1
2
3
4
5
6
7
8
9
10
|
label,sigma,n_samples,rollouts,ckpt_root,ckpt_name,perturb,noise_distribution,mean_rollout_exact,mean_rollout_token_acc,pass_at_k,all_k,correct_count_mean,correct_count_std,correct_count_q10,correct_count_q50,correct_count_q90,zero_frac,full_frac
trm_multi4_final,0.0,5000,8,/home/yurenh2/rrm/trm/checkpoints/Sudoku-extreme-1k-aug-1000-ACT-torch/pretrain_mlp_t_sudoku_official_gbs768_multi4_loguniform_repro,step_65100,both,gaussian,0.8452000021934509,0.9396741390228271,0.8452000021934509,0.8452000021934509,6.761600017547607,2.8937113285064697,0.0,8.0,8.0,0.15479999780654907,0.8452000021934509
trm_multi4_final,3e-05,5000,8,/home/yurenh2/rrm/trm/checkpoints/Sudoku-extreme-1k-aug-1000-ACT-torch/pretrain_mlp_t_sudoku_official_gbs768_multi4_loguniform_repro,step_65100,both,gaussian,0.840499997138977,0.9371709823608398,0.9114000201225281,0.7591999769210815,6.723999977111816,2.6249237060546875,1.0,8.0,8.0,0.08860000222921371,0.7591999769210815
trm_multi4_final,0.0001,5000,8,/home/yurenh2/rrm/trm/checkpoints/Sudoku-extreme-1k-aug-1000-ACT-torch/pretrain_mlp_t_sudoku_official_gbs768_multi4_loguniform_repro,step_65100,both,gaussian,0.8407999873161316,0.9373589754104614,0.9136000275611877,0.7577999830245972,6.726399898529053,2.612344264984131,1.0,8.0,8.0,0.08640000224113464,0.7577999830245972
trm_multi4_final,0.0003,5000,8,/home/yurenh2/rrm/trm/checkpoints/Sudoku-extreme-1k-aug-1000-ACT-torch/pretrain_mlp_t_sudoku_official_gbs768_multi4_loguniform_repro,step_65100,both,gaussian,0.8399249911308289,0.9371064305305481,0.9092000126838684,0.7572000026702881,6.719399929046631,2.6245501041412354,1.0,8.0,8.0,0.09080000221729279,0.7572000026702881
trm_multi4_final,0.001,5000,8,/home/yurenh2/rrm/trm/checkpoints/Sudoku-extreme-1k-aug-1000-ACT-torch/pretrain_mlp_t_sudoku_official_gbs768_multi4_loguniform_repro,step_65100,both,gaussian,0.8385249972343445,0.9361668229103088,0.9089999794960022,0.7588000297546387,6.708199977874756,2.6401236057281494,1.0,8.0,8.0,0.09099999815225601,0.7588000297546387
trm_multi4_final,0.003,5000,8,/home/yurenh2/rrm/trm/checkpoints/Sudoku-extreme-1k-aug-1000-ACT-torch/pretrain_mlp_t_sudoku_official_gbs768_multi4_loguniform_repro,step_65100,both,gaussian,0.8406999707221985,0.9372812509536743,0.9165999889373779,0.753000020980835,6.725599765777588,2.6047465801239014,1.0,8.0,8.0,0.08340000361204147,0.753000020980835
trm_multi4_final,0.01,5000,8,/home/yurenh2/rrm/trm/checkpoints/Sudoku-extreme-1k-aug-1000-ACT-torch/pretrain_mlp_t_sudoku_official_gbs768_multi4_loguniform_repro,step_65100,both,gaussian,0.838100016117096,0.9362147450447083,0.9132000207901001,0.7477999925613403,6.704800128936768,2.612366199493408,1.0,8.0,8.0,0.0868000015616417,0.7477999925613403
trm_multi4_final,0.03,5000,8,/home/yurenh2/rrm/trm/checkpoints/Sudoku-extreme-1k-aug-1000-ACT-torch/pretrain_mlp_t_sudoku_official_gbs768_multi4_loguniform_repro,step_65100,both,gaussian,0.8380249738693237,0.9364882707595825,0.9197999835014343,0.7378000020980835,6.70419979095459,2.575946807861328,1.0,8.0,8.0,0.08020000159740448,0.7378000020980835
trm_multi4_final,0.1,5000,8,/home/yurenh2/rrm/trm/checkpoints/Sudoku-extreme-1k-aug-1000-ACT-torch/pretrain_mlp_t_sudoku_official_gbs768_multi4_loguniform_repro,step_65100,both,gaussian,0.841949999332428,0.9376598596572876,0.9305999875068665,0.7305999994277954,6.735599994659424,2.5063304901123047,2.0,8.0,8.0,0.06939999759197235,0.7305999994277954
|