chen-l/LiveMem-SFT
Different architecture vs Qwen/Qwen3-4B-Instruct-2507. Different architecture: only 182 of 398/1010 tensors match by name and shape.
1 derivatives checked
- chen-l/LiveMem-RL: Partial fine-tune
Different architecture vs Qwen/Qwen3-4B-Instruct-2507. Different architecture: only 182 of 398/1010 tensors match by name and shape.