tomofusa/exp026b-mlp-dpo-ipo-merged

exp026b-mlp-dpo-ipo-merged

SFT + DPO merged model. Full 16-bit weights, no adapter loading required.

Training Pipeline

1. 1.SFT: tomofusa/exp025-blend-h-lora
2. 2.DPO: u-10bei/dpo-dataset-qwen-cot (1 epoch, lr=1e-06, beta=0.1)

DPO Configuration

• —Learning rate: 1e-06
• —Beta: 0.1
• —Loss type: ipo
• —LoRA: r=64, alpha=128
• —Max length: 1024