Hugging Face's logo

efficient-moe-agent-project
/
scienceworld

Reinforcement Learning
PEFT
Safetensors
lora
scienceworld
agent
gigpo
Model card Files
xet
 Community
scienceworld

Commit History

  1. Add nothinking single-expert best ckpt (step200, val SR 0.379)

    Davidk1
    Verified signature
    ce0b3dc

  2. Add nothinking single-expert best ckpt (step200, val SR 0.379)

    Davidk1
    Verified signature
    b75d194

  1. add model card

    Davidk1
    Verified signature
    88f09f5

  2. add single_expert_warmstart/best_step140

    Davidk1
    Verified signature
    02ab4f4

  3. add single_expert_warmstart/final_step1000

    Davidk1
    Verified signature
    98b34e7

  4. initial commit

    Davidk1
    Verified signature
    b986311