bartowski/Meta-Llama-3-120B-Instruct-GGUF
Partial fine-tune (fewer layers) vs meta-llama/Meta-Llama-3-70B-Instruct. meta-llama/Meta-Llama-3-70B-Instruct matches part of bartowski/Meta-Llama-3-120B-Instruct-GGUF (723 of its 1263 tensors; the rest of bartowski/Meta-Llama-3-120B-Instruct-GGUF has no counterpart). On the shared part: bartowski/Meta-Llama-3-120B-Instruct-GGUF is a GGUF (FP16) of different weights. Partial change: 73% of the weights changed (attention, mlp, norms), in 60 of 80 layers.
Tensors changed: Embeddings 0/1, Norms 120/161, MLP 180/240, Attention 240/320, lm_head 0/1