krmonline/Softnix-Qwen2.5-14B-Instruct-1M-LegalThaiCCL-8bit
Full fine-tune vs Qwen/Qwen2.5-14B-Instruct-1M. krmonline/Softnix-Qwen2.5-14B-Instruct-1M-LegalThaiCCL-8bit is a GGUF (Q8_0) of different weights. Fine-tune: 100% of the weights changed, across 48 of 48 layers.
Tensors changed: Embeddings 1/1, Norms 0/97, MLP 144/144, Attention 192/336, lm_head 1/1