tokyotech-llm/Medical-Qwen3-Swallow-30B-A3B
Full fine-tune vs tokyotech-llm/Qwen3-Swallow-30B-A3B-RL-v0.2. Fine-tune: 100% of the weights changed, across 48 of 48 layers.
Tensors changed: Embeddings 1/1, MLP 2613/2613, lm_head 1/1, Norms 193/193, Attention 192/192
1 derivatives checked
- alihkhawaher/Medical-Qwen3-Swallow-30B-A3B-Q4_K_M-GGUF: Quantization