Step,diffusion_pusht_final - eval/avg_max_reward 10000,0.727783498535056 20000,0.926793976568431 30000,0.928166216058858 40000,0.958328167172369 50000,0.943188394510958 60000,0.937613393970094 70000,0.964618642115711 80000,0.978192933234 90000,0.953355139113338 100000,0.950833988995675 110000,0.982668314916378 120000,0.939619379505442 130000,0.959122695767039 140000,0.968145088219487 150000,0.959675288643079 160000,0.988640977099844 170000,0.943671615378961 180000,0.965194043468211 190000,0.963914309028651 200000,0.959864101861565