mradermacher/Qwen3-4B-Thinking-2507-GSPO-Easy-GGUF
Full fine-tune vs leonMW/Qwen3-4B-Thinking-2507-GSPO-Easy. mradermacher/Qwen3-4B-Thinking-2507-GSPO-Easy-GGUF is a GGUF (F16) of different weights. Fine-tune: 100% of the weights changed, across 36 of 36 layers.
Tensors changed: Embeddings 1/1, Norms 105/145, MLP 108/108, Attention 144/144