3035 lines
82 KiB
JSON
3035 lines
82 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.01611984567934403,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.742156791687012,
|
|
"epoch": 5.3732818931146765e-05,
|
|
"grad_norm": 4.5625,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.7975,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 12012.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.74215326309204,
|
|
"epoch": 0.00010746563786229353,
|
|
"grad_norm": 5.375,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7849,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 23548.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.742180919647216,
|
|
"epoch": 0.0001611984567934403,
|
|
"grad_norm": 4.96875,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.763,
|
|
"mean_token_accuracy": 0.00020668564829975366,
|
|
"num_tokens": 35237.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.742225646972656,
|
|
"epoch": 0.00021493127572458706,
|
|
"grad_norm": 5.03125,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.7267,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 47773.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.742192363739013,
|
|
"epoch": 0.00026866409465573383,
|
|
"grad_norm": 5.21875,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.6027,
|
|
"mean_token_accuracy": 0.0023876232909969984,
|
|
"num_tokens": 59398.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.741855812072753,
|
|
"epoch": 0.0003223969135868806,
|
|
"grad_norm": 4.0625,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.473,
|
|
"mean_token_accuracy": 0.020790282264351846,
|
|
"num_tokens": 70474.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.740978240966797,
|
|
"epoch": 0.00037612973251802736,
|
|
"grad_norm": 3.109375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.3535,
|
|
"mean_token_accuracy": 0.03107255753129721,
|
|
"num_tokens": 81441.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.739280223846436,
|
|
"epoch": 0.0004298625514491741,
|
|
"grad_norm": 2.53125,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.23,
|
|
"mean_token_accuracy": 0.032152241468429564,
|
|
"num_tokens": 92819.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.736611366271973,
|
|
"epoch": 0.0004835953703803209,
|
|
"grad_norm": 2.484375,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.0679,
|
|
"mean_token_accuracy": 0.0433739360421896,
|
|
"num_tokens": 105565.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.732536792755127,
|
|
"epoch": 0.0005373281893114677,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.9911,
|
|
"mean_token_accuracy": 0.037867228500545025,
|
|
"num_tokens": 116424.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.72894058227539,
|
|
"epoch": 0.0005910610082426144,
|
|
"grad_norm": 1.984375,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.9525,
|
|
"mean_token_accuracy": 0.03848287593573332,
|
|
"num_tokens": 127807.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.727017021179199,
|
|
"epoch": 0.0006447938271737612,
|
|
"grad_norm": 2.109375,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.8619,
|
|
"mean_token_accuracy": 0.03968206029385328,
|
|
"num_tokens": 139177.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.726418113708496,
|
|
"epoch": 0.000698526646104908,
|
|
"grad_norm": 1.9765625,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.8295,
|
|
"mean_token_accuracy": 0.03538134638220072,
|
|
"num_tokens": 151296.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.725299739837647,
|
|
"epoch": 0.0007522594650360547,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.7022,
|
|
"mean_token_accuracy": 0.03797216359525919,
|
|
"num_tokens": 160488.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.723019790649413,
|
|
"epoch": 0.0008059922839672015,
|
|
"grad_norm": 1.984375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.6965,
|
|
"mean_token_accuracy": 0.03595883902162313,
|
|
"num_tokens": 171627.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.718975067138672,
|
|
"epoch": 0.0008597251028983482,
|
|
"grad_norm": 2.015625,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.6005,
|
|
"mean_token_accuracy": 0.03552283551543951,
|
|
"num_tokens": 182978.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.713298988342284,
|
|
"epoch": 0.000913457921829495,
|
|
"grad_norm": 2.125,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.5237,
|
|
"mean_token_accuracy": 0.03787758443504572,
|
|
"num_tokens": 193362.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.705536365509033,
|
|
"epoch": 0.0009671907407606418,
|
|
"grad_norm": 1.9609375,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.4601,
|
|
"mean_token_accuracy": 0.037437847815454006,
|
|
"num_tokens": 204896.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.695236206054688,
|
|
"epoch": 0.0010209235596917885,
|
|
"grad_norm": 2.03125,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.3647,
|
|
"mean_token_accuracy": 0.03763183522969484,
|
|
"num_tokens": 215828.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.683992767333985,
|
|
"epoch": 0.0010746563786229353,
|
|
"grad_norm": 2.203125,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.2553,
|
|
"mean_token_accuracy": 0.04709151349961758,
|
|
"num_tokens": 226311.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.66820011138916,
|
|
"epoch": 0.001128389197554082,
|
|
"grad_norm": 1.96875,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.1779,
|
|
"mean_token_accuracy": 0.04788613952696323,
|
|
"num_tokens": 237817.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.641796970367432,
|
|
"epoch": 0.0011821220164852288,
|
|
"grad_norm": 1.9375,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.0389,
|
|
"mean_token_accuracy": 0.05182615555822849,
|
|
"num_tokens": 248778.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.600088024139405,
|
|
"epoch": 0.0012358548354163756,
|
|
"grad_norm": 1.890625,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.9776,
|
|
"mean_token_accuracy": 0.048598403483629225,
|
|
"num_tokens": 260638.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.544377613067628,
|
|
"epoch": 0.0012895876543475224,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.8365,
|
|
"mean_token_accuracy": 0.049195655062794684,
|
|
"num_tokens": 272223.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.488489818572997,
|
|
"epoch": 0.0013433204732786691,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.6543,
|
|
"mean_token_accuracy": 0.06183730475604534,
|
|
"num_tokens": 282883.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.427958393096924,
|
|
"epoch": 0.001397053292209816,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.5628,
|
|
"mean_token_accuracy": 0.05788363590836525,
|
|
"num_tokens": 293318.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.361266994476319,
|
|
"epoch": 0.0014507861111409627,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.3999,
|
|
"mean_token_accuracy": 0.06325617916882038,
|
|
"num_tokens": 303563.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.266337871551514,
|
|
"epoch": 0.0015045189300721094,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.2536,
|
|
"mean_token_accuracy": 0.07024482376873493,
|
|
"num_tokens": 313968.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.139619827270508,
|
|
"epoch": 0.0015582517490032562,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.2142,
|
|
"mean_token_accuracy": 0.06537894494831561,
|
|
"num_tokens": 325261.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 10.008739376068116,
|
|
"epoch": 0.001611984567934403,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.9979,
|
|
"mean_token_accuracy": 0.07281787917017937,
|
|
"num_tokens": 335803.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.818781661987305,
|
|
"epoch": 0.0016657173868655497,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.8841,
|
|
"mean_token_accuracy": 0.0677342727780342,
|
|
"num_tokens": 346582.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.64435920715332,
|
|
"epoch": 0.0017194502057966965,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.8699,
|
|
"mean_token_accuracy": 0.07075726985931396,
|
|
"num_tokens": 357779.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.435050868988037,
|
|
"epoch": 0.0017731830247278433,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.6867,
|
|
"mean_token_accuracy": 0.07693751528859138,
|
|
"num_tokens": 369191.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.177427005767822,
|
|
"epoch": 0.00182691584365899,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.6553,
|
|
"mean_token_accuracy": 0.07154584787786007,
|
|
"num_tokens": 381374.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.965550804138184,
|
|
"epoch": 0.0018806486625901368,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.5673,
|
|
"mean_token_accuracy": 0.07786881327629089,
|
|
"num_tokens": 392152.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.720534420013427,
|
|
"epoch": 0.0019343814815212836,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.5062,
|
|
"mean_token_accuracy": 0.07432210743427277,
|
|
"num_tokens": 404121.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.521537494659423,
|
|
"epoch": 0.0019881143004524303,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.4599,
|
|
"mean_token_accuracy": 0.0757918406277895,
|
|
"num_tokens": 416144.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.401978397369385,
|
|
"epoch": 0.002041847119383577,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.3406,
|
|
"mean_token_accuracy": 0.08547781854867935,
|
|
"num_tokens": 427141.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.218191432952882,
|
|
"epoch": 0.002095579938314724,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.2438,
|
|
"mean_token_accuracy": 0.07949326783418656,
|
|
"num_tokens": 439113.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.094591236114502,
|
|
"epoch": 0.0021493127572458706,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.3,
|
|
"mean_token_accuracy": 0.08100042939186096,
|
|
"num_tokens": 450595.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.087145900726318,
|
|
"epoch": 0.0022030455761770174,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.3575,
|
|
"mean_token_accuracy": 0.08150793164968491,
|
|
"num_tokens": 462811.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.998901224136352,
|
|
"epoch": 0.002256778395108164,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.3188,
|
|
"mean_token_accuracy": 0.08503394201397896,
|
|
"num_tokens": 473629.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.012935066223145,
|
|
"epoch": 0.002310511214039311,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.2985,
|
|
"mean_token_accuracy": 0.08433945029973984,
|
|
"num_tokens": 484731.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.911737585067749,
|
|
"epoch": 0.0023642440329704577,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.2521,
|
|
"mean_token_accuracy": 0.08716428875923157,
|
|
"num_tokens": 496309.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.943577146530151,
|
|
"epoch": 0.0024179768519016044,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.2935,
|
|
"mean_token_accuracy": 0.08446456864476204,
|
|
"num_tokens": 508715.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.8531383037567135,
|
|
"epoch": 0.002471709670832751,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.1316,
|
|
"mean_token_accuracy": 0.0886182926595211,
|
|
"num_tokens": 519791.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.81738133430481,
|
|
"epoch": 0.002525442489763898,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.2519,
|
|
"mean_token_accuracy": 0.08779674172401428,
|
|
"num_tokens": 531897.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.859956073760986,
|
|
"epoch": 0.0025791753086950447,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.2388,
|
|
"mean_token_accuracy": 0.0916181556880474,
|
|
"num_tokens": 543243.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.7477397441864015,
|
|
"epoch": 0.0026329081276261915,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.0962,
|
|
"mean_token_accuracy": 0.09303439110517502,
|
|
"num_tokens": 554139.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.774733972549439,
|
|
"epoch": 0.0026866409465573383,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.1076,
|
|
"mean_token_accuracy": 0.09551994875073433,
|
|
"num_tokens": 565378.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.715104246139527,
|
|
"epoch": 0.002740373765488485,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.0605,
|
|
"mean_token_accuracy": 0.09414524212479591,
|
|
"num_tokens": 576628.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.768790674209595,
|
|
"epoch": 0.002794106584419632,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.1575,
|
|
"mean_token_accuracy": 0.09156634658575058,
|
|
"num_tokens": 588760.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.687872219085693,
|
|
"epoch": 0.0028478394033507786,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.0791,
|
|
"mean_token_accuracy": 0.10132888033986091,
|
|
"num_tokens": 600024.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.683035945892334,
|
|
"epoch": 0.0029015722222819253,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.0399,
|
|
"mean_token_accuracy": 0.09685612693428994,
|
|
"num_tokens": 613150.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.63238935470581,
|
|
"epoch": 0.002955305041213072,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.0179,
|
|
"mean_token_accuracy": 0.09851899445056915,
|
|
"num_tokens": 624762.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.715113353729248,
|
|
"epoch": 0.003009037860144219,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.0639,
|
|
"mean_token_accuracy": 0.09616550207138061,
|
|
"num_tokens": 635761.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.620885419845581,
|
|
"epoch": 0.0030627706790753656,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.0092,
|
|
"mean_token_accuracy": 0.09458725675940513,
|
|
"num_tokens": 647882.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.5718300342559814,
|
|
"epoch": 0.0031165034980065124,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.0483,
|
|
"mean_token_accuracy": 0.09655040949583053,
|
|
"num_tokens": 659749.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.516621065139771,
|
|
"epoch": 0.003170236316937659,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000147,
|
|
"loss": 6.9872,
|
|
"mean_token_accuracy": 0.0997149519622326,
|
|
"num_tokens": 672002.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.547021913528442,
|
|
"epoch": 0.003223969135868806,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.9284,
|
|
"mean_token_accuracy": 0.10656724348664284,
|
|
"num_tokens": 682312.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.456842422485352,
|
|
"epoch": 0.0032777019547999527,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.965,
|
|
"mean_token_accuracy": 0.09908917099237442,
|
|
"num_tokens": 692188.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.561665821075439,
|
|
"epoch": 0.0033314347737310995,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 6.9484,
|
|
"mean_token_accuracy": 0.10072302371263504,
|
|
"num_tokens": 703789.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.566508626937866,
|
|
"epoch": 0.0033851675926622462,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.0206,
|
|
"mean_token_accuracy": 0.10419768020510674,
|
|
"num_tokens": 714752.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.47095513343811,
|
|
"epoch": 0.003438900411593393,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.9571,
|
|
"mean_token_accuracy": 0.10431606322526932,
|
|
"num_tokens": 726716.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.433886337280273,
|
|
"epoch": 0.0034926332305245398,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.9813,
|
|
"mean_token_accuracy": 0.1021119087934494,
|
|
"num_tokens": 737267.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.5219165802001955,
|
|
"epoch": 0.0035463660494556865,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.0025,
|
|
"mean_token_accuracy": 0.10279990583658219,
|
|
"num_tokens": 750180.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.45853362083435,
|
|
"epoch": 0.0036000988683868333,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.8296,
|
|
"mean_token_accuracy": 0.10508297532796859,
|
|
"num_tokens": 761337.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.402671909332275,
|
|
"epoch": 0.00365383168731798,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.8015,
|
|
"mean_token_accuracy": 0.1093854360282421,
|
|
"num_tokens": 771379.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.37822003364563,
|
|
"epoch": 0.003707564506249127,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.8888,
|
|
"mean_token_accuracy": 0.10919447690248489,
|
|
"num_tokens": 782508.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.4207008361816404,
|
|
"epoch": 0.0037612973251802736,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.9096,
|
|
"mean_token_accuracy": 0.10041842758655548,
|
|
"num_tokens": 795541.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.383993768692017,
|
|
"epoch": 0.0038150301441114204,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.8261,
|
|
"mean_token_accuracy": 0.10757922530174255,
|
|
"num_tokens": 805665.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.3851542472839355,
|
|
"epoch": 0.003868762963042567,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.9727,
|
|
"mean_token_accuracy": 0.1046277977526188,
|
|
"num_tokens": 817393.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.4332190990448,
|
|
"epoch": 0.003922495781973714,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.9061,
|
|
"mean_token_accuracy": 0.1035635381937027,
|
|
"num_tokens": 828436.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.383777904510498,
|
|
"epoch": 0.003976228600904861,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.7789,
|
|
"mean_token_accuracy": 0.10867122486233712,
|
|
"num_tokens": 839223.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.291258668899536,
|
|
"epoch": 0.004029961419836007,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.7478,
|
|
"mean_token_accuracy": 0.11749055162072182,
|
|
"num_tokens": 848953.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.338685035705566,
|
|
"epoch": 0.004083694238767154,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.8829,
|
|
"mean_token_accuracy": 0.10381969884037971,
|
|
"num_tokens": 860349.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.314834403991699,
|
|
"epoch": 0.004137427057698301,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.8047,
|
|
"mean_token_accuracy": 0.10881992653012276,
|
|
"num_tokens": 870993.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.348164367675781,
|
|
"epoch": 0.004191159876629448,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.8523,
|
|
"mean_token_accuracy": 0.10951408222317696,
|
|
"num_tokens": 880950.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.376342344284057,
|
|
"epoch": 0.0042448926955605945,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.8578,
|
|
"mean_token_accuracy": 0.11057165712118149,
|
|
"num_tokens": 892583.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.21990008354187,
|
|
"epoch": 0.004298625514491741,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.6711,
|
|
"mean_token_accuracy": 0.1182975672185421,
|
|
"num_tokens": 902191.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.244998121261597,
|
|
"epoch": 0.004352358333422888,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.7288,
|
|
"mean_token_accuracy": 0.11365634202957153,
|
|
"num_tokens": 914274.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.147398805618286,
|
|
"epoch": 0.004406091152354035,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.6684,
|
|
"mean_token_accuracy": 0.10877410843968391,
|
|
"num_tokens": 925253.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.30414662361145,
|
|
"epoch": 0.0044598239712851815,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.8047,
|
|
"mean_token_accuracy": 0.11197239682078361,
|
|
"num_tokens": 935520.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.248277902603149,
|
|
"epoch": 0.004513556790216328,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.7966,
|
|
"mean_token_accuracy": 0.11800143346190453,
|
|
"num_tokens": 947817.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.1241779804229735,
|
|
"epoch": 0.004567289609147475,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.7073,
|
|
"mean_token_accuracy": 0.1153214804828167,
|
|
"num_tokens": 959474.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.291252708435058,
|
|
"epoch": 0.004621022428078622,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.7,
|
|
"mean_token_accuracy": 0.11340035423636437,
|
|
"num_tokens": 971271.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.1269285678863525,
|
|
"epoch": 0.004674755247009769,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.7119,
|
|
"mean_token_accuracy": 0.11621161103248596,
|
|
"num_tokens": 981995.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.202797365188599,
|
|
"epoch": 0.004728488065940915,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.7741,
|
|
"mean_token_accuracy": 0.11647263541817665,
|
|
"num_tokens": 994346.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.202062654495239,
|
|
"epoch": 0.004782220884872062,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.7803,
|
|
"mean_token_accuracy": 0.11048730537295341,
|
|
"num_tokens": 1005552.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.2215375900268555,
|
|
"epoch": 0.004835953703803209,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.7147,
|
|
"mean_token_accuracy": 0.10997768640518188,
|
|
"num_tokens": 1017202.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.02616319656372,
|
|
"epoch": 0.004889686522734356,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.6729,
|
|
"mean_token_accuracy": 0.11669943779706955,
|
|
"num_tokens": 1027450.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.129533100128174,
|
|
"epoch": 0.004943419341665502,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.6599,
|
|
"mean_token_accuracy": 0.11500771567225457,
|
|
"num_tokens": 1038990.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.127964210510254,
|
|
"epoch": 0.004997152160596649,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.6597,
|
|
"mean_token_accuracy": 0.11586256548762322,
|
|
"num_tokens": 1050643.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.037399816513061,
|
|
"epoch": 0.005050884979527796,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.5917,
|
|
"mean_token_accuracy": 0.12084505632519722,
|
|
"num_tokens": 1061182.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.155901622772217,
|
|
"epoch": 0.005104617798458943,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.7526,
|
|
"mean_token_accuracy": 0.11448139399290085,
|
|
"num_tokens": 1073074.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.101831865310669,
|
|
"epoch": 0.0051583506173900895,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.7746,
|
|
"mean_token_accuracy": 0.1180665597319603,
|
|
"num_tokens": 1085061.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.091040754318238,
|
|
"epoch": 0.005212083436321236,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.6384,
|
|
"mean_token_accuracy": 0.11752296015620231,
|
|
"num_tokens": 1096148.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.014005517959594,
|
|
"epoch": 0.005265816255252383,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.5661,
|
|
"mean_token_accuracy": 0.11984377726912498,
|
|
"num_tokens": 1106159.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.034162092208862,
|
|
"epoch": 0.00531954907418353,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.7162,
|
|
"mean_token_accuracy": 0.11838811710476875,
|
|
"num_tokens": 1117159.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.145173740386963,
|
|
"epoch": 0.0053732818931146765,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.6168,
|
|
"mean_token_accuracy": 0.12091756090521813,
|
|
"num_tokens": 1129633.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.200956392288208,
|
|
"epoch": 0.005427014712045823,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.7907,
|
|
"mean_token_accuracy": 0.115936279296875,
|
|
"num_tokens": 1140814.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.03369836807251,
|
|
"epoch": 0.00548074753097697,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.6819,
|
|
"mean_token_accuracy": 0.1204757682979107,
|
|
"num_tokens": 1152963.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 6.96367130279541,
|
|
"epoch": 0.005534480349908117,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.5086,
|
|
"mean_token_accuracy": 0.12356813177466393,
|
|
"num_tokens": 1165092.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.031797409057617,
|
|
"epoch": 0.005588213168839264,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.5961,
|
|
"mean_token_accuracy": 0.12022339180111885,
|
|
"num_tokens": 1177403.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.019965028762817,
|
|
"epoch": 0.00564194598777041,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.5984,
|
|
"mean_token_accuracy": 0.1161931224167347,
|
|
"num_tokens": 1187436.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.0072669506073,
|
|
"epoch": 0.005695678806701557,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.6886,
|
|
"mean_token_accuracy": 0.11582138538360595,
|
|
"num_tokens": 1199906.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.0678855895996096,
|
|
"epoch": 0.005749411625632704,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.6592,
|
|
"mean_token_accuracy": 0.11758691444993019,
|
|
"num_tokens": 1211411.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.979720735549927,
|
|
"epoch": 0.005803144444563851,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.609,
|
|
"mean_token_accuracy": 0.11972267776727677,
|
|
"num_tokens": 1223669.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.941835403442383,
|
|
"epoch": 0.0058568772634949974,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.5478,
|
|
"mean_token_accuracy": 0.11847051307559013,
|
|
"num_tokens": 1234931.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.954667425155639,
|
|
"epoch": 0.005910610082426144,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.5731,
|
|
"mean_token_accuracy": 0.1192402996122837,
|
|
"num_tokens": 1245731.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.920712327957153,
|
|
"epoch": 0.005964342901357291,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.5466,
|
|
"mean_token_accuracy": 0.1234957829117775,
|
|
"num_tokens": 1257441.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.959730100631714,
|
|
"epoch": 0.006018075720288438,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.5742,
|
|
"mean_token_accuracy": 0.12339851334691047,
|
|
"num_tokens": 1269169.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.96678729057312,
|
|
"epoch": 0.0060718085392195845,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.557,
|
|
"mean_token_accuracy": 0.12321766391396523,
|
|
"num_tokens": 1280636.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.821154022216797,
|
|
"epoch": 0.006125541358150731,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.5018,
|
|
"mean_token_accuracy": 0.1237683854997158,
|
|
"num_tokens": 1292266.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.918811464309693,
|
|
"epoch": 0.006179274177081878,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.5705,
|
|
"mean_token_accuracy": 0.12514246329665185,
|
|
"num_tokens": 1304090.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.919731712341308,
|
|
"epoch": 0.006233006996013025,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.6274,
|
|
"mean_token_accuracy": 0.12263328060507775,
|
|
"num_tokens": 1314758.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.972955799102783,
|
|
"epoch": 0.0062867398149441716,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.5953,
|
|
"mean_token_accuracy": 0.12088563293218613,
|
|
"num_tokens": 1326085.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.0157567977905275,
|
|
"epoch": 0.006340472633875318,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.524,
|
|
"mean_token_accuracy": 0.12394717037677765,
|
|
"num_tokens": 1336766.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.809007263183593,
|
|
"epoch": 0.006394205452806465,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.4435,
|
|
"mean_token_accuracy": 0.1222992442548275,
|
|
"num_tokens": 1346864.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.922023057937622,
|
|
"epoch": 0.006447938271737612,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.555,
|
|
"mean_token_accuracy": 0.12275824695825577,
|
|
"num_tokens": 1357429.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.899812126159668,
|
|
"epoch": 0.006501671090668759,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.4479,
|
|
"mean_token_accuracy": 0.12394843995571136,
|
|
"num_tokens": 1367080.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.854321146011353,
|
|
"epoch": 0.006555403909599905,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.5873,
|
|
"mean_token_accuracy": 0.12046407908201218,
|
|
"num_tokens": 1379104.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.856927537918091,
|
|
"epoch": 0.006609136728531052,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.4104,
|
|
"mean_token_accuracy": 0.1328809566795826,
|
|
"num_tokens": 1390346.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.752168655395508,
|
|
"epoch": 0.006662869547462199,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.5313,
|
|
"mean_token_accuracy": 0.11913828998804092,
|
|
"num_tokens": 1401512.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.89538083076477,
|
|
"epoch": 0.006716602366393346,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.5226,
|
|
"mean_token_accuracy": 0.12425552606582642,
|
|
"num_tokens": 1412410.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.8058703422546385,
|
|
"epoch": 0.0067703351853244925,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.4158,
|
|
"mean_token_accuracy": 0.12984914779663087,
|
|
"num_tokens": 1424028.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.92828893661499,
|
|
"epoch": 0.006824068004255639,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.5536,
|
|
"mean_token_accuracy": 0.1245744988322258,
|
|
"num_tokens": 1435275.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.855579996109009,
|
|
"epoch": 0.006877800823186786,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.5365,
|
|
"mean_token_accuracy": 0.12668125703930855,
|
|
"num_tokens": 1447336.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.864982080459595,
|
|
"epoch": 0.006931533642117933,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.4817,
|
|
"mean_token_accuracy": 0.12454084455966949,
|
|
"num_tokens": 1459224.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.926883172988892,
|
|
"epoch": 0.0069852664610490795,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.5983,
|
|
"mean_token_accuracy": 0.12362459078431129,
|
|
"num_tokens": 1472077.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.854774618148804,
|
|
"epoch": 0.007038999279980226,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.5727,
|
|
"mean_token_accuracy": 0.12581925988197326,
|
|
"num_tokens": 1484296.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.821399688720703,
|
|
"epoch": 0.007092732098911373,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.5249,
|
|
"mean_token_accuracy": 0.12466432079672814,
|
|
"num_tokens": 1495723.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.840889120101929,
|
|
"epoch": 0.00714646491784252,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.4137,
|
|
"mean_token_accuracy": 0.13339626044034958,
|
|
"num_tokens": 1506610.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.906989669799804,
|
|
"epoch": 0.007200197736773667,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.5218,
|
|
"mean_token_accuracy": 0.12840427160263063,
|
|
"num_tokens": 1519216.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.711476755142212,
|
|
"epoch": 0.007253930555704813,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.4914,
|
|
"mean_token_accuracy": 0.1274320252239704,
|
|
"num_tokens": 1529925.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.791730213165283,
|
|
"epoch": 0.00730766337463596,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.4405,
|
|
"mean_token_accuracy": 0.13115567862987518,
|
|
"num_tokens": 1541329.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.745653057098389,
|
|
"epoch": 0.007361396193567107,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.4713,
|
|
"mean_token_accuracy": 0.12333054170012474,
|
|
"num_tokens": 1552716.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.778638029098511,
|
|
"epoch": 0.007415129012498254,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.4131,
|
|
"mean_token_accuracy": 0.12723663300275803,
|
|
"num_tokens": 1563170.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.834349298477173,
|
|
"epoch": 0.0074688618314294,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.5494,
|
|
"mean_token_accuracy": 0.12827260047197342,
|
|
"num_tokens": 1574431.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.828717660903931,
|
|
"epoch": 0.007522594650360547,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.4955,
|
|
"mean_token_accuracy": 0.12269587889313698,
|
|
"num_tokens": 1586173.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.696405744552612,
|
|
"epoch": 0.007576327469291694,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.4586,
|
|
"mean_token_accuracy": 0.13177242800593375,
|
|
"num_tokens": 1596809.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.771217346191406,
|
|
"epoch": 0.007630060288222841,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.4083,
|
|
"mean_token_accuracy": 0.13248787596821784,
|
|
"num_tokens": 1608033.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.784906053543091,
|
|
"epoch": 0.0076837931071539875,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.4543,
|
|
"mean_token_accuracy": 0.13260905370116233,
|
|
"num_tokens": 1619135.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.735342168807984,
|
|
"epoch": 0.007737525926085134,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.498,
|
|
"mean_token_accuracy": 0.1327671304345131,
|
|
"num_tokens": 1630162.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.833858489990234,
|
|
"epoch": 0.007791258745016281,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.541,
|
|
"mean_token_accuracy": 0.12662407904863357,
|
|
"num_tokens": 1641439.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.7514866352081295,
|
|
"epoch": 0.007844991563947428,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.448,
|
|
"mean_token_accuracy": 0.12918282598257064,
|
|
"num_tokens": 1653615.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.690993213653565,
|
|
"epoch": 0.007898724382878575,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.337,
|
|
"mean_token_accuracy": 0.13188016936182975,
|
|
"num_tokens": 1663262.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.9024210453033445,
|
|
"epoch": 0.007952457201809721,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.5751,
|
|
"mean_token_accuracy": 0.11772055700421333,
|
|
"num_tokens": 1676012.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.826274967193603,
|
|
"epoch": 0.008006190020740869,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.4727,
|
|
"mean_token_accuracy": 0.12126590088009834,
|
|
"num_tokens": 1688145.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.65394024848938,
|
|
"epoch": 0.008059922839672015,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.4,
|
|
"mean_token_accuracy": 0.13146659955382348,
|
|
"num_tokens": 1698710.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.672468805313111,
|
|
"epoch": 0.008113655658603162,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.4149,
|
|
"mean_token_accuracy": 0.13393419086933137,
|
|
"num_tokens": 1710956.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.82416672706604,
|
|
"epoch": 0.008167388477534308,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.4814,
|
|
"mean_token_accuracy": 0.1282752551138401,
|
|
"num_tokens": 1722915.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.652000999450683,
|
|
"epoch": 0.008221121296465456,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.377,
|
|
"mean_token_accuracy": 0.13255829513072967,
|
|
"num_tokens": 1734474.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.735916709899902,
|
|
"epoch": 0.008274854115396602,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.4846,
|
|
"mean_token_accuracy": 0.1270184561610222,
|
|
"num_tokens": 1747871.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.716990184783936,
|
|
"epoch": 0.00832858693432775,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.41,
|
|
"mean_token_accuracy": 0.12982381954789163,
|
|
"num_tokens": 1759368.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.751021957397461,
|
|
"epoch": 0.008382319753258895,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.3701,
|
|
"mean_token_accuracy": 0.12724083960056304,
|
|
"num_tokens": 1770077.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.627581453323364,
|
|
"epoch": 0.008436052572190043,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.3565,
|
|
"mean_token_accuracy": 0.12726172357797622,
|
|
"num_tokens": 1781866.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.664901685714722,
|
|
"epoch": 0.008489785391121189,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.3501,
|
|
"mean_token_accuracy": 0.12985118851065636,
|
|
"num_tokens": 1792495.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.677159643173217,
|
|
"epoch": 0.008543518210052337,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.3551,
|
|
"mean_token_accuracy": 0.13435045927762984,
|
|
"num_tokens": 1803615.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.595445966720581,
|
|
"epoch": 0.008597251028983482,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.3478,
|
|
"mean_token_accuracy": 0.13019294515252114,
|
|
"num_tokens": 1814302.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.664141750335693,
|
|
"epoch": 0.00865098384791463,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.375,
|
|
"mean_token_accuracy": 0.13435978665947915,
|
|
"num_tokens": 1825883.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.707999420166016,
|
|
"epoch": 0.008704716666845776,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.4584,
|
|
"mean_token_accuracy": 0.1291785128414631,
|
|
"num_tokens": 1837863.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.616083335876465,
|
|
"epoch": 0.008758449485776924,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.345,
|
|
"mean_token_accuracy": 0.12927156910300255,
|
|
"num_tokens": 1848566.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.642962121963501,
|
|
"epoch": 0.00881218230470807,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.4351,
|
|
"mean_token_accuracy": 0.13251445367932319,
|
|
"num_tokens": 1862068.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.711829996109008,
|
|
"epoch": 0.008865915123639217,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.4152,
|
|
"mean_token_accuracy": 0.13048165887594224,
|
|
"num_tokens": 1873985.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.599660491943359,
|
|
"epoch": 0.008919647942570363,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.3885,
|
|
"mean_token_accuracy": 0.13085729926824569,
|
|
"num_tokens": 1885805.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.55834698677063,
|
|
"epoch": 0.00897338076150151,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.3623,
|
|
"mean_token_accuracy": 0.1338193781673908,
|
|
"num_tokens": 1897520.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.635346221923828,
|
|
"epoch": 0.009027113580432657,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.3417,
|
|
"mean_token_accuracy": 0.1300429657101631,
|
|
"num_tokens": 1908645.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.607639932632447,
|
|
"epoch": 0.009080846399363804,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.3419,
|
|
"mean_token_accuracy": 0.13699555322527884,
|
|
"num_tokens": 1919933.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.721379947662354,
|
|
"epoch": 0.00913457921829495,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.4167,
|
|
"mean_token_accuracy": 0.13236463963985443,
|
|
"num_tokens": 1932150.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.618501901626587,
|
|
"epoch": 0.009188312037226098,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.3661,
|
|
"mean_token_accuracy": 0.13121916949748993,
|
|
"num_tokens": 1944379.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.6225792407989506,
|
|
"epoch": 0.009242044856157244,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.3634,
|
|
"mean_token_accuracy": 0.13575408905744552,
|
|
"num_tokens": 1956156.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.704810047149659,
|
|
"epoch": 0.009295777675088391,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.4815,
|
|
"mean_token_accuracy": 0.12164724692702293,
|
|
"num_tokens": 1968585.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.561042642593383,
|
|
"epoch": 0.009349510494019537,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.3241,
|
|
"mean_token_accuracy": 0.13110409080982208,
|
|
"num_tokens": 1979627.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.66269817352295,
|
|
"epoch": 0.009403243312950685,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.4118,
|
|
"mean_token_accuracy": 0.12834282219409943,
|
|
"num_tokens": 1993368.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.666582441329956,
|
|
"epoch": 0.00945697613188183,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.4453,
|
|
"mean_token_accuracy": 0.12781335040926933,
|
|
"num_tokens": 2004334.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.532254314422607,
|
|
"epoch": 0.009510708950812978,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.2593,
|
|
"mean_token_accuracy": 0.1311105139553547,
|
|
"num_tokens": 2015328.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.537533617019653,
|
|
"epoch": 0.009564441769744124,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.3633,
|
|
"mean_token_accuracy": 0.13210451900959014,
|
|
"num_tokens": 2027341.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.504496908187866,
|
|
"epoch": 0.009618174588675272,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.2276,
|
|
"mean_token_accuracy": 0.13667807802557946,
|
|
"num_tokens": 2036774.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.616758346557617,
|
|
"epoch": 0.009671907407606418,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.2844,
|
|
"mean_token_accuracy": 0.13679326474666595,
|
|
"num_tokens": 2047320.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.426760816574097,
|
|
"epoch": 0.009725640226537565,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.2331,
|
|
"mean_token_accuracy": 0.13850077539682387,
|
|
"num_tokens": 2058336.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.521706295013428,
|
|
"epoch": 0.009779373045468711,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.2774,
|
|
"mean_token_accuracy": 0.13604633510112762,
|
|
"num_tokens": 2069767.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.505149173736572,
|
|
"epoch": 0.009833105864399859,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.2513,
|
|
"mean_token_accuracy": 0.14015916734933853,
|
|
"num_tokens": 2080018.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.545533227920532,
|
|
"epoch": 0.009886838683331005,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.3207,
|
|
"mean_token_accuracy": 0.1339274115860462,
|
|
"num_tokens": 2091274.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.638880681991577,
|
|
"epoch": 0.009940571502262152,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.4585,
|
|
"mean_token_accuracy": 0.12505250200629234,
|
|
"num_tokens": 2104780.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.619468021392822,
|
|
"epoch": 0.009994304321193298,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.4593,
|
|
"mean_token_accuracy": 0.12707706540822983,
|
|
"num_tokens": 2118181.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.606321144104004,
|
|
"epoch": 0.010048037140124446,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.3766,
|
|
"mean_token_accuracy": 0.13838007897138596,
|
|
"num_tokens": 2128371.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.5294524192810055,
|
|
"epoch": 0.010101769959055592,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.2047,
|
|
"mean_token_accuracy": 0.14117706716060638,
|
|
"num_tokens": 2140385.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.588686227798462,
|
|
"epoch": 0.01015550277798674,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.4255,
|
|
"mean_token_accuracy": 0.1322489708662033,
|
|
"num_tokens": 2151030.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.435458135604859,
|
|
"epoch": 0.010209235596917885,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.1656,
|
|
"mean_token_accuracy": 0.14035440534353255,
|
|
"num_tokens": 2162977.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.6304816722869875,
|
|
"epoch": 0.010262968415849033,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.4356,
|
|
"mean_token_accuracy": 0.1257287710905075,
|
|
"num_tokens": 2175320.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.527151346206665,
|
|
"epoch": 0.010316701234780179,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.287,
|
|
"mean_token_accuracy": 0.1363198533654213,
|
|
"num_tokens": 2187401.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.432109069824219,
|
|
"epoch": 0.010370434053711327,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.2711,
|
|
"mean_token_accuracy": 0.13861160352826118,
|
|
"num_tokens": 2198475.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.65678300857544,
|
|
"epoch": 0.010424166872642473,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.3047,
|
|
"mean_token_accuracy": 0.1330752246081829,
|
|
"num_tokens": 2209401.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.332471513748169,
|
|
"epoch": 0.01047789969157362,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.2283,
|
|
"mean_token_accuracy": 0.13588522896170616,
|
|
"num_tokens": 2220687.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.645492458343506,
|
|
"epoch": 0.010531632510504766,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.3946,
|
|
"mean_token_accuracy": 0.13446979448199273,
|
|
"num_tokens": 2232876.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.5025159358978275,
|
|
"epoch": 0.010585365329435914,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.2048,
|
|
"mean_token_accuracy": 0.13936366885900497,
|
|
"num_tokens": 2243683.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.3703827381134035,
|
|
"epoch": 0.01063909814836706,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.2209,
|
|
"mean_token_accuracy": 0.13601144254207612,
|
|
"num_tokens": 2254919.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.555957651138305,
|
|
"epoch": 0.010692830967298207,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.271,
|
|
"mean_token_accuracy": 0.13698455616831778,
|
|
"num_tokens": 2267583.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.436920642852783,
|
|
"epoch": 0.010746563786229353,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.3389,
|
|
"mean_token_accuracy": 0.13568416684865953,
|
|
"num_tokens": 2279228.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.481776905059815,
|
|
"epoch": 0.0108002966051605,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000499998026082006,
|
|
"loss": 6.2045,
|
|
"mean_token_accuracy": 0.14247661232948303,
|
|
"num_tokens": 2291213.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.456072664260864,
|
|
"epoch": 0.010854029424091647,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999900070995136,
|
|
"loss": 6.365,
|
|
"mean_token_accuracy": 0.12945087924599646,
|
|
"num_tokens": 2304080.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.436254262924194,
|
|
"epoch": 0.010907762243022794,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999758199023239,
|
|
"loss": 6.1827,
|
|
"mean_token_accuracy": 0.1411418817937374,
|
|
"num_tokens": 2315266.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.41214075088501,
|
|
"epoch": 0.01096149506195394,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999554648793858,
|
|
"loss": 6.2316,
|
|
"mean_token_accuracy": 0.13700917810201646,
|
|
"num_tokens": 2327333.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.507096195220948,
|
|
"epoch": 0.011015227880885088,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999289425887425,
|
|
"loss": 6.2534,
|
|
"mean_token_accuracy": 0.13917383402585984,
|
|
"num_tokens": 2338356.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.328510856628418,
|
|
"epoch": 0.011068960699816234,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004998962537575161,
|
|
"loss": 6.2425,
|
|
"mean_token_accuracy": 0.13840818852186204,
|
|
"num_tokens": 2348935.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.51852068901062,
|
|
"epoch": 0.011122693518747381,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998573992818874,
|
|
"loss": 6.21,
|
|
"mean_token_accuracy": 0.13913707360625266,
|
|
"num_tokens": 2359588.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.315857791900635,
|
|
"epoch": 0.011176426337678527,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004998123802270715,
|
|
"loss": 6.1941,
|
|
"mean_token_accuracy": 0.1444079250097275,
|
|
"num_tokens": 2369463.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.451612567901611,
|
|
"epoch": 0.011230159156609675,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004997611978272886,
|
|
"loss": 6.1782,
|
|
"mean_token_accuracy": 0.14179054498672486,
|
|
"num_tokens": 2381900.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.408647871017456,
|
|
"epoch": 0.01128389197554082,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004997038534857298,
|
|
"loss": 6.1465,
|
|
"mean_token_accuracy": 0.14524738639593124,
|
|
"num_tokens": 2393709.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.469882774353027,
|
|
"epoch": 0.011337624794471968,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004996403487745194,
|
|
"loss": 6.3334,
|
|
"mean_token_accuracy": 0.14204481840133668,
|
|
"num_tokens": 2405084.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.475049352645874,
|
|
"epoch": 0.011391357613403114,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000499570685434671,
|
|
"loss": 6.2438,
|
|
"mean_token_accuracy": 0.13890639692544937,
|
|
"num_tokens": 2416174.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.432177257537842,
|
|
"epoch": 0.011445090432334262,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004994948653760405,
|
|
"loss": 6.241,
|
|
"mean_token_accuracy": 0.13499629348516465,
|
|
"num_tokens": 2427222.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.369433784484864,
|
|
"epoch": 0.011498823251265408,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004994128906772729,
|
|
"loss": 6.1785,
|
|
"mean_token_accuracy": 0.1440369337797165,
|
|
"num_tokens": 2437796.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.392252683639526,
|
|
"epoch": 0.011552556070196555,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000499324763585746,
|
|
"loss": 6.2227,
|
|
"mean_token_accuracy": 0.14073971509933472,
|
|
"num_tokens": 2449065.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.409636116027832,
|
|
"epoch": 0.011606288889127701,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004992304865175085,
|
|
"loss": 6.2569,
|
|
"mean_token_accuracy": 0.14365285485982895,
|
|
"num_tokens": 2460652.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.43699197769165,
|
|
"epoch": 0.011660021708058849,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004991300620572138,
|
|
"loss": 6.1729,
|
|
"mean_token_accuracy": 0.14646220952272415,
|
|
"num_tokens": 2472364.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.4865803718566895,
|
|
"epoch": 0.011713754526989995,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004990234929580494,
|
|
"loss": 6.3948,
|
|
"mean_token_accuracy": 0.13162244707345963,
|
|
"num_tokens": 2485395.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.4109639644622805,
|
|
"epoch": 0.011767487345921143,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004989107821416609,
|
|
"loss": 6.286,
|
|
"mean_token_accuracy": 0.13713820427656173,
|
|
"num_tokens": 2498081.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.264999532699585,
|
|
"epoch": 0.011821220164852288,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004987919326980723,
|
|
"loss": 6.1171,
|
|
"mean_token_accuracy": 0.1371752992272377,
|
|
"num_tokens": 2508453.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.571272945404052,
|
|
"epoch": 0.011874952983783436,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004986669478856011,
|
|
"loss": 6.326,
|
|
"mean_token_accuracy": 0.12960056066513062,
|
|
"num_tokens": 2520474.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.443903303146362,
|
|
"epoch": 0.011928685802714582,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004985358311307688,
|
|
"loss": 6.158,
|
|
"mean_token_accuracy": 0.13977293744683267,
|
|
"num_tokens": 2531809.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.353152847290039,
|
|
"epoch": 0.01198241862164573,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004983985860282081,
|
|
"loss": 6.2273,
|
|
"mean_token_accuracy": 0.14189963564276695,
|
|
"num_tokens": 2542561.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.420229101181031,
|
|
"epoch": 0.012036151440576875,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004982552163405623,
|
|
"loss": 6.1896,
|
|
"mean_token_accuracy": 0.14108152464032173,
|
|
"num_tokens": 2554386.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.431586265563965,
|
|
"epoch": 0.012089884259508023,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004981057259983839,
|
|
"loss": 6.2606,
|
|
"mean_token_accuracy": 0.13941885083913802,
|
|
"num_tokens": 2565765.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.474301815032959,
|
|
"epoch": 0.012143617078439169,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004979501191000262,
|
|
"loss": 6.309,
|
|
"mean_token_accuracy": 0.13312820792198182,
|
|
"num_tokens": 2577496.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.4574237823486325,
|
|
"epoch": 0.012197349897370317,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004977883999115311,
|
|
"loss": 6.2035,
|
|
"mean_token_accuracy": 0.14595086351037026,
|
|
"num_tokens": 2588301.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.379302215576172,
|
|
"epoch": 0.012251082716301463,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004976205728665113,
|
|
"loss": 6.2631,
|
|
"mean_token_accuracy": 0.13971091583371162,
|
|
"num_tokens": 2599201.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.404904079437256,
|
|
"epoch": 0.01230481553523261,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004974466425660307,
|
|
"loss": 6.1603,
|
|
"mean_token_accuracy": 0.13958672881126405,
|
|
"num_tokens": 2611402.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.436859178543091,
|
|
"epoch": 0.012358548354163756,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004972666137784759,
|
|
"loss": 6.2561,
|
|
"mean_token_accuracy": 0.1411876492202282,
|
|
"num_tokens": 2623976.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.373582601547241,
|
|
"epoch": 0.012412281173094904,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004970804914394271,
|
|
"loss": 6.2529,
|
|
"mean_token_accuracy": 0.13747993111610413,
|
|
"num_tokens": 2635093.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.433339548110962,
|
|
"epoch": 0.01246601399202605,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004968882806515225,
|
|
"loss": 6.1381,
|
|
"mean_token_accuracy": 0.15079879462718965,
|
|
"num_tokens": 2646262.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.345254898071289,
|
|
"epoch": 0.012519746810957197,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004966899866843177,
|
|
"loss": 6.2442,
|
|
"mean_token_accuracy": 0.1461564227938652,
|
|
"num_tokens": 2657325.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.337249803543091,
|
|
"epoch": 0.012573479629888343,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000496485614974142,
|
|
"loss": 6.1118,
|
|
"mean_token_accuracy": 0.14194193556904794,
|
|
"num_tokens": 2669218.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.406280469894409,
|
|
"epoch": 0.01262721244881949,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004962751711239492,
|
|
"loss": 6.1312,
|
|
"mean_token_accuracy": 0.14137721359729766,
|
|
"num_tokens": 2681034.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.346201705932617,
|
|
"epoch": 0.012680945267750637,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004960586609031636,
|
|
"loss": 6.2434,
|
|
"mean_token_accuracy": 0.13876248747110367,
|
|
"num_tokens": 2692853.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.3197790622711185,
|
|
"epoch": 0.012734678086681784,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004958360902475224,
|
|
"loss": 5.9986,
|
|
"mean_token_accuracy": 0.15109058767557143,
|
|
"num_tokens": 2703470.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.242659568786621,
|
|
"epoch": 0.01278841090561293,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004956074652589125,
|
|
"loss": 6.1118,
|
|
"mean_token_accuracy": 0.14718493372201918,
|
|
"num_tokens": 2713740.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.295391702651978,
|
|
"epoch": 0.012842143724544078,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004953727922052035,
|
|
"loss": 6.0638,
|
|
"mean_token_accuracy": 0.15268367379903794,
|
|
"num_tokens": 2724951.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.261331224441529,
|
|
"epoch": 0.012895876543475224,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004951320775200756,
|
|
"loss": 6.0954,
|
|
"mean_token_accuracy": 0.15390754938125611,
|
|
"num_tokens": 2735545.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.3905986785888675,
|
|
"epoch": 0.012949609362406371,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004948853278028436,
|
|
"loss": 6.1651,
|
|
"mean_token_accuracy": 0.1437719851732254,
|
|
"num_tokens": 2747176.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.268735885620117,
|
|
"epoch": 0.013003342181337517,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004946325498182755,
|
|
"loss": 6.0428,
|
|
"mean_token_accuracy": 0.14660073667764664,
|
|
"num_tokens": 2757439.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.2571423053741455,
|
|
"epoch": 0.013057075000268665,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004943737504964076,
|
|
"loss": 6.0858,
|
|
"mean_token_accuracy": 0.14328044056892394,
|
|
"num_tokens": 2768832.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.417108821868896,
|
|
"epoch": 0.01311080781919981,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000494108936932354,
|
|
"loss": 6.2178,
|
|
"mean_token_accuracy": 0.13913026005029677,
|
|
"num_tokens": 2780557.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.278023815155029,
|
|
"epoch": 0.013164540638130958,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004938381163861124,
|
|
"loss": 6.1701,
|
|
"mean_token_accuracy": 0.1409448765218258,
|
|
"num_tokens": 2791630.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.337775039672851,
|
|
"epoch": 0.013218273457062104,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004935612962823645,
|
|
"loss": 6.1783,
|
|
"mean_token_accuracy": 0.14574046805500984,
|
|
"num_tokens": 2802920.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.382103633880615,
|
|
"epoch": 0.013272006275993252,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004932784842102739,
|
|
"loss": 6.2238,
|
|
"mean_token_accuracy": 0.13276997581124306,
|
|
"num_tokens": 2815119.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.2876763343811035,
|
|
"epoch": 0.013325739094924398,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004929896879232758,
|
|
"loss": 6.0663,
|
|
"mean_token_accuracy": 0.1527392864227295,
|
|
"num_tokens": 2825475.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.328493213653564,
|
|
"epoch": 0.013379471913855545,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004926949153388668,
|
|
"loss": 6.0968,
|
|
"mean_token_accuracy": 0.14614279344677925,
|
|
"num_tokens": 2836751.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.252760934829712,
|
|
"epoch": 0.013433204732786691,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004923941745383859,
|
|
"loss": 6.139,
|
|
"mean_token_accuracy": 0.14770944342017173,
|
|
"num_tokens": 2848396.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.351215839385986,
|
|
"epoch": 0.013486937551717839,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000492087473766794,
|
|
"loss": 6.1295,
|
|
"mean_token_accuracy": 0.14621034264564514,
|
|
"num_tokens": 2859635.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.2123987674713135,
|
|
"epoch": 0.013540670370648985,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000491774821432448,
|
|
"loss": 6.0881,
|
|
"mean_token_accuracy": 0.1439559243619442,
|
|
"num_tokens": 2871294.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.327570629119873,
|
|
"epoch": 0.013594403189580133,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004914562261068693,
|
|
"loss": 6.1526,
|
|
"mean_token_accuracy": 0.1407077744603157,
|
|
"num_tokens": 2881877.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.277802896499634,
|
|
"epoch": 0.013648136008511278,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004911316965245098,
|
|
"loss": 6.0476,
|
|
"mean_token_accuracy": 0.15187419354915618,
|
|
"num_tokens": 2893084.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.234827136993408,
|
|
"epoch": 0.013701868827442426,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000490801241582512,
|
|
"loss": 6.0462,
|
|
"mean_token_accuracy": 0.1518764987587929,
|
|
"num_tokens": 2903884.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.328769969940185,
|
|
"epoch": 0.013755601646373572,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000490464870340465,
|
|
"loss": 6.1729,
|
|
"mean_token_accuracy": 0.1456258215010166,
|
|
"num_tokens": 2916302.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.20207462310791,
|
|
"epoch": 0.01380933446530472,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004901225920201563,
|
|
"loss": 6.0262,
|
|
"mean_token_accuracy": 0.1456386551260948,
|
|
"num_tokens": 2926945.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.190553951263428,
|
|
"epoch": 0.013863067284235865,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000489774416005319,
|
|
"loss": 6.028,
|
|
"mean_token_accuracy": 0.15670164972543715,
|
|
"num_tokens": 2938305.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.295749282836914,
|
|
"epoch": 0.013916800103167013,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004894203518413742,
|
|
"loss": 6.1881,
|
|
"mean_token_accuracy": 0.14524212405085563,
|
|
"num_tokens": 2949652.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.41364369392395,
|
|
"epoch": 0.013970532922098159,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004890604092351701,
|
|
"loss": 6.0934,
|
|
"mean_token_accuracy": 0.15407448559999465,
|
|
"num_tokens": 2960237.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.0873411178588865,
|
|
"epoch": 0.014024265741029307,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000488694598054715,
|
|
"loss": 6.0066,
|
|
"mean_token_accuracy": 0.14816917181015016,
|
|
"num_tokens": 2971360.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.388800096511841,
|
|
"epoch": 0.014077998559960453,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004883229283289071,
|
|
"loss": 6.1488,
|
|
"mean_token_accuracy": 0.1438701219856739,
|
|
"num_tokens": 2981751.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.234357833862305,
|
|
"epoch": 0.0141317313788916,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00048794541024725993,
|
|
"loss": 6.1687,
|
|
"mean_token_accuracy": 0.13734994381666182,
|
|
"num_tokens": 2993284.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.4042885303497314,
|
|
"epoch": 0.014185464197822746,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004875620541596221,
|
|
"loss": 6.1477,
|
|
"mean_token_accuracy": 0.14683955833315848,
|
|
"num_tokens": 3004853.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.319836950302124,
|
|
"epoch": 0.014239197016753894,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00048717287057589454,
|
|
"loss": 6.136,
|
|
"mean_token_accuracy": 0.14580863192677498,
|
|
"num_tokens": 3015875.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.145222330093384,
|
|
"epoch": 0.01429292983568504,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004867778701657417,
|
|
"loss": 6.0159,
|
|
"mean_token_accuracy": 0.15017836838960646,
|
|
"num_tokens": 3026180.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.153919506072998,
|
|
"epoch": 0.014346662654616187,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00048637706375829955,
|
|
"loss": 5.9849,
|
|
"mean_token_accuracy": 0.1454961195588112,
|
|
"num_tokens": 3038647.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.223771286010742,
|
|
"epoch": 0.014400395473547333,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000485970462341878,
|
|
"loss": 5.9984,
|
|
"mean_token_accuracy": 0.14801558256149291,
|
|
"num_tokens": 3050087.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.259019470214843,
|
|
"epoch": 0.01445412829247848,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00048555807706366044,
|
|
"loss": 6.183,
|
|
"mean_token_accuracy": 0.15022045597434044,
|
|
"num_tokens": 3061962.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.317928457260132,
|
|
"epoch": 0.014507861111409627,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048513991922939756,
|
|
"loss": 6.1854,
|
|
"mean_token_accuracy": 0.13924189805984497,
|
|
"num_tokens": 3074375.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.198589038848877,
|
|
"epoch": 0.014561593930340774,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00048471600030309744,
|
|
"loss": 5.9442,
|
|
"mean_token_accuracy": 0.15290747433900834,
|
|
"num_tokens": 3085760.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.198614454269409,
|
|
"epoch": 0.01461532674927192,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048428633190671186,
|
|
"loss": 6.1072,
|
|
"mean_token_accuracy": 0.15340364277362822,
|
|
"num_tokens": 3096894.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.294735717773437,
|
|
"epoch": 0.014669059568203068,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004838509258198167,
|
|
"loss": 6.1162,
|
|
"mean_token_accuracy": 0.14734745621681214,
|
|
"num_tokens": 3108485.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.298273134231567,
|
|
"epoch": 0.014722792387134214,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048340979397929,
|
|
"loss": 6.0534,
|
|
"mean_token_accuracy": 0.14786504060029984,
|
|
"num_tokens": 3119499.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.128656005859375,
|
|
"epoch": 0.014776525206065361,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00048296294847898386,
|
|
"loss": 5.9904,
|
|
"mean_token_accuracy": 0.1504951611161232,
|
|
"num_tokens": 3130389.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.2860222339630125,
|
|
"epoch": 0.014830258024996507,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004825104015693934,
|
|
"loss": 6.1335,
|
|
"mean_token_accuracy": 0.1482733502984047,
|
|
"num_tokens": 3141034.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.265560674667358,
|
|
"epoch": 0.014883990843927655,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004820521656573208,
|
|
"loss": 6.0837,
|
|
"mean_token_accuracy": 0.14638752341270447,
|
|
"num_tokens": 3152438.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.274151802062988,
|
|
"epoch": 0.0149377236628588,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00048158825330553505,
|
|
"loss": 6.0729,
|
|
"mean_token_accuracy": 0.1501833215355873,
|
|
"num_tokens": 3163994.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.260838508605957,
|
|
"epoch": 0.014991456481789948,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00048111867723242763,
|
|
"loss": 6.0426,
|
|
"mean_token_accuracy": 0.1526077687740326,
|
|
"num_tokens": 3175330.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.147992849349976,
|
|
"epoch": 0.015045189300721094,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004806434503116637,
|
|
"loss": 5.9983,
|
|
"mean_token_accuracy": 0.15310876518487931,
|
|
"num_tokens": 3185981.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.192197990417481,
|
|
"epoch": 0.015098922119652242,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004801625855718296,
|
|
"loss": 5.979,
|
|
"mean_token_accuracy": 0.1544252395629883,
|
|
"num_tokens": 3198267.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.26663064956665,
|
|
"epoch": 0.015152654938583388,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00047967609619607477,
|
|
"loss": 6.0358,
|
|
"mean_token_accuracy": 0.14738833606243135,
|
|
"num_tokens": 3210405.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.139094734191895,
|
|
"epoch": 0.015206387757514536,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004791839955217513,
|
|
"loss": 6.0653,
|
|
"mean_token_accuracy": 0.14450788348913193,
|
|
"num_tokens": 3220528.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.196222591400146,
|
|
"epoch": 0.015260120576445681,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047868629704004786,
|
|
"loss": 6.0286,
|
|
"mean_token_accuracy": 0.14643459096550943,
|
|
"num_tokens": 3231623.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.109784889221191,
|
|
"epoch": 0.015313853395376829,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047818301439561965,
|
|
"loss": 6.0073,
|
|
"mean_token_accuracy": 0.1485434204339981,
|
|
"num_tokens": 3242067.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.153221464157104,
|
|
"epoch": 0.015367586214307975,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047767416138621454,
|
|
"loss": 5.9263,
|
|
"mean_token_accuracy": 0.1522674724459648,
|
|
"num_tokens": 3252898.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.196251726150512,
|
|
"epoch": 0.015421319033239123,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000477159751962295,
|
|
"loss": 6.1578,
|
|
"mean_token_accuracy": 0.14196772649884223,
|
|
"num_tokens": 3265128.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.312836074829102,
|
|
"epoch": 0.015475051852170268,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00047663980022665507,
|
|
"loss": 6.0614,
|
|
"mean_token_accuracy": 0.1504521295428276,
|
|
"num_tokens": 3275808.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.115658760070801,
|
|
"epoch": 0.015528784671101416,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047611432043403437,
|
|
"loss": 6.0512,
|
|
"mean_token_accuracy": 0.15013080686330796,
|
|
"num_tokens": 3286682.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.259523916244507,
|
|
"epoch": 0.015582517490032562,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004755833269907267,
|
|
"loss": 6.0835,
|
|
"mean_token_accuracy": 0.15509197413921355,
|
|
"num_tokens": 3299402.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.286191463470459,
|
|
"epoch": 0.01563625030896371,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004750468344541857,
|
|
"loss": 6.1418,
|
|
"mean_token_accuracy": 0.14176064357161522,
|
|
"num_tokens": 3311185.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.329611492156983,
|
|
"epoch": 0.015689983127894856,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047450485753262525,
|
|
"loss": 6.0735,
|
|
"mean_token_accuracy": 0.14907490611076354,
|
|
"num_tokens": 3322489.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.127014255523681,
|
|
"epoch": 0.015743715946826,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00047395741108461633,
|
|
"loss": 6.0965,
|
|
"mean_token_accuracy": 0.1506562888622284,
|
|
"num_tokens": 3334731.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.229137897491455,
|
|
"epoch": 0.01579744876575715,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00047340451011867985,
|
|
"loss": 6.0128,
|
|
"mean_token_accuracy": 0.15023027807474137,
|
|
"num_tokens": 3347107.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.205422401428223,
|
|
"epoch": 0.015851181584688297,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00047284616979287515,
|
|
"loss": 6.0012,
|
|
"mean_token_accuracy": 0.15306572914123534,
|
|
"num_tokens": 3358640.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.24063081741333,
|
|
"epoch": 0.015904914403619443,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00047228240541438433,
|
|
"loss": 6.1422,
|
|
"mean_token_accuracy": 0.14401476681232453,
|
|
"num_tokens": 3369743.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.308213472366333,
|
|
"epoch": 0.01595864722255059,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047171323243909257,
|
|
"loss": 6.1041,
|
|
"mean_token_accuracy": 0.1444757252931595,
|
|
"num_tokens": 3381413.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.189971303939819,
|
|
"epoch": 0.016012380041481738,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00047113866647116457,
|
|
"loss": 5.9555,
|
|
"mean_token_accuracy": 0.15134403705596924,
|
|
"num_tokens": 3391347.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.308747673034668,
|
|
"epoch": 0.016066112860412884,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004705587232626164,
|
|
"loss": 6.1736,
|
|
"mean_token_accuracy": 0.1430051103234291,
|
|
"num_tokens": 3403348.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.152447462081909,
|
|
"epoch": 0.01611984567934403,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00046997341871288424,
|
|
"loss": 5.9481,
|
|
"mean_token_accuracy": 0.15207927227020263,
|
|
"num_tokens": 3414845.0,
|
|
"step": 1500
|
|
}
|
|
],
|
|
"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": 791493223219200.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|