| method,run_name,stage,global_train_steps,eval_accuracy,eval_greedy_accuracy,eval_sampled_pass_at_1,eval_sampled_pass_at_4,eval_avg_completion_length,eval_num_samples,eval_temperature,learning_rate,beta,temperature,max_completion_length,num_generations,train_reward_mean,train_reward_std,kl_mean,entropy_mean,avg_completion_length,completion_length_clip_ratio,grad_norm,last_loss,end_learning_rate,wall_clock_seconds |
| llm_controller,llm_seed42,-1,0,0.225,0.225,0.22,0.31,19.185,4,1.0,1e-06,0.04,1.0,512,16,,,,,,,,,, |
| llm_controller,llm_seed42,0,100,0.24,0.24,0.215,0.31,18.855,4,1.0,1e-06,0.04,1.0,512,16,0.5534133762121201,0.3936914769653231,0.0038538010309275,0.0573228343762457,19.9303125,0.0,3.161412366069853,0.102,2.6221547724253333e-10,78.17542266845703 |
| continuous_grpo,continuous_long_seed42,-1,0,0.225,0.225,0.22,0.31,19.185,4,1.0,1e-06,0.04,1.0,512,8,,,,,,,,,, |
| continuous_grpo,continuous_long_seed42,0,200,0.255,0.255,0.245,0.325,17.88,4,1.0,1e-06,0.04,1.0,512,8,0.5683742192387581,0.4761904725804925,0.019905953802372,0.0570064028352499,19.43875,0.0,2.534748571543023,-0.0074,9.269534157077176e-07,736.0401890277863 |
| continuous_grpo,continuous_long_seed42,1,400,0.25,0.25,0.255,0.34,17.585,4,1.0,1e-06,0.04,1.0,512,8,0.6293063132464886,0.4732742319442332,0.0578753219591453,0.0488568475656211,18.8446875,0.0,2.588927525356412,-0.0353,6.834598583458861e-07,736.0401890277863 |
| continuous_grpo,continuous_long_seed42,2,600,0.26,0.26,0.275,0.345,17.85,4,1.0,1e-06,0.04,1.0,512,8,0.6279417244344949,0.4787439277069643,0.0811165439139585,0.0476155159482732,18.9290625,0.0,2.8439276323188096,-0.0248,3.656440744218464e-07,736.0401890277863 |
| continuous_grpo,continuous_long_seed42,3,800,0.255,0.255,0.29,0.34,17.79,4,1.0,1e-06,0.04,1.0,512,8,0.636554714217782,0.4907465049438178,0.0867256929987343,0.050090711209923,19.14921875,0.0,3.523692013407126,-0.0905,1.0225780783651689e-07,736.0401890277863 |
| continuous_grpo,continuous_long_seed42,4,1000,0.265,0.265,0.31,0.355,17.83,4,1.0,1e-06,0.04,1.0,512,8,0.6424943773448467,0.4787482109293342,0.0795430200407281,0.0468580761831253,19.0875,0.0,2.7047918717749417,0.0519,2.622381702066523e-12,736.0401890277863 |
| fixed_grpo,staged_fixed_long_seed42,-1,0,0.225,0.225,0.22,0.31,19.185,4,1.0,1e-06,0.04,1.0,512,8,,,,,,,,,, |
| fixed_grpo,staged_fixed_long_seed42,0,200,0.24,0.24,0.22,0.32,18.26,4,1.0,1e-06,0.04,1.0,512,8,0.5643152790144086,0.481897279350087,0.0140436457589385,0.0582492147944867,19.48546875,0.0,2.4682861060230064,-0.0152,6.555816718389895e-11,151.95827078819275 |
| fixed_grpo,staged_fixed_long_seed42,1,400,0.28,0.28,0.265,0.335,17.24,4,1.0,1e-06,0.04,1.0,512,8,0.6174122439324856,0.4736468594521284,0.0249295311356763,0.0455461079161614,18.63484375,0.0,2.441798182804778,-0.0582,6.555816718389895e-11,154.1927354335785 |
| fixed_grpo,staged_fixed_long_seed42,2,600,0.305,0.305,0.285,0.355,17.19,4,1.0,1e-06,0.04,1.0,512,8,0.65621238283813,0.4776016463385895,0.0106469797887439,0.0384739167476072,18.3840625,0.0,2.1399336876044983,-0.0364,6.555816718389895e-11,152.39425325393677 |
| fixed_grpo,staged_fixed_long_seed42,3,800,0.325,0.325,0.32,0.365,17.135,4,1.0,1e-06,0.04,1.0,512,8,0.6746103132516146,0.4680243648821488,0.0103098346880506,0.0322844655648805,18.28046875,0.0,1.852596173787024,-0.0204,6.555816718389895e-11,151.4703402519226 |
| fixed_grpo,staged_fixed_long_seed42,4,1000,0.345,0.345,0.345,0.395,17.215,4,1.0,1e-06,0.04,1.0,512,8,0.6946119498461485,0.4774691890459507,0.0048778887826824,0.0295818717731162,18.26640625,0.0,1.721839446695021,-0.0187,6.555816718389895e-11,150.67903304100037 |
| continuous_grpo,continuous_seed42,-1,0,0.225,0.225,0.22,0.31,19.185,4,1.0,1e-06,0.04,1.0,512,16,,,,,,,,,, |
| continuous_grpo,continuous_seed42,0,100,0.23,0.23,0.21,0.33,18.4,4,1.0,1e-06,0.04,1.0,512,16,0.5592820857558399,0.3936004642350599,0.0055253497094963,0.0571631756611168,19.9084375,0.0,3.031685498040169,0.1504,9.277939410481504e-07,367.81041526794434 |
| continuous_grpo,continuous_seed42,1,200,0.245,0.245,0.24,0.315,17.575,4,1.0,1e-06,0.04,1.0,512,16,0.5603432653844357,0.3768988569499925,0.0434578162952675,0.0516500638332217,18.62375,0.0,3.0826165729481727,0.0004,6.84965323802926e-07,367.81041526794434 |
| continuous_grpo,continuous_seed42,2,300,0.25,0.25,0.255,0.33,17.45,4,1.0,1e-06,0.04,1.0,512,16,0.6014958855509758,0.3721435946645215,0.0705563474417431,0.0473552478384226,18.836875,0.0,2.995308451645542,-0.1043,3.6720459429194737e-07,367.81041526794434 |
| continuous_grpo,continuous_seed42,3,400,0.245,0.245,0.27,0.335,17.46,4,1.0,1e-06,0.04,1.0,512,16,0.6533837493509054,0.4060600923700258,0.0949894714378751,0.0499761152034625,18.7440625,0.0,3.511686566862627,0.0017,1.0324119308178164e-07,367.81041526794434 |
| continuous_grpo,continuous_seed42,4,500,0.255,0.255,0.265,0.335,17.495,4,1.0,1e-06,0.04,1.0,512,16,0.6201340295374393,0.3578025915240869,0.0849548888468416,0.0464471467025578,18.6953125,0.0,3.5394887678115627,0.0163,1.0489499300603276e-11,367.81041526794434 |
| llm_controller,llm_long_seed42,-1,0,0.225,0.225,0.22,0.31,19.185,4,1.0,1e-06,0.04,1.0,512,8,,,,,,,,,, |
| llm_controller,llm_long_seed42,0,200,0.245,0.245,0.215,0.32,18.055,4,1.0,1e-06,0.04,1.0,512,8,0.5587043429911137,0.4799258411489427,0.0176558254804695,0.0577014622092247,19.3825,0.0,2.5069461597222835,-0.027,6.555816718389895e-11,154.20320773124695 |
| llm_controller,llm_long_seed42,1,400,0.27,0.27,0.28,0.345,17.24,4,1.0,1.5e-06,0.04,1.1,512,8,0.6161690094321967,0.4686581644113176,0.0221556870336644,0.0487054003076627,18.64453125,0.0,2.187809831183404,0.0457,9.833725077584843e-11,157.2362425327301 |
| llm_controller,llm_long_seed42,2,600,0.31,0.31,0.305,0.365,17.21,4,1.0,2.25e-06,0.04,1.1,512,8,0.6659581090509892,0.4776299386844039,0.0209678140091818,0.0372779573895968,18.424375,0.0,2.01415314522892,0.111,1.4750587616377262e-10,152.71643114089966 |
| llm_controller,llm_long_seed42,3,800,0.34,0.34,0.355,0.4,17.19,4,1.0,2.25e-06,0.04,1.1,512,8,0.6902341424673796,0.4828688650066033,0.0100686291360761,0.0319678850285708,18.3790625,0.0,1.766346096615889,0.1076,1.4750587616377262e-10,152.10105562210083 |
| llm_controller,llm_long_seed42,4,1000,0.365,0.365,0.375,0.405,17.205,4,1.0,2.8125e-06,0.02,1.1,512,8,0.727456141859293,0.4916196739207953,0.0241176683301023,0.0270785682462155,18.3275,0.0,1.5239346842560872,0.1116,1.8438234520471582e-10,151.04308915138245 |
| fixed_grpo,smoke_staged,-1,0,0.2083333333333333,0.2083333333333333,0.1666666666666666,0.25,19.125,4,1.0,1e-06,0.04,1.0,512,8,,,,,,,,,, |
| fixed_grpo,smoke_staged,0,20,0.2083333333333333,0.2083333333333333,0.2083333333333333,0.25,18.95833333333333,4,1.0,1e-06,0.04,1.0,512,8,0.5500518798828125,0.5041159600019455,0.0012562903342768,0.061837051808834,20.2359375,0.0,2.210408035712317,-0.0239,6.8193482986388395e-09,20.10961961746216 |
| continuous_grpo,smoke_continuous,-1,0,0.2083333333333333,0.2083333333333333,0.1666666666666666,0.25,19.125,4,1.0,1e-06,0.04,1.0,512,16,,,,,,,,,, |
| continuous_grpo,smoke_continuous,0,20,0.2083333333333333,0.2083333333333333,0.25,0.25,19.08333333333333,4,1.0,1e-06,0.04,1.0,512,16,0.5614409944042563,0.4356262283399701,0.0013832212032866,0.0711977096274495,20.796875,0.0,3.41826611161232,-0.1065,5.82297295140367e-07,37.57142972946167 |
| continuous_grpo,smoke_continuous,1,40,0.2083333333333333,0.2083333333333333,0.2083333333333333,0.25,19.08333333333333,4,1.0,1e-06,0.04,1.0,512,16,0.535277726687491,0.3741546288132668,0.0013061563418887,0.0555895517580211,20.0140625,0.0,2.409226834657602,-0.0239,1.7077534966650763e-09,37.57142972946167 |
| fixed_grpo,staged_fixed_seed42,-1,0,0.225,0.225,0.22,0.31,19.185,4,1.0,1e-06,0.04,1.0,512,16,,,,,,,,,, |
| fixed_grpo,staged_fixed_seed42,0,100,0.24,0.24,0.215,0.31,18.855,4,1.0,1e-06,0.04,1.0,512,16,0.5534133762121201,0.3936914769653231,0.0038538010309275,0.0573228343762457,19.9303125,0.0,3.161412366069853,0.102,2.6221547724253333e-10,77.97688889503479 |
|
|
