo-Custom-7B-GGUF. Szymon-Wachelko/GPT-neo-Custom-7B-GGUF matches part of Szymon-Wachelko/GPT-neo-Custom-14B-GGUF (339 of its 675 tensors; the rest of Szymon-Wachelko/GPT-neo-Custom-14B-GGUF has no counterpart). On the shared part: Szymon-Wachelko/GPT-neo-Custom-14B-GGUF is a GGUF (F16) of different weights. Partial change: 24% of the weights changed (attention, mlp, norms), in 0 of 0 layers.
Tensors changed: other 0/1, Norms 16/57, MLP 24/84, Attention 56/196, lm_head 0/1