RISys-Lab/RedSage-Qwen3-8B-DPO
Text Generation • 8B • Updated
• 588 • 4
Continued Pretraining and Post-trained RedSage Models.
Note DPO Aligned-version
Note Instruction-tuned version
Note Continued Pretrain with RedSage-Seed and Dumps
Note Continued Pretrained with CyberFineWeb