3035 lines
81 KiB
JSON
3035 lines
81 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.09808409076047865,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.691962432861327,
|
|
"epoch": 0.0003269469692015955,
|
|
"grad_norm": 12.5625,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.8118,
|
|
"mean_token_accuracy": 0.0002268856973387301,
|
|
"num_tokens": 10275.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.691912364959716,
|
|
"epoch": 0.000653893938403191,
|
|
"grad_norm": 12.0,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7548,
|
|
"mean_token_accuracy": 0.0008593271079007537,
|
|
"num_tokens": 20293.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.690659618377685,
|
|
"epoch": 0.0009808409076047864,
|
|
"grad_norm": 9.375,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.5796,
|
|
"mean_token_accuracy": 0.027759611373767257,
|
|
"num_tokens": 30769.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.671999549865722,
|
|
"epoch": 0.001307787876806382,
|
|
"grad_norm": 5.5625,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.333,
|
|
"mean_token_accuracy": 0.04241157788783312,
|
|
"num_tokens": 41351.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.554566478729248,
|
|
"epoch": 0.0016347348460079775,
|
|
"grad_norm": 4.625,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.128,
|
|
"mean_token_accuracy": 0.04083434790372849,
|
|
"num_tokens": 51137.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.48172836303711,
|
|
"epoch": 0.001961681815209573,
|
|
"grad_norm": 2.84375,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.0286,
|
|
"mean_token_accuracy": 0.04095044266432524,
|
|
"num_tokens": 61184.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.52005319595337,
|
|
"epoch": 0.0022886287844111684,
|
|
"grad_norm": 2.890625,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.9428,
|
|
"mean_token_accuracy": 0.04355649538338184,
|
|
"num_tokens": 71499.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.493383026123047,
|
|
"epoch": 0.002615575753612764,
|
|
"grad_norm": 2.390625,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.9404,
|
|
"mean_token_accuracy": 0.042642039619386195,
|
|
"num_tokens": 81761.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.462896347045898,
|
|
"epoch": 0.0029425227228143595,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.9134,
|
|
"mean_token_accuracy": 0.044699825905263425,
|
|
"num_tokens": 93149.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.480470275878906,
|
|
"epoch": 0.003269469692015955,
|
|
"grad_norm": 2.40625,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.8578,
|
|
"mean_token_accuracy": 0.04424864072352648,
|
|
"num_tokens": 103201.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.50320463180542,
|
|
"epoch": 0.0035964166612175506,
|
|
"grad_norm": 2.21875,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.8204,
|
|
"mean_token_accuracy": 0.05240613967180252,
|
|
"num_tokens": 114340.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.402948760986328,
|
|
"epoch": 0.003923363630419146,
|
|
"grad_norm": 2.078125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.7316,
|
|
"mean_token_accuracy": 0.05212643444538116,
|
|
"num_tokens": 125783.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.340407657623292,
|
|
"epoch": 0.004250310599620741,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.6209,
|
|
"mean_token_accuracy": 0.05632065013051033,
|
|
"num_tokens": 135351.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.26957654953003,
|
|
"epoch": 0.004577257568822337,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.6186,
|
|
"mean_token_accuracy": 0.056864064931869504,
|
|
"num_tokens": 146942.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.386287784576416,
|
|
"epoch": 0.004904204538023932,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.5645,
|
|
"mean_token_accuracy": 0.056648611277341845,
|
|
"num_tokens": 157676.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.315990447998047,
|
|
"epoch": 0.005231151507225528,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.4324,
|
|
"mean_token_accuracy": 0.056685902178287506,
|
|
"num_tokens": 168608.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.201674747467042,
|
|
"epoch": 0.0055580984764271235,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.3533,
|
|
"mean_token_accuracy": 0.05901941917836666,
|
|
"num_tokens": 179268.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.28765230178833,
|
|
"epoch": 0.005885045445628719,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.3399,
|
|
"mean_token_accuracy": 0.05402175635099411,
|
|
"num_tokens": 189279.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.194643878936768,
|
|
"epoch": 0.006211992414830315,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.196,
|
|
"mean_token_accuracy": 0.06014312915503979,
|
|
"num_tokens": 198791.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.213055515289307,
|
|
"epoch": 0.00653893938403191,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.1506,
|
|
"mean_token_accuracy": 0.05811347998678684,
|
|
"num_tokens": 208262.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.124658584594727,
|
|
"epoch": 0.006865886353233506,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.0631,
|
|
"mean_token_accuracy": 0.05890461802482605,
|
|
"num_tokens": 219359.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.020441341400147,
|
|
"epoch": 0.007192833322435101,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.9009,
|
|
"mean_token_accuracy": 0.061437368765473364,
|
|
"num_tokens": 229582.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.00902338027954,
|
|
"epoch": 0.007519780291636697,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.9367,
|
|
"mean_token_accuracy": 0.057738688215613364,
|
|
"num_tokens": 239856.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.907579135894775,
|
|
"epoch": 0.007846727260838291,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.7953,
|
|
"mean_token_accuracy": 0.05762846954166889,
|
|
"num_tokens": 249620.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.752235412597656,
|
|
"epoch": 0.008173674230039887,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.7378,
|
|
"mean_token_accuracy": 0.05578424893319607,
|
|
"num_tokens": 260081.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.781640815734864,
|
|
"epoch": 0.008500621199241483,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.5809,
|
|
"mean_token_accuracy": 0.06519303396344185,
|
|
"num_tokens": 269889.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.525392723083495,
|
|
"epoch": 0.008827568168443078,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.6158,
|
|
"mean_token_accuracy": 0.06163802072405815,
|
|
"num_tokens": 279884.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.505074214935302,
|
|
"epoch": 0.009154515137644674,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.5592,
|
|
"mean_token_accuracy": 0.05903047285974026,
|
|
"num_tokens": 290532.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.370058345794678,
|
|
"epoch": 0.00948146210684627,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.4303,
|
|
"mean_token_accuracy": 0.058588751778006556,
|
|
"num_tokens": 300384.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 9.222196769714355,
|
|
"epoch": 0.009808409076047865,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.3183,
|
|
"mean_token_accuracy": 0.06703893281519413,
|
|
"num_tokens": 310486.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.067441463470459,
|
|
"epoch": 0.01013535604524946,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.369,
|
|
"mean_token_accuracy": 0.06330439783632755,
|
|
"num_tokens": 320405.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.02968339920044,
|
|
"epoch": 0.010462303014451056,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.3285,
|
|
"mean_token_accuracy": 0.06537889093160629,
|
|
"num_tokens": 331041.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.945148849487305,
|
|
"epoch": 0.010789249983652651,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 8.3496,
|
|
"mean_token_accuracy": 0.06229307875037193,
|
|
"num_tokens": 341014.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.904346561431884,
|
|
"epoch": 0.011116196952854247,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 8.299,
|
|
"mean_token_accuracy": 0.06237541511654854,
|
|
"num_tokens": 352840.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.780626392364502,
|
|
"epoch": 0.011443143922055843,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 8.2693,
|
|
"mean_token_accuracy": 0.06663689725100994,
|
|
"num_tokens": 362829.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.713014221191406,
|
|
"epoch": 0.011770090891257438,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 8.1825,
|
|
"mean_token_accuracy": 0.06633488647639751,
|
|
"num_tokens": 373330.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.689865589141846,
|
|
"epoch": 0.012097037860459034,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 8.1874,
|
|
"mean_token_accuracy": 0.06622433178126812,
|
|
"num_tokens": 383488.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.68079071044922,
|
|
"epoch": 0.01242398482966063,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 8.1841,
|
|
"mean_token_accuracy": 0.06749865673482418,
|
|
"num_tokens": 393936.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.624857902526855,
|
|
"epoch": 0.012750931798862225,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 8.2894,
|
|
"mean_token_accuracy": 0.06525838598608971,
|
|
"num_tokens": 404013.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.582938766479492,
|
|
"epoch": 0.01307787876806382,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 8.1825,
|
|
"mean_token_accuracy": 0.06989333964884281,
|
|
"num_tokens": 414009.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.596421909332275,
|
|
"epoch": 0.013404825737265416,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000102,
|
|
"loss": 8.2087,
|
|
"mean_token_accuracy": 0.06921766847372054,
|
|
"num_tokens": 424011.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.601021480560302,
|
|
"epoch": 0.013731772706467011,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 8.1777,
|
|
"mean_token_accuracy": 0.06699340268969536,
|
|
"num_tokens": 434985.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.557740116119385,
|
|
"epoch": 0.014058719675668607,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000107,
|
|
"loss": 8.1858,
|
|
"mean_token_accuracy": 0.06825171709060669,
|
|
"num_tokens": 446276.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.430389785766602,
|
|
"epoch": 0.014385666644870202,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 8.0834,
|
|
"mean_token_accuracy": 0.07472717836499214,
|
|
"num_tokens": 455307.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.491849899291992,
|
|
"epoch": 0.014712613614071798,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000112,
|
|
"loss": 8.1734,
|
|
"mean_token_accuracy": 0.07013615183532237,
|
|
"num_tokens": 465253.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.517255306243896,
|
|
"epoch": 0.015039560583273394,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 8.1761,
|
|
"mean_token_accuracy": 0.06887588053941726,
|
|
"num_tokens": 475588.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.475957870483398,
|
|
"epoch": 0.015366507552474989,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 8.0487,
|
|
"mean_token_accuracy": 0.07346096634864807,
|
|
"num_tokens": 485356.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.470558166503906,
|
|
"epoch": 0.015693454521676583,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 8.0824,
|
|
"mean_token_accuracy": 0.0730382464826107,
|
|
"num_tokens": 494798.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.372940826416016,
|
|
"epoch": 0.01602040149087818,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000122,
|
|
"loss": 8.1423,
|
|
"mean_token_accuracy": 0.07283038049936294,
|
|
"num_tokens": 504907.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.429611682891846,
|
|
"epoch": 0.016347348460079774,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 8.153,
|
|
"mean_token_accuracy": 0.07119653932750225,
|
|
"num_tokens": 515219.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.42165298461914,
|
|
"epoch": 0.01667429542928137,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000127,
|
|
"loss": 8.0908,
|
|
"mean_token_accuracy": 0.07306713648140431,
|
|
"num_tokens": 524467.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.492340278625488,
|
|
"epoch": 0.017001242398482965,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 8.1286,
|
|
"mean_token_accuracy": 0.07200958207249641,
|
|
"num_tokens": 533573.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.316695594787598,
|
|
"epoch": 0.01732818936768456,
|
|
"grad_norm": 1.90625,
|
|
"learning_rate": 0.000132,
|
|
"loss": 8.1454,
|
|
"mean_token_accuracy": 0.07354601062834262,
|
|
"num_tokens": 543373.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.469814777374268,
|
|
"epoch": 0.017655136336886156,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 8.0899,
|
|
"mean_token_accuracy": 0.0713756311684847,
|
|
"num_tokens": 554398.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 8.352188110351562,
|
|
"epoch": 0.017982083306087752,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 8.063,
|
|
"mean_token_accuracy": 0.0770417533814907,
|
|
"num_tokens": 564362.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.408161735534668,
|
|
"epoch": 0.018309030275289347,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 8.1399,
|
|
"mean_token_accuracy": 0.07096323706209659,
|
|
"num_tokens": 575658.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 8.40747356414795,
|
|
"epoch": 0.018635977244490943,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 8.1079,
|
|
"mean_token_accuracy": 0.07521827667951583,
|
|
"num_tokens": 585744.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 8.309930896759033,
|
|
"epoch": 0.01896292421369254,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.983,
|
|
"mean_token_accuracy": 0.07047860436141491,
|
|
"num_tokens": 595435.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 8.389182662963867,
|
|
"epoch": 0.019289871182894134,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000147,
|
|
"loss": 8.0106,
|
|
"mean_token_accuracy": 0.0744885340332985,
|
|
"num_tokens": 605374.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 8.33745756149292,
|
|
"epoch": 0.01961681815209573,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 8.1112,
|
|
"mean_token_accuracy": 0.07103150337934494,
|
|
"num_tokens": 616418.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 8.340550994873047,
|
|
"epoch": 0.019943765121297325,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.9849,
|
|
"mean_token_accuracy": 0.08127295225858688,
|
|
"num_tokens": 625857.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 8.38352565765381,
|
|
"epoch": 0.02027071209049892,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 8.0612,
|
|
"mean_token_accuracy": 0.07608035206794739,
|
|
"num_tokens": 636469.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 8.316315841674804,
|
|
"epoch": 0.020597659059700516,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000157,
|
|
"loss": 8.0281,
|
|
"mean_token_accuracy": 0.0735818523913622,
|
|
"num_tokens": 648105.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 8.296995544433594,
|
|
"epoch": 0.02092460602890211,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.9422,
|
|
"mean_token_accuracy": 0.0793572224676609,
|
|
"num_tokens": 657214.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 8.264441871643067,
|
|
"epoch": 0.021251552998103707,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000162,
|
|
"loss": 8.0265,
|
|
"mean_token_accuracy": 0.073749577999115,
|
|
"num_tokens": 667228.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 8.35250301361084,
|
|
"epoch": 0.021578499967305303,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.9781,
|
|
"mean_token_accuracy": 0.07626760490238667,
|
|
"num_tokens": 677412.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 8.250011539459228,
|
|
"epoch": 0.0219054469365069,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.9378,
|
|
"mean_token_accuracy": 0.07601722851395606,
|
|
"num_tokens": 687893.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 8.302884387969971,
|
|
"epoch": 0.022232393905708494,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 8.0661,
|
|
"mean_token_accuracy": 0.07206159345805645,
|
|
"num_tokens": 698497.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 8.278049850463868,
|
|
"epoch": 0.02255934087491009,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.9492,
|
|
"mean_token_accuracy": 0.07396932058036328,
|
|
"num_tokens": 708593.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 8.230413436889648,
|
|
"epoch": 0.022886287844111685,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.9611,
|
|
"mean_token_accuracy": 0.07906975001096725,
|
|
"num_tokens": 719781.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 8.322347927093507,
|
|
"epoch": 0.02321323481331328,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.9722,
|
|
"mean_token_accuracy": 0.07742859311401844,
|
|
"num_tokens": 729946.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 8.199689579010009,
|
|
"epoch": 0.023540181782514876,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.9721,
|
|
"mean_token_accuracy": 0.07622929476201534,
|
|
"num_tokens": 740542.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 8.228047943115234,
|
|
"epoch": 0.02386712875171647,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.8555,
|
|
"mean_token_accuracy": 0.08382938280701638,
|
|
"num_tokens": 750387.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 8.169698715209961,
|
|
"epoch": 0.024194075720918067,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.9124,
|
|
"mean_token_accuracy": 0.07616014927625656,
|
|
"num_tokens": 759708.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 8.173001480102538,
|
|
"epoch": 0.024521022690119663,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.861,
|
|
"mean_token_accuracy": 0.08310946524143219,
|
|
"num_tokens": 770655.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 8.247628688812256,
|
|
"epoch": 0.02484796965932126,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.9962,
|
|
"mean_token_accuracy": 0.07753782495856285,
|
|
"num_tokens": 780263.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 8.120156478881835,
|
|
"epoch": 0.025174916628522854,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.8174,
|
|
"mean_token_accuracy": 0.08559585362672806,
|
|
"num_tokens": 789667.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 8.231314945220948,
|
|
"epoch": 0.02550186359772445,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.8745,
|
|
"mean_token_accuracy": 0.07828151807188988,
|
|
"num_tokens": 800099.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 8.123421001434327,
|
|
"epoch": 0.025828810566926045,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.8562,
|
|
"mean_token_accuracy": 0.08755742833018303,
|
|
"num_tokens": 809347.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 8.190375709533692,
|
|
"epoch": 0.02615575753612764,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.8895,
|
|
"mean_token_accuracy": 0.0774234376847744,
|
|
"num_tokens": 819721.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 8.068334150314332,
|
|
"epoch": 0.026482704505329236,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.8147,
|
|
"mean_token_accuracy": 0.08247106224298477,
|
|
"num_tokens": 830278.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 8.158495712280274,
|
|
"epoch": 0.02680965147453083,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.8692,
|
|
"mean_token_accuracy": 0.07925740331411361,
|
|
"num_tokens": 839760.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 8.091846561431884,
|
|
"epoch": 0.027136598443732427,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.8936,
|
|
"mean_token_accuracy": 0.08189868628978729,
|
|
"num_tokens": 850778.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 8.150493049621582,
|
|
"epoch": 0.027463545412934023,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.8127,
|
|
"mean_token_accuracy": 0.08337080627679824,
|
|
"num_tokens": 860419.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 8.17796688079834,
|
|
"epoch": 0.027790492382135618,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.9115,
|
|
"mean_token_accuracy": 0.08478068634867668,
|
|
"num_tokens": 869939.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 8.120305299758911,
|
|
"epoch": 0.028117439351337214,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.8569,
|
|
"mean_token_accuracy": 0.08082921504974365,
|
|
"num_tokens": 880331.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 8.22661657333374,
|
|
"epoch": 0.02844438632053881,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.9214,
|
|
"mean_token_accuracy": 0.07378799393773079,
|
|
"num_tokens": 891342.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 8.110522651672364,
|
|
"epoch": 0.028771333289740405,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.8951,
|
|
"mean_token_accuracy": 0.07804202027618885,
|
|
"num_tokens": 902253.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 8.053891658782959,
|
|
"epoch": 0.029098280258942,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.8047,
|
|
"mean_token_accuracy": 0.08209186047315598,
|
|
"num_tokens": 913086.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 8.072042274475098,
|
|
"epoch": 0.029425227228143596,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.8269,
|
|
"mean_token_accuracy": 0.0826948344707489,
|
|
"num_tokens": 925082.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 8.100944375991821,
|
|
"epoch": 0.02975217419734519,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.822,
|
|
"mean_token_accuracy": 0.07624491900205613,
|
|
"num_tokens": 933955.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 8.06613826751709,
|
|
"epoch": 0.030079121166546787,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.7756,
|
|
"mean_token_accuracy": 0.08377034105360508,
|
|
"num_tokens": 943951.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 8.096276569366456,
|
|
"epoch": 0.030406068135748383,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.7703,
|
|
"mean_token_accuracy": 0.08263073042035103,
|
|
"num_tokens": 955297.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.890885972976685,
|
|
"epoch": 0.030733015104949978,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.8169,
|
|
"mean_token_accuracy": 0.08191272169351578,
|
|
"num_tokens": 967012.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 8.025649976730346,
|
|
"epoch": 0.031059962074151574,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.7327,
|
|
"mean_token_accuracy": 0.08218838796019554,
|
|
"num_tokens": 977956.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 8.034643745422363,
|
|
"epoch": 0.031386909043353166,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.7506,
|
|
"mean_token_accuracy": 0.08425655737519264,
|
|
"num_tokens": 987270.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.962606287002563,
|
|
"epoch": 0.03171385601255476,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.7517,
|
|
"mean_token_accuracy": 0.08468098640441894,
|
|
"num_tokens": 998010.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.962759017944336,
|
|
"epoch": 0.03204080298175636,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.7869,
|
|
"mean_token_accuracy": 0.09083530306816101,
|
|
"num_tokens": 1008173.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 8.062295961380006,
|
|
"epoch": 0.03236774995095795,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.7972,
|
|
"mean_token_accuracy": 0.08286057338118553,
|
|
"num_tokens": 1019037.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 8.059563636779785,
|
|
"epoch": 0.03269469692015955,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.7878,
|
|
"mean_token_accuracy": 0.08349623754620553,
|
|
"num_tokens": 1028822.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.930562353134155,
|
|
"epoch": 0.033021643889361144,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.6614,
|
|
"mean_token_accuracy": 0.08791738487780094,
|
|
"num_tokens": 1038749.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 8.014323568344116,
|
|
"epoch": 0.03334859085856274,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.8018,
|
|
"mean_token_accuracy": 0.07863838225603104,
|
|
"num_tokens": 1049087.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.964654541015625,
|
|
"epoch": 0.033675537827764335,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.7015,
|
|
"mean_token_accuracy": 0.08479668125510216,
|
|
"num_tokens": 1060711.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.885542392730713,
|
|
"epoch": 0.03400248479696593,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.7692,
|
|
"mean_token_accuracy": 0.08220863975584507,
|
|
"num_tokens": 1070643.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 8.0744797706604,
|
|
"epoch": 0.034329431766167526,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.7278,
|
|
"mean_token_accuracy": 0.08255242258310318,
|
|
"num_tokens": 1080970.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.896595048904419,
|
|
"epoch": 0.03465637873536912,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.6582,
|
|
"mean_token_accuracy": 0.08374542146921157,
|
|
"num_tokens": 1090833.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.986835241317749,
|
|
"epoch": 0.03498332570457072,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.7316,
|
|
"mean_token_accuracy": 0.0834172960370779,
|
|
"num_tokens": 1100160.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.924727725982666,
|
|
"epoch": 0.03531027267377231,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.7039,
|
|
"mean_token_accuracy": 0.08192352429032326,
|
|
"num_tokens": 1110846.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.879991912841797,
|
|
"epoch": 0.03563721964297391,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.5956,
|
|
"mean_token_accuracy": 0.09090287759900093,
|
|
"num_tokens": 1119529.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.930740880966186,
|
|
"epoch": 0.035964166612175504,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.6682,
|
|
"mean_token_accuracy": 0.08713096231222153,
|
|
"num_tokens": 1129798.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.884546184539795,
|
|
"epoch": 0.0362911135813771,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.6967,
|
|
"mean_token_accuracy": 0.08679120689630508,
|
|
"num_tokens": 1141689.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.8929304599761965,
|
|
"epoch": 0.036618060550578695,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.6573,
|
|
"mean_token_accuracy": 0.08585221692919731,
|
|
"num_tokens": 1152522.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.8179162502288815,
|
|
"epoch": 0.03694500751978029,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.635,
|
|
"mean_token_accuracy": 0.09055891707539558,
|
|
"num_tokens": 1163135.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.820132493972778,
|
|
"epoch": 0.037271954488981886,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.5863,
|
|
"mean_token_accuracy": 0.08870974853634835,
|
|
"num_tokens": 1172866.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.876567792892456,
|
|
"epoch": 0.03759890145818348,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.6747,
|
|
"mean_token_accuracy": 0.085811897367239,
|
|
"num_tokens": 1182924.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.812422180175782,
|
|
"epoch": 0.03792584842738508,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.677,
|
|
"mean_token_accuracy": 0.08168176487088204,
|
|
"num_tokens": 1194036.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.8082019805908205,
|
|
"epoch": 0.03825279539658667,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.6098,
|
|
"mean_token_accuracy": 0.09360664561390877,
|
|
"num_tokens": 1205162.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.891174364089966,
|
|
"epoch": 0.03857974236578827,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.6325,
|
|
"mean_token_accuracy": 0.08774000033736229,
|
|
"num_tokens": 1214953.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.7473228931427,
|
|
"epoch": 0.038906689334989863,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.6528,
|
|
"mean_token_accuracy": 0.08504325672984123,
|
|
"num_tokens": 1225745.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.927584314346314,
|
|
"epoch": 0.03923363630419146,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.634,
|
|
"mean_token_accuracy": 0.08675257638096809,
|
|
"num_tokens": 1236076.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.775005245208741,
|
|
"epoch": 0.039560583273393055,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.6204,
|
|
"mean_token_accuracy": 0.08933146819472312,
|
|
"num_tokens": 1245843.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.827740716934204,
|
|
"epoch": 0.03988753024259465,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.6253,
|
|
"mean_token_accuracy": 0.08472589626908303,
|
|
"num_tokens": 1255723.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.858044815063477,
|
|
"epoch": 0.040214477211796246,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.5126,
|
|
"mean_token_accuracy": 0.09874000698328018,
|
|
"num_tokens": 1265289.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.757145357131958,
|
|
"epoch": 0.04054142418099784,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.6338,
|
|
"mean_token_accuracy": 0.08969606682658196,
|
|
"num_tokens": 1275495.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.7039237976074215,
|
|
"epoch": 0.04086837115019944,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.6506,
|
|
"mean_token_accuracy": 0.0886372298002243,
|
|
"num_tokens": 1286473.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.858344173431396,
|
|
"epoch": 0.04119531811940103,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.6054,
|
|
"mean_token_accuracy": 0.08685552179813386,
|
|
"num_tokens": 1296460.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.811800909042359,
|
|
"epoch": 0.04152226508860263,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.621,
|
|
"mean_token_accuracy": 0.08496971726417542,
|
|
"num_tokens": 1306497.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.738426733016968,
|
|
"epoch": 0.04184921205780422,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.5677,
|
|
"mean_token_accuracy": 0.09172414764761924,
|
|
"num_tokens": 1318420.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.7946553230285645,
|
|
"epoch": 0.04217615902700582,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.5636,
|
|
"mean_token_accuracy": 0.0850396528840065,
|
|
"num_tokens": 1327612.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.676682567596435,
|
|
"epoch": 0.042503105996207415,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.4902,
|
|
"mean_token_accuracy": 0.09149853065609932,
|
|
"num_tokens": 1337450.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.803067302703857,
|
|
"epoch": 0.04283005296540901,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.5868,
|
|
"mean_token_accuracy": 0.08667276427149773,
|
|
"num_tokens": 1347953.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.681396389007569,
|
|
"epoch": 0.043156999934610606,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.5154,
|
|
"mean_token_accuracy": 0.08922559693455696,
|
|
"num_tokens": 1358009.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.745648193359375,
|
|
"epoch": 0.0434839469038122,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.5665,
|
|
"mean_token_accuracy": 0.08471445068717003,
|
|
"num_tokens": 1367725.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.802916955947876,
|
|
"epoch": 0.0438108938730138,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.6316,
|
|
"mean_token_accuracy": 0.08761190436780453,
|
|
"num_tokens": 1379334.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.655632400512696,
|
|
"epoch": 0.04413784084221539,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.5457,
|
|
"mean_token_accuracy": 0.0921371579170227,
|
|
"num_tokens": 1389991.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.734749984741211,
|
|
"epoch": 0.04446478781141699,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.531,
|
|
"mean_token_accuracy": 0.08393392376601697,
|
|
"num_tokens": 1400797.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.68169674873352,
|
|
"epoch": 0.04479173478061858,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.5403,
|
|
"mean_token_accuracy": 0.09117294400930405,
|
|
"num_tokens": 1411805.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.709935283660888,
|
|
"epoch": 0.04511868174982018,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.4793,
|
|
"mean_token_accuracy": 0.09437239617109298,
|
|
"num_tokens": 1421539.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.54593505859375,
|
|
"epoch": 0.045445628719021774,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.4307,
|
|
"mean_token_accuracy": 0.09856887757778168,
|
|
"num_tokens": 1432136.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.710141134262085,
|
|
"epoch": 0.04577257568822337,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.5314,
|
|
"mean_token_accuracy": 0.09457024410367013,
|
|
"num_tokens": 1443471.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.645555210113526,
|
|
"epoch": 0.046099522657424966,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.4682,
|
|
"mean_token_accuracy": 0.09427132233977317,
|
|
"num_tokens": 1453503.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.64605302810669,
|
|
"epoch": 0.04642646962662656,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.4646,
|
|
"mean_token_accuracy": 0.09701912105083466,
|
|
"num_tokens": 1463254.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.713723421096802,
|
|
"epoch": 0.04675341659582816,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.5837,
|
|
"mean_token_accuracy": 0.08762456104159355,
|
|
"num_tokens": 1472890.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.548030757904053,
|
|
"epoch": 0.04708036356502975,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.435,
|
|
"mean_token_accuracy": 0.1046062558889389,
|
|
"num_tokens": 1483583.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.713066864013672,
|
|
"epoch": 0.04740731053423135,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.5893,
|
|
"mean_token_accuracy": 0.0898235484957695,
|
|
"num_tokens": 1494278.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.6250792980194095,
|
|
"epoch": 0.04773425750343294,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.4813,
|
|
"mean_token_accuracy": 0.0951371893286705,
|
|
"num_tokens": 1504638.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.6247305393219,
|
|
"epoch": 0.04806120447263454,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.4745,
|
|
"mean_token_accuracy": 0.09250791445374489,
|
|
"num_tokens": 1514581.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.616424989700318,
|
|
"epoch": 0.048388151441836134,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.5399,
|
|
"mean_token_accuracy": 0.09155056923627854,
|
|
"num_tokens": 1525455.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.652921056747436,
|
|
"epoch": 0.04871509841103773,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.481,
|
|
"mean_token_accuracy": 0.09119962081313134,
|
|
"num_tokens": 1537267.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.599986028671265,
|
|
"epoch": 0.049042045380239326,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.3972,
|
|
"mean_token_accuracy": 0.09648198038339614,
|
|
"num_tokens": 1547181.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.520318412780762,
|
|
"epoch": 0.04936899234944092,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.3361,
|
|
"mean_token_accuracy": 0.0945985570549965,
|
|
"num_tokens": 1557267.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.518356227874756,
|
|
"epoch": 0.04969593931864252,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.4096,
|
|
"mean_token_accuracy": 0.09483497068285943,
|
|
"num_tokens": 1567762.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.686395645141602,
|
|
"epoch": 0.05002288628784411,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000382,
|
|
"loss": 7.545,
|
|
"mean_token_accuracy": 0.08728384301066398,
|
|
"num_tokens": 1579312.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.513725757598877,
|
|
"epoch": 0.05034983325704571,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 7.3851,
|
|
"mean_token_accuracy": 0.10093086138367653,
|
|
"num_tokens": 1588752.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.5739607334136965,
|
|
"epoch": 0.0506767802262473,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.3855,
|
|
"mean_token_accuracy": 0.0969861201941967,
|
|
"num_tokens": 1599510.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.501125860214233,
|
|
"epoch": 0.0510037271954489,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.3525,
|
|
"mean_token_accuracy": 0.09426184669137001,
|
|
"num_tokens": 1609651.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.587576103210449,
|
|
"epoch": 0.051330674164650494,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.4996,
|
|
"mean_token_accuracy": 0.08769121989607812,
|
|
"num_tokens": 1619840.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.6007983684539795,
|
|
"epoch": 0.05165762113385209,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.4702,
|
|
"mean_token_accuracy": 0.08686918206512928,
|
|
"num_tokens": 1630725.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.448578500747681,
|
|
"epoch": 0.051984568103053685,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.346,
|
|
"mean_token_accuracy": 0.09301474839448928,
|
|
"num_tokens": 1640535.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.5419182777404785,
|
|
"epoch": 0.05231151507225528,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.3466,
|
|
"mean_token_accuracy": 0.09515140354633331,
|
|
"num_tokens": 1650714.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.457037878036499,
|
|
"epoch": 0.05263846204145688,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000402,
|
|
"loss": 7.4376,
|
|
"mean_token_accuracy": 0.0899427741765976,
|
|
"num_tokens": 1661879.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.522365570068359,
|
|
"epoch": 0.05296540901065847,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 7.3543,
|
|
"mean_token_accuracy": 0.09633980542421341,
|
|
"num_tokens": 1672343.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.500396776199341,
|
|
"epoch": 0.05329235597986007,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 7.3541,
|
|
"mean_token_accuracy": 0.0942266747355461,
|
|
"num_tokens": 1682197.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.4255210876464846,
|
|
"epoch": 0.05361930294906166,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 7.3543,
|
|
"mean_token_accuracy": 0.09494220539927482,
|
|
"num_tokens": 1692182.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.456794881820679,
|
|
"epoch": 0.05394624991826326,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000412,
|
|
"loss": 7.3824,
|
|
"mean_token_accuracy": 0.08804743736982346,
|
|
"num_tokens": 1702646.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.501739406585694,
|
|
"epoch": 0.054273196887464854,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 7.3653,
|
|
"mean_token_accuracy": 0.092994274944067,
|
|
"num_tokens": 1712950.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.439680671691894,
|
|
"epoch": 0.05460014385666645,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000417,
|
|
"loss": 7.281,
|
|
"mean_token_accuracy": 0.10278869196772575,
|
|
"num_tokens": 1722799.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.510075855255127,
|
|
"epoch": 0.054927090825868045,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 7.3868,
|
|
"mean_token_accuracy": 0.09211011007428169,
|
|
"num_tokens": 1732687.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.429882335662842,
|
|
"epoch": 0.05525403779506964,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000422,
|
|
"loss": 7.3804,
|
|
"mean_token_accuracy": 0.09936274141073227,
|
|
"num_tokens": 1743709.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.479153442382812,
|
|
"epoch": 0.055580984764271237,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 7.3367,
|
|
"mean_token_accuracy": 0.09895628616213799,
|
|
"num_tokens": 1753129.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.420769166946411,
|
|
"epoch": 0.05590793173347283,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000427,
|
|
"loss": 7.3216,
|
|
"mean_token_accuracy": 0.09391170740127563,
|
|
"num_tokens": 1764072.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.441389846801758,
|
|
"epoch": 0.05623487870267443,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 7.2937,
|
|
"mean_token_accuracy": 0.10005690380930901,
|
|
"num_tokens": 1774015.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.4574812889099125,
|
|
"epoch": 0.05656182567187602,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000432,
|
|
"loss": 7.3564,
|
|
"mean_token_accuracy": 0.09394175931811333,
|
|
"num_tokens": 1784257.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.392144346237183,
|
|
"epoch": 0.05688877264107762,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 7.2862,
|
|
"mean_token_accuracy": 0.09536551535129548,
|
|
"num_tokens": 1793294.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.488471984863281,
|
|
"epoch": 0.057215719610279214,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000437,
|
|
"loss": 7.3619,
|
|
"mean_token_accuracy": 0.09285857379436493,
|
|
"num_tokens": 1804034.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.381966066360474,
|
|
"epoch": 0.05754266657948081,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 7.2703,
|
|
"mean_token_accuracy": 0.10443824678659439,
|
|
"num_tokens": 1812571.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.317243957519532,
|
|
"epoch": 0.057869613548682405,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000442,
|
|
"loss": 7.3304,
|
|
"mean_token_accuracy": 0.0973969504237175,
|
|
"num_tokens": 1822824.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.547090101242065,
|
|
"epoch": 0.058196560517884,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 7.3785,
|
|
"mean_token_accuracy": 0.09615707248449326,
|
|
"num_tokens": 1833237.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.352640581130982,
|
|
"epoch": 0.058523507487085596,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000447,
|
|
"loss": 7.3567,
|
|
"mean_token_accuracy": 0.0972889244556427,
|
|
"num_tokens": 1843920.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.37256441116333,
|
|
"epoch": 0.05885045445628719,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 7.3819,
|
|
"mean_token_accuracy": 0.093709996342659,
|
|
"num_tokens": 1854599.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.463394737243652,
|
|
"epoch": 0.05917740142548879,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 7.2866,
|
|
"mean_token_accuracy": 0.09582863971590996,
|
|
"num_tokens": 1863988.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.294298934936523,
|
|
"epoch": 0.05950434839469038,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 7.3274,
|
|
"mean_token_accuracy": 0.09870437532663345,
|
|
"num_tokens": 1874249.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.405703210830689,
|
|
"epoch": 0.05983129536389198,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 7.2653,
|
|
"mean_token_accuracy": 0.09711567685008049,
|
|
"num_tokens": 1884221.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.381225490570069,
|
|
"epoch": 0.060158242333093574,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 7.2881,
|
|
"mean_token_accuracy": 0.10006763488054275,
|
|
"num_tokens": 1894287.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.280563163757324,
|
|
"epoch": 0.06048518930229517,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000462,
|
|
"loss": 7.237,
|
|
"mean_token_accuracy": 0.09955689907073975,
|
|
"num_tokens": 1903474.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.411317920684814,
|
|
"epoch": 0.060812136271496765,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 7.2571,
|
|
"mean_token_accuracy": 0.10307784602046013,
|
|
"num_tokens": 1912237.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.263377666473389,
|
|
"epoch": 0.06113908324069836,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000467,
|
|
"loss": 7.3123,
|
|
"mean_token_accuracy": 0.09369613602757454,
|
|
"num_tokens": 1922501.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.432010126113892,
|
|
"epoch": 0.061466030209899956,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 7.3107,
|
|
"mean_token_accuracy": 0.09645366221666336,
|
|
"num_tokens": 1932910.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.397045373916626,
|
|
"epoch": 0.06179297717910155,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 7.4127,
|
|
"mean_token_accuracy": 0.0905526876449585,
|
|
"num_tokens": 1944408.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.305300378799439,
|
|
"epoch": 0.06211992414830315,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 7.3132,
|
|
"mean_token_accuracy": 0.09562604799866677,
|
|
"num_tokens": 1955149.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.358934783935547,
|
|
"epoch": 0.06244687111750474,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000477,
|
|
"loss": 7.2651,
|
|
"mean_token_accuracy": 0.09923698529601097,
|
|
"num_tokens": 1964777.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.284130334854126,
|
|
"epoch": 0.06277381808670633,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 7.2538,
|
|
"mean_token_accuracy": 0.10322569608688355,
|
|
"num_tokens": 1974781.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.348219013214111,
|
|
"epoch": 0.06310076505590793,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000482,
|
|
"loss": 7.2242,
|
|
"mean_token_accuracy": 0.10055499300360679,
|
|
"num_tokens": 1984540.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.290968942642212,
|
|
"epoch": 0.06342771202510952,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 7.2561,
|
|
"mean_token_accuracy": 0.0991756021976471,
|
|
"num_tokens": 1995360.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 7.307056856155396,
|
|
"epoch": 0.06375465899431113,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000487,
|
|
"loss": 7.2425,
|
|
"mean_token_accuracy": 0.0994577057659626,
|
|
"num_tokens": 2006299.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.292001819610595,
|
|
"epoch": 0.06408160596351271,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 7.2949,
|
|
"mean_token_accuracy": 0.10042648911476135,
|
|
"num_tokens": 2017482.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.321526336669922,
|
|
"epoch": 0.06440855293271432,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000492,
|
|
"loss": 7.309,
|
|
"mean_token_accuracy": 0.09644284099340439,
|
|
"num_tokens": 2027669.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.327935981750488,
|
|
"epoch": 0.0647354999019159,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 7.2268,
|
|
"mean_token_accuracy": 0.10123592019081115,
|
|
"num_tokens": 2037047.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 7.312641668319702,
|
|
"epoch": 0.06506244687111751,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000497,
|
|
"loss": 7.2666,
|
|
"mean_token_accuracy": 0.09748207852244377,
|
|
"num_tokens": 2046813.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.140112829208374,
|
|
"epoch": 0.0653893938403191,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 7.2152,
|
|
"mean_token_accuracy": 0.10107465907931328,
|
|
"num_tokens": 2056669.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 7.370557641983032,
|
|
"epoch": 0.0657163408095207,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000499998026082006,
|
|
"loss": 7.199,
|
|
"mean_token_accuracy": 0.10289056971669197,
|
|
"num_tokens": 2066786.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 7.234280109405518,
|
|
"epoch": 0.06604328777872229,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004999900070995136,
|
|
"loss": 7.2556,
|
|
"mean_token_accuracy": 0.09878846332430839,
|
|
"num_tokens": 2077248.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 7.230677461624145,
|
|
"epoch": 0.06637023474792389,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999758199023239,
|
|
"loss": 7.2683,
|
|
"mean_token_accuracy": 0.0940069980919361,
|
|
"num_tokens": 2087448.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 7.270314884185791,
|
|
"epoch": 0.06669718171712548,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999554648793858,
|
|
"loss": 7.1791,
|
|
"mean_token_accuracy": 0.1049445316195488,
|
|
"num_tokens": 2098481.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 7.187046861648559,
|
|
"epoch": 0.06702412868632708,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999289425887425,
|
|
"loss": 7.1849,
|
|
"mean_token_accuracy": 0.10531515032052993,
|
|
"num_tokens": 2108775.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 7.303104543685913,
|
|
"epoch": 0.06735107565552867,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004998962537575161,
|
|
"loss": 7.1798,
|
|
"mean_token_accuracy": 0.1043311208486557,
|
|
"num_tokens": 2118485.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 7.096952104568482,
|
|
"epoch": 0.06767802262473027,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004998573992818874,
|
|
"loss": 7.0703,
|
|
"mean_token_accuracy": 0.10433712080121041,
|
|
"num_tokens": 2127567.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 7.158503150939941,
|
|
"epoch": 0.06800496959393186,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998123802270715,
|
|
"loss": 7.1382,
|
|
"mean_token_accuracy": 0.1033934198319912,
|
|
"num_tokens": 2138449.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 7.296292161941528,
|
|
"epoch": 0.06833191656313346,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004997611978272886,
|
|
"loss": 7.2746,
|
|
"mean_token_accuracy": 0.0988951563835144,
|
|
"num_tokens": 2148644.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 7.170377254486084,
|
|
"epoch": 0.06865886353233505,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004997038534857298,
|
|
"loss": 7.2498,
|
|
"mean_token_accuracy": 0.10384517982602119,
|
|
"num_tokens": 2158412.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 7.332276487350464,
|
|
"epoch": 0.06898581050153665,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004996403487745194,
|
|
"loss": 7.2538,
|
|
"mean_token_accuracy": 0.09230363294482231,
|
|
"num_tokens": 2167762.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 7.300696897506714,
|
|
"epoch": 0.06931275747073824,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000499570685434671,
|
|
"loss": 7.3523,
|
|
"mean_token_accuracy": 0.09614870846271514,
|
|
"num_tokens": 2178848.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 7.202943801879883,
|
|
"epoch": 0.06963970443993985,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004994948653760405,
|
|
"loss": 7.1882,
|
|
"mean_token_accuracy": 0.10102253258228303,
|
|
"num_tokens": 2189422.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 7.240464353561402,
|
|
"epoch": 0.06996665140914143,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004994128906772729,
|
|
"loss": 7.1795,
|
|
"mean_token_accuracy": 0.10200649499893188,
|
|
"num_tokens": 2198982.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 7.198509359359742,
|
|
"epoch": 0.07029359837834304,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000499324763585746,
|
|
"loss": 7.2344,
|
|
"mean_token_accuracy": 0.0994394488632679,
|
|
"num_tokens": 2208568.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 7.27600884437561,
|
|
"epoch": 0.07062054534754462,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004992304865175085,
|
|
"loss": 7.1934,
|
|
"mean_token_accuracy": 0.10426802039146424,
|
|
"num_tokens": 2218892.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 7.1890027046203615,
|
|
"epoch": 0.07094749231674623,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004991300620572138,
|
|
"loss": 7.1505,
|
|
"mean_token_accuracy": 0.10170021578669548,
|
|
"num_tokens": 2229398.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 7.145445060729981,
|
|
"epoch": 0.07127443928594782,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004990234929580494,
|
|
"loss": 7.2077,
|
|
"mean_token_accuracy": 0.10364220291376114,
|
|
"num_tokens": 2239584.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 7.238229465484619,
|
|
"epoch": 0.07160138625514942,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004989107821416609,
|
|
"loss": 7.1354,
|
|
"mean_token_accuracy": 0.09976739063858986,
|
|
"num_tokens": 2250520.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 7.162268733978271,
|
|
"epoch": 0.07192833322435101,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004987919326980723,
|
|
"loss": 7.1512,
|
|
"mean_token_accuracy": 0.10593457892537117,
|
|
"num_tokens": 2260427.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 7.150348138809204,
|
|
"epoch": 0.07225528019355261,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004986669478856011,
|
|
"loss": 7.2221,
|
|
"mean_token_accuracy": 0.09651356115937233,
|
|
"num_tokens": 2270578.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 7.166797685623169,
|
|
"epoch": 0.0725822271627542,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004985358311307688,
|
|
"loss": 7.1709,
|
|
"mean_token_accuracy": 0.10436831042170525,
|
|
"num_tokens": 2280884.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 7.1626897811889645,
|
|
"epoch": 0.0729091741319558,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004983985860282081,
|
|
"loss": 7.1358,
|
|
"mean_token_accuracy": 0.10171283408999443,
|
|
"num_tokens": 2290267.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 7.164863777160645,
|
|
"epoch": 0.07323612110115739,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004982552163405623,
|
|
"loss": 7.1792,
|
|
"mean_token_accuracy": 0.09882023707032203,
|
|
"num_tokens": 2300847.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 7.224768877029419,
|
|
"epoch": 0.07356306807035899,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004981057259983839,
|
|
"loss": 7.0865,
|
|
"mean_token_accuracy": 0.10409277006983757,
|
|
"num_tokens": 2311129.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 7.075718593597412,
|
|
"epoch": 0.07389001503956058,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004979501191000262,
|
|
"loss": 7.1951,
|
|
"mean_token_accuracy": 0.10550073087215424,
|
|
"num_tokens": 2322101.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 7.191465902328491,
|
|
"epoch": 0.07421696200876218,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004977883999115311,
|
|
"loss": 7.0725,
|
|
"mean_token_accuracy": 0.10739018693566323,
|
|
"num_tokens": 2332639.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 7.127969074249267,
|
|
"epoch": 0.07454390897796377,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004976205728665113,
|
|
"loss": 7.1794,
|
|
"mean_token_accuracy": 0.10356512665748596,
|
|
"num_tokens": 2343508.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 7.175882577896118,
|
|
"epoch": 0.07487085594716537,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004974466425660307,
|
|
"loss": 7.0978,
|
|
"mean_token_accuracy": 0.10513930395245552,
|
|
"num_tokens": 2353384.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 7.064570760726928,
|
|
"epoch": 0.07519780291636696,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004972666137784759,
|
|
"loss": 7.0728,
|
|
"mean_token_accuracy": 0.11417345926165581,
|
|
"num_tokens": 2363462.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 7.086591005325317,
|
|
"epoch": 0.07552474988556857,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004970804914394271,
|
|
"loss": 7.0717,
|
|
"mean_token_accuracy": 0.10745552852749825,
|
|
"num_tokens": 2373474.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 7.076848649978638,
|
|
"epoch": 0.07585169685477015,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004968882806515225,
|
|
"loss": 7.1314,
|
|
"mean_token_accuracy": 0.1009025052189827,
|
|
"num_tokens": 2385199.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 7.271294116973877,
|
|
"epoch": 0.07617864382397176,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004966899866843177,
|
|
"loss": 7.1672,
|
|
"mean_token_accuracy": 0.09792510718107224,
|
|
"num_tokens": 2395798.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 7.1627648830413815,
|
|
"epoch": 0.07650559079317334,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000496485614974142,
|
|
"loss": 7.1479,
|
|
"mean_token_accuracy": 0.10423750802874565,
|
|
"num_tokens": 2405536.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 7.139655208587646,
|
|
"epoch": 0.07683253776237495,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004962751711239492,
|
|
"loss": 7.0947,
|
|
"mean_token_accuracy": 0.10896513685584068,
|
|
"num_tokens": 2415326.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 7.079001569747925,
|
|
"epoch": 0.07715948473157654,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004960586609031636,
|
|
"loss": 7.0992,
|
|
"mean_token_accuracy": 0.10462969392538071,
|
|
"num_tokens": 2426177.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 7.114988660812378,
|
|
"epoch": 0.07748643170077814,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004958360902475224,
|
|
"loss": 7.1049,
|
|
"mean_token_accuracy": 0.10703596845269203,
|
|
"num_tokens": 2436308.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 7.147863912582397,
|
|
"epoch": 0.07781337866997973,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004956074652589125,
|
|
"loss": 7.0982,
|
|
"mean_token_accuracy": 0.10920525342226028,
|
|
"num_tokens": 2446866.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 7.0905113697052,
|
|
"epoch": 0.07814032563918133,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004953727922052035,
|
|
"loss": 7.1053,
|
|
"mean_token_accuracy": 0.1001770630478859,
|
|
"num_tokens": 2457999.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 7.139458036422729,
|
|
"epoch": 0.07846727260838292,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004951320775200756,
|
|
"loss": 7.0647,
|
|
"mean_token_accuracy": 0.1055167481303215,
|
|
"num_tokens": 2468333.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 7.095925617218017,
|
|
"epoch": 0.07879421957758452,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004948853278028436,
|
|
"loss": 7.188,
|
|
"mean_token_accuracy": 0.10348707437515259,
|
|
"num_tokens": 2478288.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 7.189743328094482,
|
|
"epoch": 0.07912116654678611,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004946325498182755,
|
|
"loss": 7.091,
|
|
"mean_token_accuracy": 0.11169375106692314,
|
|
"num_tokens": 2487845.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 7.059930849075317,
|
|
"epoch": 0.07944811351598771,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004943737504964076,
|
|
"loss": 7.0835,
|
|
"mean_token_accuracy": 0.10219844505190849,
|
|
"num_tokens": 2498419.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 7.053397178649902,
|
|
"epoch": 0.0797750604851893,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000494108936932354,
|
|
"loss": 7.0853,
|
|
"mean_token_accuracy": 0.10752587541937828,
|
|
"num_tokens": 2509124.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 7.217895412445069,
|
|
"epoch": 0.0801020074543909,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004938381163861124,
|
|
"loss": 7.0519,
|
|
"mean_token_accuracy": 0.1040741816163063,
|
|
"num_tokens": 2519255.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 7.014905691146851,
|
|
"epoch": 0.08042895442359249,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004935612962823645,
|
|
"loss": 7.2048,
|
|
"mean_token_accuracy": 0.09748868644237518,
|
|
"num_tokens": 2530576.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 7.2198834896087645,
|
|
"epoch": 0.0807559013927941,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004932784842102739,
|
|
"loss": 7.0658,
|
|
"mean_token_accuracy": 0.10952000319957733,
|
|
"num_tokens": 2542632.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 7.004525756835937,
|
|
"epoch": 0.08108284836199568,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004929896879232758,
|
|
"loss": 7.0442,
|
|
"mean_token_accuracy": 0.11097664833068847,
|
|
"num_tokens": 2552960.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 7.162525033950805,
|
|
"epoch": 0.08140979533119728,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004926949153388668,
|
|
"loss": 7.0337,
|
|
"mean_token_accuracy": 0.10997907668352128,
|
|
"num_tokens": 2563214.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.986301422119141,
|
|
"epoch": 0.08173674230039887,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004923941745383859,
|
|
"loss": 7.0436,
|
|
"mean_token_accuracy": 0.10774937942624092,
|
|
"num_tokens": 2573712.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 7.044104719161988,
|
|
"epoch": 0.08206368926960048,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000492087473766794,
|
|
"loss": 6.9847,
|
|
"mean_token_accuracy": 0.1079469420015812,
|
|
"num_tokens": 2583024.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 7.041021776199341,
|
|
"epoch": 0.08239063623880206,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000491774821432448,
|
|
"loss": 7.0402,
|
|
"mean_token_accuracy": 0.10729892030358315,
|
|
"num_tokens": 2591816.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 7.127667951583862,
|
|
"epoch": 0.08271758320800367,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004914562261068693,
|
|
"loss": 7.0439,
|
|
"mean_token_accuracy": 0.11361931115388871,
|
|
"num_tokens": 2602895.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 7.125279426574707,
|
|
"epoch": 0.08304453017720526,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004911316965245098,
|
|
"loss": 7.086,
|
|
"mean_token_accuracy": 0.1084518425166607,
|
|
"num_tokens": 2613537.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.954402017593384,
|
|
"epoch": 0.08337147714640686,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000490801241582512,
|
|
"loss": 7.0291,
|
|
"mean_token_accuracy": 0.11241161450743675,
|
|
"num_tokens": 2623282.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 7.049486398696899,
|
|
"epoch": 0.08369842411560845,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000490464870340465,
|
|
"loss": 6.9434,
|
|
"mean_token_accuracy": 0.10828463956713677,
|
|
"num_tokens": 2634237.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 7.025067663192749,
|
|
"epoch": 0.08402537108481005,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004901225920201563,
|
|
"loss": 7.0188,
|
|
"mean_token_accuracy": 0.1099492497742176,
|
|
"num_tokens": 2643427.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 7.083580255508423,
|
|
"epoch": 0.08435231805401164,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000489774416005319,
|
|
"loss": 7.0976,
|
|
"mean_token_accuracy": 0.1054080605506897,
|
|
"num_tokens": 2653876.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 7.084479761123657,
|
|
"epoch": 0.08467926502321324,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004894203518413742,
|
|
"loss": 7.0001,
|
|
"mean_token_accuracy": 0.11094668358564377,
|
|
"num_tokens": 2663265.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 7.065930271148682,
|
|
"epoch": 0.08500621199241483,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004890604092351701,
|
|
"loss": 6.9809,
|
|
"mean_token_accuracy": 0.11362912952899933,
|
|
"num_tokens": 2673289.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.941449785232544,
|
|
"epoch": 0.08533315896161643,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000488694598054715,
|
|
"loss": 6.9276,
|
|
"mean_token_accuracy": 0.11456131264567375,
|
|
"num_tokens": 2681843.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 7.107578706741333,
|
|
"epoch": 0.08566010593081802,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004883229283289071,
|
|
"loss": 7.0564,
|
|
"mean_token_accuracy": 0.10534741431474685,
|
|
"num_tokens": 2692509.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 7.058593034744263,
|
|
"epoch": 0.08598705290001962,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048794541024725993,
|
|
"loss": 7.0761,
|
|
"mean_token_accuracy": 0.10744128301739693,
|
|
"num_tokens": 2703847.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 7.020749425888061,
|
|
"epoch": 0.08631399986922121,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004875620541596221,
|
|
"loss": 6.9588,
|
|
"mean_token_accuracy": 0.11348334029316902,
|
|
"num_tokens": 2714173.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.973781538009644,
|
|
"epoch": 0.08664094683842281,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048717287057589454,
|
|
"loss": 6.9867,
|
|
"mean_token_accuracy": 0.10945778414607048,
|
|
"num_tokens": 2724315.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 7.043337774276734,
|
|
"epoch": 0.0869678938076244,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004867778701657417,
|
|
"loss": 7.0305,
|
|
"mean_token_accuracy": 0.10321362689137459,
|
|
"num_tokens": 2734738.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.999337959289551,
|
|
"epoch": 0.087294840776826,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048637706375829955,
|
|
"loss": 6.9605,
|
|
"mean_token_accuracy": 0.11335636526346207,
|
|
"num_tokens": 2744565.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.927283859252929,
|
|
"epoch": 0.0876217877460276,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000485970462341878,
|
|
"loss": 6.9999,
|
|
"mean_token_accuracy": 0.11326926499605179,
|
|
"num_tokens": 2754474.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 7.036970615386963,
|
|
"epoch": 0.0879487347152292,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048555807706366044,
|
|
"loss": 6.8887,
|
|
"mean_token_accuracy": 0.1132916696369648,
|
|
"num_tokens": 2763350.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.958545827865601,
|
|
"epoch": 0.08827568168443078,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048513991922939756,
|
|
"loss": 6.9478,
|
|
"mean_token_accuracy": 0.1178972952067852,
|
|
"num_tokens": 2774148.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.958732652664184,
|
|
"epoch": 0.08860262865363239,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00048471600030309744,
|
|
"loss": 6.9812,
|
|
"mean_token_accuracy": 0.11009031683206558,
|
|
"num_tokens": 2784151.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 7.074668073654175,
|
|
"epoch": 0.08892957562283398,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00048428633190671186,
|
|
"loss": 6.9868,
|
|
"mean_token_accuracy": 0.10659226328134537,
|
|
"num_tokens": 2796049.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.98486852645874,
|
|
"epoch": 0.08925652259203558,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004838509258198167,
|
|
"loss": 7.0205,
|
|
"mean_token_accuracy": 0.10692472159862518,
|
|
"num_tokens": 2805739.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.9505034446716305,
|
|
"epoch": 0.08958346956123717,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048340979397929,
|
|
"loss": 6.9577,
|
|
"mean_token_accuracy": 0.10758891850709915,
|
|
"num_tokens": 2815889.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 7.083079862594604,
|
|
"epoch": 0.08991041653043877,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00048296294847898386,
|
|
"loss": 7.1401,
|
|
"mean_token_accuracy": 0.1008207194507122,
|
|
"num_tokens": 2828732.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 7.0747274398803714,
|
|
"epoch": 0.09023736349964036,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0004825104015693934,
|
|
"loss": 7.0118,
|
|
"mean_token_accuracy": 0.10586313381791115,
|
|
"num_tokens": 2839229.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.904927396774292,
|
|
"epoch": 0.09056431046884196,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004820521656573208,
|
|
"loss": 6.9334,
|
|
"mean_token_accuracy": 0.10469542071223259,
|
|
"num_tokens": 2849466.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 7.036262941360474,
|
|
"epoch": 0.09089125743804355,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048158825330553505,
|
|
"loss": 7.0066,
|
|
"mean_token_accuracy": 0.10151828527450561,
|
|
"num_tokens": 2859228.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.887896347045898,
|
|
"epoch": 0.09121820440724515,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048111867723242763,
|
|
"loss": 6.9391,
|
|
"mean_token_accuracy": 0.11997276619076729,
|
|
"num_tokens": 2868648.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.985194826126099,
|
|
"epoch": 0.09154515137644674,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004806434503116637,
|
|
"loss": 6.911,
|
|
"mean_token_accuracy": 0.11612675413489341,
|
|
"num_tokens": 2879010.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.870703172683716,
|
|
"epoch": 0.09187209834564834,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004801625855718296,
|
|
"loss": 6.8977,
|
|
"mean_token_accuracy": 0.1134621523320675,
|
|
"num_tokens": 2890088.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 7.03192720413208,
|
|
"epoch": 0.09219904531484993,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00047967609619607477,
|
|
"loss": 7.0231,
|
|
"mean_token_accuracy": 0.10255616903305054,
|
|
"num_tokens": 2900798.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 7.115162944793701,
|
|
"epoch": 0.09252599228405153,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004791839955217513,
|
|
"loss": 6.9932,
|
|
"mean_token_accuracy": 0.1080642357468605,
|
|
"num_tokens": 2912236.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.870238828659057,
|
|
"epoch": 0.09285293925325312,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00047868629704004786,
|
|
"loss": 6.8956,
|
|
"mean_token_accuracy": 0.11324675604701043,
|
|
"num_tokens": 2922199.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 7.047099208831787,
|
|
"epoch": 0.09317988622245472,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00047818301439561965,
|
|
"loss": 6.9184,
|
|
"mean_token_accuracy": 0.1117422491312027,
|
|
"num_tokens": 2932379.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.97562427520752,
|
|
"epoch": 0.09350683319165631,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00047767416138621454,
|
|
"loss": 6.9086,
|
|
"mean_token_accuracy": 0.10740035325288773,
|
|
"num_tokens": 2942046.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 7.038304662704467,
|
|
"epoch": 0.0938337801608579,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000477159751962295,
|
|
"loss": 6.9918,
|
|
"mean_token_accuracy": 0.10368636325001716,
|
|
"num_tokens": 2952468.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.943646955490112,
|
|
"epoch": 0.0941607271300595,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047663980022665507,
|
|
"loss": 6.9347,
|
|
"mean_token_accuracy": 0.10849333181977272,
|
|
"num_tokens": 2962510.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.993288040161133,
|
|
"epoch": 0.0944876740992611,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00047611432043403437,
|
|
"loss": 6.9537,
|
|
"mean_token_accuracy": 0.11526132896542549,
|
|
"num_tokens": 2972725.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.861820936203003,
|
|
"epoch": 0.0948146210684627,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004755833269907267,
|
|
"loss": 6.9089,
|
|
"mean_token_accuracy": 0.11412198022007942,
|
|
"num_tokens": 2982761.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.942057752609253,
|
|
"epoch": 0.09514156803766428,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004750468344541857,
|
|
"loss": 6.7743,
|
|
"mean_token_accuracy": 0.1132668562233448,
|
|
"num_tokens": 2992175.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.847037696838379,
|
|
"epoch": 0.09546851500686589,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047450485753262525,
|
|
"loss": 6.8506,
|
|
"mean_token_accuracy": 0.11924657598137856,
|
|
"num_tokens": 3002593.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.938974905014038,
|
|
"epoch": 0.09579546197606748,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00047395741108461633,
|
|
"loss": 6.8745,
|
|
"mean_token_accuracy": 0.11872759386897087,
|
|
"num_tokens": 3013251.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.913155460357666,
|
|
"epoch": 0.09612240894526908,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00047340451011867985,
|
|
"loss": 6.8615,
|
|
"mean_token_accuracy": 0.11367089822888374,
|
|
"num_tokens": 3023290.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.880503749847412,
|
|
"epoch": 0.09644935591447067,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047284616979287515,
|
|
"loss": 6.8184,
|
|
"mean_token_accuracy": 0.11757438033819198,
|
|
"num_tokens": 3032377.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.910917472839356,
|
|
"epoch": 0.09677630288367227,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047228240541438433,
|
|
"loss": 6.8997,
|
|
"mean_token_accuracy": 0.11520154550671577,
|
|
"num_tokens": 3043273.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.923095798492431,
|
|
"epoch": 0.09710324985287386,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00047171323243909257,
|
|
"loss": 6.9032,
|
|
"mean_token_accuracy": 0.1145704597234726,
|
|
"num_tokens": 3053696.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.816758489608764,
|
|
"epoch": 0.09743019682207546,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047113866647116457,
|
|
"loss": 6.9068,
|
|
"mean_token_accuracy": 0.11733032763004303,
|
|
"num_tokens": 3063742.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 7.046560573577881,
|
|
"epoch": 0.09775714379127705,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004705587232626164,
|
|
"loss": 6.8556,
|
|
"mean_token_accuracy": 0.11156056895852089,
|
|
"num_tokens": 3073153.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.8397290229797365,
|
|
"epoch": 0.09808409076047865,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046997341871288424,
|
|
"loss": 6.8291,
|
|
"mean_token_accuracy": 0.12191251441836357,
|
|
"num_tokens": 3082567.0,
|
|
"step": 1500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 4000,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 1,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 4853613809664000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|