8079 lines
219 KiB
JSON
8079 lines
219 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.1309018942606728,
|
|
"eval_steps": 1000,
|
|
"global_step": 4000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.74207363128662,
|
|
"epoch": 0.0014136273678258412,
|
|
"grad_norm": 6.1875,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.5511,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 4340.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.742022514343262,
|
|
"epoch": 0.0028272547356516823,
|
|
"grad_norm": 6.71875,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.5265,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 8692.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.742014980316162,
|
|
"epoch": 0.004240882103477523,
|
|
"grad_norm": 6.78125,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.4477,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 12453.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.742022132873535,
|
|
"epoch": 0.005654509471303365,
|
|
"grad_norm": 6.71875,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.4201,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 16439.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.741988182067871,
|
|
"epoch": 0.007068136839129205,
|
|
"grad_norm": 6.8125,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.295,
|
|
"mean_token_accuracy": 0.0014771121321246028,
|
|
"num_tokens": 20515.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.74176025390625,
|
|
"epoch": 0.008481764206955046,
|
|
"grad_norm": 5.5,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.1401,
|
|
"mean_token_accuracy": 0.029323107562959194,
|
|
"num_tokens": 24336.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.741047763824463,
|
|
"epoch": 0.009895391574780888,
|
|
"grad_norm": 4.5625,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.0398,
|
|
"mean_token_accuracy": 0.05071598924696445,
|
|
"num_tokens": 28465.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.739372825622558,
|
|
"epoch": 0.01130901894260673,
|
|
"grad_norm": 3.90625,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.828,
|
|
"mean_token_accuracy": 0.06147046238183975,
|
|
"num_tokens": 32217.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.73618106842041,
|
|
"epoch": 0.01272264631043257,
|
|
"grad_norm": 3.015625,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.7393,
|
|
"mean_token_accuracy": 0.05857079476118088,
|
|
"num_tokens": 36369.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.732595252990723,
|
|
"epoch": 0.01413627367825841,
|
|
"grad_norm": 2.859375,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.6251,
|
|
"mean_token_accuracy": 0.0617414329200983,
|
|
"num_tokens": 40227.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.730756187438965,
|
|
"epoch": 0.015549901046084252,
|
|
"grad_norm": 2.5,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.6041,
|
|
"mean_token_accuracy": 0.058244453743100166,
|
|
"num_tokens": 44266.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.728981113433838,
|
|
"epoch": 0.016963528413910092,
|
|
"grad_norm": 2.640625,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.5679,
|
|
"mean_token_accuracy": 0.06274866946041584,
|
|
"num_tokens": 48187.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.726116180419922,
|
|
"epoch": 0.018377155781735936,
|
|
"grad_norm": 2.453125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.4557,
|
|
"mean_token_accuracy": 0.05981031022965908,
|
|
"num_tokens": 52252.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.721884059906007,
|
|
"epoch": 0.019790783149561775,
|
|
"grad_norm": 2.53125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.4336,
|
|
"mean_token_accuracy": 0.05559282228350639,
|
|
"num_tokens": 56421.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.7160475730896,
|
|
"epoch": 0.021204410517387615,
|
|
"grad_norm": 2.421875,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.3486,
|
|
"mean_token_accuracy": 0.06220139227807522,
|
|
"num_tokens": 60341.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.70057315826416,
|
|
"epoch": 0.02261803788521346,
|
|
"grad_norm": 2.421875,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.216,
|
|
"mean_token_accuracy": 0.06268311068415641,
|
|
"num_tokens": 64465.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.682071495056153,
|
|
"epoch": 0.0240316652530393,
|
|
"grad_norm": 2.21875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.1917,
|
|
"mean_token_accuracy": 0.06404139809310436,
|
|
"num_tokens": 68514.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.654492664337159,
|
|
"epoch": 0.02544529262086514,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.1028,
|
|
"mean_token_accuracy": 0.06569066680967808,
|
|
"num_tokens": 72661.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.621818923950196,
|
|
"epoch": 0.02685891998869098,
|
|
"grad_norm": 2.203125,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.9739,
|
|
"mean_token_accuracy": 0.07454013973474502,
|
|
"num_tokens": 76698.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.587986373901368,
|
|
"epoch": 0.02827254735651682,
|
|
"grad_norm": 2.109375,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.8935,
|
|
"mean_token_accuracy": 0.07267784662544727,
|
|
"num_tokens": 80732.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.545151424407958,
|
|
"epoch": 0.029686174724342665,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.8022,
|
|
"mean_token_accuracy": 0.06954946778714657,
|
|
"num_tokens": 84797.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.516861534118652,
|
|
"epoch": 0.031099802092168505,
|
|
"grad_norm": 2.234375,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.722,
|
|
"mean_token_accuracy": 0.07358665019273758,
|
|
"num_tokens": 88758.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.459677600860596,
|
|
"epoch": 0.032513429459994345,
|
|
"grad_norm": 1.953125,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.6039,
|
|
"mean_token_accuracy": 0.07231484800577163,
|
|
"num_tokens": 92722.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.407901668548584,
|
|
"epoch": 0.033927056827820185,
|
|
"grad_norm": 2.125,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.4476,
|
|
"mean_token_accuracy": 0.07675158120691776,
|
|
"num_tokens": 96602.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.325055503845215,
|
|
"epoch": 0.035340684195646024,
|
|
"grad_norm": 1.9453125,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.3606,
|
|
"mean_token_accuracy": 0.07958476431667805,
|
|
"num_tokens": 100598.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.222711753845214,
|
|
"epoch": 0.03675431156347187,
|
|
"grad_norm": 1.875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.2593,
|
|
"mean_token_accuracy": 0.07609154582023621,
|
|
"num_tokens": 104460.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.145068645477295,
|
|
"epoch": 0.03816793893129771,
|
|
"grad_norm": 2.109375,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.1373,
|
|
"mean_token_accuracy": 0.07444728389382363,
|
|
"num_tokens": 108360.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.019375801086426,
|
|
"epoch": 0.03958156629912355,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.9991,
|
|
"mean_token_accuracy": 0.07835616618394851,
|
|
"num_tokens": 112176.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.877924728393555,
|
|
"epoch": 0.04099519366694939,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.0465,
|
|
"mean_token_accuracy": 0.07128159664571285,
|
|
"num_tokens": 116192.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 9.766163635253907,
|
|
"epoch": 0.04240882103477523,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.9125,
|
|
"mean_token_accuracy": 0.07681304663419723,
|
|
"num_tokens": 120221.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.596385955810547,
|
|
"epoch": 0.04382244840260108,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.8471,
|
|
"mean_token_accuracy": 0.07160197533667087,
|
|
"num_tokens": 124364.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.397625064849853,
|
|
"epoch": 0.04523607577042692,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.6486,
|
|
"mean_token_accuracy": 0.08006830923259259,
|
|
"num_tokens": 128266.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.197708892822266,
|
|
"epoch": 0.04664970313825276,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.5695,
|
|
"mean_token_accuracy": 0.08508073762059212,
|
|
"num_tokens": 131985.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.003838157653808,
|
|
"epoch": 0.0480633305060786,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.4586,
|
|
"mean_token_accuracy": 0.08466509208083153,
|
|
"num_tokens": 135892.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.81080961227417,
|
|
"epoch": 0.04947695787390444,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.5228,
|
|
"mean_token_accuracy": 0.075396229326725,
|
|
"num_tokens": 139937.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.659114837646484,
|
|
"epoch": 0.05089058524173028,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.3426,
|
|
"mean_token_accuracy": 0.08097245395183564,
|
|
"num_tokens": 143659.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.553722858428955,
|
|
"epoch": 0.052304212609556124,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.4259,
|
|
"mean_token_accuracy": 0.08198288567364216,
|
|
"num_tokens": 147394.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.326274490356445,
|
|
"epoch": 0.05371783997738196,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.2628,
|
|
"mean_token_accuracy": 0.08642936870455742,
|
|
"num_tokens": 151082.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.238064765930176,
|
|
"epoch": 0.0551314673452078,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.25,
|
|
"mean_token_accuracy": 0.08761365041136741,
|
|
"num_tokens": 155099.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.188647174835205,
|
|
"epoch": 0.05654509471303364,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.4052,
|
|
"mean_token_accuracy": 0.08035821095108986,
|
|
"num_tokens": 159297.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.190512943267823,
|
|
"epoch": 0.05795872208085948,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.268,
|
|
"mean_token_accuracy": 0.08253459706902504,
|
|
"num_tokens": 163337.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.12247543334961,
|
|
"epoch": 0.05937234944868533,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.3087,
|
|
"mean_token_accuracy": 0.08554964698851109,
|
|
"num_tokens": 167440.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.021088409423829,
|
|
"epoch": 0.06078597681651117,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.1661,
|
|
"mean_token_accuracy": 0.08840125799179077,
|
|
"num_tokens": 171238.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.036321210861207,
|
|
"epoch": 0.06219960418433701,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.2139,
|
|
"mean_token_accuracy": 0.0908890649676323,
|
|
"num_tokens": 175476.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.951605987548828,
|
|
"epoch": 0.06361323155216285,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.1261,
|
|
"mean_token_accuracy": 0.08661228194832801,
|
|
"num_tokens": 179398.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.849934148788452,
|
|
"epoch": 0.06502685891998869,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.1123,
|
|
"mean_token_accuracy": 0.09166858419775963,
|
|
"num_tokens": 183355.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.945307731628418,
|
|
"epoch": 0.06644048628781453,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.1135,
|
|
"mean_token_accuracy": 0.08504104316234588,
|
|
"num_tokens": 187278.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.837193965911865,
|
|
"epoch": 0.06785411365564037,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.1333,
|
|
"mean_token_accuracy": 0.09098687320947647,
|
|
"num_tokens": 191136.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.820778036117554,
|
|
"epoch": 0.06926774102346621,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.1943,
|
|
"mean_token_accuracy": 0.0956549420952797,
|
|
"num_tokens": 195096.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.859644556045533,
|
|
"epoch": 0.07068136839129205,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.1855,
|
|
"mean_token_accuracy": 0.08993390724062919,
|
|
"num_tokens": 199483.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.814686632156372,
|
|
"epoch": 0.0720949957591179,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.1139,
|
|
"mean_token_accuracy": 0.09194719940423965,
|
|
"num_tokens": 203515.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.748371124267578,
|
|
"epoch": 0.07350862312694374,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.116,
|
|
"mean_token_accuracy": 0.09589750841259956,
|
|
"num_tokens": 207451.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.774532127380371,
|
|
"epoch": 0.07492225049476958,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.0308,
|
|
"mean_token_accuracy": 0.1008292630314827,
|
|
"num_tokens": 211507.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.691170597076416,
|
|
"epoch": 0.07633587786259542,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.09,
|
|
"mean_token_accuracy": 0.09460079073905944,
|
|
"num_tokens": 215236.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.701872444152832,
|
|
"epoch": 0.07774950523042126,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.0577,
|
|
"mean_token_accuracy": 0.10146123990416527,
|
|
"num_tokens": 218934.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.6655951023101805,
|
|
"epoch": 0.0791631325982471,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 6.9994,
|
|
"mean_token_accuracy": 0.10914008393883705,
|
|
"num_tokens": 223185.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.753876638412476,
|
|
"epoch": 0.08057675996607294,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.1232,
|
|
"mean_token_accuracy": 0.09365936666727066,
|
|
"num_tokens": 227044.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.66315860748291,
|
|
"epoch": 0.08199038733389878,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.0308,
|
|
"mean_token_accuracy": 0.10147744789719582,
|
|
"num_tokens": 231359.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.710978984832764,
|
|
"epoch": 0.08340401470172462,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.0018,
|
|
"mean_token_accuracy": 0.11066402792930603,
|
|
"num_tokens": 235224.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.5690467834472654,
|
|
"epoch": 0.08481764206955046,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.8653,
|
|
"mean_token_accuracy": 0.11461045518517494,
|
|
"num_tokens": 239119.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.518357563018799,
|
|
"epoch": 0.0862312694373763,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.9653,
|
|
"mean_token_accuracy": 0.10772728994488716,
|
|
"num_tokens": 243275.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.694993162155152,
|
|
"epoch": 0.08764489680520215,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 6.9709,
|
|
"mean_token_accuracy": 0.11304818466305733,
|
|
"num_tokens": 247415.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.528787612915039,
|
|
"epoch": 0.089058524173028,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.9654,
|
|
"mean_token_accuracy": 0.10677048042416573,
|
|
"num_tokens": 251575.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.5503821849823,
|
|
"epoch": 0.09047215154085383,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.9441,
|
|
"mean_token_accuracy": 0.11381662338972091,
|
|
"num_tokens": 255686.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.599794197082519,
|
|
"epoch": 0.09188577890867967,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.9401,
|
|
"mean_token_accuracy": 0.11095880568027497,
|
|
"num_tokens": 259850.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.6137940883636475,
|
|
"epoch": 0.09329940627650551,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.9622,
|
|
"mean_token_accuracy": 0.10780008137226105,
|
|
"num_tokens": 264043.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.65163402557373,
|
|
"epoch": 0.09471303364433135,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.0215,
|
|
"mean_token_accuracy": 0.1125954583287239,
|
|
"num_tokens": 268296.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.420255088806153,
|
|
"epoch": 0.0961266610121572,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.836,
|
|
"mean_token_accuracy": 0.11534447073936463,
|
|
"num_tokens": 272062.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.355507898330688,
|
|
"epoch": 0.09754028837998303,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.8008,
|
|
"mean_token_accuracy": 0.12810793220996858,
|
|
"num_tokens": 275936.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.469290637969971,
|
|
"epoch": 0.09895391574780887,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.8213,
|
|
"mean_token_accuracy": 0.1251584582030773,
|
|
"num_tokens": 279526.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.5056859970092775,
|
|
"epoch": 0.10036754311563471,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.8874,
|
|
"mean_token_accuracy": 0.11884394437074661,
|
|
"num_tokens": 283577.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.507189226150513,
|
|
"epoch": 0.10178117048346055,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.7619,
|
|
"mean_token_accuracy": 0.112952870875597,
|
|
"num_tokens": 287377.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.379469537734986,
|
|
"epoch": 0.10319479785128641,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.9101,
|
|
"mean_token_accuracy": 0.1114804707467556,
|
|
"num_tokens": 291582.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.508454275131226,
|
|
"epoch": 0.10460842521911225,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.7891,
|
|
"mean_token_accuracy": 0.12308030501008034,
|
|
"num_tokens": 295608.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.4715001583099365,
|
|
"epoch": 0.10602205258693809,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.8013,
|
|
"mean_token_accuracy": 0.12096530124545098,
|
|
"num_tokens": 299660.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.332260084152222,
|
|
"epoch": 0.10743567995476393,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.7183,
|
|
"mean_token_accuracy": 0.13139156624674797,
|
|
"num_tokens": 303412.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.269100999832153,
|
|
"epoch": 0.10884930732258977,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.7782,
|
|
"mean_token_accuracy": 0.12529514580965043,
|
|
"num_tokens": 307352.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.428966236114502,
|
|
"epoch": 0.1102629346904156,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.8533,
|
|
"mean_token_accuracy": 0.1281038463115692,
|
|
"num_tokens": 311418.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.324939203262329,
|
|
"epoch": 0.11167656205824145,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.6783,
|
|
"mean_token_accuracy": 0.13713634759187698,
|
|
"num_tokens": 315150.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.591144180297851,
|
|
"epoch": 0.11309018942606729,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.8727,
|
|
"mean_token_accuracy": 0.11698746904730797,
|
|
"num_tokens": 318898.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.242686319351196,
|
|
"epoch": 0.11450381679389313,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.6935,
|
|
"mean_token_accuracy": 0.13000213503837585,
|
|
"num_tokens": 322495.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.333932304382325,
|
|
"epoch": 0.11591744416171897,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.8398,
|
|
"mean_token_accuracy": 0.12843888401985168,
|
|
"num_tokens": 326473.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.408268976211548,
|
|
"epoch": 0.1173310715295448,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.7863,
|
|
"mean_token_accuracy": 0.12803610786795616,
|
|
"num_tokens": 330393.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.3277123928070065,
|
|
"epoch": 0.11874469889737066,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.7054,
|
|
"mean_token_accuracy": 0.13229965120553971,
|
|
"num_tokens": 334449.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.445288515090942,
|
|
"epoch": 0.1201583262651965,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.8181,
|
|
"mean_token_accuracy": 0.11966062635183335,
|
|
"num_tokens": 338740.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.247825288772583,
|
|
"epoch": 0.12157195363302234,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.6722,
|
|
"mean_token_accuracy": 0.13701122775673866,
|
|
"num_tokens": 342429.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.355695915222168,
|
|
"epoch": 0.12298558100084818,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.6994,
|
|
"mean_token_accuracy": 0.13420747891068457,
|
|
"num_tokens": 346218.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.303587532043457,
|
|
"epoch": 0.12439920836867402,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.696,
|
|
"mean_token_accuracy": 0.1381567671895027,
|
|
"num_tokens": 350310.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.251111030578613,
|
|
"epoch": 0.12581283573649985,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.6712,
|
|
"mean_token_accuracy": 0.13844901993870734,
|
|
"num_tokens": 353952.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.367745399475098,
|
|
"epoch": 0.1272264631043257,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.851,
|
|
"mean_token_accuracy": 0.1284424804151058,
|
|
"num_tokens": 357667.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.356730890274048,
|
|
"epoch": 0.12864009047215155,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.6561,
|
|
"mean_token_accuracy": 0.138033926486969,
|
|
"num_tokens": 361309.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.101002645492554,
|
|
"epoch": 0.13005371783997738,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.6954,
|
|
"mean_token_accuracy": 0.1395917169749737,
|
|
"num_tokens": 365099.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.374086189270019,
|
|
"epoch": 0.13146734520780323,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.7496,
|
|
"mean_token_accuracy": 0.13763813823461532,
|
|
"num_tokens": 369141.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.359116935729981,
|
|
"epoch": 0.13288097257562906,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.6675,
|
|
"mean_token_accuracy": 0.1343040108680725,
|
|
"num_tokens": 373128.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.0868512153625485,
|
|
"epoch": 0.1342945999434549,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.6654,
|
|
"mean_token_accuracy": 0.14124030545353888,
|
|
"num_tokens": 377103.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.250053071975708,
|
|
"epoch": 0.13570822731128074,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.7219,
|
|
"mean_token_accuracy": 0.13121307715773584,
|
|
"num_tokens": 381573.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.313079786300659,
|
|
"epoch": 0.1371218546791066,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.7674,
|
|
"mean_token_accuracy": 0.13335683792829514,
|
|
"num_tokens": 385759.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.177998399734497,
|
|
"epoch": 0.13853548204693242,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.6308,
|
|
"mean_token_accuracy": 0.14019295275211335,
|
|
"num_tokens": 389700.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.207444953918457,
|
|
"epoch": 0.13994910941475827,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.5526,
|
|
"mean_token_accuracy": 0.1400491252541542,
|
|
"num_tokens": 393279.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.098467206954956,
|
|
"epoch": 0.1413627367825841,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.639,
|
|
"mean_token_accuracy": 0.13535916954278945,
|
|
"num_tokens": 397379.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.206504201889038,
|
|
"epoch": 0.14277636415040995,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.6984,
|
|
"mean_token_accuracy": 0.13445916026830673,
|
|
"num_tokens": 401349.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.222359657287598,
|
|
"epoch": 0.1441899915182358,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.6017,
|
|
"mean_token_accuracy": 0.1456646129488945,
|
|
"num_tokens": 405361.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.192652940750122,
|
|
"epoch": 0.14560361888606163,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.6375,
|
|
"mean_token_accuracy": 0.14064312353730202,
|
|
"num_tokens": 409614.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.041819143295288,
|
|
"epoch": 0.14701724625388748,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.5656,
|
|
"mean_token_accuracy": 0.14632121846079826,
|
|
"num_tokens": 413529.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.031795978546143,
|
|
"epoch": 0.1484308736217133,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.5605,
|
|
"mean_token_accuracy": 0.14058664664626122,
|
|
"num_tokens": 417365.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.1277703762054445,
|
|
"epoch": 0.14984450098953916,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.5435,
|
|
"mean_token_accuracy": 0.1460522949695587,
|
|
"num_tokens": 421232.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.189267683029175,
|
|
"epoch": 0.151258128357365,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.6245,
|
|
"mean_token_accuracy": 0.1368570201098919,
|
|
"num_tokens": 425503.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.121035575866699,
|
|
"epoch": 0.15267175572519084,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.6376,
|
|
"mean_token_accuracy": 0.15090529546141623,
|
|
"num_tokens": 429770.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.14065523147583,
|
|
"epoch": 0.15408538309301667,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.5419,
|
|
"mean_token_accuracy": 0.14979021847248078,
|
|
"num_tokens": 433657.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.119353485107422,
|
|
"epoch": 0.15549901046084252,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.6487,
|
|
"mean_token_accuracy": 0.1421989232301712,
|
|
"num_tokens": 437732.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.953150367736816,
|
|
"epoch": 0.15691263782866835,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.3952,
|
|
"mean_token_accuracy": 0.15552812665700913,
|
|
"num_tokens": 441866.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.980320453643799,
|
|
"epoch": 0.1583262651964942,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.4783,
|
|
"mean_token_accuracy": 0.14983998835086823,
|
|
"num_tokens": 445437.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.0185740947723385,
|
|
"epoch": 0.15973989256432006,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.4586,
|
|
"mean_token_accuracy": 0.15214325338602067,
|
|
"num_tokens": 449602.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.880740737915039,
|
|
"epoch": 0.16115351993214588,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.3774,
|
|
"mean_token_accuracy": 0.14884328693151475,
|
|
"num_tokens": 453140.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.237173557281494,
|
|
"epoch": 0.16256714729997174,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.6234,
|
|
"mean_token_accuracy": 0.13322453647851945,
|
|
"num_tokens": 457210.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.069291353225708,
|
|
"epoch": 0.16398077466779756,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.5821,
|
|
"mean_token_accuracy": 0.14147427529096604,
|
|
"num_tokens": 461534.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.033083820343018,
|
|
"epoch": 0.16539440203562342,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.4198,
|
|
"mean_token_accuracy": 0.14874927774071695,
|
|
"num_tokens": 465055.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.058513498306274,
|
|
"epoch": 0.16680802940344924,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.444,
|
|
"mean_token_accuracy": 0.14927454963326453,
|
|
"num_tokens": 469133.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.091696453094483,
|
|
"epoch": 0.1682216567712751,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.5729,
|
|
"mean_token_accuracy": 0.14638903141021728,
|
|
"num_tokens": 473078.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.0970518589019775,
|
|
"epoch": 0.16963528413910092,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.5146,
|
|
"mean_token_accuracy": 0.14790180325508118,
|
|
"num_tokens": 476988.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.953248834609985,
|
|
"epoch": 0.17104891150692678,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.3748,
|
|
"mean_token_accuracy": 0.15677656084299088,
|
|
"num_tokens": 481170.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.832110261917114,
|
|
"epoch": 0.1724625388747526,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.3622,
|
|
"mean_token_accuracy": 0.1458218924701214,
|
|
"num_tokens": 484853.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.015735912322998,
|
|
"epoch": 0.17387616624257846,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.4019,
|
|
"mean_token_accuracy": 0.15347179323434829,
|
|
"num_tokens": 488550.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.82008581161499,
|
|
"epoch": 0.1752897936104043,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.4353,
|
|
"mean_token_accuracy": 0.15502589270472528,
|
|
"num_tokens": 492621.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.970222520828247,
|
|
"epoch": 0.17670342097823014,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.4046,
|
|
"mean_token_accuracy": 0.14566168785095215,
|
|
"num_tokens": 496627.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.795285272598266,
|
|
"epoch": 0.178117048346056,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.4612,
|
|
"mean_token_accuracy": 0.14936821013689042,
|
|
"num_tokens": 500794.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.059206104278564,
|
|
"epoch": 0.17953067571388182,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.4831,
|
|
"mean_token_accuracy": 0.14905075654387473,
|
|
"num_tokens": 504672.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.794796943664551,
|
|
"epoch": 0.18094430308170767,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.4242,
|
|
"mean_token_accuracy": 0.16338785141706466,
|
|
"num_tokens": 508834.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.082453823089599,
|
|
"epoch": 0.1823579304495335,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.388,
|
|
"mean_token_accuracy": 0.15189965814352036,
|
|
"num_tokens": 512722.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.872324419021607,
|
|
"epoch": 0.18377155781735935,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.4785,
|
|
"mean_token_accuracy": 0.14856359511613845,
|
|
"num_tokens": 517270.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.005723524093628,
|
|
"epoch": 0.18518518518518517,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.4087,
|
|
"mean_token_accuracy": 0.15056324899196624,
|
|
"num_tokens": 521438.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.857723665237427,
|
|
"epoch": 0.18659881255301103,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.3936,
|
|
"mean_token_accuracy": 0.14879612773656845,
|
|
"num_tokens": 525890.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.036524677276612,
|
|
"epoch": 0.18801243992083685,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.4232,
|
|
"mean_token_accuracy": 0.149623654037714,
|
|
"num_tokens": 529775.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.84811635017395,
|
|
"epoch": 0.1894260672886627,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.4021,
|
|
"mean_token_accuracy": 0.1505955122411251,
|
|
"num_tokens": 533889.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.79489049911499,
|
|
"epoch": 0.19083969465648856,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.3423,
|
|
"mean_token_accuracy": 0.16277977973222732,
|
|
"num_tokens": 537701.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.869364881515503,
|
|
"epoch": 0.1922533220243144,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.3895,
|
|
"mean_token_accuracy": 0.15102225691080093,
|
|
"num_tokens": 542160.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.848371076583862,
|
|
"epoch": 0.19366694939214024,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.3673,
|
|
"mean_token_accuracy": 0.1594443902373314,
|
|
"num_tokens": 546317.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.84291672706604,
|
|
"epoch": 0.19508057675996607,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.2123,
|
|
"mean_token_accuracy": 0.1704620197415352,
|
|
"num_tokens": 550304.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.779688024520874,
|
|
"epoch": 0.19649420412779192,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.3247,
|
|
"mean_token_accuracy": 0.1501743510365486,
|
|
"num_tokens": 554276.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.807392883300781,
|
|
"epoch": 0.19790783149561775,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.3376,
|
|
"mean_token_accuracy": 0.15815870016813277,
|
|
"num_tokens": 558044.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.69501748085022,
|
|
"epoch": 0.1993214588634436,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.3189,
|
|
"mean_token_accuracy": 0.15602569431066513,
|
|
"num_tokens": 561667.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.884977722167969,
|
|
"epoch": 0.20073508623126943,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.3569,
|
|
"mean_token_accuracy": 0.14620595276355744,
|
|
"num_tokens": 565738.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.865601634979248,
|
|
"epoch": 0.20214871359909528,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.3868,
|
|
"mean_token_accuracy": 0.15818096846342086,
|
|
"num_tokens": 569629.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.779451847076416,
|
|
"epoch": 0.2035623409669211,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.2616,
|
|
"mean_token_accuracy": 0.1626145839691162,
|
|
"num_tokens": 573272.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.587831258773804,
|
|
"epoch": 0.20497596833474696,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.3088,
|
|
"mean_token_accuracy": 0.15946338027715684,
|
|
"num_tokens": 577224.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.784856128692627,
|
|
"epoch": 0.20638959570257281,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.2521,
|
|
"mean_token_accuracy": 0.16293586939573287,
|
|
"num_tokens": 580919.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.7658926486969,
|
|
"epoch": 0.20780322307039864,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.3651,
|
|
"mean_token_accuracy": 0.15333567410707474,
|
|
"num_tokens": 584985.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.826333856582641,
|
|
"epoch": 0.2092168504382245,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.3085,
|
|
"mean_token_accuracy": 0.16295087337493896,
|
|
"num_tokens": 588979.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.648220825195312,
|
|
"epoch": 0.21063047780605032,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.2787,
|
|
"mean_token_accuracy": 0.16390681266784668,
|
|
"num_tokens": 592930.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.663014554977417,
|
|
"epoch": 0.21204410517387617,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.3273,
|
|
"mean_token_accuracy": 0.16125326007604598,
|
|
"num_tokens": 596832.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.937385177612304,
|
|
"epoch": 0.213457732541702,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.3617,
|
|
"mean_token_accuracy": 0.15750757604837418,
|
|
"num_tokens": 600934.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.72532844543457,
|
|
"epoch": 0.21487135990952785,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.2373,
|
|
"mean_token_accuracy": 0.15954903662204742,
|
|
"num_tokens": 604868.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.742078399658203,
|
|
"epoch": 0.21628498727735368,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.3596,
|
|
"mean_token_accuracy": 0.150635027885437,
|
|
"num_tokens": 608747.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.704383659362793,
|
|
"epoch": 0.21769861464517953,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.3572,
|
|
"mean_token_accuracy": 0.15894525349140168,
|
|
"num_tokens": 612627.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.658515930175781,
|
|
"epoch": 0.21911224201300536,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.2251,
|
|
"mean_token_accuracy": 0.17111287266016006,
|
|
"num_tokens": 616455.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.6746655941009525,
|
|
"epoch": 0.2205258693808312,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.326,
|
|
"mean_token_accuracy": 0.15957299321889878,
|
|
"num_tokens": 620617.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.608477210998535,
|
|
"epoch": 0.22193949674865707,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.2692,
|
|
"mean_token_accuracy": 0.1650322839617729,
|
|
"num_tokens": 624863.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.672313022613525,
|
|
"epoch": 0.2233531241164829,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.2819,
|
|
"mean_token_accuracy": 0.165033021569252,
|
|
"num_tokens": 629073.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.696346044540405,
|
|
"epoch": 0.22476675148430875,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.3095,
|
|
"mean_token_accuracy": 0.16314611732959747,
|
|
"num_tokens": 633550.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.60796046257019,
|
|
"epoch": 0.22618037885213457,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.2888,
|
|
"mean_token_accuracy": 0.1602451756596565,
|
|
"num_tokens": 637459.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.599572229385376,
|
|
"epoch": 0.22759400621996043,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.2433,
|
|
"mean_token_accuracy": 0.15662450045347215,
|
|
"num_tokens": 641587.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.646263837814331,
|
|
"epoch": 0.22900763358778625,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.421,
|
|
"mean_token_accuracy": 0.14964068084955215,
|
|
"num_tokens": 645726.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.735714673995972,
|
|
"epoch": 0.2304212609556121,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.2004,
|
|
"mean_token_accuracy": 0.16830723136663436,
|
|
"num_tokens": 649688.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.588094997406006,
|
|
"epoch": 0.23183488832343793,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.2291,
|
|
"mean_token_accuracy": 0.15805790424346924,
|
|
"num_tokens": 653745.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.485406589508057,
|
|
"epoch": 0.23324851569126379,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.1713,
|
|
"mean_token_accuracy": 0.1704267144203186,
|
|
"num_tokens": 657650.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.624208736419678,
|
|
"epoch": 0.2346621430590896,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.25,
|
|
"mean_token_accuracy": 0.1644523784518242,
|
|
"num_tokens": 661614.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.544505929946899,
|
|
"epoch": 0.23607577042691547,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.1953,
|
|
"mean_token_accuracy": 0.1738327383995056,
|
|
"num_tokens": 665484.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.586504888534546,
|
|
"epoch": 0.23748939779474132,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.1872,
|
|
"mean_token_accuracy": 0.16504454612731934,
|
|
"num_tokens": 669718.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.625461673736572,
|
|
"epoch": 0.23890302516256715,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.2064,
|
|
"mean_token_accuracy": 0.16486725360155105,
|
|
"num_tokens": 673913.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.616315221786499,
|
|
"epoch": 0.240316652530393,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.2657,
|
|
"mean_token_accuracy": 0.15634770691394806,
|
|
"num_tokens": 677911.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.552777242660523,
|
|
"epoch": 0.24173027989821882,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.0792,
|
|
"mean_token_accuracy": 0.16897292137145997,
|
|
"num_tokens": 681629.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.523988723754883,
|
|
"epoch": 0.24314390726604468,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.1307,
|
|
"mean_token_accuracy": 0.16514982432126998,
|
|
"num_tokens": 685675.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.562773180007935,
|
|
"epoch": 0.2445575346338705,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.0325,
|
|
"mean_token_accuracy": 0.17125135958194732,
|
|
"num_tokens": 689630.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.422208642959594,
|
|
"epoch": 0.24597116200169636,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.1486,
|
|
"mean_token_accuracy": 0.17256792038679122,
|
|
"num_tokens": 693580.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.469058227539063,
|
|
"epoch": 0.24738478936952218,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.2098,
|
|
"mean_token_accuracy": 0.1618221327662468,
|
|
"num_tokens": 697531.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.530948734283447,
|
|
"epoch": 0.24879841673734804,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.1695,
|
|
"mean_token_accuracy": 0.1632336474955082,
|
|
"num_tokens": 701421.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.483053684234619,
|
|
"epoch": 0.2502120441051739,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.1415,
|
|
"mean_token_accuracy": 0.16531084030866622,
|
|
"num_tokens": 705557.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.433886241912842,
|
|
"epoch": 0.2516256714729997,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.0636,
|
|
"mean_token_accuracy": 0.1743975192308426,
|
|
"num_tokens": 709173.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.3847051620483395,
|
|
"epoch": 0.25303929884082554,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.0107,
|
|
"mean_token_accuracy": 0.1733053870499134,
|
|
"num_tokens": 713144.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.407188081741333,
|
|
"epoch": 0.2544529262086514,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.3262,
|
|
"mean_token_accuracy": 0.15888329446315766,
|
|
"num_tokens": 717187.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.68216462135315,
|
|
"epoch": 0.25586655357647725,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.1792,
|
|
"mean_token_accuracy": 0.16036764532327652,
|
|
"num_tokens": 721311.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.426914548873901,
|
|
"epoch": 0.2572801809443031,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.1407,
|
|
"mean_token_accuracy": 0.17487923800945282,
|
|
"num_tokens": 725193.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.396089220046997,
|
|
"epoch": 0.2586938083121289,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.1083,
|
|
"mean_token_accuracy": 0.16730499863624573,
|
|
"num_tokens": 729208.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.535867500305176,
|
|
"epoch": 0.26010743567995476,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.1726,
|
|
"mean_token_accuracy": 0.17124995291233064,
|
|
"num_tokens": 733364.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.3439921855926515,
|
|
"epoch": 0.2615210630477806,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.1221,
|
|
"mean_token_accuracy": 0.16099516451358795,
|
|
"num_tokens": 737549.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.488129186630249,
|
|
"epoch": 0.26293469041560646,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.1294,
|
|
"mean_token_accuracy": 0.16554402112960814,
|
|
"num_tokens": 742009.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.518905687332153,
|
|
"epoch": 0.26434831778343226,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.2303,
|
|
"mean_token_accuracy": 0.16439489349722863,
|
|
"num_tokens": 746258.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.532473945617676,
|
|
"epoch": 0.2657619451512581,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.1433,
|
|
"mean_token_accuracy": 0.1789155274629593,
|
|
"num_tokens": 750027.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.243679428100586,
|
|
"epoch": 0.26717557251908397,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.0869,
|
|
"mean_token_accuracy": 0.16812117397785187,
|
|
"num_tokens": 754030.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.291708707809448,
|
|
"epoch": 0.2685891998869098,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.0033,
|
|
"mean_token_accuracy": 0.17290493547916413,
|
|
"num_tokens": 757663.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.357456398010254,
|
|
"epoch": 0.2700028272547357,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.1225,
|
|
"mean_token_accuracy": 0.16563030779361726,
|
|
"num_tokens": 761806.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.466005516052246,
|
|
"epoch": 0.2714164546225615,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.1282,
|
|
"mean_token_accuracy": 0.16799217388033866,
|
|
"num_tokens": 765900.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.314237213134765,
|
|
"epoch": 0.27283008199038733,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.0802,
|
|
"mean_token_accuracy": 0.16931777596473693,
|
|
"num_tokens": 769842.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.361982011795044,
|
|
"epoch": 0.2742437093582132,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.2017,
|
|
"mean_token_accuracy": 0.16852450519800186,
|
|
"num_tokens": 773717.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.245587730407715,
|
|
"epoch": 0.27565733672603904,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000487,
|
|
"loss": 5.96,
|
|
"mean_token_accuracy": 0.17952348291873932,
|
|
"num_tokens": 777720.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.214972496032715,
|
|
"epoch": 0.27707096409386484,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 5.9622,
|
|
"mean_token_accuracy": 0.17511400282382966,
|
|
"num_tokens": 781482.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.306806802749634,
|
|
"epoch": 0.2784845914616907,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.0785,
|
|
"mean_token_accuracy": 0.16686865836381912,
|
|
"num_tokens": 785390.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.173373794555664,
|
|
"epoch": 0.27989821882951654,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.017,
|
|
"mean_token_accuracy": 0.18208521008491516,
|
|
"num_tokens": 789464.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.59720025062561,
|
|
"epoch": 0.2813118461973424,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.1232,
|
|
"mean_token_accuracy": 0.16400160863995553,
|
|
"num_tokens": 793229.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.182021236419677,
|
|
"epoch": 0.2827254735651682,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.0322,
|
|
"mean_token_accuracy": 0.17415720224380493,
|
|
"num_tokens": 797490.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.2827254735651682,
|
|
"eval_entropy": 6.051144407963093,
|
|
"eval_loss": 6.294586181640625,
|
|
"eval_mean_token_accuracy": 0.16992966667567475,
|
|
"eval_num_tokens": 797490.0,
|
|
"eval_runtime": 4.8981,
|
|
"eval_samples_per_second": 2303.769,
|
|
"eval_steps_per_second": 288.073,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.32871265411377,
|
|
"epoch": 0.28413910093299405,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004999999849524451,
|
|
"loss": 6.0691,
|
|
"mean_token_accuracy": 0.17283144295215608,
|
|
"num_tokens": 801242.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.356661939620972,
|
|
"epoch": 0.2855527283008199,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000499999923821757,
|
|
"loss": 6.1618,
|
|
"mean_token_accuracy": 0.16092283874750138,
|
|
"num_tokens": 805285.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.229091024398803,
|
|
"epoch": 0.28696635566864576,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000499999815667476,
|
|
"loss": 5.9852,
|
|
"mean_token_accuracy": 0.17933639883995056,
|
|
"num_tokens": 809278.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.252400302886963,
|
|
"epoch": 0.2883799830364716,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004999996604896251,
|
|
"loss": 6.1391,
|
|
"mean_token_accuracy": 0.1700467899441719,
|
|
"num_tokens": 813519.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.289431047439575,
|
|
"epoch": 0.2897936104042974,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004999994582882363,
|
|
"loss": 6.0622,
|
|
"mean_token_accuracy": 0.1710103139281273,
|
|
"num_tokens": 817549.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.267086839675903,
|
|
"epoch": 0.29120723777212326,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004999992090633522,
|
|
"loss": 6.0473,
|
|
"mean_token_accuracy": 0.17169943898916246,
|
|
"num_tokens": 821608.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.152920913696289,
|
|
"epoch": 0.2926208651399491,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004999989128150248,
|
|
"loss": 5.9653,
|
|
"mean_token_accuracy": 0.18662210255861283,
|
|
"num_tokens": 825795.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.3972742557525635,
|
|
"epoch": 0.29403449250777497,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004999985695433159,
|
|
"loss": 6.0705,
|
|
"mean_token_accuracy": 0.17289165109395982,
|
|
"num_tokens": 829693.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.2147924423217775,
|
|
"epoch": 0.29544811987560077,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004999981792482973,
|
|
"loss": 6.0918,
|
|
"mean_token_accuracy": 0.1711990311741829,
|
|
"num_tokens": 833665.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.21585865020752,
|
|
"epoch": 0.2968617472434266,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0004999977419300507,
|
|
"loss": 6.0505,
|
|
"mean_token_accuracy": 0.16072292476892472,
|
|
"num_tokens": 837107.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.245480632781982,
|
|
"epoch": 0.2982753746112525,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004999972575886673,
|
|
"loss": 6.1374,
|
|
"mean_token_accuracy": 0.16482933461666108,
|
|
"num_tokens": 841182.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.287858152389527,
|
|
"epoch": 0.29968900197907833,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004999967262242483,
|
|
"loss": 6.0854,
|
|
"mean_token_accuracy": 0.1722053959965706,
|
|
"num_tokens": 845266.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.216232633590698,
|
|
"epoch": 0.3011026293469042,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000499996147836905,
|
|
"loss": 6.037,
|
|
"mean_token_accuracy": 0.1735639214515686,
|
|
"num_tokens": 849118.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.361587762832642,
|
|
"epoch": 0.30251625671473,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000499995522426758,
|
|
"loss": 6.144,
|
|
"mean_token_accuracy": 0.16536193639039992,
|
|
"num_tokens": 853101.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.167893266677856,
|
|
"epoch": 0.30392988408255583,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0004999948499939383,
|
|
"loss": 5.9842,
|
|
"mean_token_accuracy": 0.18141428083181382,
|
|
"num_tokens": 856671.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.260188484191895,
|
|
"epoch": 0.3053435114503817,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004999941305385863,
|
|
"loss": 5.9521,
|
|
"mean_token_accuracy": 0.18240092396736146,
|
|
"num_tokens": 860639.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.1321197032928465,
|
|
"epoch": 0.30675713881820754,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0004999933640608521,
|
|
"loss": 5.984,
|
|
"mean_token_accuracy": 0.17295702844858168,
|
|
"num_tokens": 864270.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.153362846374511,
|
|
"epoch": 0.30817076618603334,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004999925505608963,
|
|
"loss": 5.9524,
|
|
"mean_token_accuracy": 0.16724486351013185,
|
|
"num_tokens": 868069.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.131827163696289,
|
|
"epoch": 0.3095843935538592,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004999916900388887,
|
|
"loss": 5.9961,
|
|
"mean_token_accuracy": 0.1748175099492073,
|
|
"num_tokens": 871943.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.255445575714111,
|
|
"epoch": 0.31099802092168505,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004999907824950093,
|
|
"loss": 6.0249,
|
|
"mean_token_accuracy": 0.17832081466913224,
|
|
"num_tokens": 876168.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.095445013046264,
|
|
"epoch": 0.3124116482895109,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0004999898279294475,
|
|
"loss": 6.0014,
|
|
"mean_token_accuracy": 0.17347852513194084,
|
|
"num_tokens": 880134.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.133862781524658,
|
|
"epoch": 0.3138252756573367,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0004999888263424031,
|
|
"loss": 6.0011,
|
|
"mean_token_accuracy": 0.17168307155370713,
|
|
"num_tokens": 883970.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.260912227630615,
|
|
"epoch": 0.31523890302516255,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0004999877777340852,
|
|
"loss": 6.0514,
|
|
"mean_token_accuracy": 0.16243282705545425,
|
|
"num_tokens": 887980.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.018231344223023,
|
|
"epoch": 0.3166525303929884,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000499986682104713,
|
|
"loss": 5.8732,
|
|
"mean_token_accuracy": 0.18358443528413773,
|
|
"num_tokens": 891867.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.121918344497681,
|
|
"epoch": 0.31806615776081426,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004999855394545157,
|
|
"loss": 5.967,
|
|
"mean_token_accuracy": 0.1776360958814621,
|
|
"num_tokens": 895769.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.1299745559692385,
|
|
"epoch": 0.3194797851286401,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004999843497837317,
|
|
"loss": 5.9112,
|
|
"mean_token_accuracy": 0.17768835574388503,
|
|
"num_tokens": 899640.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.127333307266236,
|
|
"epoch": 0.3208934124964659,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00049998311309261,
|
|
"loss": 5.9222,
|
|
"mean_token_accuracy": 0.18061408549547195,
|
|
"num_tokens": 903598.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.185299062728882,
|
|
"epoch": 0.32230703986429177,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004999818293814089,
|
|
"loss": 5.9639,
|
|
"mean_token_accuracy": 0.17317982614040375,
|
|
"num_tokens": 907720.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 5.960520887374878,
|
|
"epoch": 0.3237206672321176,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004999804986503966,
|
|
"loss": 5.9432,
|
|
"mean_token_accuracy": 0.1775066375732422,
|
|
"num_tokens": 911531.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.173041248321534,
|
|
"epoch": 0.3251342945999435,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0004999791208998513,
|
|
"loss": 5.8607,
|
|
"mean_token_accuracy": 0.18050449788570405,
|
|
"num_tokens": 915428.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.007273769378662,
|
|
"epoch": 0.3265479219677693,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004999776961300611,
|
|
"loss": 5.875,
|
|
"mean_token_accuracy": 0.17531459927558898,
|
|
"num_tokens": 919499.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.183330106735229,
|
|
"epoch": 0.3279615493355951,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004999762243413236,
|
|
"loss": 5.8994,
|
|
"mean_token_accuracy": 0.18333160728216172,
|
|
"num_tokens": 923530.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.1380363464355465,
|
|
"epoch": 0.329375176703421,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004999747055339464,
|
|
"loss": 5.846,
|
|
"mean_token_accuracy": 0.1873629868030548,
|
|
"num_tokens": 927770.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 5.994632005691528,
|
|
"epoch": 0.33078880407124683,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0004999731397082469,
|
|
"loss": 5.8898,
|
|
"mean_token_accuracy": 0.1810222864151001,
|
|
"num_tokens": 932008.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.158831453323364,
|
|
"epoch": 0.3322024314390727,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004999715268645524,
|
|
"loss": 5.9115,
|
|
"mean_token_accuracy": 0.1809718891978264,
|
|
"num_tokens": 936388.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.054045629501343,
|
|
"epoch": 0.3336160588068985,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004999698670032,
|
|
"loss": 5.9492,
|
|
"mean_token_accuracy": 0.18082664608955384,
|
|
"num_tokens": 940436.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 5.983764600753784,
|
|
"epoch": 0.33502968617472434,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004999681601245366,
|
|
"loss": 5.9589,
|
|
"mean_token_accuracy": 0.17784594148397445,
|
|
"num_tokens": 944296.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.146846437454224,
|
|
"epoch": 0.3364433135425502,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004999664062289189,
|
|
"loss": 5.989,
|
|
"mean_token_accuracy": 0.17501545697450638,
|
|
"num_tokens": 948318.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.1085621356964115,
|
|
"epoch": 0.33785694091037605,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004999646053167134,
|
|
"loss": 5.8718,
|
|
"mean_token_accuracy": 0.18369389176368714,
|
|
"num_tokens": 952147.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.098910617828369,
|
|
"epoch": 0.33927056827820185,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0004999627573882966,
|
|
"loss": 6.0154,
|
|
"mean_token_accuracy": 0.17044340819120407,
|
|
"num_tokens": 956445.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.067022705078125,
|
|
"epoch": 0.3406841956460277,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004999608624440546,
|
|
"loss": 5.8574,
|
|
"mean_token_accuracy": 0.1799109995365143,
|
|
"num_tokens": 960327.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.08806881904602,
|
|
"epoch": 0.34209782301385355,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004999589204843834,
|
|
"loss": 5.9898,
|
|
"mean_token_accuracy": 0.18153459578752518,
|
|
"num_tokens": 964387.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.120113325119019,
|
|
"epoch": 0.3435114503816794,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000499956931509689,
|
|
"loss": 5.9261,
|
|
"mean_token_accuracy": 0.18374113589525223,
|
|
"num_tokens": 968327.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.132302856445312,
|
|
"epoch": 0.3449250777495052,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000499954895520387,
|
|
"loss": 5.9704,
|
|
"mean_token_accuracy": 0.1799431711435318,
|
|
"num_tokens": 972358.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.008096218109131,
|
|
"epoch": 0.34633870511733106,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004999528125169028,
|
|
"loss": 5.8406,
|
|
"mean_token_accuracy": 0.175810743868351,
|
|
"num_tokens": 976266.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.0000334739685055,
|
|
"epoch": 0.3477523324851569,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000499950682499672,
|
|
"loss": 5.9705,
|
|
"mean_token_accuracy": 0.17845053225755692,
|
|
"num_tokens": 980521.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.040744686126709,
|
|
"epoch": 0.34916595985298277,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0004999485054691395,
|
|
"loss": 5.8551,
|
|
"mean_token_accuracy": 0.18057717978954316,
|
|
"num_tokens": 984413.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.139350652694702,
|
|
"epoch": 0.3505795872208086,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004999462814257604,
|
|
"loss": 5.9464,
|
|
"mean_token_accuracy": 0.16705757826566697,
|
|
"num_tokens": 988080.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.139582967758178,
|
|
"epoch": 0.3519932145886344,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004999440103699996,
|
|
"loss": 5.8926,
|
|
"mean_token_accuracy": 0.1826186239719391,
|
|
"num_tokens": 992010.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 5.878952169418335,
|
|
"epoch": 0.35340684195646027,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004999416923023316,
|
|
"loss": 5.7834,
|
|
"mean_token_accuracy": 0.194471637904644,
|
|
"num_tokens": 995721.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 5.920779228210449,
|
|
"epoch": 0.3548204693242861,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004999393272232409,
|
|
"loss": 5.8949,
|
|
"mean_token_accuracy": 0.18157693296670913,
|
|
"num_tokens": 999763.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.085085344314575,
|
|
"epoch": 0.356234096692112,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004999369151332218,
|
|
"loss": 5.938,
|
|
"mean_token_accuracy": 0.18311794847249985,
|
|
"num_tokens": 1003610.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.084949398040772,
|
|
"epoch": 0.3576477240599378,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004999344560327784,
|
|
"loss": 5.8817,
|
|
"mean_token_accuracy": 0.1808584988117218,
|
|
"num_tokens": 1007699.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.034257507324218,
|
|
"epoch": 0.35906135142776363,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0004999319499224247,
|
|
"loss": 5.8248,
|
|
"mean_token_accuracy": 0.19244421720504762,
|
|
"num_tokens": 1011573.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 5.997032213211059,
|
|
"epoch": 0.3604749787955895,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004999293968026843,
|
|
"loss": 5.9599,
|
|
"mean_token_accuracy": 0.17845921069383622,
|
|
"num_tokens": 1015569.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.032209300994873,
|
|
"epoch": 0.36188860616341534,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004999267966740909,
|
|
"loss": 5.8013,
|
|
"mean_token_accuracy": 0.1848297104239464,
|
|
"num_tokens": 1019451.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 5.926502704620361,
|
|
"epoch": 0.3633022335312412,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000499924149537188,
|
|
"loss": 5.7224,
|
|
"mean_token_accuracy": 0.19073605984449388,
|
|
"num_tokens": 1023084.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.071679019927979,
|
|
"epoch": 0.364715860899067,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0004999214553925287,
|
|
"loss": 5.841,
|
|
"mean_token_accuracy": 0.18661287277936936,
|
|
"num_tokens": 1026941.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.050941610336304,
|
|
"epoch": 0.36612948826689284,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004999187142406761,
|
|
"loss": 5.7795,
|
|
"mean_token_accuracy": 0.19633757770061494,
|
|
"num_tokens": 1030872.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 5.994774961471558,
|
|
"epoch": 0.3675431156347187,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000499915926082203,
|
|
"loss": 5.8838,
|
|
"mean_token_accuracy": 0.17777508050203322,
|
|
"num_tokens": 1035024.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 5.969749546051025,
|
|
"epoch": 0.36895674300254455,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004999130909176923,
|
|
"loss": 5.7872,
|
|
"mean_token_accuracy": 0.18587531447410582,
|
|
"num_tokens": 1038988.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 5.884610033035278,
|
|
"epoch": 0.37037037037037035,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004999102087477362,
|
|
"loss": 5.7024,
|
|
"mean_token_accuracy": 0.19922589361667634,
|
|
"num_tokens": 1043068.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 5.866126918792725,
|
|
"epoch": 0.3717839977381962,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004999072795729376,
|
|
"loss": 5.776,
|
|
"mean_token_accuracy": 0.18842145353555678,
|
|
"num_tokens": 1046931.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 5.8829490661621096,
|
|
"epoch": 0.37319762510602206,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004999043033939081,
|
|
"loss": 5.7317,
|
|
"mean_token_accuracy": 0.19113447070121764,
|
|
"num_tokens": 1050718.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 5.90353946685791,
|
|
"epoch": 0.3746112524738479,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004999012802112701,
|
|
"loss": 5.8071,
|
|
"mean_token_accuracy": 0.1775677904486656,
|
|
"num_tokens": 1054796.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 5.928818321228027,
|
|
"epoch": 0.3760248798416737,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004998982100256552,
|
|
"loss": 5.7504,
|
|
"mean_token_accuracy": 0.18737074881792068,
|
|
"num_tokens": 1058657.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 5.908525562286377,
|
|
"epoch": 0.37743850720949956,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004998950928377052,
|
|
"loss": 5.8693,
|
|
"mean_token_accuracy": 0.18807025104761124,
|
|
"num_tokens": 1062532.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 5.892481327056885,
|
|
"epoch": 0.3788521345773254,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004998919286480713,
|
|
"loss": 5.7168,
|
|
"mean_token_accuracy": 0.19382492154836656,
|
|
"num_tokens": 1066274.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 5.8984781265258786,
|
|
"epoch": 0.38026576194515127,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004998887174574153,
|
|
"loss": 5.8189,
|
|
"mean_token_accuracy": 0.18274522572755814,
|
|
"num_tokens": 1070179.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 5.870694589614868,
|
|
"epoch": 0.3816793893129771,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004998854592664079,
|
|
"loss": 5.7916,
|
|
"mean_token_accuracy": 0.18562129884958267,
|
|
"num_tokens": 1073892.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 5.9831644058227536,
|
|
"epoch": 0.3830930166808029,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004998821540757301,
|
|
"loss": 5.8439,
|
|
"mean_token_accuracy": 0.18298695832490922,
|
|
"num_tokens": 1078399.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 5.8720427513122555,
|
|
"epoch": 0.3845066440486288,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004998788018860728,
|
|
"loss": 5.754,
|
|
"mean_token_accuracy": 0.19132645279169083,
|
|
"num_tokens": 1082371.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 5.921624422073364,
|
|
"epoch": 0.38592027141645463,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004998754026981364,
|
|
"loss": 5.8936,
|
|
"mean_token_accuracy": 0.17152681052684784,
|
|
"num_tokens": 1086474.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 5.971354389190674,
|
|
"epoch": 0.3873338987842805,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004998719565126316,
|
|
"loss": 5.8067,
|
|
"mean_token_accuracy": 0.19351842254400253,
|
|
"num_tokens": 1090519.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 5.819832897186279,
|
|
"epoch": 0.3887475261521063,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004998684633302783,
|
|
"loss": 5.7492,
|
|
"mean_token_accuracy": 0.18426936864852905,
|
|
"num_tokens": 1094450.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.021639633178711,
|
|
"epoch": 0.39016115351993214,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004998649231518068,
|
|
"loss": 5.7939,
|
|
"mean_token_accuracy": 0.1839954972267151,
|
|
"num_tokens": 1098249.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 5.936412715911866,
|
|
"epoch": 0.391574780887758,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004998613359779568,
|
|
"loss": 5.7502,
|
|
"mean_token_accuracy": 0.1965183287858963,
|
|
"num_tokens": 1101951.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 5.852648210525513,
|
|
"epoch": 0.39298840825558384,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004998577018094782,
|
|
"loss": 5.7843,
|
|
"mean_token_accuracy": 0.1870306134223938,
|
|
"num_tokens": 1106097.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 5.901752901077271,
|
|
"epoch": 0.3944020356234097,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004998540206471304,
|
|
"loss": 5.7213,
|
|
"mean_token_accuracy": 0.1883508324623108,
|
|
"num_tokens": 1110031.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 5.926095390319825,
|
|
"epoch": 0.3958156629912355,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004998502924916826,
|
|
"loss": 5.8754,
|
|
"mean_token_accuracy": 0.1840156227350235,
|
|
"num_tokens": 1114146.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.012360382080078,
|
|
"epoch": 0.39722929035906135,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004998465173439142,
|
|
"loss": 5.8229,
|
|
"mean_token_accuracy": 0.18926550894975663,
|
|
"num_tokens": 1118161.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 5.941721773147583,
|
|
"epoch": 0.3986429177268872,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004998426952046142,
|
|
"loss": 5.6464,
|
|
"mean_token_accuracy": 0.20289266854524612,
|
|
"num_tokens": 1122014.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 5.76336145401001,
|
|
"epoch": 0.40005654509471306,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004998388260745811,
|
|
"loss": 5.8007,
|
|
"mean_token_accuracy": 0.17852477133274078,
|
|
"num_tokens": 1125929.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 5.954787492752075,
|
|
"epoch": 0.40147017246253885,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004998349099546238,
|
|
"loss": 5.7595,
|
|
"mean_token_accuracy": 0.18921004384756088,
|
|
"num_tokens": 1129781.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 5.847860431671142,
|
|
"epoch": 0.4028837998303647,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004998309468455606,
|
|
"loss": 5.6644,
|
|
"mean_token_accuracy": 0.20904712975025178,
|
|
"num_tokens": 1133591.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 5.821159505844117,
|
|
"epoch": 0.40429742719819056,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004998269367482199,
|
|
"loss": 5.7892,
|
|
"mean_token_accuracy": 0.1861359179019928,
|
|
"num_tokens": 1137406.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 5.842456293106079,
|
|
"epoch": 0.4057110545660164,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0004998228796634396,
|
|
"loss": 5.7103,
|
|
"mean_token_accuracy": 0.19130536019802094,
|
|
"num_tokens": 1141393.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 5.929346418380737,
|
|
"epoch": 0.4071246819338422,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004998187755920677,
|
|
"loss": 5.8373,
|
|
"mean_token_accuracy": 0.18991080820560455,
|
|
"num_tokens": 1145155.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 5.8391400337219235,
|
|
"epoch": 0.40853830930166807,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004998146245349621,
|
|
"loss": 5.7896,
|
|
"mean_token_accuracy": 0.18923215866088866,
|
|
"num_tokens": 1149104.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 5.861079597473145,
|
|
"epoch": 0.4099519366694939,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004998104264929901,
|
|
"loss": 5.7447,
|
|
"mean_token_accuracy": 0.18697023391723633,
|
|
"num_tokens": 1153022.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 5.918407154083252,
|
|
"epoch": 0.4113655640373198,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004998061814670291,
|
|
"loss": 5.7603,
|
|
"mean_token_accuracy": 0.184474678337574,
|
|
"num_tokens": 1156965.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 5.811736297607422,
|
|
"epoch": 0.41277919140514563,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004998018894579664,
|
|
"loss": 5.7111,
|
|
"mean_token_accuracy": 0.1976108655333519,
|
|
"num_tokens": 1160893.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 5.9467895984649655,
|
|
"epoch": 0.4141928187729714,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004997975504666989,
|
|
"loss": 5.8613,
|
|
"mean_token_accuracy": 0.17591351866722107,
|
|
"num_tokens": 1165037.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 5.983978176116944,
|
|
"epoch": 0.4156064461407973,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0004997931644941335,
|
|
"loss": 5.7535,
|
|
"mean_token_accuracy": 0.1833250492811203,
|
|
"num_tokens": 1168911.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 5.925879240036011,
|
|
"epoch": 0.41702007350862313,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004997887315411867,
|
|
"loss": 5.752,
|
|
"mean_token_accuracy": 0.1955208569765091,
|
|
"num_tokens": 1172861.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 5.804015302658081,
|
|
"epoch": 0.418433700876449,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004997842516087852,
|
|
"loss": 5.7843,
|
|
"mean_token_accuracy": 0.18648661077022552,
|
|
"num_tokens": 1177361.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 5.9768593311309814,
|
|
"epoch": 0.4198473282442748,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004997797246978651,
|
|
"loss": 5.7778,
|
|
"mean_token_accuracy": 0.17635598480701448,
|
|
"num_tokens": 1181549.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 5.921354866027832,
|
|
"epoch": 0.42126095561210064,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004997751508093724,
|
|
"loss": 5.7786,
|
|
"mean_token_accuracy": 0.18104448914527893,
|
|
"num_tokens": 1185685.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 5.8056823253631595,
|
|
"epoch": 0.4226745829799265,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004997705299442632,
|
|
"loss": 5.8771,
|
|
"mean_token_accuracy": 0.17976228892803192,
|
|
"num_tokens": 1189557.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 5.953405332565308,
|
|
"epoch": 0.42408821034775235,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004997658621035033,
|
|
"loss": 5.7987,
|
|
"mean_token_accuracy": 0.19059500545263292,
|
|
"num_tokens": 1193969.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 5.930359935760498,
|
|
"epoch": 0.42550183771557815,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004997611472880681,
|
|
"loss": 5.7775,
|
|
"mean_token_accuracy": 0.1988312780857086,
|
|
"num_tokens": 1197843.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 5.812527418136597,
|
|
"epoch": 0.426915465083404,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000499756385498943,
|
|
"loss": 5.7792,
|
|
"mean_token_accuracy": 0.18155284225940704,
|
|
"num_tokens": 1201938.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 5.8252583026885985,
|
|
"epoch": 0.42832909245122985,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004997515767371231,
|
|
"loss": 5.6347,
|
|
"mean_token_accuracy": 0.1871896967291832,
|
|
"num_tokens": 1205757.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 5.889902544021607,
|
|
"epoch": 0.4297427198190557,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004997467210036135,
|
|
"loss": 5.7705,
|
|
"mean_token_accuracy": 0.18698835819959642,
|
|
"num_tokens": 1209779.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 5.821274614334106,
|
|
"epoch": 0.43115634718688156,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000499741818299429,
|
|
"loss": 5.7323,
|
|
"mean_token_accuracy": 0.19483786523342134,
|
|
"num_tokens": 1213598.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 5.744606351852417,
|
|
"epoch": 0.43256997455470736,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004997368686255943,
|
|
"loss": 5.6435,
|
|
"mean_token_accuracy": 0.19433416724205016,
|
|
"num_tokens": 1217452.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 5.858005809783935,
|
|
"epoch": 0.4339836019225332,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004997318719831438,
|
|
"loss": 5.7862,
|
|
"mean_token_accuracy": 0.18418201208114623,
|
|
"num_tokens": 1221354.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 5.793173360824585,
|
|
"epoch": 0.43539722929035907,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004997268283731216,
|
|
"loss": 5.6892,
|
|
"mean_token_accuracy": 0.19399462640285492,
|
|
"num_tokens": 1225217.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 5.891930007934571,
|
|
"epoch": 0.4368108566581849,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004997217377965821,
|
|
"loss": 5.6947,
|
|
"mean_token_accuracy": 0.18456600606441498,
|
|
"num_tokens": 1229016.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 5.927923154830933,
|
|
"epoch": 0.4382244840260107,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000499716600254589,
|
|
"loss": 5.7846,
|
|
"mean_token_accuracy": 0.19131413698196412,
|
|
"num_tokens": 1232825.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 5.911680030822754,
|
|
"epoch": 0.4396381113938366,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000499711415748216,
|
|
"loss": 5.7284,
|
|
"mean_token_accuracy": 0.1870203584432602,
|
|
"num_tokens": 1237032.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 5.6728795051574705,
|
|
"epoch": 0.4410517387616624,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004997061842785468,
|
|
"loss": 5.6221,
|
|
"mean_token_accuracy": 0.19849239736795427,
|
|
"num_tokens": 1241043.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 5.756577777862549,
|
|
"epoch": 0.4424653661294883,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004997009058466745,
|
|
"loss": 5.7198,
|
|
"mean_token_accuracy": 0.196329565346241,
|
|
"num_tokens": 1245175.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 5.840588140487671,
|
|
"epoch": 0.44387899349731413,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004996955804537024,
|
|
"loss": 5.6716,
|
|
"mean_token_accuracy": 0.19366080611944197,
|
|
"num_tokens": 1249540.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 5.783361577987671,
|
|
"epoch": 0.44529262086513993,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004996902081007435,
|
|
"loss": 5.6577,
|
|
"mean_token_accuracy": 0.186581951379776,
|
|
"num_tokens": 1253532.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 5.89839916229248,
|
|
"epoch": 0.4467062482329658,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004996847887889205,
|
|
"loss": 5.6562,
|
|
"mean_token_accuracy": 0.18789928704500197,
|
|
"num_tokens": 1257492.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 5.844485521316528,
|
|
"epoch": 0.44811987560079164,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0004996793225193662,
|
|
"loss": 5.7554,
|
|
"mean_token_accuracy": 0.18641360402107238,
|
|
"num_tokens": 1261388.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 5.823833084106445,
|
|
"epoch": 0.4495335029686175,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004996738092932227,
|
|
"loss": 5.6579,
|
|
"mean_token_accuracy": 0.1978958413004875,
|
|
"num_tokens": 1265417.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 5.7972126483917235,
|
|
"epoch": 0.4509471303364433,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004996682491116425,
|
|
"loss": 5.6477,
|
|
"mean_token_accuracy": 0.19896170347929001,
|
|
"num_tokens": 1269454.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 5.7864405632019045,
|
|
"epoch": 0.45236075770426915,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004996626419757875,
|
|
"loss": 5.7422,
|
|
"mean_token_accuracy": 0.19003152698278428,
|
|
"num_tokens": 1273677.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 5.809814310073852,
|
|
"epoch": 0.453774385072095,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004996569878868296,
|
|
"loss": 5.5735,
|
|
"mean_token_accuracy": 0.20513303130865096,
|
|
"num_tokens": 1277683.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 5.726590728759765,
|
|
"epoch": 0.45518801243992085,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004996512868459505,
|
|
"loss": 5.725,
|
|
"mean_token_accuracy": 0.18426367342472078,
|
|
"num_tokens": 1281945.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 5.715907955169678,
|
|
"epoch": 0.45660163980774665,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004996455388543416,
|
|
"loss": 5.6331,
|
|
"mean_token_accuracy": 0.1939169466495514,
|
|
"num_tokens": 1285709.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 5.965580797195434,
|
|
"epoch": 0.4580152671755725,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0004996397439132042,
|
|
"loss": 5.795,
|
|
"mean_token_accuracy": 0.17828181833028794,
|
|
"num_tokens": 1289772.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 5.7739866256713865,
|
|
"epoch": 0.45942889454339836,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004996339020237496,
|
|
"loss": 5.6691,
|
|
"mean_token_accuracy": 0.19340460002422333,
|
|
"num_tokens": 1293671.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 5.850045537948608,
|
|
"epoch": 0.4608425219112242,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004996280131871985,
|
|
"loss": 5.5937,
|
|
"mean_token_accuracy": 0.19490807950496675,
|
|
"num_tokens": 1297334.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 5.788289546966553,
|
|
"epoch": 0.46225614927905007,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004996220774047816,
|
|
"loss": 5.6794,
|
|
"mean_token_accuracy": 0.1935423344373703,
|
|
"num_tokens": 1301219.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 5.798729944229126,
|
|
"epoch": 0.46366977664687586,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004996160946777398,
|
|
"loss": 5.6636,
|
|
"mean_token_accuracy": 0.19195165783166884,
|
|
"num_tokens": 1305166.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 5.729593896865845,
|
|
"epoch": 0.4650834040147017,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004996100650073229,
|
|
"loss": 5.5989,
|
|
"mean_token_accuracy": 0.20028469562530518,
|
|
"num_tokens": 1308958.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 5.800077962875366,
|
|
"epoch": 0.46649703138252757,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004996039883947915,
|
|
"loss": 5.817,
|
|
"mean_token_accuracy": 0.18166496753692626,
|
|
"num_tokens": 1313289.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 5.939286613464356,
|
|
"epoch": 0.4679106587503534,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004995978648414154,
|
|
"loss": 5.6835,
|
|
"mean_token_accuracy": 0.19870763123035431,
|
|
"num_tokens": 1317366.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 5.767664241790771,
|
|
"epoch": 0.4693242861181792,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004995916943484745,
|
|
"loss": 5.6659,
|
|
"mean_token_accuracy": 0.1947951689362526,
|
|
"num_tokens": 1321362.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 5.746976137161255,
|
|
"epoch": 0.4707379134860051,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004995854769172583,
|
|
"loss": 5.6213,
|
|
"mean_token_accuracy": 0.19329093098640443,
|
|
"num_tokens": 1325147.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 5.845152759552002,
|
|
"epoch": 0.47215154085383093,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004995792125490662,
|
|
"loss": 5.6702,
|
|
"mean_token_accuracy": 0.1898410737514496,
|
|
"num_tokens": 1329072.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 5.709325075149536,
|
|
"epoch": 0.4735651682216568,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004995729012452074,
|
|
"loss": 5.6052,
|
|
"mean_token_accuracy": 0.198711159825325,
|
|
"num_tokens": 1332947.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 5.627379941940307,
|
|
"epoch": 0.47497879558948264,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004995665430070009,
|
|
"loss": 5.621,
|
|
"mean_token_accuracy": 0.19632248133420943,
|
|
"num_tokens": 1336903.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 5.819957590103149,
|
|
"epoch": 0.47639242295730844,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0004995601378357757,
|
|
"loss": 5.6884,
|
|
"mean_token_accuracy": 0.19369249790906906,
|
|
"num_tokens": 1340989.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 5.899555778503418,
|
|
"epoch": 0.4778060503251343,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004995536857328702,
|
|
"loss": 5.7288,
|
|
"mean_token_accuracy": 0.17993859052658082,
|
|
"num_tokens": 1345251.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 5.822159147262573,
|
|
"epoch": 0.47921967769296014,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004995471866996332,
|
|
"loss": 5.8217,
|
|
"mean_token_accuracy": 0.18204482644796371,
|
|
"num_tokens": 1349167.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 5.724724149703979,
|
|
"epoch": 0.480633305060786,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004995406407374226,
|
|
"loss": 5.5332,
|
|
"mean_token_accuracy": 0.2037088915705681,
|
|
"num_tokens": 1353309.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 5.69287486076355,
|
|
"epoch": 0.4820469324286118,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004995340478476067,
|
|
"loss": 5.656,
|
|
"mean_token_accuracy": 0.20406957417726518,
|
|
"num_tokens": 1357129.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 5.6713849067687985,
|
|
"epoch": 0.48346055979643765,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004995274080315631,
|
|
"loss": 5.6198,
|
|
"mean_token_accuracy": 0.19456420093774796,
|
|
"num_tokens": 1360824.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 5.721224403381347,
|
|
"epoch": 0.4848741871642635,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004995207212906798,
|
|
"loss": 5.6495,
|
|
"mean_token_accuracy": 0.1971166342496872,
|
|
"num_tokens": 1364688.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 5.750689172744751,
|
|
"epoch": 0.48628781453208936,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000499513987626354,
|
|
"loss": 5.5703,
|
|
"mean_token_accuracy": 0.20018759667873381,
|
|
"num_tokens": 1368889.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 5.674837827682495,
|
|
"epoch": 0.48770144189991516,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004995072070399933,
|
|
"loss": 5.6836,
|
|
"mean_token_accuracy": 0.2023005336523056,
|
|
"num_tokens": 1372912.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 5.6823383331298825,
|
|
"epoch": 0.489115069267741,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004995003795330146,
|
|
"loss": 5.5365,
|
|
"mean_token_accuracy": 0.20185312926769255,
|
|
"num_tokens": 1376566.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 5.636135816574097,
|
|
"epoch": 0.49052869663556686,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004994935051068447,
|
|
"loss": 5.5587,
|
|
"mean_token_accuracy": 0.1983194187283516,
|
|
"num_tokens": 1380710.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 5.698077821731568,
|
|
"epoch": 0.4919423240033927,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004994865837629206,
|
|
"loss": 5.6284,
|
|
"mean_token_accuracy": 0.2035332128405571,
|
|
"num_tokens": 1384485.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 5.785005521774292,
|
|
"epoch": 0.49335595137121857,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004994796155026887,
|
|
"loss": 5.6135,
|
|
"mean_token_accuracy": 0.1922136127948761,
|
|
"num_tokens": 1388297.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 5.679270887374878,
|
|
"epoch": 0.49476957873904437,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004994726003276053,
|
|
"loss": 5.6053,
|
|
"mean_token_accuracy": 0.20150333791971206,
|
|
"num_tokens": 1392390.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 5.657969236373901,
|
|
"epoch": 0.4961832061068702,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004994655382391365,
|
|
"loss": 5.641,
|
|
"mean_token_accuracy": 0.19058468639850618,
|
|
"num_tokens": 1396561.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 5.740747547149658,
|
|
"epoch": 0.4975968334746961,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004994584292387582,
|
|
"loss": 5.6525,
|
|
"mean_token_accuracy": 0.19100899398326873,
|
|
"num_tokens": 1400556.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.835249662399292,
|
|
"epoch": 0.49901046084252193,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004994512733279562,
|
|
"loss": 5.7011,
|
|
"mean_token_accuracy": 0.18798956274986267,
|
|
"num_tokens": 1404503.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 5.733709383010864,
|
|
"epoch": 0.5004240882103478,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004994440705082261,
|
|
"loss": 5.6529,
|
|
"mean_token_accuracy": 0.1851186528801918,
|
|
"num_tokens": 1408504.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 5.705524873733521,
|
|
"epoch": 0.5018377155781736,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004994368207810731,
|
|
"loss": 5.4874,
|
|
"mean_token_accuracy": 0.2035467341542244,
|
|
"num_tokens": 1412429.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 5.629953145980835,
|
|
"epoch": 0.5032513429459994,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004994295241480126,
|
|
"loss": 5.5837,
|
|
"mean_token_accuracy": 0.19919728189706803,
|
|
"num_tokens": 1416466.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 5.678175735473633,
|
|
"epoch": 0.5046649703138253,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004994221806105692,
|
|
"loss": 5.5531,
|
|
"mean_token_accuracy": 0.20242667198181152,
|
|
"num_tokens": 1420226.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.559377288818359,
|
|
"epoch": 0.5060785976816511,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000499414790170278,
|
|
"loss": 5.5673,
|
|
"mean_token_accuracy": 0.19314760267734526,
|
|
"num_tokens": 1423987.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 5.732563257217407,
|
|
"epoch": 0.507492225049477,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004994073528286833,
|
|
"loss": 5.6066,
|
|
"mean_token_accuracy": 0.2005213737487793,
|
|
"num_tokens": 1428264.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 5.717577838897705,
|
|
"epoch": 0.5089058524173028,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004993998685873395,
|
|
"loss": 5.5313,
|
|
"mean_token_accuracy": 0.1915834978222847,
|
|
"num_tokens": 1432350.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.589537763595581,
|
|
"epoch": 0.5103194797851286,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004993923374478109,
|
|
"loss": 5.5276,
|
|
"mean_token_accuracy": 0.20159524530172349,
|
|
"num_tokens": 1436298.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.542631483078003,
|
|
"epoch": 0.5117331071529545,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004993847594116713,
|
|
"loss": 5.4777,
|
|
"mean_token_accuracy": 0.20509825497865677,
|
|
"num_tokens": 1440206.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.758960771560669,
|
|
"epoch": 0.5131467345207803,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004993771344805046,
|
|
"loss": 5.6741,
|
|
"mean_token_accuracy": 0.1927124723792076,
|
|
"num_tokens": 1444566.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 5.707762908935547,
|
|
"epoch": 0.5145603618886062,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004993694626559043,
|
|
"loss": 5.6075,
|
|
"mean_token_accuracy": 0.19674081355333328,
|
|
"num_tokens": 1448792.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.755503702163696,
|
|
"epoch": 0.515973989256432,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004993617439394737,
|
|
"loss": 5.6768,
|
|
"mean_token_accuracy": 0.1900164932012558,
|
|
"num_tokens": 1452456.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.683109760284424,
|
|
"epoch": 0.5173876166242578,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004993539783328261,
|
|
"loss": 5.5949,
|
|
"mean_token_accuracy": 0.19960361272096633,
|
|
"num_tokens": 1456415.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.583623313903809,
|
|
"epoch": 0.5188012439920837,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004993461658375845,
|
|
"loss": 5.5242,
|
|
"mean_token_accuracy": 0.20874519199132918,
|
|
"num_tokens": 1460390.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.659956359863282,
|
|
"epoch": 0.5202148713599095,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004993383064553814,
|
|
"loss": 5.6479,
|
|
"mean_token_accuracy": 0.19412834346294403,
|
|
"num_tokens": 1464363.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 5.733077812194824,
|
|
"epoch": 0.5216284987277354,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004993304001878595,
|
|
"loss": 5.5266,
|
|
"mean_token_accuracy": 0.20524653643369675,
|
|
"num_tokens": 1468372.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 5.567001724243164,
|
|
"epoch": 0.5230421260955612,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004993224470366712,
|
|
"loss": 5.6421,
|
|
"mean_token_accuracy": 0.19311347603797913,
|
|
"num_tokens": 1472272.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.545955991744995,
|
|
"epoch": 0.524455753463387,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004993144470034787,
|
|
"loss": 5.4196,
|
|
"mean_token_accuracy": 0.21298958063125611,
|
|
"num_tokens": 1476195.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.790099716186523,
|
|
"epoch": 0.5258693808312129,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004993064000899538,
|
|
"loss": 5.6113,
|
|
"mean_token_accuracy": 0.1957776889204979,
|
|
"num_tokens": 1480364.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.554702568054199,
|
|
"epoch": 0.5272830081990387,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004992983062977784,
|
|
"loss": 5.6126,
|
|
"mean_token_accuracy": 0.2019720569252968,
|
|
"num_tokens": 1484523.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 5.681436681747437,
|
|
"epoch": 0.5286966355668645,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004992901656286439,
|
|
"loss": 5.6293,
|
|
"mean_token_accuracy": 0.19366944432258607,
|
|
"num_tokens": 1488456.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.638179159164428,
|
|
"epoch": 0.5301102629346904,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000499281978084252,
|
|
"loss": 5.4644,
|
|
"mean_token_accuracy": 0.20922451019287108,
|
|
"num_tokens": 1492599.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.57268214225769,
|
|
"epoch": 0.5315238903025162,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004992737436663134,
|
|
"loss": 5.4751,
|
|
"mean_token_accuracy": 0.22505437433719636,
|
|
"num_tokens": 1496373.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.473353672027588,
|
|
"epoch": 0.5329375176703421,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0004992654623765493,
|
|
"loss": 5.5507,
|
|
"mean_token_accuracy": 0.20358589738607408,
|
|
"num_tokens": 1500288.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.735796165466309,
|
|
"epoch": 0.5343511450381679,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004992571342166904,
|
|
"loss": 5.6745,
|
|
"mean_token_accuracy": 0.19157627820968628,
|
|
"num_tokens": 1504169.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.760134601593018,
|
|
"epoch": 0.5357647724059937,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004992487591884772,
|
|
"loss": 5.6245,
|
|
"mean_token_accuracy": 0.19079590141773223,
|
|
"num_tokens": 1508292.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.669271850585938,
|
|
"epoch": 0.5371783997738196,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00049924033729366,
|
|
"loss": 5.6237,
|
|
"mean_token_accuracy": 0.2010587587952614,
|
|
"num_tokens": 1512366.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.646955680847168,
|
|
"epoch": 0.5385920271416454,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000499231868533999,
|
|
"loss": 5.6106,
|
|
"mean_token_accuracy": 0.1991715833544731,
|
|
"num_tokens": 1516627.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.649567794799805,
|
|
"epoch": 0.5400056545094714,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004992233529112642,
|
|
"loss": 5.4597,
|
|
"mean_token_accuracy": 0.20987114906311036,
|
|
"num_tokens": 1520541.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.664584159851074,
|
|
"epoch": 0.5414192818772972,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004992147904272349,
|
|
"loss": 5.5956,
|
|
"mean_token_accuracy": 0.18337357342243193,
|
|
"num_tokens": 1524505.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.587951421737671,
|
|
"epoch": 0.542832909245123,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004992061810837012,
|
|
"loss": 5.5333,
|
|
"mean_token_accuracy": 0.20285723954439164,
|
|
"num_tokens": 1528658.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.579516839981079,
|
|
"epoch": 0.5442465366129489,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000499197524882462,
|
|
"loss": 5.4955,
|
|
"mean_token_accuracy": 0.2082669347524643,
|
|
"num_tokens": 1532620.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.605240058898926,
|
|
"epoch": 0.5456601639807747,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004991888218253265,
|
|
"loss": 5.5917,
|
|
"mean_token_accuracy": 0.1849711701273918,
|
|
"num_tokens": 1536636.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.720357370376587,
|
|
"epoch": 0.5470737913486005,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004991800719141136,
|
|
"loss": 5.6327,
|
|
"mean_token_accuracy": 0.1938754215836525,
|
|
"num_tokens": 1540821.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.5586694240570065,
|
|
"epoch": 0.5484874187164264,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004991712751506518,
|
|
"loss": 5.5268,
|
|
"mean_token_accuracy": 0.21065980643033982,
|
|
"num_tokens": 1544700.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.649634504318238,
|
|
"epoch": 0.5499010460842522,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004991624315367798,
|
|
"loss": 5.5222,
|
|
"mean_token_accuracy": 0.20534564256668092,
|
|
"num_tokens": 1548770.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.693707847595215,
|
|
"epoch": 0.5513146734520781,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004991535410743459,
|
|
"loss": 5.6152,
|
|
"mean_token_accuracy": 0.1894238770008087,
|
|
"num_tokens": 1553015.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.646745252609253,
|
|
"epoch": 0.5527283008199039,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004991446037652079,
|
|
"loss": 5.5949,
|
|
"mean_token_accuracy": 0.1843682825565338,
|
|
"num_tokens": 1556849.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.533147239685059,
|
|
"epoch": 0.5541419281877297,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004991356196112339,
|
|
"loss": 5.5687,
|
|
"mean_token_accuracy": 0.20423490107059478,
|
|
"num_tokens": 1560407.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.639888763427734,
|
|
"epoch": 0.5555555555555556,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004991265886143014,
|
|
"loss": 5.4596,
|
|
"mean_token_accuracy": 0.20686074197292328,
|
|
"num_tokens": 1564473.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.58202748298645,
|
|
"epoch": 0.5569691829233814,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004991175107762978,
|
|
"loss": 5.6361,
|
|
"mean_token_accuracy": 0.19261953085660935,
|
|
"num_tokens": 1568141.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.683831691741943,
|
|
"epoch": 0.5583828102912073,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004991083860991204,
|
|
"loss": 5.6186,
|
|
"mean_token_accuracy": 0.20144907236099244,
|
|
"num_tokens": 1572130.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.693622064590454,
|
|
"epoch": 0.5597964376590331,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000499099214584676,
|
|
"loss": 5.6006,
|
|
"mean_token_accuracy": 0.20176017433404922,
|
|
"num_tokens": 1576243.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.718212985992432,
|
|
"epoch": 0.5612100650268589,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004990899962348817,
|
|
"loss": 5.5458,
|
|
"mean_token_accuracy": 0.19472669214010238,
|
|
"num_tokens": 1579916.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.483148622512817,
|
|
"epoch": 0.5626236923946848,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004990807310516638,
|
|
"loss": 5.3971,
|
|
"mean_token_accuracy": 0.22061461508274077,
|
|
"num_tokens": 1584075.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.628428792953491,
|
|
"epoch": 0.5640373197625106,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004990714190369588,
|
|
"loss": 5.5766,
|
|
"mean_token_accuracy": 0.1951795771718025,
|
|
"num_tokens": 1588026.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.706118106842041,
|
|
"epoch": 0.5654509471303364,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000499062060192713,
|
|
"loss": 5.5487,
|
|
"mean_token_accuracy": 0.19893766939640045,
|
|
"num_tokens": 1592420.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 0.5654509471303364,
|
|
"eval_entropy": 5.4442686896050425,
|
|
"eval_loss": 5.730021953582764,
|
|
"eval_mean_token_accuracy": 0.20073206712175648,
|
|
"eval_num_tokens": 1592420.0,
|
|
"eval_runtime": 4.8588,
|
|
"eval_samples_per_second": 2322.381,
|
|
"eval_steps_per_second": 290.401,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 5.5566449642181395,
|
|
"epoch": 0.5668645744981623,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000499052654520882,
|
|
"loss": 5.4582,
|
|
"mean_token_accuracy": 0.21683346033096312,
|
|
"num_tokens": 1596409.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.625777530670166,
|
|
"epoch": 0.5682782018659881,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004990432020234317,
|
|
"loss": 5.5315,
|
|
"mean_token_accuracy": 0.20139055997133254,
|
|
"num_tokens": 1600549.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.454067277908325,
|
|
"epoch": 0.569691829233814,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004990337027023378,
|
|
"loss": 5.4343,
|
|
"mean_token_accuracy": 0.20616025179624559,
|
|
"num_tokens": 1604145.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.501142358779907,
|
|
"epoch": 0.5711054566016398,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004990241565595853,
|
|
"loss": 5.5455,
|
|
"mean_token_accuracy": 0.19608652591705322,
|
|
"num_tokens": 1608078.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.545711851119995,
|
|
"epoch": 0.5725190839694656,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004990145635971694,
|
|
"loss": 5.4784,
|
|
"mean_token_accuracy": 0.19842927604913713,
|
|
"num_tokens": 1612038.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.568185329437256,
|
|
"epoch": 0.5739327113372915,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004990049238170949,
|
|
"loss": 5.5643,
|
|
"mean_token_accuracy": 0.18239825963974,
|
|
"num_tokens": 1616059.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.484708166122436,
|
|
"epoch": 0.5753463387051173,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004989952372213767,
|
|
"loss": 5.4215,
|
|
"mean_token_accuracy": 0.217547644674778,
|
|
"num_tokens": 1619653.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.592620325088501,
|
|
"epoch": 0.5767599660729432,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004989855038120389,
|
|
"loss": 5.4529,
|
|
"mean_token_accuracy": 0.210155089199543,
|
|
"num_tokens": 1623820.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.673013305664062,
|
|
"epoch": 0.578173593440769,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000498975723591116,
|
|
"loss": 5.6388,
|
|
"mean_token_accuracy": 0.2011185497045517,
|
|
"num_tokens": 1627606.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.710173654556274,
|
|
"epoch": 0.5795872208085948,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004989658965606519,
|
|
"loss": 5.6078,
|
|
"mean_token_accuracy": 0.19209033399820327,
|
|
"num_tokens": 1631581.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.637240791320801,
|
|
"epoch": 0.5810008481764207,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004989560227227003,
|
|
"loss": 5.5653,
|
|
"mean_token_accuracy": 0.19754677414894103,
|
|
"num_tokens": 1635670.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.533895635604859,
|
|
"epoch": 0.5824144755442465,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004989461020793248,
|
|
"loss": 5.5605,
|
|
"mean_token_accuracy": 0.2057570233941078,
|
|
"num_tokens": 1639860.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.632710218429565,
|
|
"epoch": 0.5838281029120724,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004989361346325989,
|
|
"loss": 5.5747,
|
|
"mean_token_accuracy": 0.1988660603761673,
|
|
"num_tokens": 1644119.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.631876420974732,
|
|
"epoch": 0.5852417302798982,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004989261203846056,
|
|
"loss": 5.5129,
|
|
"mean_token_accuracy": 0.19564798325300217,
|
|
"num_tokens": 1648106.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.628188610076904,
|
|
"epoch": 0.586655357647724,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004989160593374379,
|
|
"loss": 5.5768,
|
|
"mean_token_accuracy": 0.1943550795316696,
|
|
"num_tokens": 1652430.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.649709939956665,
|
|
"epoch": 0.5880689850155499,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004989059514931985,
|
|
"loss": 5.533,
|
|
"mean_token_accuracy": 0.20396776497364044,
|
|
"num_tokens": 1656775.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.654207420349121,
|
|
"epoch": 0.5894826123833757,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004988957968539997,
|
|
"loss": 5.6254,
|
|
"mean_token_accuracy": 0.19971878081560135,
|
|
"num_tokens": 1661158.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.573638296127319,
|
|
"epoch": 0.5908962397512015,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000498885595421964,
|
|
"loss": 5.4981,
|
|
"mean_token_accuracy": 0.206602144241333,
|
|
"num_tokens": 1664839.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.506303167343139,
|
|
"epoch": 0.5923098671190274,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004988753471992233,
|
|
"loss": 5.3477,
|
|
"mean_token_accuracy": 0.21307512819767,
|
|
"num_tokens": 1668493.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.5169202327728275,
|
|
"epoch": 0.5937234944868532,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004988650521879195,
|
|
"loss": 5.5681,
|
|
"mean_token_accuracy": 0.1938401862978935,
|
|
"num_tokens": 1672376.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.6174476623535154,
|
|
"epoch": 0.5951371218546792,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000498854710390204,
|
|
"loss": 5.4617,
|
|
"mean_token_accuracy": 0.20711683630943298,
|
|
"num_tokens": 1676315.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.556337833404541,
|
|
"epoch": 0.596550749222505,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004988443218082384,
|
|
"loss": 5.5785,
|
|
"mean_token_accuracy": 0.20104349106550218,
|
|
"num_tokens": 1680195.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.4387730121612545,
|
|
"epoch": 0.5979643765903307,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004988338864441937,
|
|
"loss": 5.4152,
|
|
"mean_token_accuracy": 0.20739189386367798,
|
|
"num_tokens": 1684194.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.642705869674683,
|
|
"epoch": 0.5993780039581567,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004988234043002508,
|
|
"loss": 5.4491,
|
|
"mean_token_accuracy": 0.2025332361459732,
|
|
"num_tokens": 1688303.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.581110000610352,
|
|
"epoch": 0.6007916313259825,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004988128753786006,
|
|
"loss": 5.6538,
|
|
"mean_token_accuracy": 0.18712435364723207,
|
|
"num_tokens": 1692806.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.687484979629517,
|
|
"epoch": 0.6022052586938084,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004988022996814435,
|
|
"loss": 5.573,
|
|
"mean_token_accuracy": 0.1949217587709427,
|
|
"num_tokens": 1697028.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.558356523513794,
|
|
"epoch": 0.6036188860616342,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004987916772109897,
|
|
"loss": 5.5124,
|
|
"mean_token_accuracy": 0.20491410195827484,
|
|
"num_tokens": 1701117.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.507727575302124,
|
|
"epoch": 0.60503251342946,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004987810079694591,
|
|
"loss": 5.546,
|
|
"mean_token_accuracy": 0.20128675401210785,
|
|
"num_tokens": 1705157.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.534308624267578,
|
|
"epoch": 0.6064461407972859,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004987702919590819,
|
|
"loss": 5.5511,
|
|
"mean_token_accuracy": 0.210813270509243,
|
|
"num_tokens": 1709476.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.639748430252075,
|
|
"epoch": 0.6078597681651117,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004987595291820972,
|
|
"loss": 5.5881,
|
|
"mean_token_accuracy": 0.19741401970386505,
|
|
"num_tokens": 1713700.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.590295839309692,
|
|
"epoch": 0.6092733955329375,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004987487196407547,
|
|
"loss": 5.4362,
|
|
"mean_token_accuracy": 0.2136401653289795,
|
|
"num_tokens": 1717628.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.4449460983276365,
|
|
"epoch": 0.6106870229007634,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004987378633373134,
|
|
"loss": 5.351,
|
|
"mean_token_accuracy": 0.2092408284544945,
|
|
"num_tokens": 1721618.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.513040208816529,
|
|
"epoch": 0.6121006502685892,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004987269602740423,
|
|
"loss": 5.3616,
|
|
"mean_token_accuracy": 0.21337435394525528,
|
|
"num_tokens": 1725527.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.4719398498535154,
|
|
"epoch": 0.6135142776364151,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004987160104532198,
|
|
"loss": 5.4253,
|
|
"mean_token_accuracy": 0.20791638344526292,
|
|
"num_tokens": 1729562.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.54003643989563,
|
|
"epoch": 0.6149279050042409,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004987050138771347,
|
|
"loss": 5.5033,
|
|
"mean_token_accuracy": 0.20253001302480697,
|
|
"num_tokens": 1733491.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.54003643989563,
|
|
"epoch": 0.6163415323720667,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000498693970548085,
|
|
"loss": 5.3609,
|
|
"mean_token_accuracy": 0.21518508344888687,
|
|
"num_tokens": 1737309.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.562676477432251,
|
|
"epoch": 0.6177551597398926,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004986828804683787,
|
|
"loss": 5.6043,
|
|
"mean_token_accuracy": 0.19376907795667647,
|
|
"num_tokens": 1741514.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.620116949081421,
|
|
"epoch": 0.6191687871077184,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004986717436403337,
|
|
"loss": 5.4595,
|
|
"mean_token_accuracy": 0.21228516697883607,
|
|
"num_tokens": 1745249.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.419637966156006,
|
|
"epoch": 0.6205824144755443,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004986605600662773,
|
|
"loss": 5.47,
|
|
"mean_token_accuracy": 0.20662501752376555,
|
|
"num_tokens": 1749318.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.462861013412476,
|
|
"epoch": 0.6219960418433701,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004986493297485469,
|
|
"loss": 5.5437,
|
|
"mean_token_accuracy": 0.19386183023452758,
|
|
"num_tokens": 1753419.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.6106353282928465,
|
|
"epoch": 0.6234096692111959,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004986380526894896,
|
|
"loss": 5.4658,
|
|
"mean_token_accuracy": 0.20832172632217408,
|
|
"num_tokens": 1757337.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.447007656097412,
|
|
"epoch": 0.6248232965790218,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004986267288914623,
|
|
"loss": 5.4496,
|
|
"mean_token_accuracy": 0.20539478212594986,
|
|
"num_tokens": 1761335.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.396089172363281,
|
|
"epoch": 0.6262369239468476,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004986153583568315,
|
|
"loss": 5.429,
|
|
"mean_token_accuracy": 0.2138877660036087,
|
|
"num_tokens": 1765370.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.696418952941895,
|
|
"epoch": 0.6276505513146734,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004986039410879737,
|
|
"loss": 5.5815,
|
|
"mean_token_accuracy": 0.1938244104385376,
|
|
"num_tokens": 1769272.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.5408367156982425,
|
|
"epoch": 0.6290641786824993,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004985924770872748,
|
|
"loss": 5.4517,
|
|
"mean_token_accuracy": 0.20318606048822402,
|
|
"num_tokens": 1773329.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.4070816993713375,
|
|
"epoch": 0.6304778060503251,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004985809663571309,
|
|
"loss": 5.4137,
|
|
"mean_token_accuracy": 0.22061175405979155,
|
|
"num_tokens": 1777223.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.562509918212891,
|
|
"epoch": 0.631891433418151,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004985694088999476,
|
|
"loss": 5.5138,
|
|
"mean_token_accuracy": 0.2007150322198868,
|
|
"num_tokens": 1781359.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.520141315460205,
|
|
"epoch": 0.6333050607859768,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004985578047181404,
|
|
"loss": 5.4645,
|
|
"mean_token_accuracy": 0.20261273682117462,
|
|
"num_tokens": 1785605.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.530501174926758,
|
|
"epoch": 0.6347186881538026,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004985461538141344,
|
|
"loss": 5.4674,
|
|
"mean_token_accuracy": 0.2011677071452141,
|
|
"num_tokens": 1789187.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.574052000045777,
|
|
"epoch": 0.6361323155216285,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004985344561903646,
|
|
"loss": 5.5618,
|
|
"mean_token_accuracy": 0.20011836439371108,
|
|
"num_tokens": 1792971.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.608518457412719,
|
|
"epoch": 0.6375459428894543,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004985227118492758,
|
|
"loss": 5.5191,
|
|
"mean_token_accuracy": 0.20964549332857133,
|
|
"num_tokens": 1796938.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.61181960105896,
|
|
"epoch": 0.6389595702572802,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004985109207933224,
|
|
"loss": 5.5303,
|
|
"mean_token_accuracy": 0.20179047882556916,
|
|
"num_tokens": 1801017.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.59629225730896,
|
|
"epoch": 0.640373197625106,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004984990830249687,
|
|
"loss": 5.5526,
|
|
"mean_token_accuracy": 0.1967478796839714,
|
|
"num_tokens": 1804819.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.543248891830444,
|
|
"epoch": 0.6417868249929318,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004984871985466888,
|
|
"loss": 5.2833,
|
|
"mean_token_accuracy": 0.22129721641540528,
|
|
"num_tokens": 1808570.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.519654655456543,
|
|
"epoch": 0.6432004523607577,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004984752673609663,
|
|
"loss": 5.5729,
|
|
"mean_token_accuracy": 0.19660354554653167,
|
|
"num_tokens": 1812762.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.472464466094971,
|
|
"epoch": 0.6446140797285835,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004984632894702949,
|
|
"loss": 5.4289,
|
|
"mean_token_accuracy": 0.20329504907131196,
|
|
"num_tokens": 1816836.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.550273275375366,
|
|
"epoch": 0.6460277070964093,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004984512648771779,
|
|
"loss": 5.3748,
|
|
"mean_token_accuracy": 0.2208165168762207,
|
|
"num_tokens": 1820663.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.413545417785644,
|
|
"epoch": 0.6474413344642352,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004984391935841283,
|
|
"loss": 5.5055,
|
|
"mean_token_accuracy": 0.2082822859287262,
|
|
"num_tokens": 1825018.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.4688355922698975,
|
|
"epoch": 0.648854961832061,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000498427075593669,
|
|
"loss": 5.4444,
|
|
"mean_token_accuracy": 0.2056533843278885,
|
|
"num_tokens": 1829068.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.58225564956665,
|
|
"epoch": 0.650268589199887,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004984149109083325,
|
|
"loss": 5.406,
|
|
"mean_token_accuracy": 0.21179707050323487,
|
|
"num_tokens": 1832971.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.427003192901611,
|
|
"epoch": 0.6516822165677127,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000498402699530661,
|
|
"loss": 5.4303,
|
|
"mean_token_accuracy": 0.20738290697336198,
|
|
"num_tokens": 1837021.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.402704811096191,
|
|
"epoch": 0.6530958439355385,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000498390441463207,
|
|
"loss": 5.3714,
|
|
"mean_token_accuracy": 0.21459479480981827,
|
|
"num_tokens": 1841161.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.4311668395996096,
|
|
"epoch": 0.6545094713033645,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000498378136708532,
|
|
"loss": 5.4507,
|
|
"mean_token_accuracy": 0.20626376271247865,
|
|
"num_tokens": 1845219.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.5389612197875975,
|
|
"epoch": 0.6559230986711903,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004983657852692079,
|
|
"loss": 5.4873,
|
|
"mean_token_accuracy": 0.20499810874462127,
|
|
"num_tokens": 1849101.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.606302070617676,
|
|
"epoch": 0.6573367260390162,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004983533871478159,
|
|
"loss": 5.5586,
|
|
"mean_token_accuracy": 0.19572011679410933,
|
|
"num_tokens": 1853160.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.409348201751709,
|
|
"epoch": 0.658750353406842,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000498340942346947,
|
|
"loss": 5.4022,
|
|
"mean_token_accuracy": 0.2098221644759178,
|
|
"num_tokens": 1857002.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.392949056625366,
|
|
"epoch": 0.6601639807746678,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004983284508692025,
|
|
"loss": 5.364,
|
|
"mean_token_accuracy": 0.21255889534950256,
|
|
"num_tokens": 1860393.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.473573160171509,
|
|
"epoch": 0.6615776081424937,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004983159127171926,
|
|
"loss": 5.4675,
|
|
"mean_token_accuracy": 0.20936144888401031,
|
|
"num_tokens": 1864240.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.47200608253479,
|
|
"epoch": 0.6629912355103195,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004983033278935379,
|
|
"loss": 5.4102,
|
|
"mean_token_accuracy": 0.20349428206682205,
|
|
"num_tokens": 1868158.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.3972259044647215,
|
|
"epoch": 0.6644048628781454,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004982906964008686,
|
|
"loss": 5.3302,
|
|
"mean_token_accuracy": 0.21207149028778077,
|
|
"num_tokens": 1872136.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.50235013961792,
|
|
"epoch": 0.6658184902459712,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004982780182418245,
|
|
"loss": 5.4773,
|
|
"mean_token_accuracy": 0.20729459077119827,
|
|
"num_tokens": 1876278.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.567453908920288,
|
|
"epoch": 0.667232117613797,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004982652934190553,
|
|
"loss": 5.486,
|
|
"mean_token_accuracy": 0.20608954280614852,
|
|
"num_tokens": 1880477.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.323032474517822,
|
|
"epoch": 0.6686457449816229,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004982525219352204,
|
|
"loss": 5.2575,
|
|
"mean_token_accuracy": 0.22075929641723632,
|
|
"num_tokens": 1884636.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.422608804702759,
|
|
"epoch": 0.6700593723494487,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004982397037929891,
|
|
"loss": 5.4597,
|
|
"mean_token_accuracy": 0.2153452843427658,
|
|
"num_tokens": 1888473.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.483827829360962,
|
|
"epoch": 0.6714729997172745,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004982268389950399,
|
|
"loss": 5.412,
|
|
"mean_token_accuracy": 0.20630247592926027,
|
|
"num_tokens": 1892235.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.440791273117066,
|
|
"epoch": 0.6728866270851004,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000498213927544062,
|
|
"loss": 5.4818,
|
|
"mean_token_accuracy": 0.21008707135915755,
|
|
"num_tokens": 1896294.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.5512748718261715,
|
|
"epoch": 0.6743002544529262,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004982009694427535,
|
|
"loss": 5.4943,
|
|
"mean_token_accuracy": 0.2023972362279892,
|
|
"num_tokens": 1900517.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.477317762374878,
|
|
"epoch": 0.6757138818207521,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004981879646938225,
|
|
"loss": 5.406,
|
|
"mean_token_accuracy": 0.21320468336343765,
|
|
"num_tokens": 1904385.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.405479955673218,
|
|
"epoch": 0.6771275091885779,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004981749132999872,
|
|
"loss": 5.5958,
|
|
"mean_token_accuracy": 0.19751414954662322,
|
|
"num_tokens": 1908378.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.559717321395874,
|
|
"epoch": 0.6785411365564037,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004981618152639749,
|
|
"loss": 5.5465,
|
|
"mean_token_accuracy": 0.2008774310350418,
|
|
"num_tokens": 1912287.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.542268133163452,
|
|
"epoch": 0.6799547639242296,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004981486705885234,
|
|
"loss": 5.3446,
|
|
"mean_token_accuracy": 0.21260107159614564,
|
|
"num_tokens": 1916489.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.443733596801758,
|
|
"epoch": 0.6813683912920554,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004981354792763796,
|
|
"loss": 5.3265,
|
|
"mean_token_accuracy": 0.21611038148403167,
|
|
"num_tokens": 1920443.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.47453498840332,
|
|
"epoch": 0.6827820186598813,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004981222413303004,
|
|
"loss": 5.4629,
|
|
"mean_token_accuracy": 0.21295766681432723,
|
|
"num_tokens": 1924483.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.448499298095703,
|
|
"epoch": 0.6841956460277071,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004981089567530525,
|
|
"loss": 5.4585,
|
|
"mean_token_accuracy": 0.20760513991117477,
|
|
"num_tokens": 1928307.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.524633836746216,
|
|
"epoch": 0.6856092733955329,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004980956255474124,
|
|
"loss": 5.4341,
|
|
"mean_token_accuracy": 0.21599441021680832,
|
|
"num_tokens": 1932279.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.503360843658447,
|
|
"epoch": 0.6870229007633588,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000498082247716166,
|
|
"loss": 5.3663,
|
|
"mean_token_accuracy": 0.20277948826551437,
|
|
"num_tokens": 1936420.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.411725807189941,
|
|
"epoch": 0.6884365281311846,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004980688232621095,
|
|
"loss": 5.3858,
|
|
"mean_token_accuracy": 0.2142704486846924,
|
|
"num_tokens": 1940252.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.480589294433594,
|
|
"epoch": 0.6898501554990104,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004980553521880482,
|
|
"loss": 5.4548,
|
|
"mean_token_accuracy": 0.19498221576213837,
|
|
"num_tokens": 1944346.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.4414145946502686,
|
|
"epoch": 0.6912637828668363,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004980418344967978,
|
|
"loss": 5.3983,
|
|
"mean_token_accuracy": 0.2134234771132469,
|
|
"num_tokens": 1948330.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.435844564437867,
|
|
"epoch": 0.6926774102346621,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000498028270191183,
|
|
"loss": 5.402,
|
|
"mean_token_accuracy": 0.20972336530685426,
|
|
"num_tokens": 1952489.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.479365015029908,
|
|
"epoch": 0.694091037602488,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004980146592740391,
|
|
"loss": 5.397,
|
|
"mean_token_accuracy": 0.20905091166496276,
|
|
"num_tokens": 1956850.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.433628702163697,
|
|
"epoch": 0.6955046649703138,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004980010017482104,
|
|
"loss": 5.4489,
|
|
"mean_token_accuracy": 0.2088480368256569,
|
|
"num_tokens": 1960901.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.33675332069397,
|
|
"epoch": 0.6969182923381396,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004979872976165513,
|
|
"loss": 5.4547,
|
|
"mean_token_accuracy": 0.2013069435954094,
|
|
"num_tokens": 1964730.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.7374622344970705,
|
|
"epoch": 0.6983319197059655,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000497973546881926,
|
|
"loss": 5.4922,
|
|
"mean_token_accuracy": 0.19955756217241288,
|
|
"num_tokens": 1968489.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.62591814994812,
|
|
"epoch": 0.6997455470737913,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004979597495472081,
|
|
"loss": 5.6041,
|
|
"mean_token_accuracy": 0.19394168257713318,
|
|
"num_tokens": 1972536.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.542226314544678,
|
|
"epoch": 0.7011591744416172,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004979459056152814,
|
|
"loss": 5.5563,
|
|
"mean_token_accuracy": 0.1934622511267662,
|
|
"num_tokens": 1976410.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.546902847290039,
|
|
"epoch": 0.702572801809443,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000497932015089039,
|
|
"loss": 5.3735,
|
|
"mean_token_accuracy": 0.2068051278591156,
|
|
"num_tokens": 1980294.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.400882911682129,
|
|
"epoch": 0.7039864291772688,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004979180779713841,
|
|
"loss": 5.3773,
|
|
"mean_token_accuracy": 0.21438602060079576,
|
|
"num_tokens": 1984096.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.387070846557617,
|
|
"epoch": 0.7054000565450947,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004979040942652294,
|
|
"loss": 5.3429,
|
|
"mean_token_accuracy": 0.21412564367055892,
|
|
"num_tokens": 1987714.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.4379284381866455,
|
|
"epoch": 0.7068136839129205,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004978900639734974,
|
|
"loss": 5.4291,
|
|
"mean_token_accuracy": 0.2055034726858139,
|
|
"num_tokens": 1991695.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.602289438247681,
|
|
"epoch": 0.7082273112807463,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004978759870991203,
|
|
"loss": 5.5607,
|
|
"mean_token_accuracy": 0.18915494084358214,
|
|
"num_tokens": 1995735.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.417237043380737,
|
|
"epoch": 0.7096409386485723,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004978618636450401,
|
|
"loss": 5.3789,
|
|
"mean_token_accuracy": 0.2134407415986061,
|
|
"num_tokens": 1999642.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.299266576766968,
|
|
"epoch": 0.711054566016398,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004978476936142085,
|
|
"loss": 5.4159,
|
|
"mean_token_accuracy": 0.2092890202999115,
|
|
"num_tokens": 2003374.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.344558954238892,
|
|
"epoch": 0.712468193384224,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000497833477009587,
|
|
"loss": 5.3025,
|
|
"mean_token_accuracy": 0.2184336155653,
|
|
"num_tokens": 2007286.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.4438793659210205,
|
|
"epoch": 0.7138818207520498,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004978192138341468,
|
|
"loss": 5.3299,
|
|
"mean_token_accuracy": 0.2091567873954773,
|
|
"num_tokens": 2011245.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.371488094329834,
|
|
"epoch": 0.7152954481198756,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004978049040908688,
|
|
"loss": 5.3136,
|
|
"mean_token_accuracy": 0.21774005740880967,
|
|
"num_tokens": 2015535.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.462795925140381,
|
|
"epoch": 0.7167090754877015,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004977905477827437,
|
|
"loss": 5.4923,
|
|
"mean_token_accuracy": 0.19267662912607192,
|
|
"num_tokens": 2019531.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.458331871032715,
|
|
"epoch": 0.7181227028555273,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004977761449127716,
|
|
"loss": 5.3604,
|
|
"mean_token_accuracy": 0.20466733276844024,
|
|
"num_tokens": 2023665.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.399077320098877,
|
|
"epoch": 0.7195363302233532,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000497761695483963,
|
|
"loss": 5.3461,
|
|
"mean_token_accuracy": 0.2058019906282425,
|
|
"num_tokens": 2027718.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.425813341140747,
|
|
"epoch": 0.720949957591179,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004977471994993375,
|
|
"loss": 5.4704,
|
|
"mean_token_accuracy": 0.1954423427581787,
|
|
"num_tokens": 2031752.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.447065830230713,
|
|
"epoch": 0.7223635849590048,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004977326569619247,
|
|
"loss": 5.4989,
|
|
"mean_token_accuracy": 0.20671794414520264,
|
|
"num_tokens": 2035678.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.500088357925415,
|
|
"epoch": 0.7237772123268307,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004977180678747639,
|
|
"loss": 5.4082,
|
|
"mean_token_accuracy": 0.1994786888360977,
|
|
"num_tokens": 2039659.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.377178764343261,
|
|
"epoch": 0.7251908396946565,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004977034322409043,
|
|
"loss": 5.3022,
|
|
"mean_token_accuracy": 0.22200674265623094,
|
|
"num_tokens": 2043451.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.408650875091553,
|
|
"epoch": 0.7266044670624824,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004976887500634044,
|
|
"loss": 5.3769,
|
|
"mean_token_accuracy": 0.20982315689325332,
|
|
"num_tokens": 2047388.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.363976049423218,
|
|
"epoch": 0.7280180944303082,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004976740213453329,
|
|
"loss": 5.3214,
|
|
"mean_token_accuracy": 0.21539135426282882,
|
|
"num_tokens": 2051153.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.343084049224854,
|
|
"epoch": 0.729431721798134,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004976592460897678,
|
|
"loss": 5.3005,
|
|
"mean_token_accuracy": 0.22188264429569243,
|
|
"num_tokens": 2055150.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.382208204269409,
|
|
"epoch": 0.7308453491659599,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004976444242997973,
|
|
"loss": 5.2374,
|
|
"mean_token_accuracy": 0.2150484710931778,
|
|
"num_tokens": 2059065.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.382781219482422,
|
|
"epoch": 0.7322589765337857,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004976295559785187,
|
|
"loss": 5.3745,
|
|
"mean_token_accuracy": 0.198026043176651,
|
|
"num_tokens": 2063137.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.454232454299927,
|
|
"epoch": 0.7336726039016115,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004976146411290398,
|
|
"loss": 5.4954,
|
|
"mean_token_accuracy": 0.20750914961099626,
|
|
"num_tokens": 2067112.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.412012767791748,
|
|
"epoch": 0.7350862312694374,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004975996797544774,
|
|
"loss": 5.4636,
|
|
"mean_token_accuracy": 0.2137315034866333,
|
|
"num_tokens": 2071336.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.416234493255615,
|
|
"epoch": 0.7364998586372632,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004975846718579585,
|
|
"loss": 5.3089,
|
|
"mean_token_accuracy": 0.2125646710395813,
|
|
"num_tokens": 2075072.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.363396263122558,
|
|
"epoch": 0.7379134860050891,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004975696174426196,
|
|
"loss": 5.3686,
|
|
"mean_token_accuracy": 0.21096211373806,
|
|
"num_tokens": 2079177.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.520514678955078,
|
|
"epoch": 0.7393271133729149,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004975545165116069,
|
|
"loss": 5.4923,
|
|
"mean_token_accuracy": 0.20160481184720994,
|
|
"num_tokens": 2083010.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.39557147026062,
|
|
"epoch": 0.7407407407407407,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004975393690680765,
|
|
"loss": 5.3944,
|
|
"mean_token_accuracy": 0.21329910457134246,
|
|
"num_tokens": 2087225.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.355341625213623,
|
|
"epoch": 0.7421543681085666,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004975241751151942,
|
|
"loss": 5.3517,
|
|
"mean_token_accuracy": 0.2054790213704109,
|
|
"num_tokens": 2091053.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.476370191574096,
|
|
"epoch": 0.7435679954763924,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004975089346561354,
|
|
"loss": 5.4972,
|
|
"mean_token_accuracy": 0.20132934898138047,
|
|
"num_tokens": 2095152.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.485976266860962,
|
|
"epoch": 0.7449816228442183,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004974936476940851,
|
|
"loss": 5.3576,
|
|
"mean_token_accuracy": 0.21123178601264953,
|
|
"num_tokens": 2099282.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.487665700912475,
|
|
"epoch": 0.7463952502120441,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004974783142322383,
|
|
"loss": 5.4251,
|
|
"mean_token_accuracy": 0.20355846732854843,
|
|
"num_tokens": 2103358.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.468116903305054,
|
|
"epoch": 0.7478088775798699,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004974629342737995,
|
|
"loss": 5.4438,
|
|
"mean_token_accuracy": 0.2065952703356743,
|
|
"num_tokens": 2107180.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.247257423400879,
|
|
"epoch": 0.7492225049476958,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004974475078219833,
|
|
"loss": 5.3647,
|
|
"mean_token_accuracy": 0.21316832154989243,
|
|
"num_tokens": 2111002.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.357695865631103,
|
|
"epoch": 0.7506361323155216,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004974320348800134,
|
|
"loss": 5.2408,
|
|
"mean_token_accuracy": 0.20852022171020507,
|
|
"num_tokens": 2114925.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.451169013977051,
|
|
"epoch": 0.7520497596833474,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004974165154511237,
|
|
"loss": 5.3472,
|
|
"mean_token_accuracy": 0.21330751180648805,
|
|
"num_tokens": 2119039.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.410284042358398,
|
|
"epoch": 0.7534633870511733,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004974009495385578,
|
|
"loss": 5.353,
|
|
"mean_token_accuracy": 0.21694692373275756,
|
|
"num_tokens": 2123215.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.412848854064942,
|
|
"epoch": 0.7548770144189991,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004973853371455686,
|
|
"loss": 5.3483,
|
|
"mean_token_accuracy": 0.20225527435541152,
|
|
"num_tokens": 2127013.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.468294858932495,
|
|
"epoch": 0.756290641786825,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004973696782754191,
|
|
"loss": 5.4464,
|
|
"mean_token_accuracy": 0.20311361104249953,
|
|
"num_tokens": 2131310.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.42818603515625,
|
|
"epoch": 0.7577042691546508,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004973539729313819,
|
|
"loss": 5.4026,
|
|
"mean_token_accuracy": 0.2121235892176628,
|
|
"num_tokens": 2135203.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.377758741378784,
|
|
"epoch": 0.7591178965224766,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004973382211167394,
|
|
"loss": 5.3306,
|
|
"mean_token_accuracy": 0.21791625618934632,
|
|
"num_tokens": 2139212.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.377042150497436,
|
|
"epoch": 0.7605315238903025,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004973224228347837,
|
|
"loss": 5.3246,
|
|
"mean_token_accuracy": 0.20743905156850814,
|
|
"num_tokens": 2143215.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.336807489395142,
|
|
"epoch": 0.7619451512581283,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004973065780888164,
|
|
"loss": 5.3244,
|
|
"mean_token_accuracy": 0.2237856090068817,
|
|
"num_tokens": 2146984.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.466645050048828,
|
|
"epoch": 0.7633587786259542,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004972906868821489,
|
|
"loss": 5.4005,
|
|
"mean_token_accuracy": 0.21173439621925355,
|
|
"num_tokens": 2151107.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.389848041534424,
|
|
"epoch": 0.76477240599378,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004972747492181024,
|
|
"loss": 5.3846,
|
|
"mean_token_accuracy": 0.20756343454122544,
|
|
"num_tokens": 2155036.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.4115753173828125,
|
|
"epoch": 0.7661860333616058,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000497258765100008,
|
|
"loss": 5.3651,
|
|
"mean_token_accuracy": 0.20176379829645158,
|
|
"num_tokens": 2159043.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.4078697681427,
|
|
"epoch": 0.7675996607294318,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000497242734531206,
|
|
"loss": 5.381,
|
|
"mean_token_accuracy": 0.2030152916908264,
|
|
"num_tokens": 2163089.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.416038370132446,
|
|
"epoch": 0.7690132880972576,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004972266575150468,
|
|
"loss": 5.312,
|
|
"mean_token_accuracy": 0.2068483904004097,
|
|
"num_tokens": 2167089.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.413974142074585,
|
|
"epoch": 0.7704269154650834,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0004972105340548903,
|
|
"loss": 5.3581,
|
|
"mean_token_accuracy": 0.21817459911108017,
|
|
"num_tokens": 2170899.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.337532997131348,
|
|
"epoch": 0.7718405428329093,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004971943641541065,
|
|
"loss": 5.4063,
|
|
"mean_token_accuracy": 0.2055883526802063,
|
|
"num_tokens": 2174820.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.404606342315674,
|
|
"epoch": 0.7732541702007351,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004971781478160743,
|
|
"loss": 5.3472,
|
|
"mean_token_accuracy": 0.21686217188835144,
|
|
"num_tokens": 2179240.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.41935715675354,
|
|
"epoch": 0.774667797568561,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004971618850441833,
|
|
"loss": 5.4785,
|
|
"mean_token_accuracy": 0.2156490206718445,
|
|
"num_tokens": 2183302.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.568381071090698,
|
|
"epoch": 0.7760814249363868,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004971455758418322,
|
|
"loss": 5.405,
|
|
"mean_token_accuracy": 0.1920212149620056,
|
|
"num_tokens": 2187397.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.414394855499268,
|
|
"epoch": 0.7774950523042126,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004971292202124293,
|
|
"loss": 5.3526,
|
|
"mean_token_accuracy": 0.21126374155282973,
|
|
"num_tokens": 2191475.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.378679180145264,
|
|
"epoch": 0.7789086796720385,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004971128181593931,
|
|
"loss": 5.4867,
|
|
"mean_token_accuracy": 0.21251560300588607,
|
|
"num_tokens": 2195952.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.4398126125335695,
|
|
"epoch": 0.7803223070398643,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004970963696861512,
|
|
"loss": 5.3682,
|
|
"mean_token_accuracy": 0.20482446253299713,
|
|
"num_tokens": 2200555.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.51300368309021,
|
|
"epoch": 0.7817359344076902,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004970798747961417,
|
|
"loss": 5.3824,
|
|
"mean_token_accuracy": 0.2078801453113556,
|
|
"num_tokens": 2204477.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.400747537612915,
|
|
"epoch": 0.783149561775516,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0004970633334928116,
|
|
"loss": 5.4376,
|
|
"mean_token_accuracy": 0.19922342598438264,
|
|
"num_tokens": 2208252.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.42206335067749,
|
|
"epoch": 0.7845631891433418,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004970467457796179,
|
|
"loss": 5.3334,
|
|
"mean_token_accuracy": 0.20905719548463822,
|
|
"num_tokens": 2211933.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.404748868942261,
|
|
"epoch": 0.7859768165111677,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004970301116600274,
|
|
"loss": 5.3501,
|
|
"mean_token_accuracy": 0.21212174743413925,
|
|
"num_tokens": 2215966.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.34576644897461,
|
|
"epoch": 0.7873904438789935,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004970134311375168,
|
|
"loss": 5.369,
|
|
"mean_token_accuracy": 0.20176939517259598,
|
|
"num_tokens": 2219976.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.504501152038574,
|
|
"epoch": 0.7888040712468194,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004969967042155717,
|
|
"loss": 5.5131,
|
|
"mean_token_accuracy": 0.20712943375110626,
|
|
"num_tokens": 2224236.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.545808506011963,
|
|
"epoch": 0.7902176986146452,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004969799308976883,
|
|
"loss": 5.3114,
|
|
"mean_token_accuracy": 0.22004663199186325,
|
|
"num_tokens": 2228137.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.274337720870972,
|
|
"epoch": 0.791631325982471,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000496963111187372,
|
|
"loss": 5.1913,
|
|
"mean_token_accuracy": 0.21963750272989274,
|
|
"num_tokens": 2232186.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.311163997650146,
|
|
"epoch": 0.7930449533502969,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004969462450881382,
|
|
"loss": 5.3606,
|
|
"mean_token_accuracy": 0.20587599128484727,
|
|
"num_tokens": 2236183.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.4265374660491945,
|
|
"epoch": 0.7944585807181227,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004969293326035115,
|
|
"loss": 5.3162,
|
|
"mean_token_accuracy": 0.21735673695802687,
|
|
"num_tokens": 2240087.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.384152603149414,
|
|
"epoch": 0.7958722080859485,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004969123737370267,
|
|
"loss": 5.3452,
|
|
"mean_token_accuracy": 0.21657740622758864,
|
|
"num_tokens": 2243905.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.386789321899414,
|
|
"epoch": 0.7972858354537744,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004968953684922281,
|
|
"loss": 5.3324,
|
|
"mean_token_accuracy": 0.21099380999803544,
|
|
"num_tokens": 2248190.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.51895809173584,
|
|
"epoch": 0.7986994628216002,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004968783168726697,
|
|
"loss": 5.4215,
|
|
"mean_token_accuracy": 0.20733469277620314,
|
|
"num_tokens": 2252375.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.399526977539063,
|
|
"epoch": 0.8001130901894261,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004968612188819149,
|
|
"loss": 5.2524,
|
|
"mean_token_accuracy": 0.2202279359102249,
|
|
"num_tokens": 2256152.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.323817729949951,
|
|
"epoch": 0.8015267175572519,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004968440745235375,
|
|
"loss": 5.4054,
|
|
"mean_token_accuracy": 0.20645098388195038,
|
|
"num_tokens": 2260023.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.362152194976806,
|
|
"epoch": 0.8029403449250777,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004968268838011203,
|
|
"loss": 5.2363,
|
|
"mean_token_accuracy": 0.22098526656627654,
|
|
"num_tokens": 2264116.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.393295431137085,
|
|
"epoch": 0.8043539722929036,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004968096467182561,
|
|
"loss": 5.331,
|
|
"mean_token_accuracy": 0.21994270533323287,
|
|
"num_tokens": 2268113.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.323547506332398,
|
|
"epoch": 0.8057675996607294,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004967923632785473,
|
|
"loss": 5.3022,
|
|
"mean_token_accuracy": 0.2200825721025467,
|
|
"num_tokens": 2271781.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.337121534347534,
|
|
"epoch": 0.8071812270285553,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004967750334856063,
|
|
"loss": 5.2763,
|
|
"mean_token_accuracy": 0.21979268193244933,
|
|
"num_tokens": 2275662.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.371746683120728,
|
|
"epoch": 0.8085948543963811,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004967576573430546,
|
|
"loss": 5.3663,
|
|
"mean_token_accuracy": 0.21294018477201462,
|
|
"num_tokens": 2279753.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.373874664306641,
|
|
"epoch": 0.8100084817642069,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004967402348545237,
|
|
"loss": 5.3664,
|
|
"mean_token_accuracy": 0.21907803565263748,
|
|
"num_tokens": 2283684.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.5256712436676025,
|
|
"epoch": 0.8114221091320328,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004967227660236551,
|
|
"loss": 5.4241,
|
|
"mean_token_accuracy": 0.1930916875600815,
|
|
"num_tokens": 2287894.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.345724725723267,
|
|
"epoch": 0.8128357364998586,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004967052508540995,
|
|
"loss": 5.2391,
|
|
"mean_token_accuracy": 0.21856693029403687,
|
|
"num_tokens": 2291833.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.253313446044922,
|
|
"epoch": 0.8142493638676844,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004966876893495174,
|
|
"loss": 5.233,
|
|
"mean_token_accuracy": 0.2188587084412575,
|
|
"num_tokens": 2295800.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.382266187667847,
|
|
"epoch": 0.8156629912355103,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000496670081513579,
|
|
"loss": 5.3665,
|
|
"mean_token_accuracy": 0.20460723340511322,
|
|
"num_tokens": 2299704.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.46207594871521,
|
|
"epoch": 0.8170766186033361,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004966524273499644,
|
|
"loss": 5.4699,
|
|
"mean_token_accuracy": 0.1964840516448021,
|
|
"num_tokens": 2303646.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.35508623123169,
|
|
"epoch": 0.818490245971162,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004966347268623632,
|
|
"loss": 5.2776,
|
|
"mean_token_accuracy": 0.22075464427471161,
|
|
"num_tokens": 2307587.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.351888513565063,
|
|
"epoch": 0.8199038733389878,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004966169800544746,
|
|
"loss": 5.3383,
|
|
"mean_token_accuracy": 0.2160826101899147,
|
|
"num_tokens": 2311372.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.394636487960815,
|
|
"epoch": 0.8213175007068136,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004965991869300076,
|
|
"loss": 5.4191,
|
|
"mean_token_accuracy": 0.21245351135730745,
|
|
"num_tokens": 2315804.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.418462181091309,
|
|
"epoch": 0.8227311280746396,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004965813474926809,
|
|
"loss": 5.3683,
|
|
"mean_token_accuracy": 0.20156540721654892,
|
|
"num_tokens": 2320095.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.445220184326172,
|
|
"epoch": 0.8241447554424653,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004965634617462227,
|
|
"loss": 5.5311,
|
|
"mean_token_accuracy": 0.19646522551774978,
|
|
"num_tokens": 2324324.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.497080945968628,
|
|
"epoch": 0.8255583828102913,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004965455296943712,
|
|
"loss": 5.4288,
|
|
"mean_token_accuracy": 0.21110617220401764,
|
|
"num_tokens": 2328419.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.508669900894165,
|
|
"epoch": 0.8269720101781171,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004965275513408739,
|
|
"loss": 5.3289,
|
|
"mean_token_accuracy": 0.21663168221712112,
|
|
"num_tokens": 2332338.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.248326587677002,
|
|
"epoch": 0.8283856375459429,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004965095266894884,
|
|
"loss": 5.2622,
|
|
"mean_token_accuracy": 0.20836566537618637,
|
|
"num_tokens": 2336139.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.3068788051605225,
|
|
"epoch": 0.8297992649137688,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004964914557439815,
|
|
"loss": 5.4313,
|
|
"mean_token_accuracy": 0.20343007743358613,
|
|
"num_tokens": 2340222.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.519611406326294,
|
|
"epoch": 0.8312128922815946,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004964733385081299,
|
|
"loss": 5.4636,
|
|
"mean_token_accuracy": 0.18783076256513595,
|
|
"num_tokens": 2343895.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.369992256164551,
|
|
"epoch": 0.8326265196494204,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004964551749857203,
|
|
"loss": 5.196,
|
|
"mean_token_accuracy": 0.22638077437877654,
|
|
"num_tokens": 2347924.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.372833156585694,
|
|
"epoch": 0.8340401470172463,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004964369651805484,
|
|
"loss": 5.4979,
|
|
"mean_token_accuracy": 0.19390685856342316,
|
|
"num_tokens": 2352261.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.355490064620971,
|
|
"epoch": 0.8354537743850721,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004964187090964202,
|
|
"loss": 5.389,
|
|
"mean_token_accuracy": 0.2094727098941803,
|
|
"num_tokens": 2356127.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.4388844013214115,
|
|
"epoch": 0.836867401752898,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004964004067371509,
|
|
"loss": 5.2777,
|
|
"mean_token_accuracy": 0.2158604696393013,
|
|
"num_tokens": 2360318.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.473987436294555,
|
|
"epoch": 0.8382810291207238,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004963820581065659,
|
|
"loss": 5.3526,
|
|
"mean_token_accuracy": 0.20864890217781068,
|
|
"num_tokens": 2364421.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.406693792343139,
|
|
"epoch": 0.8396946564885496,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004963636632084996,
|
|
"loss": 5.3444,
|
|
"mean_token_accuracy": 0.21230589747428893,
|
|
"num_tokens": 2368360.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.381004810333252,
|
|
"epoch": 0.8411082838563755,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004963452220467966,
|
|
"loss": 5.2351,
|
|
"mean_token_accuracy": 0.2133149892091751,
|
|
"num_tokens": 2372233.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.3749185562133786,
|
|
"epoch": 0.8425219112242013,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004963267346253108,
|
|
"loss": 5.228,
|
|
"mean_token_accuracy": 0.2158984884619713,
|
|
"num_tokens": 2376175.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.213108348846435,
|
|
"epoch": 0.8439355385920272,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004963082009479063,
|
|
"loss": 5.3595,
|
|
"mean_token_accuracy": 0.21129231303930282,
|
|
"num_tokens": 2380052.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.3238255977630615,
|
|
"epoch": 0.845349165959853,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004962896210184563,
|
|
"loss": 5.4534,
|
|
"mean_token_accuracy": 0.2019223690032959,
|
|
"num_tokens": 2384273.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.401165962219238,
|
|
"epoch": 0.8467627933276788,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004962709948408439,
|
|
"loss": 5.3274,
|
|
"mean_token_accuracy": 0.2195882886648178,
|
|
"num_tokens": 2388113.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.442725038528442,
|
|
"epoch": 0.8481764206955047,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004962523224189618,
|
|
"loss": 5.4501,
|
|
"mean_token_accuracy": 0.20411702394485473,
|
|
"num_tokens": 2392345.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 0.8481764206955047,
|
|
"eval_entropy": 5.174652101299088,
|
|
"eval_loss": 5.490718364715576,
|
|
"eval_mean_token_accuracy": 0.2128728753355781,
|
|
"eval_num_tokens": 2392345.0,
|
|
"eval_runtime": 4.8375,
|
|
"eval_samples_per_second": 2332.633,
|
|
"eval_steps_per_second": 291.682,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.367189741134643,
|
|
"epoch": 0.8495900480633305,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004962336037567126,
|
|
"loss": 5.2325,
|
|
"mean_token_accuracy": 0.2248583048582077,
|
|
"num_tokens": 2396304.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.238517618179321,
|
|
"epoch": 0.8510036754311563,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004962148388580083,
|
|
"loss": 5.2524,
|
|
"mean_token_accuracy": 0.2163602665066719,
|
|
"num_tokens": 2400137.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.343601322174072,
|
|
"epoch": 0.8524173027989822,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004961960277267705,
|
|
"loss": 5.3324,
|
|
"mean_token_accuracy": 0.21407525837421418,
|
|
"num_tokens": 2404047.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.330806875228882,
|
|
"epoch": 0.853830930166808,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004961771703669308,
|
|
"loss": 5.3214,
|
|
"mean_token_accuracy": 0.2130121484398842,
|
|
"num_tokens": 2408212.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.411522912979126,
|
|
"epoch": 0.8552445575346339,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004961582667824303,
|
|
"loss": 5.3369,
|
|
"mean_token_accuracy": 0.2133579060435295,
|
|
"num_tokens": 2412073.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.462206506729126,
|
|
"epoch": 0.8566581849024597,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004961393169772196,
|
|
"loss": 5.4226,
|
|
"mean_token_accuracy": 0.21138060837984085,
|
|
"num_tokens": 2416432.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.330096626281739,
|
|
"epoch": 0.8580718122702855,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004961203209552591,
|
|
"loss": 5.3764,
|
|
"mean_token_accuracy": 0.21394240856170654,
|
|
"num_tokens": 2420331.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.407485389709473,
|
|
"epoch": 0.8594854396381114,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004961012787205188,
|
|
"loss": 5.3311,
|
|
"mean_token_accuracy": 0.20742064863443374,
|
|
"num_tokens": 2424082.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.260201835632325,
|
|
"epoch": 0.8608990670059372,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004960821902769786,
|
|
"loss": 5.2247,
|
|
"mean_token_accuracy": 0.22140268832445145,
|
|
"num_tokens": 2427702.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.364960336685181,
|
|
"epoch": 0.8623126943737631,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004960630556286278,
|
|
"loss": 5.3153,
|
|
"mean_token_accuracy": 0.21259585320949553,
|
|
"num_tokens": 2431589.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.382180643081665,
|
|
"epoch": 0.8637263217415889,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004960438747794654,
|
|
"loss": 5.3648,
|
|
"mean_token_accuracy": 0.21003350019454955,
|
|
"num_tokens": 2435741.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.294457149505615,
|
|
"epoch": 0.8651399491094147,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004960246477334998,
|
|
"loss": 5.4555,
|
|
"mean_token_accuracy": 0.21593728959560393,
|
|
"num_tokens": 2439529.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.353067779541016,
|
|
"epoch": 0.8665535764772406,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004960053744947499,
|
|
"loss": 5.2017,
|
|
"mean_token_accuracy": 0.22102957218885422,
|
|
"num_tokens": 2443511.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.446825456619263,
|
|
"epoch": 0.8679672038450664,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004959860550672433,
|
|
"loss": 5.4239,
|
|
"mean_token_accuracy": 0.2073673814535141,
|
|
"num_tokens": 2447369.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.3776548385620115,
|
|
"epoch": 0.8693808312128923,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004959666894550177,
|
|
"loss": 5.4509,
|
|
"mean_token_accuracy": 0.21306181699037552,
|
|
"num_tokens": 2451458.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.325544977188111,
|
|
"epoch": 0.8707944585807181,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004959472776621205,
|
|
"loss": 5.2568,
|
|
"mean_token_accuracy": 0.2175483986735344,
|
|
"num_tokens": 2455681.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.348993444442749,
|
|
"epoch": 0.8722080859485439,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004959278196926085,
|
|
"loss": 5.2386,
|
|
"mean_token_accuracy": 0.22383360266685487,
|
|
"num_tokens": 2459887.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.301842069625854,
|
|
"epoch": 0.8736217133163698,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004959083155505483,
|
|
"loss": 5.2391,
|
|
"mean_token_accuracy": 0.22726237922906875,
|
|
"num_tokens": 2463837.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.395781373977661,
|
|
"epoch": 0.8750353406841956,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004958887652400163,
|
|
"loss": 5.3609,
|
|
"mean_token_accuracy": 0.20157104581594468,
|
|
"num_tokens": 2467714.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.279039478302002,
|
|
"epoch": 0.8764489680520214,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004958691687650983,
|
|
"loss": 5.3171,
|
|
"mean_token_accuracy": 0.21099349111318588,
|
|
"num_tokens": 2472016.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.311031198501587,
|
|
"epoch": 0.8778625954198473,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004958495261298898,
|
|
"loss": 5.2758,
|
|
"mean_token_accuracy": 0.22611490041017532,
|
|
"num_tokens": 2475898.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.400108575820923,
|
|
"epoch": 0.8792762227876731,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004958298373384961,
|
|
"loss": 5.3629,
|
|
"mean_token_accuracy": 0.2152768462896347,
|
|
"num_tokens": 2480144.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.273153829574585,
|
|
"epoch": 0.880689850155499,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004958101023950319,
|
|
"loss": 5.3021,
|
|
"mean_token_accuracy": 0.214484603703022,
|
|
"num_tokens": 2484127.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.267424821853638,
|
|
"epoch": 0.8821034775233249,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004957903213036217,
|
|
"loss": 5.3343,
|
|
"mean_token_accuracy": 0.22128357440233232,
|
|
"num_tokens": 2488107.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.193635702133179,
|
|
"epoch": 0.8835171048911507,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004957704940683998,
|
|
"loss": 5.2381,
|
|
"mean_token_accuracy": 0.22014840990304946,
|
|
"num_tokens": 2491981.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.306534147262573,
|
|
"epoch": 0.8849307322589766,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004957506206935097,
|
|
"loss": 5.2006,
|
|
"mean_token_accuracy": 0.23508219867944719,
|
|
"num_tokens": 2496240.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.378411054611206,
|
|
"epoch": 0.8863443596268024,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004957307011831051,
|
|
"loss": 5.1466,
|
|
"mean_token_accuracy": 0.22752396166324615,
|
|
"num_tokens": 2500175.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.322124004364014,
|
|
"epoch": 0.8877579869946283,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004957107355413489,
|
|
"loss": 5.3099,
|
|
"mean_token_accuracy": 0.2118844509124756,
|
|
"num_tokens": 2503950.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.301897192001343,
|
|
"epoch": 0.8891716143624541,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004956907237724137,
|
|
"loss": 5.309,
|
|
"mean_token_accuracy": 0.21629729270935058,
|
|
"num_tokens": 2507690.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.285606002807617,
|
|
"epoch": 0.8905852417302799,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000495670665880482,
|
|
"loss": 5.3293,
|
|
"mean_token_accuracy": 0.2038235917687416,
|
|
"num_tokens": 2511645.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.364981412887573,
|
|
"epoch": 0.8919988690981058,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004956505618697456,
|
|
"loss": 5.4133,
|
|
"mean_token_accuracy": 0.1998341739177704,
|
|
"num_tokens": 2515465.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.366808462142944,
|
|
"epoch": 0.8934124964659316,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004956304117444064,
|
|
"loss": 5.3124,
|
|
"mean_token_accuracy": 0.21686449348926545,
|
|
"num_tokens": 2519428.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.2982847690582275,
|
|
"epoch": 0.8948261238337574,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004956102155086754,
|
|
"loss": 5.22,
|
|
"mean_token_accuracy": 0.21951276957988738,
|
|
"num_tokens": 2523346.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.305537414550781,
|
|
"epoch": 0.8962397512015833,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004955899731667737,
|
|
"loss": 5.3176,
|
|
"mean_token_accuracy": 0.2135823741555214,
|
|
"num_tokens": 2527299.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.402263402938843,
|
|
"epoch": 0.8976533785694091,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004955696847229315,
|
|
"loss": 5.3584,
|
|
"mean_token_accuracy": 0.21289705336093903,
|
|
"num_tokens": 2531492.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.349612045288086,
|
|
"epoch": 0.899067005937235,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004955493501813892,
|
|
"loss": 5.2509,
|
|
"mean_token_accuracy": 0.21643438786268235,
|
|
"num_tokens": 2535641.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.3515684604644775,
|
|
"epoch": 0.9004806333050608,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004955289695463967,
|
|
"loss": 5.3845,
|
|
"mean_token_accuracy": 0.21191352903842925,
|
|
"num_tokens": 2539674.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.308618640899658,
|
|
"epoch": 0.9018942606728866,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004955085428222131,
|
|
"loss": 5.2995,
|
|
"mean_token_accuracy": 0.2106427475810051,
|
|
"num_tokens": 2543574.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.2738823890686035,
|
|
"epoch": 0.9033078880407125,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004954880700131077,
|
|
"loss": 5.3746,
|
|
"mean_token_accuracy": 0.20454230010509492,
|
|
"num_tokens": 2547381.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.357921981811524,
|
|
"epoch": 0.9047215154085383,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004954675511233592,
|
|
"loss": 5.3033,
|
|
"mean_token_accuracy": 0.2107496216893196,
|
|
"num_tokens": 2551545.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.306631946563721,
|
|
"epoch": 0.9061351427763642,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004954469861572557,
|
|
"loss": 5.2813,
|
|
"mean_token_accuracy": 0.21209642738103868,
|
|
"num_tokens": 2555336.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.221210670471192,
|
|
"epoch": 0.90754877014419,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004954263751190954,
|
|
"loss": 5.2695,
|
|
"mean_token_accuracy": 0.2048604115843773,
|
|
"num_tokens": 2559301.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.324595499038696,
|
|
"epoch": 0.9089623975120158,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004954057180131857,
|
|
"loss": 5.3095,
|
|
"mean_token_accuracy": 0.20961526334285735,
|
|
"num_tokens": 2563185.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.433643341064453,
|
|
"epoch": 0.9103760248798417,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004953850148438439,
|
|
"loss": 5.3546,
|
|
"mean_token_accuracy": 0.21090478003025054,
|
|
"num_tokens": 2567110.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.271013450622559,
|
|
"epoch": 0.9117896522476675,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004953642656153968,
|
|
"loss": 5.2673,
|
|
"mean_token_accuracy": 0.210799939930439,
|
|
"num_tokens": 2571350.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.34548921585083,
|
|
"epoch": 0.9132032796154933,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.000495343470332181,
|
|
"loss": 5.2692,
|
|
"mean_token_accuracy": 0.20986772030591966,
|
|
"num_tokens": 2575139.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.3623948097229,
|
|
"epoch": 0.9146169069833192,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004953226289985425,
|
|
"loss": 5.2447,
|
|
"mean_token_accuracy": 0.20734981894493104,
|
|
"num_tokens": 2579124.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.396227598190308,
|
|
"epoch": 0.916030534351145,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004953017416188367,
|
|
"loss": 5.2993,
|
|
"mean_token_accuracy": 0.21398847103118895,
|
|
"num_tokens": 2583130.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.31501259803772,
|
|
"epoch": 0.9174441617189709,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004952808081974294,
|
|
"loss": 5.3502,
|
|
"mean_token_accuracy": 0.20861060917377472,
|
|
"num_tokens": 2587210.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.350300121307373,
|
|
"epoch": 0.9188577890867967,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004952598287386954,
|
|
"loss": 5.2156,
|
|
"mean_token_accuracy": 0.2098054587841034,
|
|
"num_tokens": 2591372.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.358992052078247,
|
|
"epoch": 0.9202714164546225,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004952388032470191,
|
|
"loss": 5.3332,
|
|
"mean_token_accuracy": 0.21697407364845275,
|
|
"num_tokens": 2595683.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.265797424316406,
|
|
"epoch": 0.9216850438224484,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004952177317267949,
|
|
"loss": 5.1931,
|
|
"mean_token_accuracy": 0.2227411612868309,
|
|
"num_tokens": 2599472.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.270275402069092,
|
|
"epoch": 0.9230986711902742,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004951966141824265,
|
|
"loss": 5.1879,
|
|
"mean_token_accuracy": 0.21839166730642318,
|
|
"num_tokens": 2603509.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.207490396499634,
|
|
"epoch": 0.9245122985581001,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004951754506183275,
|
|
"loss": 5.2003,
|
|
"mean_token_accuracy": 0.2220302402973175,
|
|
"num_tokens": 2607191.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.3171838283538815,
|
|
"epoch": 0.9259259259259259,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004951542410389208,
|
|
"loss": 5.2894,
|
|
"mean_token_accuracy": 0.2121696650981903,
|
|
"num_tokens": 2611698.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.328981685638428,
|
|
"epoch": 0.9273395532937517,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004951329854486392,
|
|
"loss": 5.2769,
|
|
"mean_token_accuracy": 0.22239854633808137,
|
|
"num_tokens": 2615340.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.251132154464722,
|
|
"epoch": 0.9287531806615776,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004951116838519248,
|
|
"loss": 5.2566,
|
|
"mean_token_accuracy": 0.21864542216062546,
|
|
"num_tokens": 2619465.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.2473451614379885,
|
|
"epoch": 0.9301668080294034,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004950903362532296,
|
|
"loss": 5.1314,
|
|
"mean_token_accuracy": 0.23277716040611268,
|
|
"num_tokens": 2623625.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.268678426742554,
|
|
"epoch": 0.9315804353972293,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004950689426570151,
|
|
"loss": 5.1579,
|
|
"mean_token_accuracy": 0.2243214190006256,
|
|
"num_tokens": 2627368.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.337188482284546,
|
|
"epoch": 0.9329940627650551,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004950475030677525,
|
|
"loss": 5.2888,
|
|
"mean_token_accuracy": 0.22068092375993728,
|
|
"num_tokens": 2631280.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.185908603668213,
|
|
"epoch": 0.9344076901328809,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004950260174899225,
|
|
"loss": 5.2118,
|
|
"mean_token_accuracy": 0.2284255415201187,
|
|
"num_tokens": 2635420.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.215357923507691,
|
|
"epoch": 0.9358213175007069,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004950044859280154,
|
|
"loss": 5.183,
|
|
"mean_token_accuracy": 0.22036392837762833,
|
|
"num_tokens": 2639370.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.429097461700439,
|
|
"epoch": 0.9372349448685326,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004949829083865312,
|
|
"loss": 5.272,
|
|
"mean_token_accuracy": 0.220812326669693,
|
|
"num_tokens": 2643195.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.314804220199585,
|
|
"epoch": 0.9386485722363584,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004949612848699795,
|
|
"loss": 5.3496,
|
|
"mean_token_accuracy": 0.20861096680164337,
|
|
"num_tokens": 2647144.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.267690372467041,
|
|
"epoch": 0.9400621996041844,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004949396153828794,
|
|
"loss": 5.1689,
|
|
"mean_token_accuracy": 0.22159842997789383,
|
|
"num_tokens": 2651201.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.304390716552734,
|
|
"epoch": 0.9414758269720102,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004949178999297599,
|
|
"loss": 5.2784,
|
|
"mean_token_accuracy": 0.2142605498433113,
|
|
"num_tokens": 2655281.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.276190900802613,
|
|
"epoch": 0.9428894543398361,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004948961385151592,
|
|
"loss": 5.1167,
|
|
"mean_token_accuracy": 0.2349545329809189,
|
|
"num_tokens": 2659226.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.247615575790405,
|
|
"epoch": 0.9443030817076619,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004948743311436254,
|
|
"loss": 5.1881,
|
|
"mean_token_accuracy": 0.2129216194152832,
|
|
"num_tokens": 2663168.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.2464855194091795,
|
|
"epoch": 0.9457167090754877,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004948524778197159,
|
|
"loss": 5.3619,
|
|
"mean_token_accuracy": 0.20938823372125626,
|
|
"num_tokens": 2667258.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.354072713851929,
|
|
"epoch": 0.9471303364433136,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004948305785479982,
|
|
"loss": 5.2519,
|
|
"mean_token_accuracy": 0.20965614020824433,
|
|
"num_tokens": 2671149.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.336177778244019,
|
|
"epoch": 0.9485439638111394,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004948086333330488,
|
|
"loss": 5.3275,
|
|
"mean_token_accuracy": 0.2187487006187439,
|
|
"num_tokens": 2675287.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.295772695541382,
|
|
"epoch": 0.9499575911789653,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004947866421794545,
|
|
"loss": 5.1894,
|
|
"mean_token_accuracy": 0.21324728280305863,
|
|
"num_tokens": 2679148.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.281546545028687,
|
|
"epoch": 0.9513712185467911,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004947646050918109,
|
|
"loss": 5.1983,
|
|
"mean_token_accuracy": 0.219826440513134,
|
|
"num_tokens": 2683125.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.169071054458618,
|
|
"epoch": 0.9527848459146169,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000494742522074724,
|
|
"loss": 5.2337,
|
|
"mean_token_accuracy": 0.22465005964040757,
|
|
"num_tokens": 2687079.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.372869873046875,
|
|
"epoch": 0.9541984732824428,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004947203931328087,
|
|
"loss": 5.3773,
|
|
"mean_token_accuracy": 0.20540621131658554,
|
|
"num_tokens": 2691065.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.290035629272461,
|
|
"epoch": 0.9556121006502686,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004946982182706901,
|
|
"loss": 5.2072,
|
|
"mean_token_accuracy": 0.22314609438180924,
|
|
"num_tokens": 2694844.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.273801517486572,
|
|
"epoch": 0.9570257280180944,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004946759974930023,
|
|
"loss": 5.3036,
|
|
"mean_token_accuracy": 0.22169433981180192,
|
|
"num_tokens": 2698932.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.1581202983856205,
|
|
"epoch": 0.9584393553859203,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004946537308043895,
|
|
"loss": 5.1332,
|
|
"mean_token_accuracy": 0.21580587178468705,
|
|
"num_tokens": 2702810.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.3055212020874025,
|
|
"epoch": 0.9598529827537461,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004946314182095053,
|
|
"loss": 5.3593,
|
|
"mean_token_accuracy": 0.20648064315319062,
|
|
"num_tokens": 2706701.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.261645174026489,
|
|
"epoch": 0.961266610121572,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004946090597130128,
|
|
"loss": 5.3104,
|
|
"mean_token_accuracy": 0.20457247644662857,
|
|
"num_tokens": 2710533.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.338420963287353,
|
|
"epoch": 0.9626802374893978,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004945866553195848,
|
|
"loss": 5.3013,
|
|
"mean_token_accuracy": 0.2202003166079521,
|
|
"num_tokens": 2714586.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.2334400653839115,
|
|
"epoch": 0.9640938648572236,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004945642050339038,
|
|
"loss": 5.3166,
|
|
"mean_token_accuracy": 0.21974583864212036,
|
|
"num_tokens": 2718357.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.491139984130859,
|
|
"epoch": 0.9655074922250495,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004945417088606617,
|
|
"loss": 5.5482,
|
|
"mean_token_accuracy": 0.1914077877998352,
|
|
"num_tokens": 2722811.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.349028253555298,
|
|
"epoch": 0.9669211195928753,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004945191668045601,
|
|
"loss": 5.2235,
|
|
"mean_token_accuracy": 0.22175128310918807,
|
|
"num_tokens": 2726815.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.342743015289306,
|
|
"epoch": 0.9683347469607012,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004944965788703101,
|
|
"loss": 5.3457,
|
|
"mean_token_accuracy": 0.19939375072717666,
|
|
"num_tokens": 2730683.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.289308977127075,
|
|
"epoch": 0.969748374328527,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004944739450626323,
|
|
"loss": 5.1528,
|
|
"mean_token_accuracy": 0.22440018355846406,
|
|
"num_tokens": 2734950.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.2037581443786625,
|
|
"epoch": 0.9711620016963528,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004944512653862572,
|
|
"loss": 5.1708,
|
|
"mean_token_accuracy": 0.23219923079013824,
|
|
"num_tokens": 2738894.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.291185712814331,
|
|
"epoch": 0.9725756290641787,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004944285398459247,
|
|
"loss": 5.2892,
|
|
"mean_token_accuracy": 0.2185189440846443,
|
|
"num_tokens": 2742848.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.288788318634033,
|
|
"epoch": 0.9739892564320045,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004944057684463844,
|
|
"loss": 5.2676,
|
|
"mean_token_accuracy": 0.21666843295097352,
|
|
"num_tokens": 2746809.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.231118106842041,
|
|
"epoch": 0.9754028837998303,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004943829511923952,
|
|
"loss": 5.2465,
|
|
"mean_token_accuracy": 0.21768148243427277,
|
|
"num_tokens": 2750450.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.1876380920410154,
|
|
"epoch": 0.9768165111676562,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004943600880887258,
|
|
"loss": 5.2598,
|
|
"mean_token_accuracy": 0.21461962908506393,
|
|
"num_tokens": 2754572.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.342851734161377,
|
|
"epoch": 0.978230138535482,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004943371791401544,
|
|
"loss": 5.3819,
|
|
"mean_token_accuracy": 0.20070458799600602,
|
|
"num_tokens": 2758614.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.406234884262085,
|
|
"epoch": 0.9796437659033079,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000494314224351469,
|
|
"loss": 5.2735,
|
|
"mean_token_accuracy": 0.21284016817808152,
|
|
"num_tokens": 2763031.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.389446210861206,
|
|
"epoch": 0.9810573932711337,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004942912237274668,
|
|
"loss": 5.4384,
|
|
"mean_token_accuracy": 0.20352917462587355,
|
|
"num_tokens": 2766953.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.326390171051026,
|
|
"epoch": 0.9824710206389595,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000494268177272955,
|
|
"loss": 5.2986,
|
|
"mean_token_accuracy": 0.21028649359941481,
|
|
"num_tokens": 2770822.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.322681999206543,
|
|
"epoch": 0.9838846480067854,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00049424508499275,
|
|
"loss": 5.2714,
|
|
"mean_token_accuracy": 0.2100103810429573,
|
|
"num_tokens": 2774599.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.227937269210815,
|
|
"epoch": 0.9852982753746112,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000494221946891678,
|
|
"loss": 5.2126,
|
|
"mean_token_accuracy": 0.21849976629018783,
|
|
"num_tokens": 2778508.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.324561738967896,
|
|
"epoch": 0.9867119027424371,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004941987629745747,
|
|
"loss": 5.3229,
|
|
"mean_token_accuracy": 0.20360010415315627,
|
|
"num_tokens": 2782535.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.3999138355255125,
|
|
"epoch": 0.9881255301102629,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004941755332462856,
|
|
"loss": 5.3643,
|
|
"mean_token_accuracy": 0.21058977842330934,
|
|
"num_tokens": 2786339.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.385817527770996,
|
|
"epoch": 0.9895391574780887,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004941522577116652,
|
|
"loss": 5.3058,
|
|
"mean_token_accuracy": 0.2122144103050232,
|
|
"num_tokens": 2790466.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.302184438705444,
|
|
"epoch": 0.9909527848459146,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004941289363755782,
|
|
"loss": 5.2854,
|
|
"mean_token_accuracy": 0.2152460917830467,
|
|
"num_tokens": 2794267.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.341381645202636,
|
|
"epoch": 0.9923664122137404,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004941055692428987,
|
|
"loss": 5.3159,
|
|
"mean_token_accuracy": 0.20648507475852967,
|
|
"num_tokens": 2798394.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.291337776184082,
|
|
"epoch": 0.9937800395815664,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00049408215631851,
|
|
"loss": 5.2363,
|
|
"mean_token_accuracy": 0.21810831874608994,
|
|
"num_tokens": 2802429.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.31103663444519,
|
|
"epoch": 0.9951936669493922,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004940586976073054,
|
|
"loss": 5.2944,
|
|
"mean_token_accuracy": 0.20695877224206924,
|
|
"num_tokens": 2806303.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.2627636909484865,
|
|
"epoch": 0.996607294317218,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004940351931141875,
|
|
"loss": 5.2031,
|
|
"mean_token_accuracy": 0.22894581407308578,
|
|
"num_tokens": 2810123.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.210841369628906,
|
|
"epoch": 0.9980209216850439,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004940116428440688,
|
|
"loss": 5.2844,
|
|
"mean_token_accuracy": 0.2093127340078354,
|
|
"num_tokens": 2813991.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.245090055465698,
|
|
"epoch": 0.9994345490528697,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004939880468018712,
|
|
"loss": 5.2436,
|
|
"mean_token_accuracy": 0.21966125816106796,
|
|
"num_tokens": 2818203.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.314150047302246,
|
|
"epoch": 1.0008481764206956,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004939644049925259,
|
|
"loss": 5.1244,
|
|
"mean_token_accuracy": 0.21864545345306396,
|
|
"num_tokens": 2822334.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.473085403442383,
|
|
"epoch": 1.0022618037885214,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000493940717420974,
|
|
"loss": 5.2669,
|
|
"mean_token_accuracy": 0.21420305520296096,
|
|
"num_tokens": 2826277.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.315029287338257,
|
|
"epoch": 1.0036754311563472,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000493916984092166,
|
|
"loss": 5.0561,
|
|
"mean_token_accuracy": 0.22078744620084761,
|
|
"num_tokens": 2830093.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.174912023544311,
|
|
"epoch": 1.005089058524173,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004938932050110621,
|
|
"loss": 5.1113,
|
|
"mean_token_accuracy": 0.23046851456165313,
|
|
"num_tokens": 2833961.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.188733339309692,
|
|
"epoch": 1.0065026858919988,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000493869380182632,
|
|
"loss": 5.0074,
|
|
"mean_token_accuracy": 0.23168223947286606,
|
|
"num_tokens": 2837923.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.145092868804932,
|
|
"epoch": 1.0079163132598248,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004938455096118548,
|
|
"loss": 4.8874,
|
|
"mean_token_accuracy": 0.23700931966304778,
|
|
"num_tokens": 2841820.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.0993650436401365,
|
|
"epoch": 1.0093299406276506,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004938215933037195,
|
|
"loss": 5.1002,
|
|
"mean_token_accuracy": 0.22340761125087738,
|
|
"num_tokens": 2845901.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.212271213531494,
|
|
"epoch": 1.0107435679954764,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004937976312632242,
|
|
"loss": 5.0403,
|
|
"mean_token_accuracy": 0.22256288677453995,
|
|
"num_tokens": 2849841.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.203208684921265,
|
|
"epoch": 1.0121571953633022,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000493773623495377,
|
|
"loss": 4.9169,
|
|
"mean_token_accuracy": 0.2299910753965378,
|
|
"num_tokens": 2853601.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.274860095977783,
|
|
"epoch": 1.013570822731128,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004937495700051955,
|
|
"loss": 5.0008,
|
|
"mean_token_accuracy": 0.2231670841574669,
|
|
"num_tokens": 2857967.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.164321994781494,
|
|
"epoch": 1.014984450098954,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004937254707977065,
|
|
"loss": 5.0466,
|
|
"mean_token_accuracy": 0.23029865324497223,
|
|
"num_tokens": 2862009.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.130255174636841,
|
|
"epoch": 1.0163980774667798,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004937013258779465,
|
|
"loss": 4.9976,
|
|
"mean_token_accuracy": 0.22396016120910645,
|
|
"num_tokens": 2866078.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.2307158470153805,
|
|
"epoch": 1.0178117048346056,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000493677135250962,
|
|
"loss": 5.0401,
|
|
"mean_token_accuracy": 0.22082847207784653,
|
|
"num_tokens": 2870041.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.201313018798828,
|
|
"epoch": 1.0192253322024314,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004936528989218084,
|
|
"loss": 4.9773,
|
|
"mean_token_accuracy": 0.2307869702577591,
|
|
"num_tokens": 2873901.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.157670402526856,
|
|
"epoch": 1.0206389595702572,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004936286168955511,
|
|
"loss": 5.019,
|
|
"mean_token_accuracy": 0.22578337490558625,
|
|
"num_tokens": 2877916.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.15597505569458,
|
|
"epoch": 1.0220525869380832,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004936042891772647,
|
|
"loss": 5.0217,
|
|
"mean_token_accuracy": 0.22585746347904206,
|
|
"num_tokens": 2882207.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.224420547485352,
|
|
"epoch": 1.023466214305909,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004935799157720338,
|
|
"loss": 5.0045,
|
|
"mean_token_accuracy": 0.23484410494565963,
|
|
"num_tokens": 2886211.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.239304113388061,
|
|
"epoch": 1.0248798416737348,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004935554966849522,
|
|
"loss": 5.2128,
|
|
"mean_token_accuracy": 0.22152378410100937,
|
|
"num_tokens": 2890167.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.135939455032348,
|
|
"epoch": 1.0262934690415606,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004935310319211231,
|
|
"loss": 4.9612,
|
|
"mean_token_accuracy": 0.22932442277669907,
|
|
"num_tokens": 2894214.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.174535322189331,
|
|
"epoch": 1.0277070964093864,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004935065214856599,
|
|
"loss": 5.103,
|
|
"mean_token_accuracy": 0.2289595425128937,
|
|
"num_tokens": 2898423.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.129387426376343,
|
|
"epoch": 1.0291207237772124,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004934819653836848,
|
|
"loss": 5.0081,
|
|
"mean_token_accuracy": 0.23242045640945436,
|
|
"num_tokens": 2902728.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.096001625061035,
|
|
"epoch": 1.0305343511450382,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004934573636203299,
|
|
"loss": 5.016,
|
|
"mean_token_accuracy": 0.22506662905216218,
|
|
"num_tokens": 2906687.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.163566017150879,
|
|
"epoch": 1.031947978512864,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004934327162007371,
|
|
"loss": 5.0078,
|
|
"mean_token_accuracy": 0.21606790870428086,
|
|
"num_tokens": 2910708.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.2279486656188965,
|
|
"epoch": 1.0333616058806898,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004934080231300573,
|
|
"loss": 5.0734,
|
|
"mean_token_accuracy": 0.22245530635118485,
|
|
"num_tokens": 2914416.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.254195880889893,
|
|
"epoch": 1.0347752332485156,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004933832844134511,
|
|
"loss": 5.0802,
|
|
"mean_token_accuracy": 0.230830354988575,
|
|
"num_tokens": 2918738.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.142400598526001,
|
|
"epoch": 1.0361888606163416,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004933585000560891,
|
|
"loss": 5.0644,
|
|
"mean_token_accuracy": 0.23004631847143173,
|
|
"num_tokens": 2922575.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.303077745437622,
|
|
"epoch": 1.0376024879841674,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004933336700631508,
|
|
"loss": 5.027,
|
|
"mean_token_accuracy": 0.22817589342594147,
|
|
"num_tokens": 2926707.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.243477392196655,
|
|
"epoch": 1.0390161153519932,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004933087944398256,
|
|
"loss": 5.0686,
|
|
"mean_token_accuracy": 0.22908651530742646,
|
|
"num_tokens": 2930785.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.0108260154724125,
|
|
"epoch": 1.040429742719819,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004932838731913124,
|
|
"loss": 4.8579,
|
|
"mean_token_accuracy": 0.2431721419095993,
|
|
"num_tokens": 2934893.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.130848550796509,
|
|
"epoch": 1.0418433700876448,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004932589063228194,
|
|
"loss": 5.0961,
|
|
"mean_token_accuracy": 0.21659439504146577,
|
|
"num_tokens": 2939034.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.22723536491394,
|
|
"epoch": 1.0432569974554706,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004932338938395647,
|
|
"loss": 4.9822,
|
|
"mean_token_accuracy": 0.22529290467500687,
|
|
"num_tokens": 2943089.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.192731094360352,
|
|
"epoch": 1.0446706248232966,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004932088357467757,
|
|
"loss": 5.0573,
|
|
"mean_token_accuracy": 0.22995467334985734,
|
|
"num_tokens": 2947128.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.1997706413269045,
|
|
"epoch": 1.0460842521911224,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004931837320496894,
|
|
"loss": 5.1278,
|
|
"mean_token_accuracy": 0.21710651665925979,
|
|
"num_tokens": 2951295.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.152848958969116,
|
|
"epoch": 1.0474978795589482,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004931585827535524,
|
|
"loss": 5.1319,
|
|
"mean_token_accuracy": 0.21446435153484344,
|
|
"num_tokens": 2955522.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.167764568328858,
|
|
"epoch": 1.048911506926774,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004931333878636205,
|
|
"loss": 5.0052,
|
|
"mean_token_accuracy": 0.22012700587511064,
|
|
"num_tokens": 2959410.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.125832176208496,
|
|
"epoch": 1.0503251342945998,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004931081473851595,
|
|
"loss": 5.0386,
|
|
"mean_token_accuracy": 0.2211052656173706,
|
|
"num_tokens": 2963411.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.104986000061035,
|
|
"epoch": 1.0517387616624259,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004930828613234446,
|
|
"loss": 5.0096,
|
|
"mean_token_accuracy": 0.2301958531141281,
|
|
"num_tokens": 2967623.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.2205822467803955,
|
|
"epoch": 1.0531523890302517,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00049305752968376,
|
|
"loss": 4.9666,
|
|
"mean_token_accuracy": 0.23770354986190795,
|
|
"num_tokens": 2971708.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.120276355743409,
|
|
"epoch": 1.0545660163980775,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004930321524714002,
|
|
"loss": 5.0071,
|
|
"mean_token_accuracy": 0.24287337809801102,
|
|
"num_tokens": 2975491.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.174483919143677,
|
|
"epoch": 1.0559796437659033,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004930067296916689,
|
|
"loss": 5.0891,
|
|
"mean_token_accuracy": 0.22007248550653458,
|
|
"num_tokens": 2979368.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.120765161514282,
|
|
"epoch": 1.057393271133729,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004929812613498791,
|
|
"loss": 5.015,
|
|
"mean_token_accuracy": 0.23333019763231277,
|
|
"num_tokens": 2983939.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.077562236785889,
|
|
"epoch": 1.058806898501555,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004929557474513537,
|
|
"loss": 4.9594,
|
|
"mean_token_accuracy": 0.22762384116649628,
|
|
"num_tokens": 2987895.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.128376579284668,
|
|
"epoch": 1.0602205258693809,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004929301880014249,
|
|
"loss": 5.0219,
|
|
"mean_token_accuracy": 0.22629786878824235,
|
|
"num_tokens": 2992017.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.119693613052368,
|
|
"epoch": 1.0616341532372067,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004929045830054344,
|
|
"loss": 5.0029,
|
|
"mean_token_accuracy": 0.2173638328909874,
|
|
"num_tokens": 2995957.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.0717541694641115,
|
|
"epoch": 1.0630477806050325,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004928789324687335,
|
|
"loss": 4.8264,
|
|
"mean_token_accuracy": 0.2484455719590187,
|
|
"num_tokens": 2999816.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.144631576538086,
|
|
"epoch": 1.0644614079728583,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004928532363966831,
|
|
"loss": 5.0167,
|
|
"mean_token_accuracy": 0.22890725880861282,
|
|
"num_tokens": 3003801.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.1410760402679445,
|
|
"epoch": 1.0658750353406843,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004928274947946534,
|
|
"loss": 5.117,
|
|
"mean_token_accuracy": 0.22192999273538588,
|
|
"num_tokens": 3007710.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.215381050109864,
|
|
"epoch": 1.06728866270851,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004928017076680244,
|
|
"loss": 5.0468,
|
|
"mean_token_accuracy": 0.2226067766547203,
|
|
"num_tokens": 3011573.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.158493137359619,
|
|
"epoch": 1.0687022900763359,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004927758750221852,
|
|
"loss": 5.0654,
|
|
"mean_token_accuracy": 0.2298080265522003,
|
|
"num_tokens": 3015832.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.057914161682129,
|
|
"epoch": 1.0701159174441617,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004927499968625349,
|
|
"loss": 4.9694,
|
|
"mean_token_accuracy": 0.23105335980653763,
|
|
"num_tokens": 3019878.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.060720205307007,
|
|
"epoch": 1.0715295448119875,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004927240731944818,
|
|
"loss": 4.8629,
|
|
"mean_token_accuracy": 0.2461293011903763,
|
|
"num_tokens": 3023725.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.2345055103302,
|
|
"epoch": 1.0729431721798135,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004926981040234437,
|
|
"loss": 5.15,
|
|
"mean_token_accuracy": 0.21904803365468978,
|
|
"num_tokens": 3028055.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.071320819854736,
|
|
"epoch": 1.0743567995476393,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004926720893548482,
|
|
"loss": 4.9807,
|
|
"mean_token_accuracy": 0.2409299835562706,
|
|
"num_tokens": 3032214.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.14586877822876,
|
|
"epoch": 1.075770426915465,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004926460291941319,
|
|
"loss": 4.979,
|
|
"mean_token_accuracy": 0.23053666800260544,
|
|
"num_tokens": 3036507.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.149288845062256,
|
|
"epoch": 1.077184054283291,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004926199235467414,
|
|
"loss": 5.1175,
|
|
"mean_token_accuracy": 0.22399256825447084,
|
|
"num_tokens": 3040720.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.175496530532837,
|
|
"epoch": 1.0785976816511167,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004925937724181326,
|
|
"loss": 5.0237,
|
|
"mean_token_accuracy": 0.22914376705884934,
|
|
"num_tokens": 3044826.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.172716474533081,
|
|
"epoch": 1.0800113090189427,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004925675758137711,
|
|
"loss": 5.0337,
|
|
"mean_token_accuracy": 0.22899835407733918,
|
|
"num_tokens": 3048840.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.114548110961914,
|
|
"epoch": 1.0814249363867685,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004925413337391315,
|
|
"loss": 4.9248,
|
|
"mean_token_accuracy": 0.23229692578315736,
|
|
"num_tokens": 3052716.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.117950773239135,
|
|
"epoch": 1.0828385637545943,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004925150461996984,
|
|
"loss": 5.0077,
|
|
"mean_token_accuracy": 0.22668319940567017,
|
|
"num_tokens": 3056662.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.154039525985718,
|
|
"epoch": 1.08425219112242,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004924887132009656,
|
|
"loss": 5.0063,
|
|
"mean_token_accuracy": 0.2286270946264267,
|
|
"num_tokens": 3060638.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.152283954620361,
|
|
"epoch": 1.085665818490246,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004924623347484369,
|
|
"loss": 5.0265,
|
|
"mean_token_accuracy": 0.22818633615970613,
|
|
"num_tokens": 3064438.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.098384666442871,
|
|
"epoch": 1.087079445858072,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004924359108476248,
|
|
"loss": 4.9041,
|
|
"mean_token_accuracy": 0.23847402185201644,
|
|
"num_tokens": 3068099.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.026648664474488,
|
|
"epoch": 1.0884930732258977,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004924094415040519,
|
|
"loss": 4.8304,
|
|
"mean_token_accuracy": 0.23882860988378524,
|
|
"num_tokens": 3072039.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 4.981844711303711,
|
|
"epoch": 1.0899067005937235,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004923829267232503,
|
|
"loss": 5.0414,
|
|
"mean_token_accuracy": 0.23245490938425065,
|
|
"num_tokens": 3076110.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.15345573425293,
|
|
"epoch": 1.0913203279615493,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004923563665107612,
|
|
"loss": 4.9482,
|
|
"mean_token_accuracy": 0.23124676197767258,
|
|
"num_tokens": 3080514.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.18023567199707,
|
|
"epoch": 1.0927339553293751,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004923297608721355,
|
|
"loss": 5.0185,
|
|
"mean_token_accuracy": 0.22580058872699738,
|
|
"num_tokens": 3084537.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.120248746871948,
|
|
"epoch": 1.094147582697201,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004923031098129338,
|
|
"loss": 5.0458,
|
|
"mean_token_accuracy": 0.23244117200374603,
|
|
"num_tokens": 3088107.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.083855247497558,
|
|
"epoch": 1.095561210065027,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004922764133387259,
|
|
"loss": 5.0165,
|
|
"mean_token_accuracy": 0.22794615626335143,
|
|
"num_tokens": 3091790.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.110710859298706,
|
|
"epoch": 1.0969748374328527,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004922496714550912,
|
|
"loss": 4.897,
|
|
"mean_token_accuracy": 0.23963625431060792,
|
|
"num_tokens": 3095757.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.155925369262695,
|
|
"epoch": 1.0983884648006785,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004922228841676186,
|
|
"loss": 5.0613,
|
|
"mean_token_accuracy": 0.22538823187351226,
|
|
"num_tokens": 3100172.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.141041469573975,
|
|
"epoch": 1.0998020921685043,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004921960514819065,
|
|
"loss": 5.1735,
|
|
"mean_token_accuracy": 0.20952586680650712,
|
|
"num_tokens": 3103857.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.1645832538604735,
|
|
"epoch": 1.1012157195363301,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004921691734035627,
|
|
"loss": 5.0788,
|
|
"mean_token_accuracy": 0.22629565447568895,
|
|
"num_tokens": 3107817.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.159869527816772,
|
|
"epoch": 1.1026293469041561,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004921422499382046,
|
|
"loss": 4.9976,
|
|
"mean_token_accuracy": 0.23880043029785156,
|
|
"num_tokens": 3111655.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.2506067752838135,
|
|
"epoch": 1.104042974271982,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000492115281091459,
|
|
"loss": 5.1255,
|
|
"mean_token_accuracy": 0.2335188254714012,
|
|
"num_tokens": 3115959.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.076710176467896,
|
|
"epoch": 1.1054566016398077,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004920882668689623,
|
|
"loss": 4.858,
|
|
"mean_token_accuracy": 0.23754922300577164,
|
|
"num_tokens": 3119930.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.011963653564453,
|
|
"epoch": 1.1068702290076335,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004920612072763603,
|
|
"loss": 5.0115,
|
|
"mean_token_accuracy": 0.2260516107082367,
|
|
"num_tokens": 3123806.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.127112245559692,
|
|
"epoch": 1.1082838563754593,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004920341023193082,
|
|
"loss": 4.9968,
|
|
"mean_token_accuracy": 0.2268049016594887,
|
|
"num_tokens": 3127942.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.10136833190918,
|
|
"epoch": 1.1096974837432854,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004920069520034708,
|
|
"loss": 5.052,
|
|
"mean_token_accuracy": 0.22814164012670518,
|
|
"num_tokens": 3132014.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.146014165878296,
|
|
"epoch": 1.1111111111111112,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004919797563345224,
|
|
"loss": 5.0784,
|
|
"mean_token_accuracy": 0.23491773158311843,
|
|
"num_tokens": 3135854.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.228779458999634,
|
|
"epoch": 1.112524738478937,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004919525153181466,
|
|
"loss": 5.1213,
|
|
"mean_token_accuracy": 0.2265205755829811,
|
|
"num_tokens": 3139804.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.204964113235474,
|
|
"epoch": 1.1139383658467628,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004919252289600367,
|
|
"loss": 5.1843,
|
|
"mean_token_accuracy": 0.22286111414432525,
|
|
"num_tokens": 3143473.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.157012414932251,
|
|
"epoch": 1.1153519932145886,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004918978972658954,
|
|
"loss": 5.0037,
|
|
"mean_token_accuracy": 0.22547006607055664,
|
|
"num_tokens": 3147626.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.1330376148223875,
|
|
"epoch": 1.1167656205824146,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004918705202414348,
|
|
"loss": 5.08,
|
|
"mean_token_accuracy": 0.21345378905534745,
|
|
"num_tokens": 3151758.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.0832582950592045,
|
|
"epoch": 1.1181792479502404,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004918430978923768,
|
|
"loss": 4.969,
|
|
"mean_token_accuracy": 0.23154980689287186,
|
|
"num_tokens": 3155490.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.252908706665039,
|
|
"epoch": 1.1195928753180662,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000491815630224452,
|
|
"loss": 5.144,
|
|
"mean_token_accuracy": 0.22490071952342988,
|
|
"num_tokens": 3159385.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.264437961578369,
|
|
"epoch": 1.121006502685892,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004917881172434012,
|
|
"loss": 5.1495,
|
|
"mean_token_accuracy": 0.21658464819192885,
|
|
"num_tokens": 3163198.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.203642702102661,
|
|
"epoch": 1.1224201300537178,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004917605589549746,
|
|
"loss": 5.0078,
|
|
"mean_token_accuracy": 0.22007464617490768,
|
|
"num_tokens": 3167140.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 4.9864466190338135,
|
|
"epoch": 1.1238337574215436,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004917329553649315,
|
|
"loss": 4.8914,
|
|
"mean_token_accuracy": 0.24323611706495285,
|
|
"num_tokens": 3170782.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.085910511016846,
|
|
"epoch": 1.1252473847893696,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004917053064790411,
|
|
"loss": 4.9655,
|
|
"mean_token_accuracy": 0.23082458823919297,
|
|
"num_tokens": 3174678.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.158240079879761,
|
|
"epoch": 1.1266610121571954,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004916776123030815,
|
|
"loss": 5.0579,
|
|
"mean_token_accuracy": 0.22890633344650269,
|
|
"num_tokens": 3178569.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.148614931106567,
|
|
"epoch": 1.1280746395250212,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004916498728428411,
|
|
"loss": 4.9417,
|
|
"mean_token_accuracy": 0.23426750153303147,
|
|
"num_tokens": 3182328.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.047394418716431,
|
|
"epoch": 1.129488266892847,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004916220881041168,
|
|
"loss": 4.9928,
|
|
"mean_token_accuracy": 0.2306484341621399,
|
|
"num_tokens": 3186477.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.06369194984436,
|
|
"epoch": 1.1309018942606728,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0004915942580927156,
|
|
"loss": 5.0541,
|
|
"mean_token_accuracy": 0.22101714909076692,
|
|
"num_tokens": 3190517.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 1.1309018942606728,
|
|
"eval_entropy": 4.887318427304187,
|
|
"eval_loss": 5.3592329025268555,
|
|
"eval_mean_token_accuracy": 0.2222705981426844,
|
|
"eval_num_tokens": 3190517.0,
|
|
"eval_runtime": 4.8277,
|
|
"eval_samples_per_second": 2337.345,
|
|
"eval_steps_per_second": 292.272,
|
|
"step": 4000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 35360,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 1000,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 572800841256960.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|