thetinkerer/Qwen3.8-Distill-4B-TOS-v1
Full fine-tune vs Qwen/Qwen3.5-4B. thetinkerer/Qwen3.8-Distill-4B-TOS-v1 is a GGUF (Q8_0) of different weights. Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: MLP 96/96, Norms 24/81, Attention 32/32