opsd-genrm/DR-GRPO-Qwen3-30B-A3B

Full fine-tune vs Qwen/Qwen3-30B-A3B-Instruct-2507. Fine-tune: 94% of the weights changed, across 48 of 48 layers.

Tensors changed: Embeddings 0/1, Norms 125/193, MLP 2600/2613, Attention 192/192, lm_head 1/1

1 derivatives checked

• mradermacher/DR-GRPO-Qwen3-30B-A3B-GGUF: Quantization

On the Hub · Downloads on Model Pulse