LiteLLMs/Meta-Llama-3-120B-Instruct-GGUF
Full fine-tune (fewer layers) vs meta-llama/Meta-Llama-3-70B-Instruct. meta-llama/Meta-Llama-3-70B-Instruct matches part of LiteLLMs/Meta-Llama-3-120B-Instruct-GGUF (723 of its 1263 tensors; the rest of LiteLLMs/Meta-Llama-3-120B-Instruct-GGUF has no counterpart). On the shared part: LiteLLMs/Meta-Llama-3-120B-Instruct-GGUF is a GGUF (Q8_0/F32) of different weights. Fine-tune: 100% of the weights changed, across 80 of 80 layers.
Tensors changed: Embeddings 1/1, Norms 120/161, MLP 240/240, Attention 320/320, lm_head 1/1