2035 lines
54 KiB
JSON
2035 lines
54 KiB
JSON
|
{
|
||
|
|
"best_global_step": null,
|
||
|
|
"best_metric": null,
|
||
|
|
"best_model_checkpoint": null,
|
||
|
|
"epoch": 0.05591590248266607,
|
||
|
|
"eval_steps": 500,
|
||
|
|
"global_step": 1000,
|
||
|
|
"is_hyper_param_search": false,
|
||
|
|
"is_local_process_zero": true,
|
||
|
|
"is_world_process_zero": true,
|
||
|
|
"log_history": [
|
||
|
|
{
|
||
|
|
"entropy": 10.742604160308838,
|
||
|
|
"epoch": 0.00027957951241333037,
|
||
|
|
"grad_norm": 5.21875,
|
||
|
|
"learning_rate": 2e-06,
|
||
|
|
"loss": 10.7452,
|
||
|
|
"mean_token_accuracy": 0.0002416299073956907,
|
||
|
|
"num_tokens": 8997.0,
|
||
|
|
"step": 5
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.742623710632325,
|
||
|
|
"epoch": 0.0005591590248266607,
|
||
|
|
"grad_norm": 5.46875,
|
||
|
|
"learning_rate": 4.5e-06,
|
||
|
|
"loss": 10.7347,
|
||
|
|
"mean_token_accuracy": 0.000248673016903922,
|
||
|
|
"num_tokens": 18058.0,
|
||
|
|
"step": 10
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.742646789550781,
|
||
|
|
"epoch": 0.0008387385372399911,
|
||
|
|
"grad_norm": 4.84375,
|
||
|
|
"learning_rate": 7e-06,
|
||
|
|
"loss": 10.7211,
|
||
|
|
"mean_token_accuracy": 6.249999860301614e-05,
|
||
|
|
"num_tokens": 27988.0,
|
||
|
|
"step": 15
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.742666816711425,
|
||
|
|
"epoch": 0.0011183180496533215,
|
||
|
|
"grad_norm": 5.25,
|
||
|
|
"learning_rate": 9.5e-06,
|
||
|
|
"loss": 10.6549,
|
||
|
|
"mean_token_accuracy": 0.0008288750774227083,
|
||
|
|
"num_tokens": 36654.0,
|
||
|
|
"step": 20
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.742599487304688,
|
||
|
|
"epoch": 0.0013978975620666518,
|
||
|
|
"grad_norm": 4.53125,
|
||
|
|
"learning_rate": 1.2e-05,
|
||
|
|
"loss": 10.5707,
|
||
|
|
"mean_token_accuracy": 0.00845626472728327,
|
||
|
|
"num_tokens": 45807.0,
|
||
|
|
"step": 25
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.742288589477539,
|
||
|
|
"epoch": 0.0016774770744799822,
|
||
|
|
"grad_norm": 4.40625,
|
||
|
|
"learning_rate": 1.4500000000000002e-05,
|
||
|
|
"loss": 10.4473,
|
||
|
|
"mean_token_accuracy": 0.0340017668902874,
|
||
|
|
"num_tokens": 55528.0,
|
||
|
|
"step": 30
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.74121150970459,
|
||
|
|
"epoch": 0.0019570565868933126,
|
||
|
|
"grad_norm": 3.328125,
|
||
|
|
"learning_rate": 1.7000000000000003e-05,
|
||
|
|
"loss": 10.3292,
|
||
|
|
"mean_token_accuracy": 0.03626283407211304,
|
||
|
|
"num_tokens": 66262.0,
|
||
|
|
"step": 35
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.738728046417236,
|
||
|
|
"epoch": 0.002236636099306643,
|
||
|
|
"grad_norm": 2.671875,
|
||
|
|
"learning_rate": 1.95e-05,
|
||
|
|
"loss": 10.1506,
|
||
|
|
"mean_token_accuracy": 0.03918867856264115,
|
||
|
|
"num_tokens": 75145.0,
|
||
|
|
"step": 40
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.73520622253418,
|
||
|
|
"epoch": 0.0025162156117199733,
|
||
|
|
"grad_norm": 2.328125,
|
||
|
|
"learning_rate": 2.2e-05,
|
||
|
|
"loss": 10.0379,
|
||
|
|
"mean_token_accuracy": 0.043211689591407774,
|
||
|
|
"num_tokens": 84519.0,
|
||
|
|
"step": 45
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.732088565826416,
|
||
|
|
"epoch": 0.0027957951241333037,
|
||
|
|
"grad_norm": 2.109375,
|
||
|
|
"learning_rate": 2.4500000000000003e-05,
|
||
|
|
"loss": 9.9595,
|
||
|
|
"mean_token_accuracy": 0.041159269958734514,
|
||
|
|
"num_tokens": 93968.0,
|
||
|
|
"step": 50
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.730504989624023,
|
||
|
|
"epoch": 0.003075374636546634,
|
||
|
|
"grad_norm": 2.046875,
|
||
|
|
"learning_rate": 2.7e-05,
|
||
|
|
"loss": 9.8985,
|
||
|
|
"mean_token_accuracy": 0.04068988226354122,
|
||
|
|
"num_tokens": 103228.0,
|
||
|
|
"step": 55
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.729658222198486,
|
||
|
|
"epoch": 0.0033549541489599644,
|
||
|
|
"grad_norm": 2.203125,
|
||
|
|
"learning_rate": 2.95e-05,
|
||
|
|
"loss": 9.8123,
|
||
|
|
"mean_token_accuracy": 0.04634873028844595,
|
||
|
|
"num_tokens": 112382.0,
|
||
|
|
"step": 60
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.72733097076416,
|
||
|
|
"epoch": 0.0036345336613732944,
|
||
|
|
"grad_norm": 1.8671875,
|
||
|
|
"learning_rate": 3.2e-05,
|
||
|
|
"loss": 9.8228,
|
||
|
|
"mean_token_accuracy": 0.04002052750438452,
|
||
|
|
"num_tokens": 123006.0,
|
||
|
|
"step": 65
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.724089336395263,
|
||
|
|
"epoch": 0.003914113173786625,
|
||
|
|
"grad_norm": 1.8515625,
|
||
|
|
"learning_rate": 3.4500000000000005e-05,
|
||
|
|
"loss": 9.7301,
|
||
|
|
"mean_token_accuracy": 0.03836380410939455,
|
||
|
|
"num_tokens": 132323.0,
|
||
|
|
"step": 70
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.720303344726563,
|
||
|
|
"epoch": 0.004193692686199955,
|
||
|
|
"grad_norm": 2.046875,
|
||
|
|
"learning_rate": 3.7e-05,
|
||
|
|
"loss": 9.6339,
|
||
|
|
"mean_token_accuracy": 0.0462482463568449,
|
||
|
|
"num_tokens": 142668.0,
|
||
|
|
"step": 75
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.713562965393066,
|
||
|
|
"epoch": 0.004473272198613286,
|
||
|
|
"grad_norm": 2.03125,
|
||
|
|
"learning_rate": 3.95e-05,
|
||
|
|
"loss": 9.5438,
|
||
|
|
"mean_token_accuracy": 0.04494307301938534,
|
||
|
|
"num_tokens": 152291.0,
|
||
|
|
"step": 80
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.705206394195557,
|
||
|
|
"epoch": 0.004752851711026616,
|
||
|
|
"grad_norm": 2.03125,
|
||
|
|
"learning_rate": 4.2000000000000004e-05,
|
||
|
|
"loss": 9.4727,
|
||
|
|
"mean_token_accuracy": 0.04591982215642929,
|
||
|
|
"num_tokens": 162443.0,
|
||
|
|
"step": 85
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.695830249786377,
|
||
|
|
"epoch": 0.005032431223439947,
|
||
|
|
"grad_norm": 1.9296875,
|
||
|
|
"learning_rate": 4.45e-05,
|
||
|
|
"loss": 9.4692,
|
||
|
|
"mean_token_accuracy": 0.037067833729088305,
|
||
|
|
"num_tokens": 172011.0,
|
||
|
|
"step": 90
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.689439678192139,
|
||
|
|
"epoch": 0.005312010735853277,
|
||
|
|
"grad_norm": 1.9765625,
|
||
|
|
"learning_rate": 4.7000000000000004e-05,
|
||
|
|
"loss": 9.3015,
|
||
|
|
"mean_token_accuracy": 0.04778099395334721,
|
||
|
|
"num_tokens": 180882.0,
|
||
|
|
"step": 95
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.677646255493164,
|
||
|
|
"epoch": 0.005591590248266607,
|
||
|
|
"grad_norm": 1.9375,
|
||
|
|
"learning_rate": 4.9500000000000004e-05,
|
||
|
|
"loss": 9.2454,
|
||
|
|
"mean_token_accuracy": 0.05362424850463867,
|
||
|
|
"num_tokens": 190318.0,
|
||
|
|
"step": 100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.660746574401855,
|
||
|
|
"epoch": 0.005871169760679937,
|
||
|
|
"grad_norm": 2.046875,
|
||
|
|
"learning_rate": 5.2e-05,
|
||
|
|
"loss": 9.114,
|
||
|
|
"mean_token_accuracy": 0.05184634737670422,
|
||
|
|
"num_tokens": 199369.0,
|
||
|
|
"step": 105
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.633990478515624,
|
||
|
|
"epoch": 0.006150749273093268,
|
||
|
|
"grad_norm": 1.8125,
|
||
|
|
"learning_rate": 5.45e-05,
|
||
|
|
"loss": 9.0025,
|
||
|
|
"mean_token_accuracy": 0.0516359206289053,
|
||
|
|
"num_tokens": 208848.0,
|
||
|
|
"step": 110
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.58621587753296,
|
||
|
|
"epoch": 0.006430328785506598,
|
||
|
|
"grad_norm": 1.7578125,
|
||
|
|
"learning_rate": 5.7e-05,
|
||
|
|
"loss": 8.9071,
|
||
|
|
"mean_token_accuracy": 0.05344265438616276,
|
||
|
|
"num_tokens": 218285.0,
|
||
|
|
"step": 115
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.522095870971679,
|
||
|
|
"epoch": 0.006709908297919929,
|
||
|
|
"grad_norm": 1.6953125,
|
||
|
|
"learning_rate": 5.9499999999999996e-05,
|
||
|
|
"loss": 8.747,
|
||
|
|
"mean_token_accuracy": 0.05776390843093395,
|
||
|
|
"num_tokens": 227033.0,
|
||
|
|
"step": 120
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.44651231765747,
|
||
|
|
"epoch": 0.006989487810333259,
|
||
|
|
"grad_norm": 1.7109375,
|
||
|
|
"learning_rate": 6.2e-05,
|
||
|
|
"loss": 8.5747,
|
||
|
|
"mean_token_accuracy": 0.05609895884990692,
|
||
|
|
"num_tokens": 236020.0,
|
||
|
|
"step": 125
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.386964702606202,
|
||
|
|
"epoch": 0.007269067322746589,
|
||
|
|
"grad_norm": 1.6171875,
|
||
|
|
"learning_rate": 6.450000000000001e-05,
|
||
|
|
"loss": 8.5067,
|
||
|
|
"mean_token_accuracy": 0.05812028683722019,
|
||
|
|
"num_tokens": 244996.0,
|
||
|
|
"step": 130
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.300998210906982,
|
||
|
|
"epoch": 0.0075486468351599195,
|
||
|
|
"grad_norm": 1.5546875,
|
||
|
|
"learning_rate": 6.7e-05,
|
||
|
|
"loss": 8.3979,
|
||
|
|
"mean_token_accuracy": 0.05419078357517719,
|
||
|
|
"num_tokens": 254721.0,
|
||
|
|
"step": 135
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.223974227905273,
|
||
|
|
"epoch": 0.00782822634757325,
|
||
|
|
"grad_norm": 1.4921875,
|
||
|
|
"learning_rate": 6.950000000000001e-05,
|
||
|
|
"loss": 8.1952,
|
||
|
|
"mean_token_accuracy": 0.05796922855079174,
|
||
|
|
"num_tokens": 264203.0,
|
||
|
|
"step": 140
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.118547725677491,
|
||
|
|
"epoch": 0.00810780585998658,
|
||
|
|
"grad_norm": 1.5390625,
|
||
|
|
"learning_rate": 7.2e-05,
|
||
|
|
"loss": 8.1377,
|
||
|
|
"mean_token_accuracy": 0.062206852808594705,
|
||
|
|
"num_tokens": 274136.0,
|
||
|
|
"step": 145
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 9.95664005279541,
|
||
|
|
"epoch": 0.00838738537239991,
|
||
|
|
"grad_norm": 1.4296875,
|
||
|
|
"learning_rate": 7.45e-05,
|
||
|
|
"loss": 8.0921,
|
||
|
|
"mean_token_accuracy": 0.053653810545802115,
|
||
|
|
"num_tokens": 284089.0,
|
||
|
|
"step": 150
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 9.849724769592285,
|
||
|
|
"epoch": 0.008666964884813241,
|
||
|
|
"grad_norm": 1.3671875,
|
||
|
|
"learning_rate": 7.7e-05,
|
||
|
|
"loss": 7.8983,
|
||
|
|
"mean_token_accuracy": 0.05716096460819244,
|
||
|
|
"num_tokens": 292975.0,
|
||
|
|
"step": 155
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 9.631487941741943,
|
||
|
|
"epoch": 0.008946544397226572,
|
||
|
|
"grad_norm": 1.2734375,
|
||
|
|
"learning_rate": 7.950000000000001e-05,
|
||
|
|
"loss": 7.799,
|
||
|
|
"mean_token_accuracy": 0.06275118589401245,
|
||
|
|
"num_tokens": 301900.0,
|
||
|
|
"step": 160
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 9.396039676666259,
|
||
|
|
"epoch": 0.0092261239096399,
|
||
|
|
"grad_norm": 1.140625,
|
||
|
|
"learning_rate": 8.2e-05,
|
||
|
|
"loss": 7.73,
|
||
|
|
"mean_token_accuracy": 0.06139553301036358,
|
||
|
|
"num_tokens": 312052.0,
|
||
|
|
"step": 165
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 9.197185707092284,
|
||
|
|
"epoch": 0.009505703422053232,
|
||
|
|
"grad_norm": 1.015625,
|
||
|
|
"learning_rate": 8.450000000000001e-05,
|
||
|
|
"loss": 7.4949,
|
||
|
|
"mean_token_accuracy": 0.07359966151416301,
|
||
|
|
"num_tokens": 321834.0,
|
||
|
|
"step": 170
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.953872203826904,
|
||
|
|
"epoch": 0.009785282934466562,
|
||
|
|
"grad_norm": 1.03125,
|
||
|
|
"learning_rate": 8.7e-05,
|
||
|
|
"loss": 7.4616,
|
||
|
|
"mean_token_accuracy": 0.06581225953996181,
|
||
|
|
"num_tokens": 331806.0,
|
||
|
|
"step": 175
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.738122272491456,
|
||
|
|
"epoch": 0.010064862446879893,
|
||
|
|
"grad_norm": 1.0390625,
|
||
|
|
"learning_rate": 8.95e-05,
|
||
|
|
"loss": 7.401,
|
||
|
|
"mean_token_accuracy": 0.06802928820252419,
|
||
|
|
"num_tokens": 341015.0,
|
||
|
|
"step": 180
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.514873027801514,
|
||
|
|
"epoch": 0.010344441959293222,
|
||
|
|
"grad_norm": 0.96875,
|
||
|
|
"learning_rate": 9.2e-05,
|
||
|
|
"loss": 7.4106,
|
||
|
|
"mean_token_accuracy": 0.07147527374327182,
|
||
|
|
"num_tokens": 350653.0,
|
||
|
|
"step": 185
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.432024192810058,
|
||
|
|
"epoch": 0.010624021471706553,
|
||
|
|
"grad_norm": 1.265625,
|
||
|
|
"learning_rate": 9.45e-05,
|
||
|
|
"loss": 7.4886,
|
||
|
|
"mean_token_accuracy": 0.06500204354524612,
|
||
|
|
"num_tokens": 360090.0,
|
||
|
|
"step": 190
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.343926239013673,
|
||
|
|
"epoch": 0.010903600984119884,
|
||
|
|
"grad_norm": 0.953125,
|
||
|
|
"learning_rate": 9.7e-05,
|
||
|
|
"loss": 7.2512,
|
||
|
|
"mean_token_accuracy": 0.07501669861376285,
|
||
|
|
"num_tokens": 370265.0,
|
||
|
|
"step": 195
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.184742641448974,
|
||
|
|
"epoch": 0.011183180496533215,
|
||
|
|
"grad_norm": 1.109375,
|
||
|
|
"learning_rate": 9.95e-05,
|
||
|
|
"loss": 7.3226,
|
||
|
|
"mean_token_accuracy": 0.070186922326684,
|
||
|
|
"num_tokens": 379060.0,
|
||
|
|
"step": 200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.110335540771484,
|
||
|
|
"epoch": 0.011462760008946544,
|
||
|
|
"grad_norm": 1.0078125,
|
||
|
|
"learning_rate": 0.000102,
|
||
|
|
"loss": 7.2938,
|
||
|
|
"mean_token_accuracy": 0.0759833563119173,
|
||
|
|
"num_tokens": 388621.0,
|
||
|
|
"step": 205
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.189294242858887,
|
||
|
|
"epoch": 0.011742339521359875,
|
||
|
|
"grad_norm": 1.140625,
|
||
|
|
"learning_rate": 0.00010449999999999999,
|
||
|
|
"loss": 7.2676,
|
||
|
|
"mean_token_accuracy": 0.0727431409060955,
|
||
|
|
"num_tokens": 397708.0,
|
||
|
|
"step": 210
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.10356707572937,
|
||
|
|
"epoch": 0.012021919033773205,
|
||
|
|
"grad_norm": 0.99609375,
|
||
|
|
"learning_rate": 0.000107,
|
||
|
|
"loss": 7.2602,
|
||
|
|
"mean_token_accuracy": 0.07345507219433785,
|
||
|
|
"num_tokens": 406979.0,
|
||
|
|
"step": 215
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.034816122055053,
|
||
|
|
"epoch": 0.012301498546186536,
|
||
|
|
"grad_norm": 0.9375,
|
||
|
|
"learning_rate": 0.0001095,
|
||
|
|
"loss": 7.2885,
|
||
|
|
"mean_token_accuracy": 0.07478194162249566,
|
||
|
|
"num_tokens": 416150.0,
|
||
|
|
"step": 220
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.053417491912843,
|
||
|
|
"epoch": 0.012581078058599865,
|
||
|
|
"grad_norm": 0.9765625,
|
||
|
|
"learning_rate": 0.000112,
|
||
|
|
"loss": 7.161,
|
||
|
|
"mean_token_accuracy": 0.07471481338143349,
|
||
|
|
"num_tokens": 425052.0,
|
||
|
|
"step": 225
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.964312744140625,
|
||
|
|
"epoch": 0.012860657571013196,
|
||
|
|
"grad_norm": 0.98828125,
|
||
|
|
"learning_rate": 0.0001145,
|
||
|
|
"loss": 7.2054,
|
||
|
|
"mean_token_accuracy": 0.08025330156087876,
|
||
|
|
"num_tokens": 433950.0,
|
||
|
|
"step": 230
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.9365815162658695,
|
||
|
|
"epoch": 0.013140237083426527,
|
||
|
|
"grad_norm": 1.3671875,
|
||
|
|
"learning_rate": 0.00011700000000000001,
|
||
|
|
"loss": 7.178,
|
||
|
|
"mean_token_accuracy": 0.08213647454977036,
|
||
|
|
"num_tokens": 443400.0,
|
||
|
|
"step": 235
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.868668031692505,
|
||
|
|
"epoch": 0.013419816595839858,
|
||
|
|
"grad_norm": 1.6484375,
|
||
|
|
"learning_rate": 0.00011949999999999999,
|
||
|
|
"loss": 7.1882,
|
||
|
|
"mean_token_accuracy": 0.07630038633942604,
|
||
|
|
"num_tokens": 452965.0,
|
||
|
|
"step": 240
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.8286830425262455,
|
||
|
|
"epoch": 0.013699396108253187,
|
||
|
|
"grad_norm": 1.1171875,
|
||
|
|
"learning_rate": 0.000122,
|
||
|
|
"loss": 7.0823,
|
||
|
|
"mean_token_accuracy": 0.08161357119679451,
|
||
|
|
"num_tokens": 461926.0,
|
||
|
|
"step": 245
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.832862472534179,
|
||
|
|
"epoch": 0.013978975620666518,
|
||
|
|
"grad_norm": 1.015625,
|
||
|
|
"learning_rate": 0.0001245,
|
||
|
|
"loss": 7.0678,
|
||
|
|
"mean_token_accuracy": 0.08250153660774232,
|
||
|
|
"num_tokens": 471013.0,
|
||
|
|
"step": 250
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.846067714691162,
|
||
|
|
"epoch": 0.014258555133079848,
|
||
|
|
"grad_norm": 0.98828125,
|
||
|
|
"learning_rate": 0.000127,
|
||
|
|
"loss": 7.0425,
|
||
|
|
"mean_token_accuracy": 0.08530599400401115,
|
||
|
|
"num_tokens": 480800.0,
|
||
|
|
"step": 255
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.760735368728637,
|
||
|
|
"epoch": 0.014538134645493177,
|
||
|
|
"grad_norm": 1.046875,
|
||
|
|
"learning_rate": 0.0001295,
|
||
|
|
"loss": 7.0328,
|
||
|
|
"mean_token_accuracy": 0.0859468162059784,
|
||
|
|
"num_tokens": 489410.0,
|
||
|
|
"step": 260
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.729041671752929,
|
||
|
|
"epoch": 0.014817714157906508,
|
||
|
|
"grad_norm": 1.015625,
|
||
|
|
"learning_rate": 0.000132,
|
||
|
|
"loss": 7.0624,
|
||
|
|
"mean_token_accuracy": 0.08790867775678635,
|
||
|
|
"num_tokens": 498568.0,
|
||
|
|
"step": 265
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.717026662826538,
|
||
|
|
"epoch": 0.015097293670319839,
|
||
|
|
"grad_norm": 0.99609375,
|
||
|
|
"learning_rate": 0.00013450000000000002,
|
||
|
|
"loss": 6.9952,
|
||
|
|
"mean_token_accuracy": 0.08537343889474869,
|
||
|
|
"num_tokens": 508412.0,
|
||
|
|
"step": 270
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.722399425506592,
|
||
|
|
"epoch": 0.01537687318273317,
|
||
|
|
"grad_norm": 1.0546875,
|
||
|
|
"learning_rate": 0.00013700000000000002,
|
||
|
|
"loss": 6.9036,
|
||
|
|
"mean_token_accuracy": 0.08399934843182563,
|
||
|
|
"num_tokens": 516447.0,
|
||
|
|
"step": 275
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.711771678924561,
|
||
|
|
"epoch": 0.0156564526951465,
|
||
|
|
"grad_norm": 1.09375,
|
||
|
|
"learning_rate": 0.0001395,
|
||
|
|
"loss": 6.9704,
|
||
|
|
"mean_token_accuracy": 0.08819907456636429,
|
||
|
|
"num_tokens": 525352.0,
|
||
|
|
"step": 280
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.623401927947998,
|
||
|
|
"epoch": 0.01593603220755983,
|
||
|
|
"grad_norm": 1.1796875,
|
||
|
|
"learning_rate": 0.00014199999999999998,
|
||
|
|
"loss": 6.9797,
|
||
|
|
"mean_token_accuracy": 0.09028599634766579,
|
||
|
|
"num_tokens": 534827.0,
|
||
|
|
"step": 285
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.567380142211914,
|
||
|
|
"epoch": 0.01621561171997316,
|
||
|
|
"grad_norm": 0.984375,
|
||
|
|
"learning_rate": 0.0001445,
|
||
|
|
"loss": 6.8852,
|
||
|
|
"mean_token_accuracy": 0.09008300229907036,
|
||
|
|
"num_tokens": 543666.0,
|
||
|
|
"step": 290
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.5774678707122805,
|
||
|
|
"epoch": 0.01649519123238649,
|
||
|
|
"grad_norm": 0.953125,
|
||
|
|
"learning_rate": 0.000147,
|
||
|
|
"loss": 7.0362,
|
||
|
|
"mean_token_accuracy": 0.08438777178525925,
|
||
|
|
"num_tokens": 553296.0,
|
||
|
|
"step": 295
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.5840624332427975,
|
||
|
|
"epoch": 0.01677477074479982,
|
||
|
|
"grad_norm": 1.1015625,
|
||
|
|
"learning_rate": 0.0001495,
|
||
|
|
"loss": 6.9249,
|
||
|
|
"mean_token_accuracy": 0.08579359352588653,
|
||
|
|
"num_tokens": 562053.0,
|
||
|
|
"step": 300
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.535729122161865,
|
||
|
|
"epoch": 0.017054350257213153,
|
||
|
|
"grad_norm": 1.15625,
|
||
|
|
"learning_rate": 0.000152,
|
||
|
|
"loss": 6.8026,
|
||
|
|
"mean_token_accuracy": 0.08932173550128937,
|
||
|
|
"num_tokens": 571090.0,
|
||
|
|
"step": 305
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.51135892868042,
|
||
|
|
"epoch": 0.017333929769626482,
|
||
|
|
"grad_norm": 0.921875,
|
||
|
|
"learning_rate": 0.00015450000000000001,
|
||
|
|
"loss": 6.949,
|
||
|
|
"mean_token_accuracy": 0.08511770591139793,
|
||
|
|
"num_tokens": 580432.0,
|
||
|
|
"step": 310
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.45130934715271,
|
||
|
|
"epoch": 0.01761350928203981,
|
||
|
|
"grad_norm": 0.984375,
|
||
|
|
"learning_rate": 0.000157,
|
||
|
|
"loss": 6.7722,
|
||
|
|
"mean_token_accuracy": 0.09298585653305054,
|
||
|
|
"num_tokens": 589977.0,
|
||
|
|
"step": 315
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.5484930038452145,
|
||
|
|
"epoch": 0.017893088794453144,
|
||
|
|
"grad_norm": 1.140625,
|
||
|
|
"learning_rate": 0.0001595,
|
||
|
|
"loss": 6.9432,
|
||
|
|
"mean_token_accuracy": 0.08774355575442314,
|
||
|
|
"num_tokens": 599676.0,
|
||
|
|
"step": 320
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.364548540115356,
|
||
|
|
"epoch": 0.018172668306866473,
|
||
|
|
"grad_norm": 1.0234375,
|
||
|
|
"learning_rate": 0.000162,
|
||
|
|
"loss": 6.746,
|
||
|
|
"mean_token_accuracy": 0.09561119675636291,
|
||
|
|
"num_tokens": 609037.0,
|
||
|
|
"step": 325
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.475339651107788,
|
||
|
|
"epoch": 0.0184522478192798,
|
||
|
|
"grad_norm": 0.91796875,
|
||
|
|
"learning_rate": 0.00016450000000000001,
|
||
|
|
"loss": 6.8463,
|
||
|
|
"mean_token_accuracy": 0.09227285087108612,
|
||
|
|
"num_tokens": 618191.0,
|
||
|
|
"step": 330
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.462693548202514,
|
||
|
|
"epoch": 0.018731827331693134,
|
||
|
|
"grad_norm": 1.390625,
|
||
|
|
"learning_rate": 0.00016700000000000002,
|
||
|
|
"loss": 6.8617,
|
||
|
|
"mean_token_accuracy": 0.08824969157576561,
|
||
|
|
"num_tokens": 627463.0,
|
||
|
|
"step": 335
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.467849254608154,
|
||
|
|
"epoch": 0.019011406844106463,
|
||
|
|
"grad_norm": 1.1171875,
|
||
|
|
"learning_rate": 0.00016950000000000003,
|
||
|
|
"loss": 6.848,
|
||
|
|
"mean_token_accuracy": 0.09017552435398102,
|
||
|
|
"num_tokens": 636421.0,
|
||
|
|
"step": 340
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.368214416503906,
|
||
|
|
"epoch": 0.019290986356519796,
|
||
|
|
"grad_norm": 0.95703125,
|
||
|
|
"learning_rate": 0.00017199999999999998,
|
||
|
|
"loss": 6.848,
|
||
|
|
"mean_token_accuracy": 0.08877173736691475,
|
||
|
|
"num_tokens": 646378.0,
|
||
|
|
"step": 345
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.436206865310669,
|
||
|
|
"epoch": 0.019570565868933125,
|
||
|
|
"grad_norm": 1.2265625,
|
||
|
|
"learning_rate": 0.00017449999999999999,
|
||
|
|
"loss": 6.8486,
|
||
|
|
"mean_token_accuracy": 0.08993291035294533,
|
||
|
|
"num_tokens": 655706.0,
|
||
|
|
"step": 350
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.3630943775177,
|
||
|
|
"epoch": 0.019850145381346454,
|
||
|
|
"grad_norm": 1.109375,
|
||
|
|
"learning_rate": 0.000177,
|
||
|
|
"loss": 6.8057,
|
||
|
|
"mean_token_accuracy": 0.09233427047729492,
|
||
|
|
"num_tokens": 664903.0,
|
||
|
|
"step": 355
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.301582717895508,
|
||
|
|
"epoch": 0.020129724893759787,
|
||
|
|
"grad_norm": 1.0625,
|
||
|
|
"learning_rate": 0.0001795,
|
||
|
|
"loss": 6.7755,
|
||
|
|
"mean_token_accuracy": 0.09259092882275581,
|
||
|
|
"num_tokens": 674144.0,
|
||
|
|
"step": 360
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.440686845779419,
|
||
|
|
"epoch": 0.020409304406173116,
|
||
|
|
"grad_norm": 1.203125,
|
||
|
|
"learning_rate": 0.000182,
|
||
|
|
"loss": 6.7231,
|
||
|
|
"mean_token_accuracy": 0.09879153519868851,
|
||
|
|
"num_tokens": 682427.0,
|
||
|
|
"step": 365
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.2801062107086185,
|
||
|
|
"epoch": 0.020688883918586445,
|
||
|
|
"grad_norm": 1.0703125,
|
||
|
|
"learning_rate": 0.0001845,
|
||
|
|
"loss": 6.8196,
|
||
|
|
"mean_token_accuracy": 0.09161863029003144,
|
||
|
|
"num_tokens": 691388.0,
|
||
|
|
"step": 370
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.400664806365967,
|
||
|
|
"epoch": 0.020968463430999777,
|
||
|
|
"grad_norm": 1.2109375,
|
||
|
|
"learning_rate": 0.000187,
|
||
|
|
"loss": 6.7844,
|
||
|
|
"mean_token_accuracy": 0.09717175960540772,
|
||
|
|
"num_tokens": 701182.0,
|
||
|
|
"step": 375
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.355081558227539,
|
||
|
|
"epoch": 0.021248042943413106,
|
||
|
|
"grad_norm": 1.09375,
|
||
|
|
"learning_rate": 0.0001895,
|
||
|
|
"loss": 6.7637,
|
||
|
|
"mean_token_accuracy": 0.0951042465865612,
|
||
|
|
"num_tokens": 710348.0,
|
||
|
|
"step": 380
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.235377836227417,
|
||
|
|
"epoch": 0.021527622455826435,
|
||
|
|
"grad_norm": 0.9921875,
|
||
|
|
"learning_rate": 0.000192,
|
||
|
|
"loss": 6.7512,
|
||
|
|
"mean_token_accuracy": 0.0942377932369709,
|
||
|
|
"num_tokens": 720718.0,
|
||
|
|
"step": 385
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.2696562767028805,
|
||
|
|
"epoch": 0.021807201968239768,
|
||
|
|
"grad_norm": 1.25,
|
||
|
|
"learning_rate": 0.0001945,
|
||
|
|
"loss": 6.7213,
|
||
|
|
"mean_token_accuracy": 0.10067062899470329,
|
||
|
|
"num_tokens": 729450.0,
|
||
|
|
"step": 390
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.3472309589385985,
|
||
|
|
"epoch": 0.022086781480653097,
|
||
|
|
"grad_norm": 1.15625,
|
||
|
|
"learning_rate": 0.00019700000000000002,
|
||
|
|
"loss": 6.7288,
|
||
|
|
"mean_token_accuracy": 0.09962123259902,
|
||
|
|
"num_tokens": 738474.0,
|
||
|
|
"step": 395
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.205544471740723,
|
||
|
|
"epoch": 0.02236636099306643,
|
||
|
|
"grad_norm": 1.046875,
|
||
|
|
"learning_rate": 0.00019950000000000002,
|
||
|
|
"loss": 6.7218,
|
||
|
|
"mean_token_accuracy": 0.09991101995110512,
|
||
|
|
"num_tokens": 748305.0,
|
||
|
|
"step": 400
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.273484086990356,
|
||
|
|
"epoch": 0.02264594050547976,
|
||
|
|
"grad_norm": 0.984375,
|
||
|
|
"learning_rate": 0.000202,
|
||
|
|
"loss": 6.7688,
|
||
|
|
"mean_token_accuracy": 0.09299318492412567,
|
||
|
|
"num_tokens": 758173.0,
|
||
|
|
"step": 405
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.260233402252197,
|
||
|
|
"epoch": 0.022925520017893088,
|
||
|
|
"grad_norm": 1.0703125,
|
||
|
|
"learning_rate": 0.00020449999999999998,
|
||
|
|
"loss": 6.806,
|
||
|
|
"mean_token_accuracy": 0.08802814856171608,
|
||
|
|
"num_tokens": 768082.0,
|
||
|
|
"step": 410
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.269196557998657,
|
||
|
|
"epoch": 0.02320509953030642,
|
||
|
|
"grad_norm": 1.09375,
|
||
|
|
"learning_rate": 0.000207,
|
||
|
|
"loss": 6.7521,
|
||
|
|
"mean_token_accuracy": 0.08867477178573609,
|
||
|
|
"num_tokens": 778575.0,
|
||
|
|
"step": 415
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.175719738006592,
|
||
|
|
"epoch": 0.02348467904271975,
|
||
|
|
"grad_norm": 1.046875,
|
||
|
|
"learning_rate": 0.0002095,
|
||
|
|
"loss": 6.6724,
|
||
|
|
"mean_token_accuracy": 0.0951958455145359,
|
||
|
|
"num_tokens": 788164.0,
|
||
|
|
"step": 420
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.268473434448242,
|
||
|
|
"epoch": 0.02376425855513308,
|
||
|
|
"grad_norm": 0.9375,
|
||
|
|
"learning_rate": 0.000212,
|
||
|
|
"loss": 6.6384,
|
||
|
|
"mean_token_accuracy": 0.10065982788801194,
|
||
|
|
"num_tokens": 798028.0,
|
||
|
|
"step": 425
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.174567365646363,
|
||
|
|
"epoch": 0.02404383806754641,
|
||
|
|
"grad_norm": 1.09375,
|
||
|
|
"learning_rate": 0.0002145,
|
||
|
|
"loss": 6.6485,
|
||
|
|
"mean_token_accuracy": 0.09755216687917709,
|
||
|
|
"num_tokens": 807540.0,
|
||
|
|
"step": 430
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.178504228591919,
|
||
|
|
"epoch": 0.02432341757995974,
|
||
|
|
"grad_norm": 1.0625,
|
||
|
|
"learning_rate": 0.00021700000000000002,
|
||
|
|
"loss": 6.6809,
|
||
|
|
"mean_token_accuracy": 0.0978812776505947,
|
||
|
|
"num_tokens": 817649.0,
|
||
|
|
"step": 435
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.214717721939087,
|
||
|
|
"epoch": 0.024602997092373072,
|
||
|
|
"grad_norm": 1.171875,
|
||
|
|
"learning_rate": 0.0002195,
|
||
|
|
"loss": 6.6933,
|
||
|
|
"mean_token_accuracy": 0.09935645759105682,
|
||
|
|
"num_tokens": 826410.0,
|
||
|
|
"step": 440
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.2163246154785154,
|
||
|
|
"epoch": 0.0248825766047864,
|
||
|
|
"grad_norm": 1.140625,
|
||
|
|
"learning_rate": 0.000222,
|
||
|
|
"loss": 6.7444,
|
||
|
|
"mean_token_accuracy": 0.09374995231628418,
|
||
|
|
"num_tokens": 836299.0,
|
||
|
|
"step": 445
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.1747071743011475,
|
||
|
|
"epoch": 0.02516215611719973,
|
||
|
|
"grad_norm": 1.15625,
|
||
|
|
"learning_rate": 0.0002245,
|
||
|
|
"loss": 6.5742,
|
||
|
|
"mean_token_accuracy": 0.1049035795032978,
|
||
|
|
"num_tokens": 845199.0,
|
||
|
|
"step": 450
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.132574033737183,
|
||
|
|
"epoch": 0.025441735629613063,
|
||
|
|
"grad_norm": 1.0234375,
|
||
|
|
"learning_rate": 0.00022700000000000002,
|
||
|
|
"loss": 6.5755,
|
||
|
|
"mean_token_accuracy": 0.09809941202402114,
|
||
|
|
"num_tokens": 854235.0,
|
||
|
|
"step": 455
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.1369490146636965,
|
||
|
|
"epoch": 0.025721315142026392,
|
||
|
|
"grad_norm": 1.25,
|
||
|
|
"learning_rate": 0.00022950000000000002,
|
||
|
|
"loss": 6.6791,
|
||
|
|
"mean_token_accuracy": 0.09706456884741783,
|
||
|
|
"num_tokens": 862540.0,
|
||
|
|
"step": 460
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.143648719787597,
|
||
|
|
"epoch": 0.02600089465443972,
|
||
|
|
"grad_norm": 1.0234375,
|
||
|
|
"learning_rate": 0.00023200000000000003,
|
||
|
|
"loss": 6.5617,
|
||
|
|
"mean_token_accuracy": 0.09988364353775978,
|
||
|
|
"num_tokens": 871352.0,
|
||
|
|
"step": 465
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.129497766494751,
|
||
|
|
"epoch": 0.026280474166853054,
|
||
|
|
"grad_norm": 1.0703125,
|
||
|
|
"learning_rate": 0.00023449999999999998,
|
||
|
|
"loss": 6.5662,
|
||
|
|
"mean_token_accuracy": 0.10278114005923271,
|
||
|
|
"num_tokens": 880004.0,
|
||
|
|
"step": 470
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.0246419429779055,
|
||
|
|
"epoch": 0.026560053679266383,
|
||
|
|
"grad_norm": 0.94140625,
|
||
|
|
"learning_rate": 0.000237,
|
||
|
|
"loss": 6.5808,
|
||
|
|
"mean_token_accuracy": 0.09790191277861596,
|
||
|
|
"num_tokens": 890392.0,
|
||
|
|
"step": 475
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.201833772659302,
|
||
|
|
"epoch": 0.026839633191679715,
|
||
|
|
"grad_norm": 1.1015625,
|
||
|
|
"learning_rate": 0.0002395,
|
||
|
|
"loss": 6.6187,
|
||
|
|
"mean_token_accuracy": 0.09889682978391648,
|
||
|
|
"num_tokens": 899442.0,
|
||
|
|
"step": 480
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.035550785064697,
|
||
|
|
"epoch": 0.027119212704093044,
|
||
|
|
"grad_norm": 1.1953125,
|
||
|
|
"learning_rate": 0.000242,
|
||
|
|
"loss": 6.5621,
|
||
|
|
"mean_token_accuracy": 0.10727965980768203,
|
||
|
|
"num_tokens": 907440.0,
|
||
|
|
"step": 485
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.085067796707153,
|
||
|
|
"epoch": 0.027398792216506374,
|
||
|
|
"grad_norm": 1.0625,
|
||
|
|
"learning_rate": 0.0002445,
|
||
|
|
"loss": 6.7462,
|
||
|
|
"mean_token_accuracy": 0.08954423442482948,
|
||
|
|
"num_tokens": 917734.0,
|
||
|
|
"step": 490
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.196885871887207,
|
||
|
|
"epoch": 0.027678371728919706,
|
||
|
|
"grad_norm": 1.1328125,
|
||
|
|
"learning_rate": 0.000247,
|
||
|
|
"loss": 6.6752,
|
||
|
|
"mean_token_accuracy": 0.1043403372168541,
|
||
|
|
"num_tokens": 927717.0,
|
||
|
|
"step": 495
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.075096464157104,
|
||
|
|
"epoch": 0.027957951241333035,
|
||
|
|
"grad_norm": 0.9921875,
|
||
|
|
"learning_rate": 0.0002495,
|
||
|
|
"loss": 6.6284,
|
||
|
|
"mean_token_accuracy": 0.0997283287346363,
|
||
|
|
"num_tokens": 937385.0,
|
||
|
|
"step": 500
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.0907949924469,
|
||
|
|
"epoch": 0.028237530753746364,
|
||
|
|
"grad_norm": 1.109375,
|
||
|
|
"learning_rate": 0.000252,
|
||
|
|
"loss": 6.597,
|
||
|
|
"mean_token_accuracy": 0.09384197294712067,
|
||
|
|
"num_tokens": 946951.0,
|
||
|
|
"step": 505
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.053925037384033,
|
||
|
|
"epoch": 0.028517110266159697,
|
||
|
|
"grad_norm": 1.0390625,
|
||
|
|
"learning_rate": 0.0002545,
|
||
|
|
"loss": 6.5645,
|
||
|
|
"mean_token_accuracy": 0.09848013147711754,
|
||
|
|
"num_tokens": 957686.0,
|
||
|
|
"step": 510
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.000115585327149,
|
||
|
|
"epoch": 0.028796689778573026,
|
||
|
|
"grad_norm": 1.1484375,
|
||
|
|
"learning_rate": 0.000257,
|
||
|
|
"loss": 6.4729,
|
||
|
|
"mean_token_accuracy": 0.10326300486922264,
|
||
|
|
"num_tokens": 966610.0,
|
||
|
|
"step": 515
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.072527551651001,
|
||
|
|
"epoch": 0.029076269290986355,
|
||
|
|
"grad_norm": 1.0546875,
|
||
|
|
"learning_rate": 0.0002595,
|
||
|
|
"loss": 6.6002,
|
||
|
|
"mean_token_accuracy": 0.1036542683839798,
|
||
|
|
"num_tokens": 975073.0,
|
||
|
|
"step": 520
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.086858558654785,
|
||
|
|
"epoch": 0.029355848803399687,
|
||
|
|
"grad_norm": 1.0859375,
|
||
|
|
"learning_rate": 0.000262,
|
||
|
|
"loss": 6.5286,
|
||
|
|
"mean_token_accuracy": 0.10234683528542518,
|
||
|
|
"num_tokens": 985307.0,
|
||
|
|
"step": 525
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.068497848510742,
|
||
|
|
"epoch": 0.029635428315813016,
|
||
|
|
"grad_norm": 1.2890625,
|
||
|
|
"learning_rate": 0.00026450000000000003,
|
||
|
|
"loss": 6.4879,
|
||
|
|
"mean_token_accuracy": 0.10931713730096818,
|
||
|
|
"num_tokens": 994261.0,
|
||
|
|
"step": 530
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.861135959625244,
|
||
|
|
"epoch": 0.02991500782822635,
|
||
|
|
"grad_norm": 1.3359375,
|
||
|
|
"learning_rate": 0.00026700000000000004,
|
||
|
|
"loss": 6.5142,
|
||
|
|
"mean_token_accuracy": 0.10772426426410675,
|
||
|
|
"num_tokens": 1003321.0,
|
||
|
|
"step": 535
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.950206565856933,
|
||
|
|
"epoch": 0.030194587340639678,
|
||
|
|
"grad_norm": 1.15625,
|
||
|
|
"learning_rate": 0.00026950000000000005,
|
||
|
|
"loss": 6.5134,
|
||
|
|
"mean_token_accuracy": 0.11043959930539131,
|
||
|
|
"num_tokens": 1012712.0,
|
||
|
|
"step": 540
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.02567229270935,
|
||
|
|
"epoch": 0.030474166853053007,
|
||
|
|
"grad_norm": 1.234375,
|
||
|
|
"learning_rate": 0.00027200000000000005,
|
||
|
|
"loss": 6.508,
|
||
|
|
"mean_token_accuracy": 0.10081920623779297,
|
||
|
|
"num_tokens": 1022461.0,
|
||
|
|
"step": 545
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.014062595367432,
|
||
|
|
"epoch": 0.03075374636546634,
|
||
|
|
"grad_norm": 1.1640625,
|
||
|
|
"learning_rate": 0.0002745,
|
||
|
|
"loss": 6.5293,
|
||
|
|
"mean_token_accuracy": 0.09998132437467575,
|
||
|
|
"num_tokens": 1032857.0,
|
||
|
|
"step": 550
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.9176764488220215,
|
||
|
|
"epoch": 0.03103332587787967,
|
||
|
|
"grad_norm": 1.109375,
|
||
|
|
"learning_rate": 0.000277,
|
||
|
|
"loss": 6.4824,
|
||
|
|
"mean_token_accuracy": 0.10737242624163627,
|
||
|
|
"num_tokens": 1042492.0,
|
||
|
|
"step": 555
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.9386693954467775,
|
||
|
|
"epoch": 0.031312905390293,
|
||
|
|
"grad_norm": 1.1953125,
|
||
|
|
"learning_rate": 0.0002795,
|
||
|
|
"loss": 6.5167,
|
||
|
|
"mean_token_accuracy": 0.10647067874670028,
|
||
|
|
"num_tokens": 1052539.0,
|
||
|
|
"step": 560
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.965692949295044,
|
||
|
|
"epoch": 0.03159248490270633,
|
||
|
|
"grad_norm": 1.0859375,
|
||
|
|
"learning_rate": 0.00028199999999999997,
|
||
|
|
"loss": 6.4402,
|
||
|
|
"mean_token_accuracy": 0.10416494011878967,
|
||
|
|
"num_tokens": 1062018.0,
|
||
|
|
"step": 565
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.925234413146972,
|
||
|
|
"epoch": 0.03187206441511966,
|
||
|
|
"grad_norm": 1.390625,
|
||
|
|
"learning_rate": 0.0002845,
|
||
|
|
"loss": 6.4518,
|
||
|
|
"mean_token_accuracy": 0.11095392107963561,
|
||
|
|
"num_tokens": 1071409.0,
|
||
|
|
"step": 570
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.907749605178833,
|
||
|
|
"epoch": 0.03215164392753299,
|
||
|
|
"grad_norm": 1.1328125,
|
||
|
|
"learning_rate": 0.000287,
|
||
|
|
"loss": 6.4843,
|
||
|
|
"mean_token_accuracy": 0.10628700330853462,
|
||
|
|
"num_tokens": 1081928.0,
|
||
|
|
"step": 575
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.892273759841919,
|
||
|
|
"epoch": 0.03243122343994632,
|
||
|
|
"grad_norm": 1.046875,
|
||
|
|
"learning_rate": 0.0002895,
|
||
|
|
"loss": 6.4707,
|
||
|
|
"mean_token_accuracy": 0.10453314185142518,
|
||
|
|
"num_tokens": 1091161.0,
|
||
|
|
"step": 580
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.891652774810791,
|
||
|
|
"epoch": 0.032710802952359654,
|
||
|
|
"grad_norm": 1.2109375,
|
||
|
|
"learning_rate": 0.000292,
|
||
|
|
"loss": 6.3781,
|
||
|
|
"mean_token_accuracy": 0.1103638917207718,
|
||
|
|
"num_tokens": 1100256.0,
|
||
|
|
"step": 585
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.913065385818482,
|
||
|
|
"epoch": 0.03299038246477298,
|
||
|
|
"grad_norm": 1.546875,
|
||
|
|
"learning_rate": 0.0002945,
|
||
|
|
"loss": 6.5215,
|
||
|
|
"mean_token_accuracy": 0.10347851365804672,
|
||
|
|
"num_tokens": 1109052.0,
|
||
|
|
"step": 590
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.035890197753906,
|
||
|
|
"epoch": 0.03326996197718631,
|
||
|
|
"grad_norm": 1.21875,
|
||
|
|
"learning_rate": 0.000297,
|
||
|
|
"loss": 6.5428,
|
||
|
|
"mean_token_accuracy": 0.10616497248411179,
|
||
|
|
"num_tokens": 1118659.0,
|
||
|
|
"step": 595
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.874794387817383,
|
||
|
|
"epoch": 0.03354954148959964,
|
||
|
|
"grad_norm": 1.1015625,
|
||
|
|
"learning_rate": 0.0002995,
|
||
|
|
"loss": 6.4279,
|
||
|
|
"mean_token_accuracy": 0.1102647602558136,
|
||
|
|
"num_tokens": 1128273.0,
|
||
|
|
"step": 600
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.909904146194458,
|
||
|
|
"epoch": 0.03382912100201297,
|
||
|
|
"grad_norm": 1.1328125,
|
||
|
|
"learning_rate": 0.000302,
|
||
|
|
"loss": 6.4761,
|
||
|
|
"mean_token_accuracy": 0.10916157588362693,
|
||
|
|
"num_tokens": 1138036.0,
|
||
|
|
"step": 605
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.87423677444458,
|
||
|
|
"epoch": 0.034108700514426306,
|
||
|
|
"grad_norm": 1.2421875,
|
||
|
|
"learning_rate": 0.0003045,
|
||
|
|
"loss": 6.3715,
|
||
|
|
"mean_token_accuracy": 0.11365670934319497,
|
||
|
|
"num_tokens": 1147266.0,
|
||
|
|
"step": 610
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.87387900352478,
|
||
|
|
"epoch": 0.034388280026839635,
|
||
|
|
"grad_norm": 1.3046875,
|
||
|
|
"learning_rate": 0.000307,
|
||
|
|
"loss": 6.3922,
|
||
|
|
"mean_token_accuracy": 0.11108144894242286,
|
||
|
|
"num_tokens": 1156303.0,
|
||
|
|
"step": 615
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.776626634597778,
|
||
|
|
"epoch": 0.034667859539252964,
|
||
|
|
"grad_norm": 1.3046875,
|
||
|
|
"learning_rate": 0.0003095,
|
||
|
|
"loss": 6.4298,
|
||
|
|
"mean_token_accuracy": 0.11272333487868309,
|
||
|
|
"num_tokens": 1166127.0,
|
||
|
|
"step": 620
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.8758259296417235,
|
||
|
|
"epoch": 0.03494743905166629,
|
||
|
|
"grad_norm": 1.0859375,
|
||
|
|
"learning_rate": 0.000312,
|
||
|
|
"loss": 6.536,
|
||
|
|
"mean_token_accuracy": 0.10809575021266937,
|
||
|
|
"num_tokens": 1175540.0,
|
||
|
|
"step": 625
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.904007148742676,
|
||
|
|
"epoch": 0.03522701856407962,
|
||
|
|
"grad_norm": 1.2109375,
|
||
|
|
"learning_rate": 0.0003145,
|
||
|
|
"loss": 6.4,
|
||
|
|
"mean_token_accuracy": 0.11190460100769997,
|
||
|
|
"num_tokens": 1183883.0,
|
||
|
|
"step": 630
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.798130846023559,
|
||
|
|
"epoch": 0.03550659807649295,
|
||
|
|
"grad_norm": 1.1640625,
|
||
|
|
"learning_rate": 0.000317,
|
||
|
|
"loss": 6.3991,
|
||
|
|
"mean_token_accuracy": 0.1127168245613575,
|
||
|
|
"num_tokens": 1193327.0,
|
||
|
|
"step": 635
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.776614427566528,
|
||
|
|
"epoch": 0.03578617758890629,
|
||
|
|
"grad_norm": 1.171875,
|
||
|
|
"learning_rate": 0.0003195,
|
||
|
|
"loss": 6.3957,
|
||
|
|
"mean_token_accuracy": 0.11016259118914604,
|
||
|
|
"num_tokens": 1202477.0,
|
||
|
|
"step": 640
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.844618463516236,
|
||
|
|
"epoch": 0.036065757101319616,
|
||
|
|
"grad_norm": 1.09375,
|
||
|
|
"learning_rate": 0.000322,
|
||
|
|
"loss": 6.4907,
|
||
|
|
"mean_token_accuracy": 0.10283268168568611,
|
||
|
|
"num_tokens": 1213100.0,
|
||
|
|
"step": 645
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.734366703033447,
|
||
|
|
"epoch": 0.036345336613732945,
|
||
|
|
"grad_norm": 1.140625,
|
||
|
|
"learning_rate": 0.00032450000000000003,
|
||
|
|
"loss": 6.3229,
|
||
|
|
"mean_token_accuracy": 0.11083219647407531,
|
||
|
|
"num_tokens": 1222386.0,
|
||
|
|
"step": 650
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.851327800750733,
|
||
|
|
"epoch": 0.036624916126146274,
|
||
|
|
"grad_norm": 1.171875,
|
||
|
|
"learning_rate": 0.00032700000000000003,
|
||
|
|
"loss": 6.2995,
|
||
|
|
"mean_token_accuracy": 0.1127958782017231,
|
||
|
|
"num_tokens": 1233172.0,
|
||
|
|
"step": 655
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.731534004211426,
|
||
|
|
"epoch": 0.0369044956385596,
|
||
|
|
"grad_norm": 1.2109375,
|
||
|
|
"learning_rate": 0.00032950000000000004,
|
||
|
|
"loss": 6.2837,
|
||
|
|
"mean_token_accuracy": 0.11103828400373458,
|
||
|
|
"num_tokens": 1242477.0,
|
||
|
|
"step": 660
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.9175632953643795,
|
||
|
|
"epoch": 0.03718407515097294,
|
||
|
|
"grad_norm": 1.1953125,
|
||
|
|
"learning_rate": 0.00033200000000000005,
|
||
|
|
"loss": 6.5809,
|
||
|
|
"mean_token_accuracy": 0.11029751524329186,
|
||
|
|
"num_tokens": 1252952.0,
|
||
|
|
"step": 665
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.712823104858399,
|
||
|
|
"epoch": 0.03746365466338627,
|
||
|
|
"grad_norm": 1.2734375,
|
||
|
|
"learning_rate": 0.00033450000000000005,
|
||
|
|
"loss": 6.35,
|
||
|
|
"mean_token_accuracy": 0.11600286141037941,
|
||
|
|
"num_tokens": 1262523.0,
|
||
|
|
"step": 670
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.853010416030884,
|
||
|
|
"epoch": 0.0377432341757996,
|
||
|
|
"grad_norm": 1.21875,
|
||
|
|
"learning_rate": 0.000337,
|
||
|
|
"loss": 6.3884,
|
||
|
|
"mean_token_accuracy": 0.11264139488339424,
|
||
|
|
"num_tokens": 1271537.0,
|
||
|
|
"step": 675
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.746140670776367,
|
||
|
|
"epoch": 0.03802281368821293,
|
||
|
|
"grad_norm": 1.296875,
|
||
|
|
"learning_rate": 0.0003395,
|
||
|
|
"loss": 6.3091,
|
||
|
|
"mean_token_accuracy": 0.11562162712216377,
|
||
|
|
"num_tokens": 1280411.0,
|
||
|
|
"step": 680
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.73434042930603,
|
||
|
|
"epoch": 0.038302393200626256,
|
||
|
|
"grad_norm": 1.078125,
|
||
|
|
"learning_rate": 0.000342,
|
||
|
|
"loss": 6.3937,
|
||
|
|
"mean_token_accuracy": 0.11497898325324059,
|
||
|
|
"num_tokens": 1289152.0,
|
||
|
|
"step": 685
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.754694557189941,
|
||
|
|
"epoch": 0.03858197271303959,
|
||
|
|
"grad_norm": 1.1328125,
|
||
|
|
"learning_rate": 0.00034449999999999997,
|
||
|
|
"loss": 6.3379,
|
||
|
|
"mean_token_accuracy": 0.1117016352713108,
|
||
|
|
"num_tokens": 1298588.0,
|
||
|
|
"step": 690
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.776911306381225,
|
||
|
|
"epoch": 0.03886155222545292,
|
||
|
|
"grad_norm": 1.1796875,
|
||
|
|
"learning_rate": 0.000347,
|
||
|
|
"loss": 6.3681,
|
||
|
|
"mean_token_accuracy": 0.10708253011107445,
|
||
|
|
"num_tokens": 1308529.0,
|
||
|
|
"step": 695
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.681609010696411,
|
||
|
|
"epoch": 0.03914113173786625,
|
||
|
|
"grad_norm": 1.1171875,
|
||
|
|
"learning_rate": 0.0003495,
|
||
|
|
"loss": 6.3484,
|
||
|
|
"mean_token_accuracy": 0.11021188572049141,
|
||
|
|
"num_tokens": 1318228.0,
|
||
|
|
"step": 700
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.822549343109131,
|
||
|
|
"epoch": 0.03942071125027958,
|
||
|
|
"grad_norm": 1.125,
|
||
|
|
"learning_rate": 0.000352,
|
||
|
|
"loss": 6.4168,
|
||
|
|
"mean_token_accuracy": 0.11342558711767196,
|
||
|
|
"num_tokens": 1327855.0,
|
||
|
|
"step": 705
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.669567632675171,
|
||
|
|
"epoch": 0.03970029076269291,
|
||
|
|
"grad_norm": 1.1640625,
|
||
|
|
"learning_rate": 0.0003545,
|
||
|
|
"loss": 6.3314,
|
||
|
|
"mean_token_accuracy": 0.11251103654503822,
|
||
|
|
"num_tokens": 1337016.0,
|
||
|
|
"step": 710
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.677871179580689,
|
||
|
|
"epoch": 0.03997987027510624,
|
||
|
|
"grad_norm": 1.09375,
|
||
|
|
"learning_rate": 0.000357,
|
||
|
|
"loss": 6.1888,
|
||
|
|
"mean_token_accuracy": 0.12071979492902755,
|
||
|
|
"num_tokens": 1346829.0,
|
||
|
|
"step": 715
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.761476230621338,
|
||
|
|
"epoch": 0.04025944978751957,
|
||
|
|
"grad_norm": 1.0078125,
|
||
|
|
"learning_rate": 0.0003595,
|
||
|
|
"loss": 6.3965,
|
||
|
|
"mean_token_accuracy": 0.11669178158044816,
|
||
|
|
"num_tokens": 1355921.0,
|
||
|
|
"step": 720
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.733816766738892,
|
||
|
|
"epoch": 0.0405390292999329,
|
||
|
|
"grad_norm": 1.0859375,
|
||
|
|
"learning_rate": 0.000362,
|
||
|
|
"loss": 6.364,
|
||
|
|
"mean_token_accuracy": 0.11375537812709809,
|
||
|
|
"num_tokens": 1365443.0,
|
||
|
|
"step": 725
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.614108419418335,
|
||
|
|
"epoch": 0.04081860881234623,
|
||
|
|
"grad_norm": 1.28125,
|
||
|
|
"learning_rate": 0.0003645,
|
||
|
|
"loss": 6.2644,
|
||
|
|
"mean_token_accuracy": 0.11659296080470086,
|
||
|
|
"num_tokens": 1374804.0,
|
||
|
|
"step": 730
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.758727979660034,
|
||
|
|
"epoch": 0.04109818832475956,
|
||
|
|
"grad_norm": 1.109375,
|
||
|
|
"learning_rate": 0.000367,
|
||
|
|
"loss": 6.3118,
|
||
|
|
"mean_token_accuracy": 0.11417160928249359,
|
||
|
|
"num_tokens": 1384468.0,
|
||
|
|
"step": 735
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.5593489646911625,
|
||
|
|
"epoch": 0.04137776783717289,
|
||
|
|
"grad_norm": 1.15625,
|
||
|
|
"learning_rate": 0.0003695,
|
||
|
|
"loss": 6.3298,
|
||
|
|
"mean_token_accuracy": 0.11806502938270569,
|
||
|
|
"num_tokens": 1393590.0,
|
||
|
|
"step": 740
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.673466634750366,
|
||
|
|
"epoch": 0.041657347349586225,
|
||
|
|
"grad_norm": 1.125,
|
||
|
|
"learning_rate": 0.000372,
|
||
|
|
"loss": 6.3422,
|
||
|
|
"mean_token_accuracy": 0.1178791120648384,
|
||
|
|
"num_tokens": 1403776.0,
|
||
|
|
"step": 745
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.7786225318908695,
|
||
|
|
"epoch": 0.041936926861999554,
|
||
|
|
"grad_norm": 1.171875,
|
||
|
|
"learning_rate": 0.0003745,
|
||
|
|
"loss": 6.2333,
|
||
|
|
"mean_token_accuracy": 0.1166193075478077,
|
||
|
|
"num_tokens": 1413793.0,
|
||
|
|
"step": 750
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.492034578323365,
|
||
|
|
"epoch": 0.042216506374412884,
|
||
|
|
"grad_norm": 1.234375,
|
||
|
|
"learning_rate": 0.000377,
|
||
|
|
"loss": 6.2877,
|
||
|
|
"mean_token_accuracy": 0.11759566813707352,
|
||
|
|
"num_tokens": 1423449.0,
|
||
|
|
"step": 755
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.713007259368896,
|
||
|
|
"epoch": 0.04249608588682621,
|
||
|
|
"grad_norm": 1.1328125,
|
||
|
|
"learning_rate": 0.0003795,
|
||
|
|
"loss": 6.2932,
|
||
|
|
"mean_token_accuracy": 0.11593074426054954,
|
||
|
|
"num_tokens": 1432711.0,
|
||
|
|
"step": 760
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.684315824508667,
|
||
|
|
"epoch": 0.04277566539923954,
|
||
|
|
"grad_norm": 1.0078125,
|
||
|
|
"learning_rate": 0.000382,
|
||
|
|
"loss": 6.3212,
|
||
|
|
"mean_token_accuracy": 0.11897616535425186,
|
||
|
|
"num_tokens": 1443261.0,
|
||
|
|
"step": 765
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.577885389328003,
|
||
|
|
"epoch": 0.04305524491165287,
|
||
|
|
"grad_norm": 0.9296875,
|
||
|
|
"learning_rate": 0.0003845,
|
||
|
|
"loss": 6.3044,
|
||
|
|
"mean_token_accuracy": 0.11995612978935241,
|
||
|
|
"num_tokens": 1453896.0,
|
||
|
|
"step": 770
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.6321838855743405,
|
||
|
|
"epoch": 0.04333482442406621,
|
||
|
|
"grad_norm": 1.1875,
|
||
|
|
"learning_rate": 0.00038700000000000003,
|
||
|
|
"loss": 6.2049,
|
||
|
|
"mean_token_accuracy": 0.12229073345661164,
|
||
|
|
"num_tokens": 1463772.0,
|
||
|
|
"step": 775
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.678547048568726,
|
||
|
|
"epoch": 0.043614403936479536,
|
||
|
|
"grad_norm": 1.1171875,
|
||
|
|
"learning_rate": 0.00038950000000000003,
|
||
|
|
"loss": 6.3328,
|
||
|
|
"mean_token_accuracy": 0.1103480413556099,
|
||
|
|
"num_tokens": 1473230.0,
|
||
|
|
"step": 780
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.661152935028076,
|
||
|
|
"epoch": 0.043893983448892865,
|
||
|
|
"grad_norm": 1.1015625,
|
||
|
|
"learning_rate": 0.00039200000000000004,
|
||
|
|
"loss": 6.2652,
|
||
|
|
"mean_token_accuracy": 0.11327114403247833,
|
||
|
|
"num_tokens": 1482585.0,
|
||
|
|
"step": 785
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.591106700897217,
|
||
|
|
"epoch": 0.044173562961306194,
|
||
|
|
"grad_norm": 1.125,
|
||
|
|
"learning_rate": 0.00039450000000000005,
|
||
|
|
"loss": 6.2406,
|
||
|
|
"mean_token_accuracy": 0.11883526220917702,
|
||
|
|
"num_tokens": 1491950.0,
|
||
|
|
"step": 790
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.607058382034301,
|
||
|
|
"epoch": 0.04445314247371952,
|
||
|
|
"grad_norm": 1.1796875,
|
||
|
|
"learning_rate": 0.00039700000000000005,
|
||
|
|
"loss": 6.1899,
|
||
|
|
"mean_token_accuracy": 0.1208645947277546,
|
||
|
|
"num_tokens": 1501156.0,
|
||
|
|
"step": 795
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.541710615158081,
|
||
|
|
"epoch": 0.04473272198613286,
|
||
|
|
"grad_norm": 1.234375,
|
||
|
|
"learning_rate": 0.0003995,
|
||
|
|
"loss": 6.2352,
|
||
|
|
"mean_token_accuracy": 0.12574194967746735,
|
||
|
|
"num_tokens": 1510358.0,
|
||
|
|
"step": 800
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.601570081710816,
|
||
|
|
"epoch": 0.04501230149854619,
|
||
|
|
"grad_norm": 1.25,
|
||
|
|
"learning_rate": 0.000402,
|
||
|
|
"loss": 6.2555,
|
||
|
|
"mean_token_accuracy": 0.12034472450613976,
|
||
|
|
"num_tokens": 1519536.0,
|
||
|
|
"step": 805
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.565725421905517,
|
||
|
|
"epoch": 0.04529188101095952,
|
||
|
|
"grad_norm": 1.0859375,
|
||
|
|
"learning_rate": 0.0004045,
|
||
|
|
"loss": 6.2339,
|
||
|
|
"mean_token_accuracy": 0.11689945459365844,
|
||
|
|
"num_tokens": 1528990.0,
|
||
|
|
"step": 810
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.537348222732544,
|
||
|
|
"epoch": 0.045571460523372846,
|
||
|
|
"grad_norm": 1.0703125,
|
||
|
|
"learning_rate": 0.00040699999999999997,
|
||
|
|
"loss": 6.214,
|
||
|
|
"mean_token_accuracy": 0.12160347327589989,
|
||
|
|
"num_tokens": 1539361.0,
|
||
|
|
"step": 815
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.605962181091309,
|
||
|
|
"epoch": 0.045851040035786175,
|
||
|
|
"grad_norm": 1.1484375,
|
||
|
|
"learning_rate": 0.0004095,
|
||
|
|
"loss": 6.2271,
|
||
|
|
"mean_token_accuracy": 0.12036370635032653,
|
||
|
|
"num_tokens": 1549625.0,
|
||
|
|
"step": 820
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.620418739318848,
|
||
|
|
"epoch": 0.04613061954819951,
|
||
|
|
"grad_norm": 1.03125,
|
||
|
|
"learning_rate": 0.000412,
|
||
|
|
"loss": 6.2666,
|
||
|
|
"mean_token_accuracy": 0.1255132868885994,
|
||
|
|
"num_tokens": 1559810.0,
|
||
|
|
"step": 825
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.527469205856323,
|
||
|
|
"epoch": 0.04641019906061284,
|
||
|
|
"grad_norm": 1.3046875,
|
||
|
|
"learning_rate": 0.0004145,
|
||
|
|
"loss": 6.3231,
|
||
|
|
"mean_token_accuracy": 0.11854067072272301,
|
||
|
|
"num_tokens": 1571214.0,
|
||
|
|
"step": 830
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.7099076271057125,
|
||
|
|
"epoch": 0.04668977857302617,
|
||
|
|
"grad_norm": 1.2265625,
|
||
|
|
"learning_rate": 0.000417,
|
||
|
|
"loss": 6.2671,
|
||
|
|
"mean_token_accuracy": 0.11671644374728203,
|
||
|
|
"num_tokens": 1580742.0,
|
||
|
|
"step": 835
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.602362537384034,
|
||
|
|
"epoch": 0.0469693580854395,
|
||
|
|
"grad_norm": 1.390625,
|
||
|
|
"learning_rate": 0.0004195,
|
||
|
|
"loss": 6.3177,
|
||
|
|
"mean_token_accuracy": 0.11618626490235329,
|
||
|
|
"num_tokens": 1591290.0,
|
||
|
|
"step": 840
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.530023574829102,
|
||
|
|
"epoch": 0.04724893759785283,
|
||
|
|
"grad_norm": 1.28125,
|
||
|
|
"learning_rate": 0.000422,
|
||
|
|
"loss": 6.2365,
|
||
|
|
"mean_token_accuracy": 0.1239740714430809,
|
||
|
|
"num_tokens": 1600267.0,
|
||
|
|
"step": 845
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.557957172393799,
|
||
|
|
"epoch": 0.04752851711026616,
|
||
|
|
"grad_norm": 1.046875,
|
||
|
|
"learning_rate": 0.0004245,
|
||
|
|
"loss": 6.2153,
|
||
|
|
"mean_token_accuracy": 0.12411479502916337,
|
||
|
|
"num_tokens": 1609948.0,
|
||
|
|
"step": 850
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.517998456954956,
|
||
|
|
"epoch": 0.04780809662267949,
|
||
|
|
"grad_norm": 0.98046875,
|
||
|
|
"learning_rate": 0.000427,
|
||
|
|
"loss": 6.1905,
|
||
|
|
"mean_token_accuracy": 0.12491159066557884,
|
||
|
|
"num_tokens": 1619333.0,
|
||
|
|
"step": 855
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.520691967010498,
|
||
|
|
"epoch": 0.04808767613509282,
|
||
|
|
"grad_norm": 1.15625,
|
||
|
|
"learning_rate": 0.0004295,
|
||
|
|
"loss": 6.2327,
|
||
|
|
"mean_token_accuracy": 0.12030576914548874,
|
||
|
|
"num_tokens": 1628756.0,
|
||
|
|
"step": 860
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.4604676246643065,
|
||
|
|
"epoch": 0.04836725564750615,
|
||
|
|
"grad_norm": 1.1484375,
|
||
|
|
"learning_rate": 0.000432,
|
||
|
|
"loss": 6.1191,
|
||
|
|
"mean_token_accuracy": 0.12915168330073357,
|
||
|
|
"num_tokens": 1638580.0,
|
||
|
|
"step": 865
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.525818300247193,
|
||
|
|
"epoch": 0.04864683515991948,
|
||
|
|
"grad_norm": 1.09375,
|
||
|
|
"learning_rate": 0.0004345,
|
||
|
|
"loss": 6.1465,
|
||
|
|
"mean_token_accuracy": 0.12202514111995696,
|
||
|
|
"num_tokens": 1648453.0,
|
||
|
|
"step": 870
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.462548685073853,
|
||
|
|
"epoch": 0.04892641467233281,
|
||
|
|
"grad_norm": 1.1796875,
|
||
|
|
"learning_rate": 0.000437,
|
||
|
|
"loss": 6.1247,
|
||
|
|
"mean_token_accuracy": 0.12454857230186463,
|
||
|
|
"num_tokens": 1657780.0,
|
||
|
|
"step": 875
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.524330520629883,
|
||
|
|
"epoch": 0.049205994184746145,
|
||
|
|
"grad_norm": 1.140625,
|
||
|
|
"learning_rate": 0.0004395,
|
||
|
|
"loss": 6.3194,
|
||
|
|
"mean_token_accuracy": 0.11488607227802276,
|
||
|
|
"num_tokens": 1667935.0,
|
||
|
|
"step": 880
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.596838188171387,
|
||
|
|
"epoch": 0.049485573697159474,
|
||
|
|
"grad_norm": 1.171875,
|
||
|
|
"learning_rate": 0.000442,
|
||
|
|
"loss": 6.212,
|
||
|
|
"mean_token_accuracy": 0.12398698702454566,
|
||
|
|
"num_tokens": 1678152.0,
|
||
|
|
"step": 885
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.493438577651977,
|
||
|
|
"epoch": 0.0497651532095728,
|
||
|
|
"grad_norm": 1.078125,
|
||
|
|
"learning_rate": 0.0004445,
|
||
|
|
"loss": 6.2097,
|
||
|
|
"mean_token_accuracy": 0.1230739638209343,
|
||
|
|
"num_tokens": 1688222.0,
|
||
|
|
"step": 890
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.532832431793213,
|
||
|
|
"epoch": 0.05004473272198613,
|
||
|
|
"grad_norm": 1.09375,
|
||
|
|
"learning_rate": 0.000447,
|
||
|
|
"loss": 6.2281,
|
||
|
|
"mean_token_accuracy": 0.11738072633743286,
|
||
|
|
"num_tokens": 1698355.0,
|
||
|
|
"step": 895
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.451116752624512,
|
||
|
|
"epoch": 0.05032431223439946,
|
||
|
|
"grad_norm": 0.8984375,
|
||
|
|
"learning_rate": 0.00044950000000000003,
|
||
|
|
"loss": 6.148,
|
||
|
|
"mean_token_accuracy": 0.12611913979053496,
|
||
|
|
"num_tokens": 1708620.0,
|
||
|
|
"step": 900
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.404505968093872,
|
||
|
|
"epoch": 0.05060389174681279,
|
||
|
|
"grad_norm": 1.3359375,
|
||
|
|
"learning_rate": 0.00045200000000000004,
|
||
|
|
"loss": 6.1793,
|
||
|
|
"mean_token_accuracy": 0.12188507243990898,
|
||
|
|
"num_tokens": 1718087.0,
|
||
|
|
"step": 905
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.498981142044068,
|
||
|
|
"epoch": 0.050883471259226126,
|
||
|
|
"grad_norm": 1.34375,
|
||
|
|
"learning_rate": 0.00045450000000000004,
|
||
|
|
"loss": 6.1367,
|
||
|
|
"mean_token_accuracy": 0.12877505272626877,
|
||
|
|
"num_tokens": 1726195.0,
|
||
|
|
"step": 910
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.539384174346924,
|
||
|
|
"epoch": 0.051163050771639455,
|
||
|
|
"grad_norm": 1.15625,
|
||
|
|
"learning_rate": 0.00045700000000000005,
|
||
|
|
"loss": 6.1787,
|
||
|
|
"mean_token_accuracy": 0.12548977583646775,
|
||
|
|
"num_tokens": 1735264.0,
|
||
|
|
"step": 915
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.4787153720855715,
|
||
|
|
"epoch": 0.051442630284052784,
|
||
|
|
"grad_norm": 1.234375,
|
||
|
|
"learning_rate": 0.00045950000000000006,
|
||
|
|
"loss": 6.1781,
|
||
|
|
"mean_token_accuracy": 0.12100374773144722,
|
||
|
|
"num_tokens": 1743914.0,
|
||
|
|
"step": 920
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.478158187866211,
|
||
|
|
"epoch": 0.05172220979646611,
|
||
|
|
"grad_norm": 1.078125,
|
||
|
|
"learning_rate": 0.000462,
|
||
|
|
"loss": 6.1354,
|
||
|
|
"mean_token_accuracy": 0.12575116008520126,
|
||
|
|
"num_tokens": 1753975.0,
|
||
|
|
"step": 925
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.352355718612671,
|
||
|
|
"epoch": 0.05200178930887944,
|
||
|
|
"grad_norm": 0.99609375,
|
||
|
|
"learning_rate": 0.0004645,
|
||
|
|
"loss": 6.2459,
|
||
|
|
"mean_token_accuracy": 0.12019256725907326,
|
||
|
|
"num_tokens": 1763464.0,
|
||
|
|
"step": 930
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.56115231513977,
|
||
|
|
"epoch": 0.05228136882129278,
|
||
|
|
"grad_norm": 1.0546875,
|
||
|
|
"learning_rate": 0.000467,
|
||
|
|
"loss": 6.2273,
|
||
|
|
"mean_token_accuracy": 0.12331901043653488,
|
||
|
|
"num_tokens": 1773493.0,
|
||
|
|
"step": 935
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.558299112319946,
|
||
|
|
"epoch": 0.05256094833370611,
|
||
|
|
"grad_norm": 1.21875,
|
||
|
|
"learning_rate": 0.0004695,
|
||
|
|
"loss": 6.196,
|
||
|
|
"mean_token_accuracy": 0.12245910018682479,
|
||
|
|
"num_tokens": 1782610.0,
|
||
|
|
"step": 940
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.361427593231201,
|
||
|
|
"epoch": 0.05284052784611944,
|
||
|
|
"grad_norm": 1.0234375,
|
||
|
|
"learning_rate": 0.000472,
|
||
|
|
"loss": 6.1001,
|
||
|
|
"mean_token_accuracy": 0.1252542532980442,
|
||
|
|
"num_tokens": 1792227.0,
|
||
|
|
"step": 945
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.528857851028443,
|
||
|
|
"epoch": 0.053120107358532766,
|
||
|
|
"grad_norm": 1.1484375,
|
||
|
|
"learning_rate": 0.0004745,
|
||
|
|
"loss": 6.1815,
|
||
|
|
"mean_token_accuracy": 0.12755777463316917,
|
||
|
|
"num_tokens": 1801201.0,
|
||
|
|
"step": 950
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.345184278488159,
|
||
|
|
"epoch": 0.053399686870946095,
|
||
|
|
"grad_norm": 1.0703125,
|
||
|
|
"learning_rate": 0.000477,
|
||
|
|
"loss": 6.101,
|
||
|
|
"mean_token_accuracy": 0.12595751732587815,
|
||
|
|
"num_tokens": 1810615.0,
|
||
|
|
"step": 955
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.572854137420654,
|
||
|
|
"epoch": 0.05367926638335943,
|
||
|
|
"grad_norm": 1.0703125,
|
||
|
|
"learning_rate": 0.0004795,
|
||
|
|
"loss": 6.2557,
|
||
|
|
"mean_token_accuracy": 0.12506696805357934,
|
||
|
|
"num_tokens": 1820502.0,
|
||
|
|
"step": 960
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.402885007858276,
|
||
|
|
"epoch": 0.05395884589577276,
|
||
|
|
"grad_norm": 1.0390625,
|
||
|
|
"learning_rate": 0.000482,
|
||
|
|
"loss": 6.2045,
|
||
|
|
"mean_token_accuracy": 0.12744859382510185,
|
||
|
|
"num_tokens": 1829691.0,
|
||
|
|
"step": 965
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.47072925567627,
|
||
|
|
"epoch": 0.05423842540818609,
|
||
|
|
"grad_norm": 1.15625,
|
||
|
|
"learning_rate": 0.0004845,
|
||
|
|
"loss": 6.0947,
|
||
|
|
"mean_token_accuracy": 0.13494734093546867,
|
||
|
|
"num_tokens": 1838725.0,
|
||
|
|
"step": 970
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.41903715133667,
|
||
|
|
"epoch": 0.05451800492059942,
|
||
|
|
"grad_norm": 1.078125,
|
||
|
|
"learning_rate": 0.000487,
|
||
|
|
"loss": 6.1757,
|
||
|
|
"mean_token_accuracy": 0.12426450625061988,
|
||
|
|
"num_tokens": 1849362.0,
|
||
|
|
"step": 975
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.597773933410645,
|
||
|
|
"epoch": 0.05479758443301275,
|
||
|
|
"grad_norm": 1.1328125,
|
||
|
|
"learning_rate": 0.0004895,
|
||
|
|
"loss": 6.1783,
|
||
|
|
"mean_token_accuracy": 0.12785449251532555,
|
||
|
|
"num_tokens": 1857989.0,
|
||
|
|
"step": 980
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.431480169296265,
|
||
|
|
"epoch": 0.055077163945426076,
|
||
|
|
"grad_norm": 1.15625,
|
||
|
|
"learning_rate": 0.000492,
|
||
|
|
"loss": 6.1765,
|
||
|
|
"mean_token_accuracy": 0.12329955548048019,
|
||
|
|
"num_tokens": 1867326.0,
|
||
|
|
"step": 985
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.2434632778167725,
|
||
|
|
"epoch": 0.05535674345783941,
|
||
|
|
"grad_norm": 1.0859375,
|
||
|
|
"learning_rate": 0.0004945,
|
||
|
|
"loss": 6.1406,
|
||
|
|
"mean_token_accuracy": 0.13350122570991516,
|
||
|
|
"num_tokens": 1877827.0,
|
||
|
|
"step": 990
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.512171459197998,
|
||
|
|
"epoch": 0.05563632297025274,
|
||
|
|
"grad_norm": 1.1171875,
|
||
|
|
"learning_rate": 0.000497,
|
||
|
|
"loss": 6.137,
|
||
|
|
"mean_token_accuracy": 0.12907352671027184,
|
||
|
|
"num_tokens": 1887109.0,
|
||
|
|
"step": 995
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.38034200668335,
|
||
|
|
"epoch": 0.05591590248266607,
|
||
|
|
"grad_norm": 1.1015625,
|
||
|
|
"learning_rate": 0.0004995,
|
||
|
|
"loss": 6.1664,
|
||
|
|
"mean_token_accuracy": 0.1256708487868309,
|
||
|
|
"num_tokens": 1896852.0,
|
||
|
|
"step": 1000
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"logging_steps": 5,
|
||
|
|
"max_steps": 4000,
|
||
|
|
"num_input_tokens_seen": 0,
|
||
|
|
"num_train_epochs": 1,
|
||
|
|
"save_steps": 500,
|
||
|
|
"stateful_callbacks": {
|
||
|
|
"TrainerControl": {
|
||
|
|
"args": {
|
||
|
|
"should_epoch_stop": false,
|
||
|
|
"should_evaluate": false,
|
||
|
|
"should_log": false,
|
||
|
|
"should_save": true,
|
||
|
|
"should_training_stop": false
|
||
|
|
},
|
||
|
|
"attributes": {}
|
||
|
|
}
|
||
|
|
},
|
||
|
|
"total_flos": 409711646638080.0,
|
||
|
|
"train_batch_size": 16,
|
||
|
|
"trial_name": null,
|
||
|
|
"trial_params": null
|
||
|
|
}
|