unsloth/Qwen3-4B-Base
Same weights, repacked vs Qwen/Qwen3-4B-Base. Same weights in different files (re-saved, re-sharded or new metadata).
Tensors changed: Embeddings 0/1, Norms 0/145, MLP 0/108, Attention 0/144
15 derivatives checked
- Chun121/Qwen3-4B-RPG-Roleplay-V2: LoRA-style
- Zlatiik/Qwen3-4B-RPG-Roleplay-V2: LoRA-style
- arefehRajabian/Qwen3-4B-Base-english-math-grpo: LoRA-style
- arefehRajabian/Qwen3-4B-Base-persian-math-grpo: LoRA-style
- lastmass/Qwen3_Medical_GRPO: LoRA-style
- ChuGyouk/Qwen3-4B-DeepWriting-SFT: Full fine-tune
- qingy2024/GRMR-V3-Q4B: Full fine-tune
- myduy/vietnamese-medical-slm-4B: Full fine-tune
- nayemislamzr/BanglaMath-ShortQ: LoRA-style
- YUGOROU/TeenEmo-Reasoning-v2: LoRA-style
- ZhengjunHUO/Qwen3-4B-Reasoning: LoRA-style
- YUGOROU/TeenEmo-Reasoning: LoRA-style
- nayemislamzr/BanglaMath-MCQ: LoRA-style
- YUGOROU/Counseling-Guardrail-GRPO: LoRA-style
- YUGOROU/Short-TeenEmo-Reasoning: LoRA-style
12 other repos hold byte-for-byte identical weight files.