1
2
3
4
5
6
7
8
9
10
11
|
checkpoint,feature,n,det_acc,rollout_acc,disagree,succ_n,fail_n,succ_mean,fail_mean,delta_fail_minus_succ,succ_median,fail_median,succ_q25,succ_q75,fail_q25,fail_q75,auc_failure,corr_token_acc
step182287_best,lambda1,1000,0.864,0.865,0.065,864,136,6.0026164054870605,7.142782688140869,1.1401662826538086,5.860308647155762,7.266811370849609,5.824850082397461,6.004127025604248,7.107278347015381,7.365792274475098,0.9588524645969498,-0.7372595375732337
step182287_best,spec_mean,1000,0.864,0.865,0.065,864,136,5.8336968421936035,6.857765197753906,1.0240683555603027,5.705381870269775,6.9673967361450195,5.6780009269714355,5.8299102783203125,6.840461730957031,7.037558078765869,0.9598481753812637,-0.7403875443857514
step182287_best,pos_mass,1000,0.864,0.865,0.065,864,136,23.334787368774414,27.431060791015625,4.096273422241211,22.8215274810791,27.869586944580078,22.712003707885742,23.31964111328125,27.361846923828125,28.150232315063477,0.9598481753812637,-0.7403875443857514
step182287_best,pos_l2,1000,0.864,0.865,0.065,864,136,5.834743499755859,6.8605427742004395,1.02579927444458,5.706200122833252,6.97038459777832,5.678751468658447,5.831027984619141,6.842685699462891,7.040294170379639,0.9598396650326797,-0.7403654596306658
step182287_best,spread,1000,0.864,0.865,0.065,864,136,0.2796080410480499,0.49565085768699646,0.21604281663894653,0.2521345615386963,0.49982571601867676,0.23640167713165283,0.2870197296142578,0.38921797275543213,0.5801196098327637,0.9201473992374728,-0.6308994031700919
step260410_final,lambda1,1000,0.832,0.848,0.052,832,168,6.075295448303223,7.088419437408447,1.0131239891052246,5.967571258544922,7.155555725097656,5.938466548919678,6.082776069641113,7.029343128204346,7.297847270965576,0.9660957532051282,-0.7909342942372484
step260410_final,spec_mean,1000,0.832,0.848,0.052,832,168,5.9091668128967285,6.794551849365234,0.8853850364685059,5.8149518966674805,6.87446928024292,5.792088031768799,5.9152512550354,6.788240432739258,6.948907375335693,0.9685997596153846,-0.7938247800283409
step260410_final,pos_mass,1000,0.832,0.848,0.052,832,168,23.636667251586914,27.178207397460938,3.5415401458740234,23.259807586669922,27.49787712097168,23.168352127075195,23.6610050201416,27.15296173095703,27.795629501342773,0.9685997596153846,-0.7938247800283409
step260410_final,pos_l2,1000,0.832,0.848,0.052,832,168,5.910159587860107,6.7974982261657715,0.8873386383056641,5.815634250640869,6.875983238220215,5.792854309082031,5.915988445281982,6.789846897125244,6.953376770019531,0.968585451007326,-0.793837702081877
step260410_final,spread,1000,0.832,0.848,0.052,832,168,0.273774653673172,0.5023994445800781,0.22862479090690613,0.2473444938659668,0.49890947341918945,0.23368358612060547,0.27810585498809814,0.39853525161743164,0.595841646194458,0.9267900068681318,-0.6832077001617288
|