18233 lines
498 KiB
JSON
18233 lines
498 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 9.730124391563008,
|
|
"eval_steps": 500,
|
|
"global_step": 9000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.742581367492676,
|
|
"epoch": 0.005408328826392645,
|
|
"grad_norm": 5.4375,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.7836,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 10942.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.742576122283936,
|
|
"epoch": 0.01081665765278529,
|
|
"grad_norm": 5.0,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7725,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 21887.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.742580318450928,
|
|
"epoch": 0.016224986479177934,
|
|
"grad_norm": 5.46875,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.7688,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 34345.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.74259843826294,
|
|
"epoch": 0.02163331530557058,
|
|
"grad_norm": 5.3125,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.6831,
|
|
"mean_token_accuracy": 0.0005558153847232461,
|
|
"num_tokens": 45445.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.742469692230225,
|
|
"epoch": 0.02704164413196322,
|
|
"grad_norm": 4.34375,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.5988,
|
|
"mean_token_accuracy": 0.007499191560782492,
|
|
"num_tokens": 58237.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.742159843444824,
|
|
"epoch": 0.03244997295835587,
|
|
"grad_norm": 4.25,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.4088,
|
|
"mean_token_accuracy": 0.031410138495266436,
|
|
"num_tokens": 68098.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.740957832336425,
|
|
"epoch": 0.03785830178474851,
|
|
"grad_norm": 3.203125,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.2937,
|
|
"mean_token_accuracy": 0.04090985879302025,
|
|
"num_tokens": 79867.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.738431835174561,
|
|
"epoch": 0.04326663061114116,
|
|
"grad_norm": 2.625,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.1592,
|
|
"mean_token_accuracy": 0.037697183154523375,
|
|
"num_tokens": 91257.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.735397911071777,
|
|
"epoch": 0.048674959437533805,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.0705,
|
|
"mean_token_accuracy": 0.04106438439339399,
|
|
"num_tokens": 103474.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.733191680908202,
|
|
"epoch": 0.05408328826392644,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.9433,
|
|
"mean_token_accuracy": 0.037430111691355704,
|
|
"num_tokens": 114830.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.732486057281495,
|
|
"epoch": 0.05949161709031909,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.8905,
|
|
"mean_token_accuracy": 0.036898509971797465,
|
|
"num_tokens": 126017.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.73141107559204,
|
|
"epoch": 0.06489994591671173,
|
|
"grad_norm": 2.03125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.8185,
|
|
"mean_token_accuracy": 0.04504128023982048,
|
|
"num_tokens": 138509.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.729285430908202,
|
|
"epoch": 0.07030827474310439,
|
|
"grad_norm": 1.96875,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.764,
|
|
"mean_token_accuracy": 0.0458514004945755,
|
|
"num_tokens": 148353.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.726380920410156,
|
|
"epoch": 0.07571660356949703,
|
|
"grad_norm": 1.9140625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.7481,
|
|
"mean_token_accuracy": 0.04072931818664074,
|
|
"num_tokens": 160269.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.723598289489747,
|
|
"epoch": 0.08112493239588967,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.6457,
|
|
"mean_token_accuracy": 0.040256797894835473,
|
|
"num_tokens": 171568.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.720184516906738,
|
|
"epoch": 0.08653326122228232,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.5709,
|
|
"mean_token_accuracy": 0.04319700971245766,
|
|
"num_tokens": 184243.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.715407657623292,
|
|
"epoch": 0.09194159004867496,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.5081,
|
|
"mean_token_accuracy": 0.044286393001675604,
|
|
"num_tokens": 194887.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.70880765914917,
|
|
"epoch": 0.09734991887506761,
|
|
"grad_norm": 1.9453125,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.4386,
|
|
"mean_token_accuracy": 0.04779103547334671,
|
|
"num_tokens": 206476.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.699271774291992,
|
|
"epoch": 0.10275824770146025,
|
|
"grad_norm": 1.9453125,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.3288,
|
|
"mean_token_accuracy": 0.052064205333590505,
|
|
"num_tokens": 216578.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.6855544090271,
|
|
"epoch": 0.10816657652785289,
|
|
"grad_norm": 1.90625,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.2367,
|
|
"mean_token_accuracy": 0.05270014926791191,
|
|
"num_tokens": 226941.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.670534229278564,
|
|
"epoch": 0.11357490535424554,
|
|
"grad_norm": 1.8828125,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.1416,
|
|
"mean_token_accuracy": 0.05506536141037941,
|
|
"num_tokens": 238060.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.650165462493897,
|
|
"epoch": 0.11898323418063818,
|
|
"grad_norm": 2.015625,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.998,
|
|
"mean_token_accuracy": 0.06617085449397564,
|
|
"num_tokens": 249137.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.614893913269043,
|
|
"epoch": 0.12439156300703083,
|
|
"grad_norm": 1.8984375,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.9011,
|
|
"mean_token_accuracy": 0.0645027995109558,
|
|
"num_tokens": 260712.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.56815938949585,
|
|
"epoch": 0.12979989183342347,
|
|
"grad_norm": 1.921875,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.7629,
|
|
"mean_token_accuracy": 0.06179735250771046,
|
|
"num_tokens": 271039.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.511663913726807,
|
|
"epoch": 0.1352082206598161,
|
|
"grad_norm": 1.984375,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.6203,
|
|
"mean_token_accuracy": 0.0636001892387867,
|
|
"num_tokens": 281353.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.432361125946045,
|
|
"epoch": 0.14061654948620878,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.5429,
|
|
"mean_token_accuracy": 0.07175168134272099,
|
|
"num_tokens": 292221.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.352851963043213,
|
|
"epoch": 0.1460248783126014,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.4081,
|
|
"mean_token_accuracy": 0.06957007274031639,
|
|
"num_tokens": 303353.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.268811798095703,
|
|
"epoch": 0.15143320713899405,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.292,
|
|
"mean_token_accuracy": 0.07166561894118786,
|
|
"num_tokens": 314165.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.154801845550537,
|
|
"epoch": 0.1568415359653867,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.1152,
|
|
"mean_token_accuracy": 0.06947772391140461,
|
|
"num_tokens": 326387.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 10.024849987030029,
|
|
"epoch": 0.16224986479177933,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.0602,
|
|
"mean_token_accuracy": 0.06997468769550323,
|
|
"num_tokens": 339155.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.84305181503296,
|
|
"epoch": 0.167658193618172,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.9106,
|
|
"mean_token_accuracy": 0.07246845588088036,
|
|
"num_tokens": 350643.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.628917694091797,
|
|
"epoch": 0.17306652244456464,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.7825,
|
|
"mean_token_accuracy": 0.07223140001296997,
|
|
"num_tokens": 361977.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.454836750030518,
|
|
"epoch": 0.17847485127095727,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.7462,
|
|
"mean_token_accuracy": 0.07521784901618958,
|
|
"num_tokens": 373604.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.20318956375122,
|
|
"epoch": 0.1838831800973499,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.6587,
|
|
"mean_token_accuracy": 0.07736964598298073,
|
|
"num_tokens": 384306.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.976179218292236,
|
|
"epoch": 0.18929150892374255,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.5228,
|
|
"mean_token_accuracy": 0.08294685482978821,
|
|
"num_tokens": 396479.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.758621406555175,
|
|
"epoch": 0.19469983775013522,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.4294,
|
|
"mean_token_accuracy": 0.08506134003400803,
|
|
"num_tokens": 408489.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.530921840667725,
|
|
"epoch": 0.20010816657652786,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.3995,
|
|
"mean_token_accuracy": 0.08729378059506417,
|
|
"num_tokens": 419339.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.382069301605224,
|
|
"epoch": 0.2055164954029205,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.3478,
|
|
"mean_token_accuracy": 0.08495231121778488,
|
|
"num_tokens": 430486.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.212819290161132,
|
|
"epoch": 0.21092482422931313,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.3377,
|
|
"mean_token_accuracy": 0.08699661865830421,
|
|
"num_tokens": 440414.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.141395664215088,
|
|
"epoch": 0.21633315305570577,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.2746,
|
|
"mean_token_accuracy": 0.0853013277053833,
|
|
"num_tokens": 451475.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.020270109176636,
|
|
"epoch": 0.22174148188209844,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.3305,
|
|
"mean_token_accuracy": 0.08908139243721962,
|
|
"num_tokens": 462053.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.034120750427245,
|
|
"epoch": 0.22714981070849108,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.2768,
|
|
"mean_token_accuracy": 0.08668714761734009,
|
|
"num_tokens": 473207.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.990876579284668,
|
|
"epoch": 0.23255813953488372,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.2574,
|
|
"mean_token_accuracy": 0.08839641213417053,
|
|
"num_tokens": 484999.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.926925563812256,
|
|
"epoch": 0.23796646836127636,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.2424,
|
|
"mean_token_accuracy": 0.08917610868811607,
|
|
"num_tokens": 496772.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.89932656288147,
|
|
"epoch": 0.24337479718766902,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.222,
|
|
"mean_token_accuracy": 0.08830738216638565,
|
|
"num_tokens": 507782.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.855933761596679,
|
|
"epoch": 0.24878312601406166,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.1208,
|
|
"mean_token_accuracy": 0.09142142534255981,
|
|
"num_tokens": 518950.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.874954509735107,
|
|
"epoch": 0.2541914548404543,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.1844,
|
|
"mean_token_accuracy": 0.09523893147706985,
|
|
"num_tokens": 530795.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.853066539764404,
|
|
"epoch": 0.25959978366684694,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.202,
|
|
"mean_token_accuracy": 0.09066410064697265,
|
|
"num_tokens": 542878.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.815495109558105,
|
|
"epoch": 0.2650081124932396,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.0976,
|
|
"mean_token_accuracy": 0.09599302783608436,
|
|
"num_tokens": 553682.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.718054342269897,
|
|
"epoch": 0.2704164413196322,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.0854,
|
|
"mean_token_accuracy": 0.10266833081841469,
|
|
"num_tokens": 564433.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.75045108795166,
|
|
"epoch": 0.27582477014602486,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.1247,
|
|
"mean_token_accuracy": 0.09944702386856079,
|
|
"num_tokens": 576001.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.712650632858276,
|
|
"epoch": 0.28123309897241755,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.0446,
|
|
"mean_token_accuracy": 0.10567303821444511,
|
|
"num_tokens": 587138.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.598682832717896,
|
|
"epoch": 0.2866414277988102,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.0066,
|
|
"mean_token_accuracy": 0.10336644649505615,
|
|
"num_tokens": 597572.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.61051869392395,
|
|
"epoch": 0.2920497566252028,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 6.9642,
|
|
"mean_token_accuracy": 0.10717165470123291,
|
|
"num_tokens": 608879.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.656619739532471,
|
|
"epoch": 0.29745808545159547,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.0022,
|
|
"mean_token_accuracy": 0.09820161312818527,
|
|
"num_tokens": 620439.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.668045330047607,
|
|
"epoch": 0.3028664142779881,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.0632,
|
|
"mean_token_accuracy": 0.10111433267593384,
|
|
"num_tokens": 632543.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.635419225692749,
|
|
"epoch": 0.30827474310438074,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.1026,
|
|
"mean_token_accuracy": 0.10522567853331566,
|
|
"num_tokens": 643369.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.584413766860962,
|
|
"epoch": 0.3136830719307734,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 6.9646,
|
|
"mean_token_accuracy": 0.1047127217054367,
|
|
"num_tokens": 655328.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.67089467048645,
|
|
"epoch": 0.319091400757166,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.0248,
|
|
"mean_token_accuracy": 0.10310934484004974,
|
|
"num_tokens": 665780.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.552512168884277,
|
|
"epoch": 0.32449972958355866,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.9293,
|
|
"mean_token_accuracy": 0.10721782594919205,
|
|
"num_tokens": 676566.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.559945297241211,
|
|
"epoch": 0.3299080584099513,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.9251,
|
|
"mean_token_accuracy": 0.1060689315199852,
|
|
"num_tokens": 687407.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.49097957611084,
|
|
"epoch": 0.335316387236344,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 6.8497,
|
|
"mean_token_accuracy": 0.11217133775353431,
|
|
"num_tokens": 698162.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.46426739692688,
|
|
"epoch": 0.34072471606273663,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.8316,
|
|
"mean_token_accuracy": 0.10524874627590179,
|
|
"num_tokens": 708743.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.558863353729248,
|
|
"epoch": 0.34613304488912927,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.0442,
|
|
"mean_token_accuracy": 0.11182164400815964,
|
|
"num_tokens": 719828.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.583437824249268,
|
|
"epoch": 0.3515413737155219,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.9937,
|
|
"mean_token_accuracy": 0.1039721980690956,
|
|
"num_tokens": 732514.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.449002027511597,
|
|
"epoch": 0.35694970254191455,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.8317,
|
|
"mean_token_accuracy": 0.11070109233260154,
|
|
"num_tokens": 742391.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.36940336227417,
|
|
"epoch": 0.3623580313683072,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.8335,
|
|
"mean_token_accuracy": 0.10628218725323677,
|
|
"num_tokens": 753638.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.5532920360565186,
|
|
"epoch": 0.3677663601946998,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.0485,
|
|
"mean_token_accuracy": 0.10574989169836044,
|
|
"num_tokens": 764850.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.4345468044281,
|
|
"epoch": 0.37317468902109246,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.8115,
|
|
"mean_token_accuracy": 0.11775769889354706,
|
|
"num_tokens": 776192.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.336619281768799,
|
|
"epoch": 0.3785830178474851,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.7691,
|
|
"mean_token_accuracy": 0.11614358723163605,
|
|
"num_tokens": 787051.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.326699829101562,
|
|
"epoch": 0.3839913466738778,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.7934,
|
|
"mean_token_accuracy": 0.11272966936230659,
|
|
"num_tokens": 799281.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.331569719314575,
|
|
"epoch": 0.38939967550027044,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.813,
|
|
"mean_token_accuracy": 0.11107392236590385,
|
|
"num_tokens": 810003.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.336561441421509,
|
|
"epoch": 0.3948080043266631,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.7821,
|
|
"mean_token_accuracy": 0.11116143241524697,
|
|
"num_tokens": 819749.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.329193019866944,
|
|
"epoch": 0.4002163331530557,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.7888,
|
|
"mean_token_accuracy": 0.11996297091245652,
|
|
"num_tokens": 830552.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.390970516204834,
|
|
"epoch": 0.40562466197944835,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.8901,
|
|
"mean_token_accuracy": 0.1121357448399067,
|
|
"num_tokens": 840495.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.275109004974365,
|
|
"epoch": 0.411032990805841,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.8114,
|
|
"mean_token_accuracy": 0.1169951744377613,
|
|
"num_tokens": 852915.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.377492761611938,
|
|
"epoch": 0.41644131963223363,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.7822,
|
|
"mean_token_accuracy": 0.11362324208021164,
|
|
"num_tokens": 862762.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.277709007263184,
|
|
"epoch": 0.42184964845862627,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.7274,
|
|
"mean_token_accuracy": 0.11747928410768509,
|
|
"num_tokens": 873023.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.28900933265686,
|
|
"epoch": 0.4272579772850189,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.8274,
|
|
"mean_token_accuracy": 0.10971275642514229,
|
|
"num_tokens": 885281.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.2772074222564695,
|
|
"epoch": 0.43266630611141155,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.8206,
|
|
"mean_token_accuracy": 0.11502003371715545,
|
|
"num_tokens": 896210.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.189112377166748,
|
|
"epoch": 0.43807463493780424,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.7468,
|
|
"mean_token_accuracy": 0.11514622867107391,
|
|
"num_tokens": 909028.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.229736566543579,
|
|
"epoch": 0.4434829637641969,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.6815,
|
|
"mean_token_accuracy": 0.11853564456105233,
|
|
"num_tokens": 919710.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.304251861572266,
|
|
"epoch": 0.4488912925905895,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.736,
|
|
"mean_token_accuracy": 0.11397984251379967,
|
|
"num_tokens": 930019.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.166671371459961,
|
|
"epoch": 0.45429962141698216,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.6524,
|
|
"mean_token_accuracy": 0.11951824501156808,
|
|
"num_tokens": 940535.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.0795973777771,
|
|
"epoch": 0.4597079502433748,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.6119,
|
|
"mean_token_accuracy": 0.12140549421310425,
|
|
"num_tokens": 951390.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.141013669967651,
|
|
"epoch": 0.46511627906976744,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.6636,
|
|
"mean_token_accuracy": 0.1212614968419075,
|
|
"num_tokens": 963010.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.112875986099243,
|
|
"epoch": 0.4705246078961601,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.6152,
|
|
"mean_token_accuracy": 0.12599240839481354,
|
|
"num_tokens": 974032.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.222216367721558,
|
|
"epoch": 0.4759329367225527,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.6586,
|
|
"mean_token_accuracy": 0.11954770833253861,
|
|
"num_tokens": 983954.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.11478705406189,
|
|
"epoch": 0.48134126554894535,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.689,
|
|
"mean_token_accuracy": 0.12196131646633149,
|
|
"num_tokens": 994135.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.173604154586792,
|
|
"epoch": 0.48674959437533805,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.6493,
|
|
"mean_token_accuracy": 0.12306504994630814,
|
|
"num_tokens": 1006275.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 6.988444137573242,
|
|
"epoch": 0.4921579232017307,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.4801,
|
|
"mean_token_accuracy": 0.129085473716259,
|
|
"num_tokens": 1017767.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.0636707782745365,
|
|
"epoch": 0.4975662520281233,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.6038,
|
|
"mean_token_accuracy": 0.1241903193295002,
|
|
"num_tokens": 1028286.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.059632682800293,
|
|
"epoch": 0.502974580854516,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.6151,
|
|
"mean_token_accuracy": 0.12029262557625771,
|
|
"num_tokens": 1040555.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.171442461013794,
|
|
"epoch": 0.5083829096809086,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.7175,
|
|
"mean_token_accuracy": 0.1209532730281353,
|
|
"num_tokens": 1053668.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.12706880569458,
|
|
"epoch": 0.5137912385073012,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.6597,
|
|
"mean_token_accuracy": 0.11960283890366555,
|
|
"num_tokens": 1066042.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 6.984622383117676,
|
|
"epoch": 0.5191995673336939,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.6701,
|
|
"mean_token_accuracy": 0.12059814557433128,
|
|
"num_tokens": 1077843.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.047416830062867,
|
|
"epoch": 0.5246078961600865,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.5605,
|
|
"mean_token_accuracy": 0.1252925641834736,
|
|
"num_tokens": 1089448.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.041959238052368,
|
|
"epoch": 0.5300162249864792,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.6615,
|
|
"mean_token_accuracy": 0.12150585278868675,
|
|
"num_tokens": 1100989.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.162778997421265,
|
|
"epoch": 0.5354245538128718,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.7004,
|
|
"mean_token_accuracy": 0.1256784491240978,
|
|
"num_tokens": 1113012.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.056042337417603,
|
|
"epoch": 0.5408328826392644,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.6356,
|
|
"mean_token_accuracy": 0.12231024205684662,
|
|
"num_tokens": 1123551.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.5408328826392644,
|
|
"eval_entropy": 7.0043568221005525,
|
|
"eval_loss": 6.780564308166504,
|
|
"eval_mean_token_accuracy": 0.12075253836810589,
|
|
"eval_num_tokens": 1123551.0,
|
|
"eval_runtime": 1.9941,
|
|
"eval_samples_per_second": 1762.721,
|
|
"eval_steps_per_second": 220.654,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.074879932403564,
|
|
"epoch": 0.5462412114656571,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.5659,
|
|
"mean_token_accuracy": 0.12208069637417793,
|
|
"num_tokens": 1133838.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 6.973394107818604,
|
|
"epoch": 0.5516495402920497,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.6027,
|
|
"mean_token_accuracy": 0.12259892076253891,
|
|
"num_tokens": 1146005.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.00295934677124,
|
|
"epoch": 0.5570578691184424,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.6885,
|
|
"mean_token_accuracy": 0.1244613103568554,
|
|
"num_tokens": 1157328.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.057001447677612,
|
|
"epoch": 0.5624661979448351,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.4842,
|
|
"mean_token_accuracy": 0.1322474293410778,
|
|
"num_tokens": 1168592.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.032931184768676,
|
|
"epoch": 0.5678745267712277,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.6161,
|
|
"mean_token_accuracy": 0.12481116801500321,
|
|
"num_tokens": 1178680.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 6.9781468391418455,
|
|
"epoch": 0.5732828555976204,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.5535,
|
|
"mean_token_accuracy": 0.127150696516037,
|
|
"num_tokens": 1188825.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.008031415939331,
|
|
"epoch": 0.578691184424013,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.5956,
|
|
"mean_token_accuracy": 0.12398400753736497,
|
|
"num_tokens": 1201375.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.072990751266479,
|
|
"epoch": 0.5840995132504057,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.7227,
|
|
"mean_token_accuracy": 0.11844412088394166,
|
|
"num_tokens": 1211915.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.957782316207886,
|
|
"epoch": 0.5895078420767983,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.6289,
|
|
"mean_token_accuracy": 0.12811450883746148,
|
|
"num_tokens": 1223151.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.041991472244263,
|
|
"epoch": 0.5949161709031909,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.5414,
|
|
"mean_token_accuracy": 0.13184184953570366,
|
|
"num_tokens": 1234229.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.023768568038941,
|
|
"epoch": 0.6003244997295836,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.5529,
|
|
"mean_token_accuracy": 0.13157471716403962,
|
|
"num_tokens": 1245298.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.8409104347229,
|
|
"epoch": 0.6057328285559762,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.5595,
|
|
"mean_token_accuracy": 0.12466495037078858,
|
|
"num_tokens": 1256154.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.922661924362183,
|
|
"epoch": 0.6111411573823688,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.4705,
|
|
"mean_token_accuracy": 0.129321701079607,
|
|
"num_tokens": 1267404.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.944332456588745,
|
|
"epoch": 0.6165494862087615,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.5185,
|
|
"mean_token_accuracy": 0.12794106900691987,
|
|
"num_tokens": 1278894.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.9090385913848875,
|
|
"epoch": 0.6219578150351541,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.502,
|
|
"mean_token_accuracy": 0.13496915996074677,
|
|
"num_tokens": 1290408.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.971935558319092,
|
|
"epoch": 0.6273661438615468,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.5806,
|
|
"mean_token_accuracy": 0.12681553587317468,
|
|
"num_tokens": 1301508.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.989264011383057,
|
|
"epoch": 0.6327744726879394,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.5447,
|
|
"mean_token_accuracy": 0.13189456537365912,
|
|
"num_tokens": 1312551.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.965842962265015,
|
|
"epoch": 0.638182801514332,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.6459,
|
|
"mean_token_accuracy": 0.12647571191191673,
|
|
"num_tokens": 1324876.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.939603042602539,
|
|
"epoch": 0.6435911303407247,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.531,
|
|
"mean_token_accuracy": 0.12691377103328705,
|
|
"num_tokens": 1335401.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.8328300476074215,
|
|
"epoch": 0.6489994591671173,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.4707,
|
|
"mean_token_accuracy": 0.127620816975832,
|
|
"num_tokens": 1348265.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.7745624542236325,
|
|
"epoch": 0.65440778799351,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.4306,
|
|
"mean_token_accuracy": 0.13703566640615464,
|
|
"num_tokens": 1359078.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.712151575088501,
|
|
"epoch": 0.6598161168199026,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.3318,
|
|
"mean_token_accuracy": 0.13108599931001663,
|
|
"num_tokens": 1369499.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.8127624034881595,
|
|
"epoch": 0.6652244456462953,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.4323,
|
|
"mean_token_accuracy": 0.1286877028644085,
|
|
"num_tokens": 1380329.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.824408864974975,
|
|
"epoch": 0.670632774472688,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.5116,
|
|
"mean_token_accuracy": 0.12648341283202172,
|
|
"num_tokens": 1393047.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.872823572158813,
|
|
"epoch": 0.6760411032990806,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.5067,
|
|
"mean_token_accuracy": 0.12922688201069832,
|
|
"num_tokens": 1405600.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.811132955551147,
|
|
"epoch": 0.6814494321254733,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.4417,
|
|
"mean_token_accuracy": 0.13407272696495057,
|
|
"num_tokens": 1416670.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.825539922714233,
|
|
"epoch": 0.6868577609518659,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.4599,
|
|
"mean_token_accuracy": 0.12846649512648584,
|
|
"num_tokens": 1428929.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.827125072479248,
|
|
"epoch": 0.6922660897782585,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.3887,
|
|
"mean_token_accuracy": 0.13251324519515037,
|
|
"num_tokens": 1439298.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.817316055297852,
|
|
"epoch": 0.6976744186046512,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.4844,
|
|
"mean_token_accuracy": 0.13544763550162314,
|
|
"num_tokens": 1449853.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.891280555725098,
|
|
"epoch": 0.7030827474310438,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.5515,
|
|
"mean_token_accuracy": 0.12688857838511466,
|
|
"num_tokens": 1460816.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.740885925292969,
|
|
"epoch": 0.7084910762574365,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.4014,
|
|
"mean_token_accuracy": 0.13262718617916108,
|
|
"num_tokens": 1473003.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.843165111541748,
|
|
"epoch": 0.7138994050838291,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.5206,
|
|
"mean_token_accuracy": 0.1229576200246811,
|
|
"num_tokens": 1484745.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.823391914367676,
|
|
"epoch": 0.7193077339102217,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.4011,
|
|
"mean_token_accuracy": 0.13305559009313583,
|
|
"num_tokens": 1495803.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.803869295120239,
|
|
"epoch": 0.7247160627366144,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.4092,
|
|
"mean_token_accuracy": 0.141731408983469,
|
|
"num_tokens": 1507504.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.852431774139404,
|
|
"epoch": 0.730124391563007,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.5033,
|
|
"mean_token_accuracy": 0.12678006961941718,
|
|
"num_tokens": 1518765.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.650970602035523,
|
|
"epoch": 0.7355327203893997,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.3839,
|
|
"mean_token_accuracy": 0.1324382781982422,
|
|
"num_tokens": 1529481.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.810516691207885,
|
|
"epoch": 0.7409410492157923,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.3986,
|
|
"mean_token_accuracy": 0.13725290223956108,
|
|
"num_tokens": 1539612.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.801828384399414,
|
|
"epoch": 0.7463493780421849,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.558,
|
|
"mean_token_accuracy": 0.12963267266750336,
|
|
"num_tokens": 1551370.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.8387431621551515,
|
|
"epoch": 0.7517577068685776,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.44,
|
|
"mean_token_accuracy": 0.1378513492643833,
|
|
"num_tokens": 1561662.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.7489661693573,
|
|
"epoch": 0.7571660356949702,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.4232,
|
|
"mean_token_accuracy": 0.13230135440826415,
|
|
"num_tokens": 1573479.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.756485414505005,
|
|
"epoch": 0.7625743645213628,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.366,
|
|
"mean_token_accuracy": 0.1389087036252022,
|
|
"num_tokens": 1583684.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.7295139789581295,
|
|
"epoch": 0.7679826933477556,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.4101,
|
|
"mean_token_accuracy": 0.13583654090762137,
|
|
"num_tokens": 1595694.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.696416282653809,
|
|
"epoch": 0.7733910221741482,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.4184,
|
|
"mean_token_accuracy": 0.13240900188684462,
|
|
"num_tokens": 1606104.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.7834718227386475,
|
|
"epoch": 0.7787993510005409,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.4803,
|
|
"mean_token_accuracy": 0.13445528894662856,
|
|
"num_tokens": 1617368.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.790578842163086,
|
|
"epoch": 0.7842076798269335,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.4893,
|
|
"mean_token_accuracy": 0.1305526874959469,
|
|
"num_tokens": 1630485.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.788717746734619,
|
|
"epoch": 0.7896160086533262,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.3357,
|
|
"mean_token_accuracy": 0.1369060456752777,
|
|
"num_tokens": 1641411.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.644323205947876,
|
|
"epoch": 0.7950243374797188,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.3465,
|
|
"mean_token_accuracy": 0.13824027627706528,
|
|
"num_tokens": 1653737.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.823546600341797,
|
|
"epoch": 0.8004326663061114,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.5281,
|
|
"mean_token_accuracy": 0.12467064261436463,
|
|
"num_tokens": 1665910.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.726630735397339,
|
|
"epoch": 0.8058409951325041,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.3094,
|
|
"mean_token_accuracy": 0.134908963739872,
|
|
"num_tokens": 1676927.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.629592800140381,
|
|
"epoch": 0.8112493239588967,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.357,
|
|
"mean_token_accuracy": 0.13599003627896308,
|
|
"num_tokens": 1688167.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.650397443771363,
|
|
"epoch": 0.8166576527852893,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.4308,
|
|
"mean_token_accuracy": 0.13610320165753365,
|
|
"num_tokens": 1699899.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.7162903308868405,
|
|
"epoch": 0.822065981611682,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.3533,
|
|
"mean_token_accuracy": 0.1412688173353672,
|
|
"num_tokens": 1709862.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.640978765487671,
|
|
"epoch": 0.8274743104380746,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.3481,
|
|
"mean_token_accuracy": 0.1376900367438793,
|
|
"num_tokens": 1719902.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.691372346878052,
|
|
"epoch": 0.8328826392644673,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.3541,
|
|
"mean_token_accuracy": 0.13510108962655068,
|
|
"num_tokens": 1730351.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.589955043792725,
|
|
"epoch": 0.8382909680908599,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.3233,
|
|
"mean_token_accuracy": 0.1341069035232067,
|
|
"num_tokens": 1741783.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.609476327896118,
|
|
"epoch": 0.8436992969172525,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.3639,
|
|
"mean_token_accuracy": 0.13918454721570014,
|
|
"num_tokens": 1752952.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.713692474365234,
|
|
"epoch": 0.8491076257436452,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.3796,
|
|
"mean_token_accuracy": 0.1377579912543297,
|
|
"num_tokens": 1764014.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.692055416107178,
|
|
"epoch": 0.8545159545700378,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.3501,
|
|
"mean_token_accuracy": 0.13647191673517228,
|
|
"num_tokens": 1776154.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.651118326187134,
|
|
"epoch": 0.8599242833964305,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.405,
|
|
"mean_token_accuracy": 0.1311624176800251,
|
|
"num_tokens": 1787292.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.63629379272461,
|
|
"epoch": 0.8653326122228231,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.3667,
|
|
"mean_token_accuracy": 0.13610742837190629,
|
|
"num_tokens": 1799469.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.567807674407959,
|
|
"epoch": 0.8707409410492158,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.2441,
|
|
"mean_token_accuracy": 0.14459661692380904,
|
|
"num_tokens": 1810091.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.723603248596191,
|
|
"epoch": 0.8761492698756085,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.3925,
|
|
"mean_token_accuracy": 0.14025203064084052,
|
|
"num_tokens": 1820835.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.633598756790161,
|
|
"epoch": 0.8815575987020011,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.3549,
|
|
"mean_token_accuracy": 0.13814282566308975,
|
|
"num_tokens": 1831868.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.587992763519287,
|
|
"epoch": 0.8869659275283938,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.3383,
|
|
"mean_token_accuracy": 0.13848361223936081,
|
|
"num_tokens": 1842040.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.723627090454102,
|
|
"epoch": 0.8923742563547864,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.4083,
|
|
"mean_token_accuracy": 0.13382988572120666,
|
|
"num_tokens": 1853765.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.673075199127197,
|
|
"epoch": 0.897782585181179,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.4141,
|
|
"mean_token_accuracy": 0.13732750713825226,
|
|
"num_tokens": 1866081.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.593461990356445,
|
|
"epoch": 0.9031909140075717,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.3155,
|
|
"mean_token_accuracy": 0.13803164809942245,
|
|
"num_tokens": 1876754.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.714284658432007,
|
|
"epoch": 0.9085992428339643,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.4263,
|
|
"mean_token_accuracy": 0.1334732621908188,
|
|
"num_tokens": 1888299.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.475223159790039,
|
|
"epoch": 0.914007571660357,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.2122,
|
|
"mean_token_accuracy": 0.1465678930282593,
|
|
"num_tokens": 1899027.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.625812816619873,
|
|
"epoch": 0.9194159004867496,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.3374,
|
|
"mean_token_accuracy": 0.14080506712198257,
|
|
"num_tokens": 1911043.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.590130615234375,
|
|
"epoch": 0.9248242293131422,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.2567,
|
|
"mean_token_accuracy": 0.13707533478736877,
|
|
"num_tokens": 1921295.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.61051025390625,
|
|
"epoch": 0.9302325581395349,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.3569,
|
|
"mean_token_accuracy": 0.13993956744670868,
|
|
"num_tokens": 1932944.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.553638172149658,
|
|
"epoch": 0.9356408869659275,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.3027,
|
|
"mean_token_accuracy": 0.13260096460580825,
|
|
"num_tokens": 1944072.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.569744491577149,
|
|
"epoch": 0.9410492157923201,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.2956,
|
|
"mean_token_accuracy": 0.13898471891880035,
|
|
"num_tokens": 1954798.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.513964128494263,
|
|
"epoch": 0.9464575446187128,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.2698,
|
|
"mean_token_accuracy": 0.1473902389407158,
|
|
"num_tokens": 1965304.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.6614869117736815,
|
|
"epoch": 0.9518658734451054,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.3094,
|
|
"mean_token_accuracy": 0.13569752722978592,
|
|
"num_tokens": 1976797.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.550220966339111,
|
|
"epoch": 0.9572742022714981,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.3248,
|
|
"mean_token_accuracy": 0.13910794109106064,
|
|
"num_tokens": 1987448.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.369886207580566,
|
|
"epoch": 0.9626825310978907,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.1508,
|
|
"mean_token_accuracy": 0.1404670625925064,
|
|
"num_tokens": 2000948.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.55863881111145,
|
|
"epoch": 0.9680908599242833,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.2643,
|
|
"mean_token_accuracy": 0.14369212165474893,
|
|
"num_tokens": 2011807.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.6163318157196045,
|
|
"epoch": 0.9734991887506761,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.3207,
|
|
"mean_token_accuracy": 0.13593853637576103,
|
|
"num_tokens": 2022835.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.3906701564788815,
|
|
"epoch": 0.9789075175770687,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.2339,
|
|
"mean_token_accuracy": 0.14182498455047607,
|
|
"num_tokens": 2034331.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.662575387954712,
|
|
"epoch": 0.9843158464034614,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.2328,
|
|
"mean_token_accuracy": 0.13782825246453284,
|
|
"num_tokens": 2047002.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.408291864395141,
|
|
"epoch": 0.989724175229854,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.1835,
|
|
"mean_token_accuracy": 0.14051275625824927,
|
|
"num_tokens": 2056299.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.519381284713745,
|
|
"epoch": 0.9951325040562466,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.3108,
|
|
"mean_token_accuracy": 0.13649527579545975,
|
|
"num_tokens": 2068031.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.533435503641765,
|
|
"epoch": 1.0,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.2559,
|
|
"mean_token_accuracy": 0.1411174734433492,
|
|
"num_tokens": 2076889.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.518666791915893,
|
|
"epoch": 1.0054083288263926,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.1187,
|
|
"mean_token_accuracy": 0.14975121915340422,
|
|
"num_tokens": 2087316.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.499506664276123,
|
|
"epoch": 1.0108166576527853,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.1223,
|
|
"mean_token_accuracy": 0.14297691583633423,
|
|
"num_tokens": 2098522.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.383229446411133,
|
|
"epoch": 1.016224986479178,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.0878,
|
|
"mean_token_accuracy": 0.15435394942760466,
|
|
"num_tokens": 2110015.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.525584554672241,
|
|
"epoch": 1.0216333153055706,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.1924,
|
|
"mean_token_accuracy": 0.1424684338271618,
|
|
"num_tokens": 2121333.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.4724767208099365,
|
|
"epoch": 1.0270416441319632,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.0838,
|
|
"mean_token_accuracy": 0.14994525909423828,
|
|
"num_tokens": 2131385.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.488972854614258,
|
|
"epoch": 1.0324499729583558,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.1275,
|
|
"mean_token_accuracy": 0.1522180952131748,
|
|
"num_tokens": 2143274.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.368765354156494,
|
|
"epoch": 1.0378583017847485,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.0124,
|
|
"mean_token_accuracy": 0.15076057612895966,
|
|
"num_tokens": 2154529.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.4038872718811035,
|
|
"epoch": 1.043266630611141,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.1603,
|
|
"mean_token_accuracy": 0.1439424842596054,
|
|
"num_tokens": 2165949.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.407582426071167,
|
|
"epoch": 1.0486749594375337,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.1335,
|
|
"mean_token_accuracy": 0.1485764428973198,
|
|
"num_tokens": 2176710.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.463790941238403,
|
|
"epoch": 1.0540832882639264,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.1074,
|
|
"mean_token_accuracy": 0.15412099212408065,
|
|
"num_tokens": 2189470.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.383792352676392,
|
|
"epoch": 1.059491617090319,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.0844,
|
|
"mean_token_accuracy": 0.1579994946718216,
|
|
"num_tokens": 2200266.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.395930194854737,
|
|
"epoch": 1.0648999459167117,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.0915,
|
|
"mean_token_accuracy": 0.14566799476742745,
|
|
"num_tokens": 2210736.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.478207111358643,
|
|
"epoch": 1.0703082747431043,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.1985,
|
|
"mean_token_accuracy": 0.14791767448186874,
|
|
"num_tokens": 2221057.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.455506134033203,
|
|
"epoch": 1.075716603569497,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.1738,
|
|
"mean_token_accuracy": 0.14685477688908577,
|
|
"num_tokens": 2232352.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.510460042953492,
|
|
"epoch": 1.0811249323958896,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.0805,
|
|
"mean_token_accuracy": 0.15254651606082917,
|
|
"num_tokens": 2244655.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 1.0811249323958896,
|
|
"eval_entropy": 6.124205052852631,
|
|
"eval_loss": 6.387869834899902,
|
|
"eval_mean_token_accuracy": 0.1444568099030717,
|
|
"eval_num_tokens": 2244655.0,
|
|
"eval_runtime": 1.9538,
|
|
"eval_samples_per_second": 1799.022,
|
|
"eval_steps_per_second": 225.198,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.2928147315979,
|
|
"epoch": 1.0865332612222822,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999997383518051,
|
|
"loss": 6.0382,
|
|
"mean_token_accuracy": 0.1600428655743599,
|
|
"num_tokens": 2255699.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.372293472290039,
|
|
"epoch": 1.0919415900486749,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999986754070562,
|
|
"loss": 6.0381,
|
|
"mean_token_accuracy": 0.15901420414447784,
|
|
"num_tokens": 2266627.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.441066265106201,
|
|
"epoch": 1.0973499188750675,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004999967948166009,
|
|
"loss": 6.2142,
|
|
"mean_token_accuracy": 0.14456040039658546,
|
|
"num_tokens": 2277814.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.525017118453979,
|
|
"epoch": 1.1027582477014604,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999940965872734,
|
|
"loss": 6.1616,
|
|
"mean_token_accuracy": 0.14813328310847282,
|
|
"num_tokens": 2287344.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.212237882614136,
|
|
"epoch": 1.1081665765278528,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999905807288788,
|
|
"loss": 5.9498,
|
|
"mean_token_accuracy": 0.1576882466673851,
|
|
"num_tokens": 2298332.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.400769948959351,
|
|
"epoch": 1.1135749053542456,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000499986247254194,
|
|
"loss": 6.0778,
|
|
"mean_token_accuracy": 0.15378758013248445,
|
|
"num_tokens": 2309161.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.346611022949219,
|
|
"epoch": 1.1189832341806383,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004999810961789667,
|
|
"loss": 6.0663,
|
|
"mean_token_accuracy": 0.15356869399547576,
|
|
"num_tokens": 2319516.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.348343276977539,
|
|
"epoch": 1.124391563007031,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000499975127521916,
|
|
"loss": 5.9687,
|
|
"mean_token_accuracy": 0.15849437564611435,
|
|
"num_tokens": 2330441.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.287055778503418,
|
|
"epoch": 1.1297998918334236,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999683413047319,
|
|
"loss": 5.9837,
|
|
"mean_token_accuracy": 0.15973089039325714,
|
|
"num_tokens": 2340524.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.308336925506592,
|
|
"epoch": 1.1352082206598162,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004999607375520759,
|
|
"loss": 5.981,
|
|
"mean_token_accuracy": 0.1552950382232666,
|
|
"num_tokens": 2353379.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.3794340133667,
|
|
"epoch": 1.1406165494862088,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999523162915794,
|
|
"loss": 6.1602,
|
|
"mean_token_accuracy": 0.1466098792850971,
|
|
"num_tokens": 2365708.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.301113653182983,
|
|
"epoch": 1.1460248783126015,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999430775538459,
|
|
"loss": 6.0682,
|
|
"mean_token_accuracy": 0.1489157885313034,
|
|
"num_tokens": 2375630.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.2834553718566895,
|
|
"epoch": 1.151433207138994,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999330213724487,
|
|
"loss": 6.0778,
|
|
"mean_token_accuracy": 0.14907022714614868,
|
|
"num_tokens": 2386885.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.3159809589385985,
|
|
"epoch": 1.1568415359653867,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999221477839319,
|
|
"loss": 6.0987,
|
|
"mean_token_accuracy": 0.1497051790356636,
|
|
"num_tokens": 2397676.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.429526996612549,
|
|
"epoch": 1.1622498647917794,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004999104568278104,
|
|
"loss": 6.1048,
|
|
"mean_token_accuracy": 0.14880598485469818,
|
|
"num_tokens": 2409039.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.388683223724366,
|
|
"epoch": 1.167658193618172,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000499897948546569,
|
|
"loss": 6.1315,
|
|
"mean_token_accuracy": 0.15224068611860275,
|
|
"num_tokens": 2420103.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.3003730297088625,
|
|
"epoch": 1.1730665224445647,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004998846229856629,
|
|
"loss": 6.0491,
|
|
"mean_token_accuracy": 0.15122028589248657,
|
|
"num_tokens": 2430792.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.4121685981750485,
|
|
"epoch": 1.1784748512709573,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998704801935171,
|
|
"loss": 6.0602,
|
|
"mean_token_accuracy": 0.15023664385080338,
|
|
"num_tokens": 2442165.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.261796760559082,
|
|
"epoch": 1.18388318009735,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004998555202215267,
|
|
"loss": 5.9712,
|
|
"mean_token_accuracy": 0.15477441847324372,
|
|
"num_tokens": 2454231.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.338596963882447,
|
|
"epoch": 1.1892915089237426,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004998397431240561,
|
|
"loss": 6.0251,
|
|
"mean_token_accuracy": 0.15698925256729127,
|
|
"num_tokens": 2464828.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.240490484237671,
|
|
"epoch": 1.1946998377501352,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004998231489584395,
|
|
"loss": 6.0689,
|
|
"mean_token_accuracy": 0.14833179712295533,
|
|
"num_tokens": 2476144.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.455478239059448,
|
|
"epoch": 1.2001081665765279,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00049980573778498,
|
|
"loss": 6.2395,
|
|
"mean_token_accuracy": 0.1458302244544029,
|
|
"num_tokens": 2489116.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.402680921554565,
|
|
"epoch": 1.2055164954029205,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004997875096669498,
|
|
"loss": 6.1098,
|
|
"mean_token_accuracy": 0.15058583766222,
|
|
"num_tokens": 2501609.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.230094385147095,
|
|
"epoch": 1.2109248242293131,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004997684646705901,
|
|
"loss": 6.0114,
|
|
"mean_token_accuracy": 0.15897049903869628,
|
|
"num_tokens": 2513338.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.354293441772461,
|
|
"epoch": 1.2163331530557058,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004997486028651103,
|
|
"loss": 6.0383,
|
|
"mean_token_accuracy": 0.16039613038301467,
|
|
"num_tokens": 2523199.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.349422979354858,
|
|
"epoch": 1.2217414818820984,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004997279243226884,
|
|
"loss": 6.0512,
|
|
"mean_token_accuracy": 0.15354744642972945,
|
|
"num_tokens": 2535136.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.264004421234131,
|
|
"epoch": 1.227149810708491,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004997064291184701,
|
|
"loss": 5.9657,
|
|
"mean_token_accuracy": 0.14787945449352263,
|
|
"num_tokens": 2546428.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.358812093734741,
|
|
"epoch": 1.2325581395348837,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004996841173305691,
|
|
"loss": 6.1636,
|
|
"mean_token_accuracy": 0.15003497749567032,
|
|
"num_tokens": 2558304.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.174890947341919,
|
|
"epoch": 1.2379664683612763,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004996609890400665,
|
|
"loss": 5.848,
|
|
"mean_token_accuracy": 0.16567099541425706,
|
|
"num_tokens": 2568717.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.297175502777099,
|
|
"epoch": 1.243374797187669,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004996370443310107,
|
|
"loss": 6.0533,
|
|
"mean_token_accuracy": 0.14965572953224182,
|
|
"num_tokens": 2578748.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.327052450180053,
|
|
"epoch": 1.2487831260140616,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004996122832904165,
|
|
"loss": 6.0212,
|
|
"mean_token_accuracy": 0.1520503729581833,
|
|
"num_tokens": 2590095.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.300268507003784,
|
|
"epoch": 1.2541914548404542,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004995867060082658,
|
|
"loss": 6.0529,
|
|
"mean_token_accuracy": 0.15308669656515123,
|
|
"num_tokens": 2600929.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.392643833160401,
|
|
"epoch": 1.2595997836668469,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004995603125775065,
|
|
"loss": 6.0596,
|
|
"mean_token_accuracy": 0.15196561813354492,
|
|
"num_tokens": 2611830.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.176296854019165,
|
|
"epoch": 1.2650081124932395,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004995331030940523,
|
|
"loss": 6.0102,
|
|
"mean_token_accuracy": 0.1498885288834572,
|
|
"num_tokens": 2623177.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.219478797912598,
|
|
"epoch": 1.2704164413196322,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004995050776567826,
|
|
"loss": 6.0001,
|
|
"mean_token_accuracy": 0.15188704282045365,
|
|
"num_tokens": 2635563.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.3707170486450195,
|
|
"epoch": 1.2758247701460248,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004994762363675421,
|
|
"loss": 5.9911,
|
|
"mean_token_accuracy": 0.15853488594293594,
|
|
"num_tokens": 2647157.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.177279806137085,
|
|
"epoch": 1.2812330989724177,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004994465793311396,
|
|
"loss": 5.9629,
|
|
"mean_token_accuracy": 0.16011573225259781,
|
|
"num_tokens": 2658175.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.249725675582885,
|
|
"epoch": 1.28664142779881,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004994161066553493,
|
|
"loss": 5.9191,
|
|
"mean_token_accuracy": 0.1570979908108711,
|
|
"num_tokens": 2669458.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.347487926483154,
|
|
"epoch": 1.292049756625203,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004993848184509086,
|
|
"loss": 6.1223,
|
|
"mean_token_accuracy": 0.1580820508301258,
|
|
"num_tokens": 2679727.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.262456750869751,
|
|
"epoch": 1.2974580854515954,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004993527148315193,
|
|
"loss": 6.092,
|
|
"mean_token_accuracy": 0.15162299424409867,
|
|
"num_tokens": 2691031.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.2036655902862545,
|
|
"epoch": 1.3028664142779882,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004993197959138458,
|
|
"loss": 5.8833,
|
|
"mean_token_accuracy": 0.16101625710725784,
|
|
"num_tokens": 2701699.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.183501052856445,
|
|
"epoch": 1.3082747431043806,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004992860618175156,
|
|
"loss": 5.9299,
|
|
"mean_token_accuracy": 0.15750151723623276,
|
|
"num_tokens": 2713038.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.1955572128295895,
|
|
"epoch": 1.3136830719307735,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004992515126651184,
|
|
"loss": 5.9293,
|
|
"mean_token_accuracy": 0.1570618689060211,
|
|
"num_tokens": 2723685.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.195014238357544,
|
|
"epoch": 1.319091400757166,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004992161485822059,
|
|
"loss": 6.0006,
|
|
"mean_token_accuracy": 0.1587825059890747,
|
|
"num_tokens": 2735341.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.2399842739105225,
|
|
"epoch": 1.3244997295835588,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004991799696972913,
|
|
"loss": 5.9313,
|
|
"mean_token_accuracy": 0.16518269777297973,
|
|
"num_tokens": 2746052.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.166414260864258,
|
|
"epoch": 1.3299080584099512,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004991429761418488,
|
|
"loss": 5.9363,
|
|
"mean_token_accuracy": 0.15754909813404083,
|
|
"num_tokens": 2757155.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.239412260055542,
|
|
"epoch": 1.335316387236344,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000499105168050313,
|
|
"loss": 5.9929,
|
|
"mean_token_accuracy": 0.15189812183380128,
|
|
"num_tokens": 2769631.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.324047040939331,
|
|
"epoch": 1.3407247160627367,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004990665455600787,
|
|
"loss": 6.0868,
|
|
"mean_token_accuracy": 0.1567138358950615,
|
|
"num_tokens": 2780797.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.166640758514404,
|
|
"epoch": 1.3461330448891293,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004990271088114999,
|
|
"loss": 5.946,
|
|
"mean_token_accuracy": 0.15337988138198852,
|
|
"num_tokens": 2792721.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.2248719215393065,
|
|
"epoch": 1.351541373715522,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004989868579478901,
|
|
"loss": 6.0004,
|
|
"mean_token_accuracy": 0.16802925169467925,
|
|
"num_tokens": 2804531.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.296425151824951,
|
|
"epoch": 1.3569497025419146,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004989457931155209,
|
|
"loss": 5.9965,
|
|
"mean_token_accuracy": 0.15395874083042144,
|
|
"num_tokens": 2816666.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.195201396942139,
|
|
"epoch": 1.3623580313683072,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004989039144636219,
|
|
"loss": 6.0242,
|
|
"mean_token_accuracy": 0.15923022627830505,
|
|
"num_tokens": 2827977.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.20063362121582,
|
|
"epoch": 1.3677663601946999,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004988612221443804,
|
|
"loss": 5.9936,
|
|
"mean_token_accuracy": 0.15627091377973557,
|
|
"num_tokens": 2839069.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.340277481079101,
|
|
"epoch": 1.3731746890210925,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004988177163129403,
|
|
"loss": 6.0808,
|
|
"mean_token_accuracy": 0.15119716450572013,
|
|
"num_tokens": 2850632.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.207772541046142,
|
|
"epoch": 1.3785830178474852,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004987733971274018,
|
|
"loss": 5.9607,
|
|
"mean_token_accuracy": 0.15018116384744645,
|
|
"num_tokens": 2862204.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.178734588623047,
|
|
"epoch": 1.3839913466738778,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004987282647488211,
|
|
"loss": 5.9197,
|
|
"mean_token_accuracy": 0.1660077005624771,
|
|
"num_tokens": 2874010.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.195025396347046,
|
|
"epoch": 1.3893996755002704,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004986823193412093,
|
|
"loss": 5.9763,
|
|
"mean_token_accuracy": 0.16176897138357163,
|
|
"num_tokens": 2884315.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.326874685287476,
|
|
"epoch": 1.394808004326663,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000498635561071532,
|
|
"loss": 6.133,
|
|
"mean_token_accuracy": 0.15456459820270538,
|
|
"num_tokens": 2896327.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.183460760116577,
|
|
"epoch": 1.4002163331530557,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000498587990109709,
|
|
"loss": 5.913,
|
|
"mean_token_accuracy": 0.16137443631887435,
|
|
"num_tokens": 2907007.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.183216190338134,
|
|
"epoch": 1.4056246619794484,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004985396066286133,
|
|
"loss": 5.9469,
|
|
"mean_token_accuracy": 0.16008795648813248,
|
|
"num_tokens": 2918852.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.222980594635009,
|
|
"epoch": 1.411032990805841,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004984904108040706,
|
|
"loss": 5.86,
|
|
"mean_token_accuracy": 0.16177986562252045,
|
|
"num_tokens": 2929948.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.102756023406982,
|
|
"epoch": 1.4164413196322336,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004984404028148585,
|
|
"loss": 5.9403,
|
|
"mean_token_accuracy": 0.15873561054468155,
|
|
"num_tokens": 2941299.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.112520980834961,
|
|
"epoch": 1.4218496484586263,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004983895828427062,
|
|
"loss": 5.8676,
|
|
"mean_token_accuracy": 0.15320820808410646,
|
|
"num_tokens": 2952283.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.400298118591309,
|
|
"epoch": 1.427257977285019,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004983379510722935,
|
|
"loss": 6.2265,
|
|
"mean_token_accuracy": 0.15123522654175758,
|
|
"num_tokens": 2962241.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.251223230361939,
|
|
"epoch": 1.4326663061114115,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004982855076912506,
|
|
"loss": 5.9829,
|
|
"mean_token_accuracy": 0.154976050555706,
|
|
"num_tokens": 2973408.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.191674470901489,
|
|
"epoch": 1.4380746349378042,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004982322528901568,
|
|
"loss": 6.0012,
|
|
"mean_token_accuracy": 0.15962311923503875,
|
|
"num_tokens": 2985877.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.265266704559326,
|
|
"epoch": 1.4434829637641968,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00049817818686254,
|
|
"loss": 6.0082,
|
|
"mean_token_accuracy": 0.15608256608247756,
|
|
"num_tokens": 2996935.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.144498634338379,
|
|
"epoch": 1.4488912925905895,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004981233098048765,
|
|
"loss": 5.8997,
|
|
"mean_token_accuracy": 0.16039617508649825,
|
|
"num_tokens": 3007932.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.250757074356079,
|
|
"epoch": 1.454299621416982,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004980676219165894,
|
|
"loss": 5.9958,
|
|
"mean_token_accuracy": 0.156630976498127,
|
|
"num_tokens": 3019368.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.234979820251465,
|
|
"epoch": 1.4597079502433747,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000498011123400049,
|
|
"loss": 6.0473,
|
|
"mean_token_accuracy": 0.15403905063867568,
|
|
"num_tokens": 3029787.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.161543560028076,
|
|
"epoch": 1.4651162790697674,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004979538144605707,
|
|
"loss": 5.994,
|
|
"mean_token_accuracy": 0.15943700075149536,
|
|
"num_tokens": 3042100.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.1883018016815186,
|
|
"epoch": 1.47052460789616,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004978956953064154,
|
|
"loss": 6.0228,
|
|
"mean_token_accuracy": 0.15444456189870834,
|
|
"num_tokens": 3054160.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.235822486877441,
|
|
"epoch": 1.4759329367225527,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004978367661487883,
|
|
"loss": 5.9728,
|
|
"mean_token_accuracy": 0.15552034378051757,
|
|
"num_tokens": 3065401.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.124010753631592,
|
|
"epoch": 1.4813412655489453,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004977770272018379,
|
|
"loss": 5.915,
|
|
"mean_token_accuracy": 0.1574919804930687,
|
|
"num_tokens": 3076137.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.223858880996704,
|
|
"epoch": 1.486749594375338,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000497716478682656,
|
|
"loss": 6.0458,
|
|
"mean_token_accuracy": 0.15904476791620253,
|
|
"num_tokens": 3087274.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.195795488357544,
|
|
"epoch": 1.4921579232017308,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004976551208112759,
|
|
"loss": 5.9187,
|
|
"mean_token_accuracy": 0.16271037310361863,
|
|
"num_tokens": 3098476.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.184743976593017,
|
|
"epoch": 1.4975662520281232,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004975929538106721,
|
|
"loss": 5.9511,
|
|
"mean_token_accuracy": 0.15797366350889205,
|
|
"num_tokens": 3110060.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.053498268127441,
|
|
"epoch": 1.502974580854516,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004975299779067598,
|
|
"loss": 5.8576,
|
|
"mean_token_accuracy": 0.1606665775179863,
|
|
"num_tokens": 3121032.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.189790344238281,
|
|
"epoch": 1.5083829096809085,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004974661933283936,
|
|
"loss": 5.8774,
|
|
"mean_token_accuracy": 0.16694698929786683,
|
|
"num_tokens": 3131999.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.027636241912842,
|
|
"epoch": 1.5137912385073014,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004974016003073667,
|
|
"loss": 5.8154,
|
|
"mean_token_accuracy": 0.1604975640773773,
|
|
"num_tokens": 3142693.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.085386133193969,
|
|
"epoch": 1.5191995673336938,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004973361990784104,
|
|
"loss": 5.767,
|
|
"mean_token_accuracy": 0.16097440123558043,
|
|
"num_tokens": 3154179.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.1582213878631595,
|
|
"epoch": 1.5246078961600866,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004972699898791927,
|
|
"loss": 5.857,
|
|
"mean_token_accuracy": 0.16564760953187943,
|
|
"num_tokens": 3165314.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.119766902923584,
|
|
"epoch": 1.530016224986479,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004972029729503182,
|
|
"loss": 5.9687,
|
|
"mean_token_accuracy": 0.1669076904654503,
|
|
"num_tokens": 3177646.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.218665695190429,
|
|
"epoch": 1.535424553812872,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004971351485353265,
|
|
"loss": 5.9586,
|
|
"mean_token_accuracy": 0.1630210280418396,
|
|
"num_tokens": 3188401.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.0617822170257565,
|
|
"epoch": 1.5408328826392643,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004970665168806917,
|
|
"loss": 5.8469,
|
|
"mean_token_accuracy": 0.15670444518327714,
|
|
"num_tokens": 3200206.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.157464456558228,
|
|
"epoch": 1.5462412114656572,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004969970782358214,
|
|
"loss": 5.9625,
|
|
"mean_token_accuracy": 0.1585618481040001,
|
|
"num_tokens": 3211981.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.088303804397583,
|
|
"epoch": 1.5516495402920496,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004969268328530557,
|
|
"loss": 5.976,
|
|
"mean_token_accuracy": 0.1560826152563095,
|
|
"num_tokens": 3221997.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.2379875659942625,
|
|
"epoch": 1.5570578691184425,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004968557809876666,
|
|
"loss": 5.9227,
|
|
"mean_token_accuracy": 0.15799657702445985,
|
|
"num_tokens": 3234333.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.089317321777344,
|
|
"epoch": 1.562466197944835,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004967839228978568,
|
|
"loss": 5.9558,
|
|
"mean_token_accuracy": 0.15238141193985938,
|
|
"num_tokens": 3245083.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.162514019012451,
|
|
"epoch": 1.5678745267712277,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004967112588447585,
|
|
"loss": 5.7954,
|
|
"mean_token_accuracy": 0.16877318769693375,
|
|
"num_tokens": 3255787.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.064874601364136,
|
|
"epoch": 1.5732828555976204,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004966377890924332,
|
|
"loss": 5.9603,
|
|
"mean_token_accuracy": 0.16299306154251098,
|
|
"num_tokens": 3267793.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.110070848464966,
|
|
"epoch": 1.578691184424013,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004965635139078704,
|
|
"loss": 5.839,
|
|
"mean_token_accuracy": 0.16499156802892684,
|
|
"num_tokens": 3279100.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.144229125976563,
|
|
"epoch": 1.5840995132504057,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004964884335609857,
|
|
"loss": 6.0038,
|
|
"mean_token_accuracy": 0.16165205985307693,
|
|
"num_tokens": 3290251.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.1137713432312015,
|
|
"epoch": 1.5895078420767983,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004964125483246218,
|
|
"loss": 5.8644,
|
|
"mean_token_accuracy": 0.16258378326892853,
|
|
"num_tokens": 3301817.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.197260141372681,
|
|
"epoch": 1.594916170903191,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004963358584745457,
|
|
"loss": 5.9241,
|
|
"mean_token_accuracy": 0.1570972815155983,
|
|
"num_tokens": 3313435.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.134345865249633,
|
|
"epoch": 1.6003244997295836,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004962583642894482,
|
|
"loss": 6.0332,
|
|
"mean_token_accuracy": 0.15797800868749617,
|
|
"num_tokens": 3325489.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.202575826644898,
|
|
"epoch": 1.6057328285559762,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004961800660509437,
|
|
"loss": 5.9061,
|
|
"mean_token_accuracy": 0.16250776648521423,
|
|
"num_tokens": 3337030.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.009425020217895,
|
|
"epoch": 1.6111411573823688,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000496100964043568,
|
|
"loss": 5.8656,
|
|
"mean_token_accuracy": 0.16609732657670975,
|
|
"num_tokens": 3347766.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.021034860610962,
|
|
"epoch": 1.6165494862087615,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000496021058554778,
|
|
"loss": 5.7761,
|
|
"mean_token_accuracy": 0.16601542234420777,
|
|
"num_tokens": 3360379.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.08537859916687,
|
|
"epoch": 1.6219578150351541,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004959403498749505,
|
|
"loss": 5.9257,
|
|
"mean_token_accuracy": 0.16757240146398544,
|
|
"num_tokens": 3370962.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.6219578150351541,
|
|
"eval_entropy": 6.020744280381636,
|
|
"eval_loss": 6.140914440155029,
|
|
"eval_mean_token_accuracy": 0.15917101580310952,
|
|
"eval_num_tokens": 3370962.0,
|
|
"eval_runtime": 1.9427,
|
|
"eval_samples_per_second": 1809.294,
|
|
"eval_steps_per_second": 226.483,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.164214992523194,
|
|
"epoch": 1.6273661438615468,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004958588382973809,
|
|
"loss": 5.8379,
|
|
"mean_token_accuracy": 0.16315794885158538,
|
|
"num_tokens": 3382648.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.041830253601074,
|
|
"epoch": 1.6327744726879394,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004957765241182826,
|
|
"loss": 5.9036,
|
|
"mean_token_accuracy": 0.16003947407007219,
|
|
"num_tokens": 3393871.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.033273839950562,
|
|
"epoch": 1.638182801514332,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004956934076367854,
|
|
"loss": 5.7917,
|
|
"mean_token_accuracy": 0.16898759454488754,
|
|
"num_tokens": 3405148.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.072603559494018,
|
|
"epoch": 1.6435911303407247,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000495609489154935,
|
|
"loss": 5.8284,
|
|
"mean_token_accuracy": 0.17306120544672013,
|
|
"num_tokens": 3416647.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.115102910995484,
|
|
"epoch": 1.6489994591671173,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004955247689776911,
|
|
"loss": 5.9205,
|
|
"mean_token_accuracy": 0.15651903450489044,
|
|
"num_tokens": 3428331.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 5.937150382995606,
|
|
"epoch": 1.65440778799351,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000495439247412927,
|
|
"loss": 5.776,
|
|
"mean_token_accuracy": 0.16460416913032533,
|
|
"num_tokens": 3440683.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.128424644470215,
|
|
"epoch": 1.6598161168199026,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004953529247714285,
|
|
"loss": 5.7777,
|
|
"mean_token_accuracy": 0.16499500274658202,
|
|
"num_tokens": 3452375.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.042714262008667,
|
|
"epoch": 1.6652244456462952,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004952658013668922,
|
|
"loss": 5.9321,
|
|
"mean_token_accuracy": 0.15212839990854263,
|
|
"num_tokens": 3463255.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.175636625289917,
|
|
"epoch": 1.670632774472688,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004951778775159246,
|
|
"loss": 5.8622,
|
|
"mean_token_accuracy": 0.15969557911157609,
|
|
"num_tokens": 3473642.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.021358013153076,
|
|
"epoch": 1.6760411032990805,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004950891535380414,
|
|
"loss": 5.8372,
|
|
"mean_token_accuracy": 0.16012888848781587,
|
|
"num_tokens": 3484877.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.014034366607666,
|
|
"epoch": 1.6814494321254734,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004949996297556656,
|
|
"loss": 5.7532,
|
|
"mean_token_accuracy": 0.1709960624575615,
|
|
"num_tokens": 3495500.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.051374912261963,
|
|
"epoch": 1.6868577609518658,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004949093064941268,
|
|
"loss": 5.8748,
|
|
"mean_token_accuracy": 0.16371863037347795,
|
|
"num_tokens": 3508057.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.128537130355835,
|
|
"epoch": 1.6922660897782587,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004948181840816601,
|
|
"loss": 6.0298,
|
|
"mean_token_accuracy": 0.15803078263998033,
|
|
"num_tokens": 3521110.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.136459016799927,
|
|
"epoch": 1.697674418604651,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004947262628494045,
|
|
"loss": 5.9607,
|
|
"mean_token_accuracy": 0.16685472428798676,
|
|
"num_tokens": 3530692.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.1080584049224855,
|
|
"epoch": 1.703082747431044,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004946335431314018,
|
|
"loss": 5.9074,
|
|
"mean_token_accuracy": 0.15569821149110794,
|
|
"num_tokens": 3542195.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.141083526611328,
|
|
"epoch": 1.7084910762574363,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004945400252645959,
|
|
"loss": 5.7995,
|
|
"mean_token_accuracy": 0.17060298025608062,
|
|
"num_tokens": 3552676.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 5.938297843933105,
|
|
"epoch": 1.7138994050838292,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004944457095888308,
|
|
"loss": 5.7628,
|
|
"mean_token_accuracy": 0.16402305364608766,
|
|
"num_tokens": 3563308.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.0929491996765135,
|
|
"epoch": 1.7193077339102216,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00049435059644685,
|
|
"loss": 5.7927,
|
|
"mean_token_accuracy": 0.16668124496936798,
|
|
"num_tokens": 3574085.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.06469030380249,
|
|
"epoch": 1.7247160627366145,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004942546861842947,
|
|
"loss": 5.8596,
|
|
"mean_token_accuracy": 0.15768925547599794,
|
|
"num_tokens": 3585816.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.0371232509613035,
|
|
"epoch": 1.730124391563007,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004941579791497032,
|
|
"loss": 5.9079,
|
|
"mean_token_accuracy": 0.16382428705692292,
|
|
"num_tokens": 3597776.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.168297672271729,
|
|
"epoch": 1.7355327203893998,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004940604756945091,
|
|
"loss": 5.9224,
|
|
"mean_token_accuracy": 0.1619665116071701,
|
|
"num_tokens": 3607991.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.001934623718261,
|
|
"epoch": 1.7409410492157922,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004939621761730399,
|
|
"loss": 5.8072,
|
|
"mean_token_accuracy": 0.16959517300128937,
|
|
"num_tokens": 3618204.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.072473382949829,
|
|
"epoch": 1.746349378042185,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004938630809425166,
|
|
"loss": 5.9081,
|
|
"mean_token_accuracy": 0.15966459661722182,
|
|
"num_tokens": 3629558.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.1015989780426025,
|
|
"epoch": 1.7517577068685775,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004937631903630513,
|
|
"loss": 5.8555,
|
|
"mean_token_accuracy": 0.1621633917093277,
|
|
"num_tokens": 3640773.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.01327805519104,
|
|
"epoch": 1.7571660356949703,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004936625047976468,
|
|
"loss": 5.7683,
|
|
"mean_token_accuracy": 0.16852811276912688,
|
|
"num_tokens": 3652186.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 5.997731494903564,
|
|
"epoch": 1.7625743645213627,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004935610246121944,
|
|
"loss": 5.8237,
|
|
"mean_token_accuracy": 0.1706862360239029,
|
|
"num_tokens": 3663068.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.129118490219116,
|
|
"epoch": 1.7679826933477556,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004934587501754736,
|
|
"loss": 5.9509,
|
|
"mean_token_accuracy": 0.1577262118458748,
|
|
"num_tokens": 3673803.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.149345588684082,
|
|
"epoch": 1.7733910221741482,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004933556818591498,
|
|
"loss": 5.9262,
|
|
"mean_token_accuracy": 0.16089432835578918,
|
|
"num_tokens": 3685482.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.019583892822266,
|
|
"epoch": 1.7787993510005409,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004932518200377736,
|
|
"loss": 5.7909,
|
|
"mean_token_accuracy": 0.17309808731079102,
|
|
"num_tokens": 3697641.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.032641172409058,
|
|
"epoch": 1.7842076798269335,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004931471650887789,
|
|
"loss": 5.8044,
|
|
"mean_token_accuracy": 0.16572408378124237,
|
|
"num_tokens": 3709560.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.101502990722656,
|
|
"epoch": 1.7896160086533262,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004930417173924822,
|
|
"loss": 5.9246,
|
|
"mean_token_accuracy": 0.15817164331674577,
|
|
"num_tokens": 3720422.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 5.942840242385865,
|
|
"epoch": 1.7950243374797188,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004929354773320806,
|
|
"loss": 5.7543,
|
|
"mean_token_accuracy": 0.16456945538520812,
|
|
"num_tokens": 3731104.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.045040225982666,
|
|
"epoch": 1.8004326663061114,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004928284452936506,
|
|
"loss": 5.9151,
|
|
"mean_token_accuracy": 0.1589358001947403,
|
|
"num_tokens": 3742766.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.057123613357544,
|
|
"epoch": 1.805840995132504,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004927206216661473,
|
|
"loss": 5.7922,
|
|
"mean_token_accuracy": 0.16667329967021943,
|
|
"num_tokens": 3753064.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.060616970062256,
|
|
"epoch": 1.8112493239588967,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004926120068414015,
|
|
"loss": 5.8339,
|
|
"mean_token_accuracy": 0.16451720893383026,
|
|
"num_tokens": 3765183.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.038819885253906,
|
|
"epoch": 1.8166576527852893,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004925026012141201,
|
|
"loss": 5.823,
|
|
"mean_token_accuracy": 0.1619720585644245,
|
|
"num_tokens": 3777902.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 5.983007335662842,
|
|
"epoch": 1.822065981611682,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004923924051818832,
|
|
"loss": 5.7891,
|
|
"mean_token_accuracy": 0.1674435928463936,
|
|
"num_tokens": 3788461.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 5.970801639556885,
|
|
"epoch": 1.8274743104380746,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004922814191451435,
|
|
"loss": 5.7166,
|
|
"mean_token_accuracy": 0.1697210729122162,
|
|
"num_tokens": 3799574.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.0467628002166744,
|
|
"epoch": 1.8328826392644673,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004921696435072245,
|
|
"loss": 5.8843,
|
|
"mean_token_accuracy": 0.16265512108802796,
|
|
"num_tokens": 3811889.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 5.9999645233154295,
|
|
"epoch": 1.83829096809086,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004920570786743191,
|
|
"loss": 5.8481,
|
|
"mean_token_accuracy": 0.16382198184728622,
|
|
"num_tokens": 3822949.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.102803945541382,
|
|
"epoch": 1.8436992969172525,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004919437250554882,
|
|
"loss": 5.9143,
|
|
"mean_token_accuracy": 0.16273610591888427,
|
|
"num_tokens": 3834734.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.029530954360962,
|
|
"epoch": 1.8491076257436452,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004918295830626592,
|
|
"loss": 5.7834,
|
|
"mean_token_accuracy": 0.17444511353969575,
|
|
"num_tokens": 3845508.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 5.965552425384521,
|
|
"epoch": 1.8545159545700378,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004917146531106242,
|
|
"loss": 5.7701,
|
|
"mean_token_accuracy": 0.16514772921800613,
|
|
"num_tokens": 3856500.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.14597716331482,
|
|
"epoch": 1.8599242833964305,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004915989356170389,
|
|
"loss": 5.8608,
|
|
"mean_token_accuracy": 0.16108423322439194,
|
|
"num_tokens": 3867877.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.003490257263183,
|
|
"epoch": 1.865332612222823,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004914824310024213,
|
|
"loss": 5.8502,
|
|
"mean_token_accuracy": 0.1659428983926773,
|
|
"num_tokens": 3878613.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 5.988573265075684,
|
|
"epoch": 1.870740941049216,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004913651396901489,
|
|
"loss": 5.7536,
|
|
"mean_token_accuracy": 0.16964653134346008,
|
|
"num_tokens": 3888418.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.0804027080535885,
|
|
"epoch": 1.8761492698756084,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000491247062106459,
|
|
"loss": 5.8341,
|
|
"mean_token_accuracy": 0.16630043238401412,
|
|
"num_tokens": 3899359.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.027860641479492,
|
|
"epoch": 1.8815575987020012,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004911281986804455,
|
|
"loss": 5.9112,
|
|
"mean_token_accuracy": 0.16525566279888154,
|
|
"num_tokens": 3911568.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.022890949249268,
|
|
"epoch": 1.8869659275283936,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004910085498440587,
|
|
"loss": 5.8767,
|
|
"mean_token_accuracy": 0.15697558373212814,
|
|
"num_tokens": 3921597.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.013006639480591,
|
|
"epoch": 1.8923742563547865,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004908881160321025,
|
|
"loss": 5.7616,
|
|
"mean_token_accuracy": 0.1611903950572014,
|
|
"num_tokens": 3933124.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.026111459732055,
|
|
"epoch": 1.897782585181179,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004907668976822338,
|
|
"loss": 5.7779,
|
|
"mean_token_accuracy": 0.17135102599859237,
|
|
"num_tokens": 3944388.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.03193097114563,
|
|
"epoch": 1.9031909140075718,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004906448952349603,
|
|
"loss": 5.7552,
|
|
"mean_token_accuracy": 0.16768674552440643,
|
|
"num_tokens": 3955024.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.874382495880127,
|
|
"epoch": 1.9085992428339642,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004905221091336392,
|
|
"loss": 5.752,
|
|
"mean_token_accuracy": 0.17219420224428178,
|
|
"num_tokens": 3966029.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 5.936220693588257,
|
|
"epoch": 1.914007571660357,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004903985398244753,
|
|
"loss": 5.72,
|
|
"mean_token_accuracy": 0.16943709850311278,
|
|
"num_tokens": 3976066.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.098024606704712,
|
|
"epoch": 1.9194159004867495,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004902741877565201,
|
|
"loss": 5.9178,
|
|
"mean_token_accuracy": 0.1621759831905365,
|
|
"num_tokens": 3986467.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.042089986801147,
|
|
"epoch": 1.9248242293131423,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000490149053381669,
|
|
"loss": 5.8772,
|
|
"mean_token_accuracy": 0.16438196301460267,
|
|
"num_tokens": 3998208.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.0255107402801515,
|
|
"epoch": 1.9302325581395348,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004900231371546608,
|
|
"loss": 5.8455,
|
|
"mean_token_accuracy": 0.17020723074674607,
|
|
"num_tokens": 4008202.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.9757099628448485,
|
|
"epoch": 1.9356408869659276,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004898964395330753,
|
|
"loss": 5.6556,
|
|
"mean_token_accuracy": 0.1747577890753746,
|
|
"num_tokens": 4019256.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 5.862748003005981,
|
|
"epoch": 1.94104921579232,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004897689609773318,
|
|
"loss": 5.7121,
|
|
"mean_token_accuracy": 0.16850950121879577,
|
|
"num_tokens": 4029738.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.027835559844971,
|
|
"epoch": 1.946457544618713,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004896407019506879,
|
|
"loss": 5.9316,
|
|
"mean_token_accuracy": 0.15625764727592467,
|
|
"num_tokens": 4042277.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.052269363403321,
|
|
"epoch": 1.9518658734451053,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004895116629192371,
|
|
"loss": 5.796,
|
|
"mean_token_accuracy": 0.17734160125255585,
|
|
"num_tokens": 4054019.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.959774589538574,
|
|
"epoch": 1.9572742022714982,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004893818443519074,
|
|
"loss": 5.8651,
|
|
"mean_token_accuracy": 0.16139881312847137,
|
|
"num_tokens": 4065562.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.076329278945923,
|
|
"epoch": 1.9626825310978906,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00048925124672046,
|
|
"loss": 5.899,
|
|
"mean_token_accuracy": 0.16672396659851074,
|
|
"num_tokens": 4076617.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 5.9678566455841064,
|
|
"epoch": 1.9680908599242835,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004891198704994868,
|
|
"loss": 5.7197,
|
|
"mean_token_accuracy": 0.17168603837490082,
|
|
"num_tokens": 4087191.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.026430892944336,
|
|
"epoch": 1.973499188750676,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004889877161664096,
|
|
"loss": 5.7438,
|
|
"mean_token_accuracy": 0.17233943641185762,
|
|
"num_tokens": 4099150.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.938602876663208,
|
|
"epoch": 1.9789075175770687,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004888547842014771,
|
|
"loss": 5.7592,
|
|
"mean_token_accuracy": 0.1716069221496582,
|
|
"num_tokens": 4110299.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.977336502075195,
|
|
"epoch": 1.9843158464034614,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004887210750877648,
|
|
"loss": 5.7691,
|
|
"mean_token_accuracy": 0.1696719065308571,
|
|
"num_tokens": 4121682.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.967773580551148,
|
|
"epoch": 1.989724175229854,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004885865893111716,
|
|
"loss": 5.7105,
|
|
"mean_token_accuracy": 0.16929261684417723,
|
|
"num_tokens": 4132914.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 5.943519020080567,
|
|
"epoch": 1.9951325040562466,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004884513273604195,
|
|
"loss": 5.817,
|
|
"mean_token_accuracy": 0.16275771856307983,
|
|
"num_tokens": 4144213.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.060820526546902,
|
|
"epoch": 2.0,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 0.0004883152897270502,
|
|
"loss": 5.8566,
|
|
"mean_token_accuracy": 0.1669033153189553,
|
|
"num_tokens": 4153778.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.90231032371521,
|
|
"epoch": 2.005408328826393,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004881784769054253,
|
|
"loss": 5.4767,
|
|
"mean_token_accuracy": 0.17754273414611815,
|
|
"num_tokens": 4165536.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.054047441482544,
|
|
"epoch": 2.0108166576527853,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004880408893927226,
|
|
"loss": 5.5205,
|
|
"mean_token_accuracy": 0.1762312263250351,
|
|
"num_tokens": 4176182.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.773842191696167,
|
|
"epoch": 2.016224986479178,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004879025276889355,
|
|
"loss": 5.5271,
|
|
"mean_token_accuracy": 0.18644132167100907,
|
|
"num_tokens": 4187592.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 5.9513507843017575,
|
|
"epoch": 2.0216333153055706,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048776339229687077,
|
|
"loss": 5.3477,
|
|
"mean_token_accuracy": 0.19126913845539092,
|
|
"num_tokens": 4198199.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.860542154312133,
|
|
"epoch": 2.0270416441319634,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048762348372214683,
|
|
"loss": 5.5766,
|
|
"mean_token_accuracy": 0.17574259787797927,
|
|
"num_tokens": 4208116.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.854842281341552,
|
|
"epoch": 2.032449972958356,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00048748280247319165,
|
|
"loss": 5.3452,
|
|
"mean_token_accuracy": 0.184936061501503,
|
|
"num_tokens": 4218569.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.84954047203064,
|
|
"epoch": 2.0378583017847487,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004873413490612412,
|
|
"loss": 5.4911,
|
|
"mean_token_accuracy": 0.18496645092964173,
|
|
"num_tokens": 4228785.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.945874834060669,
|
|
"epoch": 2.043266630611141,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048719912400033757,
|
|
"loss": 5.4494,
|
|
"mean_token_accuracy": 0.18092841953039168,
|
|
"num_tokens": 4238814.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.815927982330322,
|
|
"epoch": 2.048674959437534,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000487056127807327,
|
|
"loss": 5.4196,
|
|
"mean_token_accuracy": 0.18144138157367706,
|
|
"num_tokens": 4250544.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.868709659576416,
|
|
"epoch": 2.0540832882639264,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00048691236100185795,
|
|
"loss": 5.5407,
|
|
"mean_token_accuracy": 0.17920881062746047,
|
|
"num_tokens": 4261806.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.967426872253418,
|
|
"epoch": 2.0594916170903192,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004867678241063792,
|
|
"loss": 5.5081,
|
|
"mean_token_accuracy": 0.1819603979587555,
|
|
"num_tokens": 4272472.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.858438920974732,
|
|
"epoch": 2.0648999459167117,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048662251764613833,
|
|
"loss": 5.4782,
|
|
"mean_token_accuracy": 0.18146095871925355,
|
|
"num_tokens": 4283758.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.848507308959961,
|
|
"epoch": 2.0703082747431045,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004864764421491793,
|
|
"loss": 5.574,
|
|
"mean_token_accuracy": 0.181258824467659,
|
|
"num_tokens": 4295930.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.06169810295105,
|
|
"epoch": 2.075716603569497,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004863295981463408,
|
|
"loss": 5.498,
|
|
"mean_token_accuracy": 0.1879578188061714,
|
|
"num_tokens": 4306711.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.809031867980957,
|
|
"epoch": 2.08112493239589,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004861819861712543,
|
|
"loss": 5.4812,
|
|
"mean_token_accuracy": 0.18434513956308365,
|
|
"num_tokens": 4317694.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.958704710006714,
|
|
"epoch": 2.086533261222282,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048603360676034203,
|
|
"loss": 5.5586,
|
|
"mean_token_accuracy": 0.1770416095852852,
|
|
"num_tokens": 4329552.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.864098119735718,
|
|
"epoch": 2.091941590048675,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048588446045281505,
|
|
"loss": 5.4969,
|
|
"mean_token_accuracy": 0.18100284337997435,
|
|
"num_tokens": 4340479.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.802739858627319,
|
|
"epoch": 2.0973499188750675,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004857345477906714,
|
|
"loss": 5.3742,
|
|
"mean_token_accuracy": 0.18729505985975264,
|
|
"num_tokens": 4351147.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.726704216003418,
|
|
"epoch": 2.1027582477014604,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00048558386931869417,
|
|
"loss": 5.3876,
|
|
"mean_token_accuracy": 0.18800846487283707,
|
|
"num_tokens": 4362962.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.889808416366577,
|
|
"epoch": 2.1081665765278528,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004854324255844492,
|
|
"loss": 5.4866,
|
|
"mean_token_accuracy": 0.18484701812267304,
|
|
"num_tokens": 4373723.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.75990924835205,
|
|
"epoch": 2.1135749053542456,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00048528021713828336,
|
|
"loss": 5.4655,
|
|
"mean_token_accuracy": 0.1776862010359764,
|
|
"num_tokens": 4383434.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.841428995132446,
|
|
"epoch": 2.118983234180638,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004851272445333225,
|
|
"loss": 5.4186,
|
|
"mean_token_accuracy": 0.18994414508342744,
|
|
"num_tokens": 4394847.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.8689299583435055,
|
|
"epoch": 2.124391563007031,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048497350832546937,
|
|
"loss": 5.524,
|
|
"mean_token_accuracy": 0.18320200741291046,
|
|
"num_tokens": 4406367.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.917090654373169,
|
|
"epoch": 2.1297998918334233,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004848190090734018,
|
|
"loss": 5.5273,
|
|
"mean_token_accuracy": 0.18322407305240632,
|
|
"num_tokens": 4417636.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.863876390457153,
|
|
"epoch": 2.135208220659816,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048466374733857035,
|
|
"loss": 5.5325,
|
|
"mean_token_accuracy": 0.18607548624277115,
|
|
"num_tokens": 4428473.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.845749950408935,
|
|
"epoch": 2.1406165494862086,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004845077236851966,
|
|
"loss": 5.4622,
|
|
"mean_token_accuracy": 0.18125550746917723,
|
|
"num_tokens": 4439780.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.8221944808959964,
|
|
"epoch": 2.1460248783126015,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048435093868027076,
|
|
"loss": 5.4118,
|
|
"mean_token_accuracy": 0.19358843713998794,
|
|
"num_tokens": 4450929.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.835688257217408,
|
|
"epoch": 2.151433207138994,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048419339289354995,
|
|
"loss": 5.454,
|
|
"mean_token_accuracy": 0.18442270159721375,
|
|
"num_tokens": 4462967.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.889276933670044,
|
|
"epoch": 2.1568415359653867,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000484035086897556,
|
|
"loss": 5.5347,
|
|
"mean_token_accuracy": 0.1838492140173912,
|
|
"num_tokens": 4473970.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.863001918792724,
|
|
"epoch": 2.162249864791779,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004838760212675732,
|
|
"loss": 5.5383,
|
|
"mean_token_accuracy": 0.1813681200146675,
|
|
"num_tokens": 4485660.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 2.162249864791779,
|
|
"eval_entropy": 5.725599083033475,
|
|
"eval_loss": 6.020395755767822,
|
|
"eval_mean_token_accuracy": 0.16601560962471096,
|
|
"eval_num_tokens": 4485660.0,
|
|
"eval_runtime": 1.9373,
|
|
"eval_samples_per_second": 1814.336,
|
|
"eval_steps_per_second": 227.115,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 5.8430750370025635,
|
|
"epoch": 2.167658193618172,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004837161965816464,
|
|
"loss": 5.509,
|
|
"mean_token_accuracy": 0.1867072880268097,
|
|
"num_tokens": 4497261.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.846252346038819,
|
|
"epoch": 2.1730665224445644,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048355561342057886,
|
|
"loss": 5.3884,
|
|
"mean_token_accuracy": 0.18988969624042512,
|
|
"num_tokens": 4508120.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.839772653579712,
|
|
"epoch": 2.1784748512709573,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048339427236793024,
|
|
"loss": 5.4781,
|
|
"mean_token_accuracy": 0.18922811150550842,
|
|
"num_tokens": 4519286.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.898039484024048,
|
|
"epoch": 2.1838831800973497,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004832321740100143,
|
|
"loss": 5.5553,
|
|
"mean_token_accuracy": 0.1808038905262947,
|
|
"num_tokens": 4531912.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.885161018371582,
|
|
"epoch": 2.1892915089237426,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00048306931893589666,
|
|
"loss": 5.5665,
|
|
"mean_token_accuracy": 0.18202104717493056,
|
|
"num_tokens": 4544370.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.853564596176147,
|
|
"epoch": 2.194699837750135,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048290570773739327,
|
|
"loss": 5.5119,
|
|
"mean_token_accuracy": 0.18240146934986115,
|
|
"num_tokens": 4556013.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.764708852767944,
|
|
"epoch": 2.200108166576528,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048274134100906747,
|
|
"loss": 5.4625,
|
|
"mean_token_accuracy": 0.18581133484840393,
|
|
"num_tokens": 4566200.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.856343841552734,
|
|
"epoch": 2.2055164954029207,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048257621934822835,
|
|
"loss": 5.506,
|
|
"mean_token_accuracy": 0.18467539846897124,
|
|
"num_tokens": 4576899.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.934852552413941,
|
|
"epoch": 2.210924824229313,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004824103433549285,
|
|
"loss": 5.5254,
|
|
"mean_token_accuracy": 0.17869499921798707,
|
|
"num_tokens": 4588983.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.847223997116089,
|
|
"epoch": 2.2163331530557056,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004822437136319616,
|
|
"loss": 5.479,
|
|
"mean_token_accuracy": 0.19166221618652343,
|
|
"num_tokens": 4599758.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.81277289390564,
|
|
"epoch": 2.2217414818820984,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004820763307848604,
|
|
"loss": 5.5154,
|
|
"mean_token_accuracy": 0.1851065754890442,
|
|
"num_tokens": 4610927.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.839804220199585,
|
|
"epoch": 2.2271498107084913,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048190819542189456,
|
|
"loss": 5.5307,
|
|
"mean_token_accuracy": 0.18261638432741165,
|
|
"num_tokens": 4623719.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.836520338058472,
|
|
"epoch": 2.2325581395348837,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048173930815406847,
|
|
"loss": 5.5893,
|
|
"mean_token_accuracy": 0.17861880511045455,
|
|
"num_tokens": 4635389.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.938038301467896,
|
|
"epoch": 2.2379664683612766,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004815696695951187,
|
|
"loss": 5.5353,
|
|
"mean_token_accuracy": 0.18332355469465256,
|
|
"num_tokens": 4645093.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.799956464767456,
|
|
"epoch": 2.243374797187669,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004813992803615123,
|
|
"loss": 5.5249,
|
|
"mean_token_accuracy": 0.17770509719848632,
|
|
"num_tokens": 4656490.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.897887754440307,
|
|
"epoch": 2.248783126014062,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048122814107244414,
|
|
"loss": 5.5365,
|
|
"mean_token_accuracy": 0.18031222522258758,
|
|
"num_tokens": 4667603.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.810691595077515,
|
|
"epoch": 2.2541914548404542,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048105625234983466,
|
|
"loss": 5.4848,
|
|
"mean_token_accuracy": 0.17508939504623414,
|
|
"num_tokens": 4678669.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.735423612594604,
|
|
"epoch": 2.259599783666847,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00048088361481832807,
|
|
"loss": 5.2923,
|
|
"mean_token_accuracy": 0.19484480917453767,
|
|
"num_tokens": 4689757.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.711760234832764,
|
|
"epoch": 2.2650081124932395,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048071022910528934,
|
|
"loss": 5.4666,
|
|
"mean_token_accuracy": 0.18362647742033006,
|
|
"num_tokens": 4699332.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.8964282989501955,
|
|
"epoch": 2.2704164413196324,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048053609584080274,
|
|
"loss": 5.5409,
|
|
"mean_token_accuracy": 0.1857800304889679,
|
|
"num_tokens": 4709485.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.778720140457153,
|
|
"epoch": 2.275824770146025,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048036121565766876,
|
|
"loss": 5.4826,
|
|
"mean_token_accuracy": 0.18560403287410737,
|
|
"num_tokens": 4720935.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.745053100585937,
|
|
"epoch": 2.2812330989724177,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004801855891914026,
|
|
"loss": 5.4299,
|
|
"mean_token_accuracy": 0.18251365274190903,
|
|
"num_tokens": 4731219.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.898935747146607,
|
|
"epoch": 2.28664142779881,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048000921708023117,
|
|
"loss": 5.5586,
|
|
"mean_token_accuracy": 0.18378493636846543,
|
|
"num_tokens": 4742637.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.783941411972046,
|
|
"epoch": 2.292049756625203,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004798320999650911,
|
|
"loss": 5.4516,
|
|
"mean_token_accuracy": 0.1839448481798172,
|
|
"num_tokens": 4753341.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.984134101867676,
|
|
"epoch": 2.2974580854515954,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047965423848962636,
|
|
"loss": 5.6041,
|
|
"mean_token_accuracy": 0.18355586379766464,
|
|
"num_tokens": 4765339.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.843023586273193,
|
|
"epoch": 2.302866414277988,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000479475633300186,
|
|
"loss": 5.4551,
|
|
"mean_token_accuracy": 0.18163464665412904,
|
|
"num_tokens": 4776149.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.771022653579712,
|
|
"epoch": 2.3082747431043806,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047929628504582173,
|
|
"loss": 5.4777,
|
|
"mean_token_accuracy": 0.18727335333824158,
|
|
"num_tokens": 4788120.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.860320329666138,
|
|
"epoch": 2.3136830719307735,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004791161943782856,
|
|
"loss": 5.5576,
|
|
"mean_token_accuracy": 0.17805030047893525,
|
|
"num_tokens": 4799922.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.718722105026245,
|
|
"epoch": 2.319091400757166,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047893536195202726,
|
|
"loss": 5.4895,
|
|
"mean_token_accuracy": 0.18036262691020966,
|
|
"num_tokens": 4810501.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.817003440856934,
|
|
"epoch": 2.3244997295835588,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004787537884241923,
|
|
"loss": 5.4417,
|
|
"mean_token_accuracy": 0.1868960365653038,
|
|
"num_tokens": 4820919.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.889329147338867,
|
|
"epoch": 2.329908058409951,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047857147445461943,
|
|
"loss": 5.5355,
|
|
"mean_token_accuracy": 0.18348777890205384,
|
|
"num_tokens": 4833298.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.682007360458374,
|
|
"epoch": 2.335316387236344,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004783884207058379,
|
|
"loss": 5.4153,
|
|
"mean_token_accuracy": 0.17922882437705995,
|
|
"num_tokens": 4845199.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.795744609832764,
|
|
"epoch": 2.3407247160627365,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047820462784306555,
|
|
"loss": 5.4691,
|
|
"mean_token_accuracy": 0.18276022076606752,
|
|
"num_tokens": 4856549.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.809813499450684,
|
|
"epoch": 2.3461330448891293,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000478020096534206,
|
|
"loss": 5.5131,
|
|
"mean_token_accuracy": 0.17976404577493668,
|
|
"num_tokens": 4868446.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.885415935516358,
|
|
"epoch": 2.3515413737155217,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047783482744984644,
|
|
"loss": 5.5668,
|
|
"mean_token_accuracy": 0.1731751322746277,
|
|
"num_tokens": 4879047.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.818225002288818,
|
|
"epoch": 2.3569497025419146,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00047764882126325527,
|
|
"loss": 5.4639,
|
|
"mean_token_accuracy": 0.18213903456926345,
|
|
"num_tokens": 4891591.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.744760799407959,
|
|
"epoch": 2.362358031368307,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004774620786503793,
|
|
"loss": 5.4759,
|
|
"mean_token_accuracy": 0.18766329139471055,
|
|
"num_tokens": 4902710.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.887608385086059,
|
|
"epoch": 2.3677663601947,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004772746002898417,
|
|
"loss": 5.5475,
|
|
"mean_token_accuracy": 0.18340907692909242,
|
|
"num_tokens": 4914668.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.906754684448242,
|
|
"epoch": 2.3731746890210923,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00047708638686293903,
|
|
"loss": 5.6686,
|
|
"mean_token_accuracy": 0.17607683539390565,
|
|
"num_tokens": 4926684.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.885916709899902,
|
|
"epoch": 2.378583017847485,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00047689743905363963,
|
|
"loss": 5.5433,
|
|
"mean_token_accuracy": 0.1780522108078003,
|
|
"num_tokens": 4938577.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.711051607131958,
|
|
"epoch": 2.383991346673878,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004767077575485801,
|
|
"loss": 5.4213,
|
|
"mean_token_accuracy": 0.19123190641403198,
|
|
"num_tokens": 4949070.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.815402317047119,
|
|
"epoch": 2.3893996755002704,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004765173430370635,
|
|
"loss": 5.4939,
|
|
"mean_token_accuracy": 0.18558672815561295,
|
|
"num_tokens": 4959558.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.715311431884766,
|
|
"epoch": 2.394808004326663,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004763261962110567,
|
|
"loss": 5.3557,
|
|
"mean_token_accuracy": 0.19049575328826904,
|
|
"num_tokens": 4970310.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.725232744216919,
|
|
"epoch": 2.4002163331530557,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004761343177651875,
|
|
"loss": 5.4447,
|
|
"mean_token_accuracy": 0.18282105326652526,
|
|
"num_tokens": 4982300.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.790338802337646,
|
|
"epoch": 2.4056246619794486,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004759417083967428,
|
|
"loss": 5.4725,
|
|
"mean_token_accuracy": 0.18086550235748292,
|
|
"num_tokens": 4994425.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.776576137542724,
|
|
"epoch": 2.411032990805841,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00047574836880566544,
|
|
"loss": 5.4794,
|
|
"mean_token_accuracy": 0.19098425656557083,
|
|
"num_tokens": 5005872.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.78202633857727,
|
|
"epoch": 2.4164413196322334,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004755542996945519,
|
|
"loss": 5.5075,
|
|
"mean_token_accuracy": 0.18447682559490203,
|
|
"num_tokens": 5016903.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.836867189407348,
|
|
"epoch": 2.4218496484586263,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047535950176864976,
|
|
"loss": 5.5442,
|
|
"mean_token_accuracy": 0.18406234681606293,
|
|
"num_tokens": 5028407.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.820690727233886,
|
|
"epoch": 2.427257977285019,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047516397573585525,
|
|
"loss": 5.5518,
|
|
"mean_token_accuracy": 0.18487397730350494,
|
|
"num_tokens": 5039693.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.846407175064087,
|
|
"epoch": 2.4326663061114115,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047496772230671034,
|
|
"loss": 5.5034,
|
|
"mean_token_accuracy": 0.18430618047714234,
|
|
"num_tokens": 5051279.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.8105754375457765,
|
|
"epoch": 2.4380746349378044,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004747707421944004,
|
|
"loss": 5.4522,
|
|
"mean_token_accuracy": 0.1847480833530426,
|
|
"num_tokens": 5062698.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.7512526512146,
|
|
"epoch": 2.443482963764197,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004745730361147517,
|
|
"loss": 5.4366,
|
|
"mean_token_accuracy": 0.18818385004997254,
|
|
"num_tokens": 5074927.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.796603631973267,
|
|
"epoch": 2.4488912925905897,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004743746047862286,
|
|
"loss": 5.5697,
|
|
"mean_token_accuracy": 0.1745971441268921,
|
|
"num_tokens": 5086298.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.817357540130615,
|
|
"epoch": 2.454299621416982,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004741754489299309,
|
|
"loss": 5.5101,
|
|
"mean_token_accuracy": 0.17972173988819123,
|
|
"num_tokens": 5098161.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.774695014953613,
|
|
"epoch": 2.459707950243375,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047397556926959155,
|
|
"loss": 5.4874,
|
|
"mean_token_accuracy": 0.19200887233018876,
|
|
"num_tokens": 5110743.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.90885591506958,
|
|
"epoch": 2.4651162790697674,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047377496653157374,
|
|
"loss": 5.5304,
|
|
"mean_token_accuracy": 0.1880877986550331,
|
|
"num_tokens": 5121838.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.696847534179687,
|
|
"epoch": 2.4705246078961602,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047357364144486806,
|
|
"loss": 5.4768,
|
|
"mean_token_accuracy": 0.1817433074116707,
|
|
"num_tokens": 5135047.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.772260761260986,
|
|
"epoch": 2.4759329367225527,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047337159474109057,
|
|
"loss": 5.4649,
|
|
"mean_token_accuracy": 0.17703176736831666,
|
|
"num_tokens": 5146329.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.863223314285278,
|
|
"epoch": 2.4813412655489455,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047316882715447934,
|
|
"loss": 5.5062,
|
|
"mean_token_accuracy": 0.1872244283556938,
|
|
"num_tokens": 5157511.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.693565464019775,
|
|
"epoch": 2.486749594375338,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004729653394218922,
|
|
"loss": 5.495,
|
|
"mean_token_accuracy": 0.18242276459932327,
|
|
"num_tokens": 5168456.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.931699275970459,
|
|
"epoch": 2.492157923201731,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047276113228280427,
|
|
"loss": 5.5515,
|
|
"mean_token_accuracy": 0.1825096696615219,
|
|
"num_tokens": 5179294.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.807642078399658,
|
|
"epoch": 2.497566252028123,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004725562064793045,
|
|
"loss": 5.4993,
|
|
"mean_token_accuracy": 0.18253640085458755,
|
|
"num_tokens": 5190195.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.710082483291626,
|
|
"epoch": 2.502974580854516,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00047235056275609397,
|
|
"loss": 5.3895,
|
|
"mean_token_accuracy": 0.18921253234148025,
|
|
"num_tokens": 5201289.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.741735363006592,
|
|
"epoch": 2.5083829096809085,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004721442018604822,
|
|
"loss": 5.4177,
|
|
"mean_token_accuracy": 0.19280524998903276,
|
|
"num_tokens": 5212295.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.650020360946655,
|
|
"epoch": 2.5137912385073014,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004719371245423853,
|
|
"loss": 5.3894,
|
|
"mean_token_accuracy": 0.1905950501561165,
|
|
"num_tokens": 5223144.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.713900184631347,
|
|
"epoch": 2.5191995673336938,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047172933155432273,
|
|
"loss": 5.4097,
|
|
"mean_token_accuracy": 0.18740693479776382,
|
|
"num_tokens": 5232840.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.713705444335938,
|
|
"epoch": 2.5246078961600866,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047152082365141466,
|
|
"loss": 5.5167,
|
|
"mean_token_accuracy": 0.18210788518190385,
|
|
"num_tokens": 5243796.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.8688812255859375,
|
|
"epoch": 2.530016224986479,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004713116015913794,
|
|
"loss": 5.5962,
|
|
"mean_token_accuracy": 0.17875071614980698,
|
|
"num_tokens": 5255990.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.84153995513916,
|
|
"epoch": 2.535424553812872,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00047110166613453016,
|
|
"loss": 5.5344,
|
|
"mean_token_accuracy": 0.1801824688911438,
|
|
"num_tokens": 5268134.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.672592115402222,
|
|
"epoch": 2.5408328826392643,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047089101804377305,
|
|
"loss": 5.3767,
|
|
"mean_token_accuracy": 0.18733149021863937,
|
|
"num_tokens": 5278662.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.775094842910766,
|
|
"epoch": 2.546241211465657,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004706796580846037,
|
|
"loss": 5.4486,
|
|
"mean_token_accuracy": 0.18052454739809037,
|
|
"num_tokens": 5288382.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.735822057723999,
|
|
"epoch": 2.5516495402920496,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047046758702510465,
|
|
"loss": 5.5175,
|
|
"mean_token_accuracy": 0.17908523827791215,
|
|
"num_tokens": 5299769.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.894213485717773,
|
|
"epoch": 2.5570578691184425,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047025480563594264,
|
|
"loss": 5.4694,
|
|
"mean_token_accuracy": 0.18954675197601317,
|
|
"num_tokens": 5310662.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.68624210357666,
|
|
"epoch": 2.5624661979448353,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047004131469036576,
|
|
"loss": 5.433,
|
|
"mean_token_accuracy": 0.1791605308651924,
|
|
"num_tokens": 5321785.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.790686321258545,
|
|
"epoch": 2.5678745267712277,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046982711496420057,
|
|
"loss": 5.4881,
|
|
"mean_token_accuracy": 0.17994098961353303,
|
|
"num_tokens": 5333557.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.7890860080719,
|
|
"epoch": 2.57328285559762,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046961220723584925,
|
|
"loss": 5.5283,
|
|
"mean_token_accuracy": 0.1812247097492218,
|
|
"num_tokens": 5344767.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.777838706970215,
|
|
"epoch": 2.578691184424013,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00046939659228628725,
|
|
"loss": 5.4417,
|
|
"mean_token_accuracy": 0.18691411167383193,
|
|
"num_tokens": 5356673.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.7575544834136965,
|
|
"epoch": 2.584099513250406,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00046918027089905956,
|
|
"loss": 5.5538,
|
|
"mean_token_accuracy": 0.1779048413038254,
|
|
"num_tokens": 5367296.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.7355200290679935,
|
|
"epoch": 2.5895078420767983,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046896324386027873,
|
|
"loss": 5.4201,
|
|
"mean_token_accuracy": 0.18585974872112274,
|
|
"num_tokens": 5378081.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.821057653427124,
|
|
"epoch": 2.5949161709031907,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00046874551195862145,
|
|
"loss": 5.5102,
|
|
"mean_token_accuracy": 0.17801062613725663,
|
|
"num_tokens": 5390244.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.708388137817383,
|
|
"epoch": 2.6003244997295836,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000468527075985326,
|
|
"loss": 5.3888,
|
|
"mean_token_accuracy": 0.1883876532316208,
|
|
"num_tokens": 5400741.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.796988010406494,
|
|
"epoch": 2.6057328285559764,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046830793673418924,
|
|
"loss": 5.5663,
|
|
"mean_token_accuracy": 0.18608431220054628,
|
|
"num_tokens": 5411440.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.759448432922364,
|
|
"epoch": 2.611141157382369,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00046808809500156355,
|
|
"loss": 5.4269,
|
|
"mean_token_accuracy": 0.18147926032543182,
|
|
"num_tokens": 5422472.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.705766773223877,
|
|
"epoch": 2.6165494862087613,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004678675515863545,
|
|
"loss": 5.5039,
|
|
"mean_token_accuracy": 0.17874256670475006,
|
|
"num_tokens": 5433958.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.767446708679199,
|
|
"epoch": 2.621957815035154,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004676463072900173,
|
|
"loss": 5.4192,
|
|
"mean_token_accuracy": 0.19382819086313247,
|
|
"num_tokens": 5444343.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.681805276870728,
|
|
"epoch": 2.627366143861547,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00046742436291655415,
|
|
"loss": 5.4357,
|
|
"mean_token_accuracy": 0.18407043516635896,
|
|
"num_tokens": 5456260.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.748001480102539,
|
|
"epoch": 2.6327744726879394,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00046720171927251155,
|
|
"loss": 5.5282,
|
|
"mean_token_accuracy": 0.18753280341625214,
|
|
"num_tokens": 5468281.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.655755233764649,
|
|
"epoch": 2.638182801514332,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004669783771669769,
|
|
"loss": 5.3615,
|
|
"mean_token_accuracy": 0.19994893074035644,
|
|
"num_tokens": 5480068.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.671969795227051,
|
|
"epoch": 2.6435911303407247,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000466754337411576,
|
|
"loss": 5.3704,
|
|
"mean_token_accuracy": 0.18418589234352112,
|
|
"num_tokens": 5490932.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.653346681594849,
|
|
"epoch": 2.6489994591671175,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046652960082046985,
|
|
"loss": 5.4672,
|
|
"mean_token_accuracy": 0.18149202167987824,
|
|
"num_tokens": 5501787.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.773309183120728,
|
|
"epoch": 2.65440778799351,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046630416821035164,
|
|
"loss": 5.4492,
|
|
"mean_token_accuracy": 0.18520426899194717,
|
|
"num_tokens": 5513297.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.655526876449585,
|
|
"epoch": 2.6598161168199024,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004660780404004441,
|
|
"loss": 5.4302,
|
|
"mean_token_accuracy": 0.19071539640426635,
|
|
"num_tokens": 5523797.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.685646867752075,
|
|
"epoch": 2.6652244456462952,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046585121821249623,
|
|
"loss": 5.4664,
|
|
"mean_token_accuracy": 0.191667777299881,
|
|
"num_tokens": 5535808.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.8277747631073,
|
|
"epoch": 2.670632774472688,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00046562370247078025,
|
|
"loss": 5.5487,
|
|
"mean_token_accuracy": 0.1725021183490753,
|
|
"num_tokens": 5548179.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.832285785675049,
|
|
"epoch": 2.6760411032990805,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000465395494002089,
|
|
"loss": 5.5629,
|
|
"mean_token_accuracy": 0.17875450998544692,
|
|
"num_tokens": 5559144.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.757518625259399,
|
|
"epoch": 2.6814494321254734,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004651665936357325,
|
|
"loss": 5.4518,
|
|
"mean_token_accuracy": 0.1852279335260391,
|
|
"num_tokens": 5571103.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.690479469299317,
|
|
"epoch": 2.686857760951866,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004649370022035354,
|
|
"loss": 5.3842,
|
|
"mean_token_accuracy": 0.18641475737094879,
|
|
"num_tokens": 5583239.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.68533730506897,
|
|
"epoch": 2.6922660897782587,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004647067205398332,
|
|
"loss": 5.3563,
|
|
"mean_token_accuracy": 0.18859488219022752,
|
|
"num_tokens": 5593602.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.8065142154693605,
|
|
"epoch": 2.697674418604651,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046447574948147027,
|
|
"loss": 5.5152,
|
|
"mean_token_accuracy": 0.17736031115055084,
|
|
"num_tokens": 5604696.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.704569149017334,
|
|
"epoch": 2.703082747431044,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046424408986779587,
|
|
"loss": 5.49,
|
|
"mean_token_accuracy": 0.17822713404893875,
|
|
"num_tokens": 5615668.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.703082747431044,
|
|
"eval_entropy": 5.6619423313574355,
|
|
"eval_loss": 5.923056125640869,
|
|
"eval_mean_token_accuracy": 0.17251565199006688,
|
|
"eval_num_tokens": 5615668.0,
|
|
"eval_runtime": 1.9309,
|
|
"eval_samples_per_second": 1820.364,
|
|
"eval_steps_per_second": 227.869,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.76570315361023,
|
|
"epoch": 2.7084910762574363,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004640117425406615,
|
|
"loss": 5.4237,
|
|
"mean_token_accuracy": 0.18574815541505812,
|
|
"num_tokens": 5627344.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.701715326309204,
|
|
"epoch": 2.713899405083829,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000463778708344418,
|
|
"loss": 5.4671,
|
|
"mean_token_accuracy": 0.18088625818490983,
|
|
"num_tokens": 5637472.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.713341093063354,
|
|
"epoch": 2.7193077339102216,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046354498812591207,
|
|
"loss": 5.4502,
|
|
"mean_token_accuracy": 0.18748428374528886,
|
|
"num_tokens": 5647766.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.655908727645874,
|
|
"epoch": 2.7247160627366145,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046331058273448364,
|
|
"loss": 5.3821,
|
|
"mean_token_accuracy": 0.1837076410651207,
|
|
"num_tokens": 5658343.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.677487134933472,
|
|
"epoch": 2.730124391563007,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004630754930219624,
|
|
"loss": 5.3865,
|
|
"mean_token_accuracy": 0.19113706946372985,
|
|
"num_tokens": 5669640.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.602105569839478,
|
|
"epoch": 2.7355327203893998,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004628397198426648,
|
|
"loss": 5.3608,
|
|
"mean_token_accuracy": 0.18334166407585145,
|
|
"num_tokens": 5682421.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.6781334400177,
|
|
"epoch": 2.740941049215792,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046260326405339126,
|
|
"loss": 5.3426,
|
|
"mean_token_accuracy": 0.19388457536697387,
|
|
"num_tokens": 5692797.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.761030578613282,
|
|
"epoch": 2.746349378042185,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004623661265134226,
|
|
"loss": 5.5239,
|
|
"mean_token_accuracy": 0.18626205921173095,
|
|
"num_tokens": 5702779.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.738360166549683,
|
|
"epoch": 2.7517577068685775,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004621283080845172,
|
|
"loss": 5.5032,
|
|
"mean_token_accuracy": 0.18362185806035997,
|
|
"num_tokens": 5714147.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.640808868408203,
|
|
"epoch": 2.7571660356949703,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00046188980963090783,
|
|
"loss": 5.3324,
|
|
"mean_token_accuracy": 0.19274002760648729,
|
|
"num_tokens": 5725718.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.647945594787598,
|
|
"epoch": 2.7625743645213627,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00046165063201929825,
|
|
"loss": 5.4152,
|
|
"mean_token_accuracy": 0.1916851669549942,
|
|
"num_tokens": 5737441.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.7787117004394535,
|
|
"epoch": 2.7679826933477556,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00046141077611886074,
|
|
"loss": 5.5182,
|
|
"mean_token_accuracy": 0.1864737704396248,
|
|
"num_tokens": 5748735.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.815402412414551,
|
|
"epoch": 2.773391022174148,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004611702428012319,
|
|
"loss": 5.5785,
|
|
"mean_token_accuracy": 0.17667905241250992,
|
|
"num_tokens": 5760262.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.7044572830200195,
|
|
"epoch": 2.778799351000541,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046092903294051057,
|
|
"loss": 5.4725,
|
|
"mean_token_accuracy": 0.18194340616464616,
|
|
"num_tokens": 5771909.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.587724304199218,
|
|
"epoch": 2.7842076798269337,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046068714741325385,
|
|
"loss": 5.371,
|
|
"mean_token_accuracy": 0.1913776785135269,
|
|
"num_tokens": 5782028.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.821951818466187,
|
|
"epoch": 2.789616008653326,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004604445870984743,
|
|
"loss": 5.5162,
|
|
"mean_token_accuracy": 0.18275947123765945,
|
|
"num_tokens": 5793641.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.798354768753052,
|
|
"epoch": 2.7950243374797186,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004602013528776367,
|
|
"loss": 5.4623,
|
|
"mean_token_accuracy": 0.18469780087471008,
|
|
"num_tokens": 5805392.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.736293888092041,
|
|
"epoch": 2.8004326663061114,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004599574456346548,
|
|
"loss": 5.4405,
|
|
"mean_token_accuracy": 0.1861585184931755,
|
|
"num_tokens": 5816362.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.620948266983032,
|
|
"epoch": 2.8058409951325043,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045971286625588796,
|
|
"loss": 5.3753,
|
|
"mean_token_accuracy": 0.18987466841936113,
|
|
"num_tokens": 5827518.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.806012153625488,
|
|
"epoch": 2.8112493239588967,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004594676156301383,
|
|
"loss": 5.5029,
|
|
"mean_token_accuracy": 0.1827402338385582,
|
|
"num_tokens": 5838455.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.71817798614502,
|
|
"epoch": 2.816657652785289,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045922169464864725,
|
|
"loss": 5.389,
|
|
"mean_token_accuracy": 0.19094510227441788,
|
|
"num_tokens": 5850065.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.711734294891357,
|
|
"epoch": 2.822065981611682,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000458975104205092,
|
|
"loss": 5.5202,
|
|
"mean_token_accuracy": 0.18257656395435334,
|
|
"num_tokens": 5860444.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.784728574752807,
|
|
"epoch": 2.827474310438075,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00045872784519558304,
|
|
"loss": 5.489,
|
|
"mean_token_accuracy": 0.18090750128030778,
|
|
"num_tokens": 5873984.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.746411752700806,
|
|
"epoch": 2.8328826392644673,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004584799185186599,
|
|
"loss": 5.5212,
|
|
"mean_token_accuracy": 0.18793384730815887,
|
|
"num_tokens": 5885807.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.704473447799683,
|
|
"epoch": 2.8382909680908597,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045823132507528903,
|
|
"loss": 5.4024,
|
|
"mean_token_accuracy": 0.18671886324882508,
|
|
"num_tokens": 5897122.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.63995680809021,
|
|
"epoch": 2.8436992969172525,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004579820657688593,
|
|
"loss": 5.3991,
|
|
"mean_token_accuracy": 0.18967421501874923,
|
|
"num_tokens": 5909189.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.745748233795166,
|
|
"epoch": 2.8491076257436454,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004577321415051798,
|
|
"loss": 5.4968,
|
|
"mean_token_accuracy": 0.1795533388853073,
|
|
"num_tokens": 5920135.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.72193169593811,
|
|
"epoch": 2.854515954570038,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00045748155319247564,
|
|
"loss": 5.4737,
|
|
"mean_token_accuracy": 0.187947179377079,
|
|
"num_tokens": 5932279.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.602014493942261,
|
|
"epoch": 2.8599242833964302,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004572303017413856,
|
|
"loss": 5.4078,
|
|
"mean_token_accuracy": 0.18369463682174683,
|
|
"num_tokens": 5942082.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.815732002258301,
|
|
"epoch": 2.865332612222823,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004569783880649578,
|
|
"loss": 5.4523,
|
|
"mean_token_accuracy": 0.1847761943936348,
|
|
"num_tokens": 5952856.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.584131813049316,
|
|
"epoch": 2.870740941049216,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004567258130786474,
|
|
"loss": 5.3521,
|
|
"mean_token_accuracy": 0.1931051105260849,
|
|
"num_tokens": 5963937.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.698371505737304,
|
|
"epoch": 2.8761492698756084,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00045647257770031215,
|
|
"loss": 5.4446,
|
|
"mean_token_accuracy": 0.1882467731833458,
|
|
"num_tokens": 5974852.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.723794889450073,
|
|
"epoch": 2.8815575987020012,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045621868285021023,
|
|
"loss": 5.4307,
|
|
"mean_token_accuracy": 0.1901254966855049,
|
|
"num_tokens": 5985616.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.709406042098999,
|
|
"epoch": 2.8869659275283936,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00045596412945099615,
|
|
"loss": 5.3944,
|
|
"mean_token_accuracy": 0.19327322244644166,
|
|
"num_tokens": 5997199.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.7356147289276125,
|
|
"epoch": 2.8923742563547865,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045570891842771727,
|
|
"loss": 5.4415,
|
|
"mean_token_accuracy": 0.18972726613283158,
|
|
"num_tokens": 6008654.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.670080661773682,
|
|
"epoch": 2.897782585181179,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045545305070781135,
|
|
"loss": 5.5062,
|
|
"mean_token_accuracy": 0.18060076087713242,
|
|
"num_tokens": 6020662.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.729869079589844,
|
|
"epoch": 2.903190914007572,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045519652722110206,
|
|
"loss": 5.3768,
|
|
"mean_token_accuracy": 0.205011285841465,
|
|
"num_tokens": 6031793.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.637052536010742,
|
|
"epoch": 2.908599242833964,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045493934889979655,
|
|
"loss": 5.3676,
|
|
"mean_token_accuracy": 0.19111330062150955,
|
|
"num_tokens": 6042181.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.5539429664611815,
|
|
"epoch": 2.914007571660357,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00045468151667848116,
|
|
"loss": 5.3319,
|
|
"mean_token_accuracy": 0.19787560701370238,
|
|
"num_tokens": 6052774.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.761758470535279,
|
|
"epoch": 2.9194159004867495,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000454423031494119,
|
|
"loss": 5.3492,
|
|
"mean_token_accuracy": 0.20007340908050536,
|
|
"num_tokens": 6063273.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.633448791503906,
|
|
"epoch": 2.9248242293131423,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004541638942860458,
|
|
"loss": 5.3516,
|
|
"mean_token_accuracy": 0.1881837248802185,
|
|
"num_tokens": 6075121.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.552972984313965,
|
|
"epoch": 2.9302325581395348,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004539041059959667,
|
|
"loss": 5.3285,
|
|
"mean_token_accuracy": 0.19073624461889266,
|
|
"num_tokens": 6085666.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.6588763236999515,
|
|
"epoch": 2.9356408869659276,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045364366756795295,
|
|
"loss": 5.3503,
|
|
"mean_token_accuracy": 0.19836849421262742,
|
|
"num_tokens": 6096619.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.739606475830078,
|
|
"epoch": 2.94104921579232,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00045338257994843845,
|
|
"loss": 5.5105,
|
|
"mean_token_accuracy": 0.1829600602388382,
|
|
"num_tokens": 6108800.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.694767618179322,
|
|
"epoch": 2.946457544618713,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004531208440862162,
|
|
"loss": 5.4299,
|
|
"mean_token_accuracy": 0.184509801864624,
|
|
"num_tokens": 6119789.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.6493161678314205,
|
|
"epoch": 2.9518658734451053,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045285846093243485,
|
|
"loss": 5.4109,
|
|
"mean_token_accuracy": 0.18593295961618422,
|
|
"num_tokens": 6131786.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.778415393829346,
|
|
"epoch": 2.957274202271498,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045259543144059557,
|
|
"loss": 5.4955,
|
|
"mean_token_accuracy": 0.18639069199562072,
|
|
"num_tokens": 6143012.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.680844163894653,
|
|
"epoch": 2.9626825310978906,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004523317565665479,
|
|
"loss": 5.3744,
|
|
"mean_token_accuracy": 0.19005105644464493,
|
|
"num_tokens": 6153972.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.6828240871429445,
|
|
"epoch": 2.9680908599242835,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00045206743726848716,
|
|
"loss": 5.5606,
|
|
"mean_token_accuracy": 0.18326651751995088,
|
|
"num_tokens": 6166202.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.646218872070312,
|
|
"epoch": 2.973499188750676,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004518024745069502,
|
|
"loss": 5.2738,
|
|
"mean_token_accuracy": 0.19796675592660903,
|
|
"num_tokens": 6177253.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.706516933441162,
|
|
"epoch": 2.9789075175770687,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004515368692448123,
|
|
"loss": 5.5423,
|
|
"mean_token_accuracy": 0.17865623086690902,
|
|
"num_tokens": 6188857.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.7656073570251465,
|
|
"epoch": 2.9843158464034616,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045127062244728353,
|
|
"loss": 5.495,
|
|
"mean_token_accuracy": 0.17895126789808274,
|
|
"num_tokens": 6199945.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.680928325653076,
|
|
"epoch": 2.989724175229854,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004510037350819055,
|
|
"loss": 5.4886,
|
|
"mean_token_accuracy": 0.18967449814081191,
|
|
"num_tokens": 6209866.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.743233966827392,
|
|
"epoch": 2.9951325040562464,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.00045073620811854744,
|
|
"loss": 5.4464,
|
|
"mean_token_accuracy": 0.1788336917757988,
|
|
"num_tokens": 6221636.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.629910839928521,
|
|
"epoch": 3.0,
|
|
"grad_norm": 2.328125,
|
|
"learning_rate": 0.00045046804252940294,
|
|
"loss": 5.3297,
|
|
"mean_token_accuracy": 0.18830747074551052,
|
|
"num_tokens": 6230667.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.69707932472229,
|
|
"epoch": 3.005408328826393,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045019923928898644,
|
|
"loss": 5.118,
|
|
"mean_token_accuracy": 0.21316049247980118,
|
|
"num_tokens": 6242242.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.7481467723846436,
|
|
"epoch": 3.0108166576527853,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004499297993741294,
|
|
"loss": 5.1797,
|
|
"mean_token_accuracy": 0.20156483650207518,
|
|
"num_tokens": 6253337.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.593620777130127,
|
|
"epoch": 3.016224986479178,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004496597237639773,
|
|
"loss": 5.0569,
|
|
"mean_token_accuracy": 0.20687109231948853,
|
|
"num_tokens": 6264582.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.660362005233765,
|
|
"epoch": 3.0216333153055706,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044938901343998536,
|
|
"loss": 5.0992,
|
|
"mean_token_accuracy": 0.21259990483522415,
|
|
"num_tokens": 6276172.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.587935733795166,
|
|
"epoch": 3.0270416441319634,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004491176693859156,
|
|
"loss": 4.9704,
|
|
"mean_token_accuracy": 0.22129097431898118,
|
|
"num_tokens": 6286038.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.540070295333862,
|
|
"epoch": 3.032449972958356,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004488456925878329,
|
|
"loss": 5.0489,
|
|
"mean_token_accuracy": 0.21043426394462586,
|
|
"num_tokens": 6296546.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.669938659667968,
|
|
"epoch": 3.0378583017847487,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044857308403410163,
|
|
"loss": 5.1373,
|
|
"mean_token_accuracy": 0.21281687468290328,
|
|
"num_tokens": 6308001.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.629265689849854,
|
|
"epoch": 3.043266630611141,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00044829984471538183,
|
|
"loss": 5.1439,
|
|
"mean_token_accuracy": 0.20406004339456557,
|
|
"num_tokens": 6320896.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.689284133911133,
|
|
"epoch": 3.048674959437534,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004480259756246258,
|
|
"loss": 5.2108,
|
|
"mean_token_accuracy": 0.2020654261112213,
|
|
"num_tokens": 6332710.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.629602384567261,
|
|
"epoch": 3.0540832882639264,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00044775147775707465,
|
|
"loss": 5.0681,
|
|
"mean_token_accuracy": 0.21166091561317443,
|
|
"num_tokens": 6343563.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.615605401992798,
|
|
"epoch": 3.0594916170903192,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000447476352110254,
|
|
"loss": 5.1731,
|
|
"mean_token_accuracy": 0.20783162117004395,
|
|
"num_tokens": 6354064.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.519037961959839,
|
|
"epoch": 3.0648999459167117,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004472005996839712,
|
|
"loss": 5.1062,
|
|
"mean_token_accuracy": 0.20571049004793168,
|
|
"num_tokens": 6366551.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.60793662071228,
|
|
"epoch": 3.0703082747431045,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004469242214803112,
|
|
"loss": 5.0352,
|
|
"mean_token_accuracy": 0.212779301404953,
|
|
"num_tokens": 6377086.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.6298433303833,
|
|
"epoch": 3.075716603569497,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004466472185036331,
|
|
"loss": 5.1482,
|
|
"mean_token_accuracy": 0.20868430733680726,
|
|
"num_tokens": 6388795.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.626946496963501,
|
|
"epoch": 3.08112493239589,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004463695917605663,
|
|
"loss": 5.1296,
|
|
"mean_token_accuracy": 0.2113844335079193,
|
|
"num_tokens": 6400174.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.570234155654907,
|
|
"epoch": 3.086533261222282,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000446091342260007,
|
|
"loss": 5.0936,
|
|
"mean_token_accuracy": 0.20676784962415695,
|
|
"num_tokens": 6411297.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.585538625717163,
|
|
"epoch": 3.091941590048675,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004458124710131145,
|
|
"loss": 5.1134,
|
|
"mean_token_accuracy": 0.20946162492036818,
|
|
"num_tokens": 6423487.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.724478244781494,
|
|
"epoch": 3.0973499188750675,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00044553297903330766,
|
|
"loss": 5.292,
|
|
"mean_token_accuracy": 0.1967707097530365,
|
|
"num_tokens": 6435460.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.722478771209717,
|
|
"epoch": 3.1027582477014604,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00044525286733626085,
|
|
"loss": 5.2367,
|
|
"mean_token_accuracy": 0.20661808997392656,
|
|
"num_tokens": 6445664.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.563872766494751,
|
|
"epoch": 3.1081665765278528,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004449721369399008,
|
|
"loss": 5.0886,
|
|
"mean_token_accuracy": 0.20634377598762513,
|
|
"num_tokens": 6457182.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.501190280914306,
|
|
"epoch": 3.1135749053542456,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044469078886440227,
|
|
"loss": 4.9532,
|
|
"mean_token_accuracy": 0.22265869826078416,
|
|
"num_tokens": 6468378.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.541498279571533,
|
|
"epoch": 3.118983234180638,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00044440882413218484,
|
|
"loss": 5.187,
|
|
"mean_token_accuracy": 0.20512945502996444,
|
|
"num_tokens": 6479625.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.571180772781372,
|
|
"epoch": 3.124391563007031,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00044412624376790905,
|
|
"loss": 5.0993,
|
|
"mean_token_accuracy": 0.20619929283857347,
|
|
"num_tokens": 6490713.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.489297723770141,
|
|
"epoch": 3.1297998918334233,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004438430487984726,
|
|
"loss": 5.1367,
|
|
"mean_token_accuracy": 0.1967697337269783,
|
|
"num_tokens": 6502161.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.6030020236969,
|
|
"epoch": 3.135208220659816,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004435592402530066,
|
|
"loss": 5.0473,
|
|
"mean_token_accuracy": 0.21011342406272887,
|
|
"num_tokens": 6513639.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.564647388458252,
|
|
"epoch": 3.1406165494862086,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000443274819162872,
|
|
"loss": 5.0812,
|
|
"mean_token_accuracy": 0.20863993614912033,
|
|
"num_tokens": 6524893.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.48593111038208,
|
|
"epoch": 3.1460248783126015,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004429897865616557,
|
|
"loss": 5.0748,
|
|
"mean_token_accuracy": 0.2106839030981064,
|
|
"num_tokens": 6535797.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.594881916046143,
|
|
"epoch": 3.151433207138994,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004427041434851668,
|
|
"loss": 5.167,
|
|
"mean_token_accuracy": 0.20228979587554932,
|
|
"num_tokens": 6547117.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.646280956268311,
|
|
"epoch": 3.1568415359653867,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044241789097143293,
|
|
"loss": 5.1957,
|
|
"mean_token_accuracy": 0.20873761922121048,
|
|
"num_tokens": 6558771.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.559993886947632,
|
|
"epoch": 3.162249864791779,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00044213103006069635,
|
|
"loss": 5.1634,
|
|
"mean_token_accuracy": 0.20020761936903,
|
|
"num_tokens": 6569922.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.6437593460083,
|
|
"epoch": 3.167658193618172,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00044184356179541035,
|
|
"loss": 5.1619,
|
|
"mean_token_accuracy": 0.20496343672275544,
|
|
"num_tokens": 6580561.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.561740875244141,
|
|
"epoch": 3.1730665224445644,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004415554872202352,
|
|
"loss": 5.0106,
|
|
"mean_token_accuracy": 0.20989638566970825,
|
|
"num_tokens": 6591636.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.564278173446655,
|
|
"epoch": 3.1784748512709573,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00044126680738203466,
|
|
"loss": 5.2208,
|
|
"mean_token_accuracy": 0.20207992047071457,
|
|
"num_tokens": 6603736.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.697408962249756,
|
|
"epoch": 3.1838831800973497,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004409775233298718,
|
|
"loss": 5.1527,
|
|
"mean_token_accuracy": 0.2086891621351242,
|
|
"num_tokens": 6615415.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.55486102104187,
|
|
"epoch": 3.1892915089237426,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004406876361150056,
|
|
"loss": 5.1808,
|
|
"mean_token_accuracy": 0.20389774590730667,
|
|
"num_tokens": 6625970.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.594521570205688,
|
|
"epoch": 3.194699837750135,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004403971467908869,
|
|
"loss": 5.1384,
|
|
"mean_token_accuracy": 0.20423722565174102,
|
|
"num_tokens": 6636326.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.557396173477173,
|
|
"epoch": 3.200108166576528,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004401060564131545,
|
|
"loss": 5.1705,
|
|
"mean_token_accuracy": 0.20355864316225053,
|
|
"num_tokens": 6646825.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.588989734649658,
|
|
"epoch": 3.2055164954029207,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004398143660396315,
|
|
"loss": 5.1703,
|
|
"mean_token_accuracy": 0.20732996463775635,
|
|
"num_tokens": 6657748.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.5528037548065186,
|
|
"epoch": 3.210924824229313,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004395220767303215,
|
|
"loss": 5.0564,
|
|
"mean_token_accuracy": 0.21476534754037857,
|
|
"num_tokens": 6668422.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.615345191955567,
|
|
"epoch": 3.2163331530557056,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004392291895474045,
|
|
"loss": 5.1522,
|
|
"mean_token_accuracy": 0.2010224476456642,
|
|
"num_tokens": 6678898.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.608639144897461,
|
|
"epoch": 3.2217414818820984,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004389357055552331,
|
|
"loss": 5.1839,
|
|
"mean_token_accuracy": 0.21051690131425857,
|
|
"num_tokens": 6688691.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.552885484695435,
|
|
"epoch": 3.2271498107084913,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00043864162582032897,
|
|
"loss": 5.075,
|
|
"mean_token_accuracy": 0.21406530737876892,
|
|
"num_tokens": 6700279.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.483924531936646,
|
|
"epoch": 3.2325581395348837,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004383469514113784,
|
|
"loss": 5.0118,
|
|
"mean_token_accuracy": 0.2145751088857651,
|
|
"num_tokens": 6710440.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.53403868675232,
|
|
"epoch": 3.2379664683612766,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004380516833992289,
|
|
"loss": 5.1962,
|
|
"mean_token_accuracy": 0.1990383207798004,
|
|
"num_tokens": 6722347.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.590388536453247,
|
|
"epoch": 3.243374797187669,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000437755822856885,
|
|
"loss": 5.2089,
|
|
"mean_token_accuracy": 0.20684486180543898,
|
|
"num_tokens": 6732936.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 3.243374797187669,
|
|
"eval_entropy": 5.37777090397748,
|
|
"eval_loss": 5.91184663772583,
|
|
"eval_mean_token_accuracy": 0.17475627675991168,
|
|
"eval_num_tokens": 6732936.0,
|
|
"eval_runtime": 1.9364,
|
|
"eval_samples_per_second": 1815.19,
|
|
"eval_steps_per_second": 227.222,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.573069620132446,
|
|
"epoch": 3.248783126014062,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004374593708595049,
|
|
"loss": 5.2147,
|
|
"mean_token_accuracy": 0.2031557783484459,
|
|
"num_tokens": 6744552.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.652343273162842,
|
|
"epoch": 3.2541914548404542,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00043716232848439555,
|
|
"loss": 5.1481,
|
|
"mean_token_accuracy": 0.20604717880487441,
|
|
"num_tokens": 6755749.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.569940423965454,
|
|
"epoch": 3.259599783666847,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00043686469681100985,
|
|
"loss": 5.0899,
|
|
"mean_token_accuracy": 0.20888536125421525,
|
|
"num_tokens": 6766119.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.460270977020263,
|
|
"epoch": 3.2650081124932395,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00043656647692094186,
|
|
"loss": 5.1313,
|
|
"mean_token_accuracy": 0.20296620875597,
|
|
"num_tokens": 6776491.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.69976396560669,
|
|
"epoch": 3.2704164413196324,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004362676698979233,
|
|
"loss": 5.239,
|
|
"mean_token_accuracy": 0.20118464827537536,
|
|
"num_tokens": 6788795.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.628712368011475,
|
|
"epoch": 3.275824770146025,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00043596827682781974,
|
|
"loss": 5.233,
|
|
"mean_token_accuracy": 0.1996966302394867,
|
|
"num_tokens": 6799706.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.597583246231079,
|
|
"epoch": 3.2812330989724177,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004356682987986262,
|
|
"loss": 5.1885,
|
|
"mean_token_accuracy": 0.20046019405126572,
|
|
"num_tokens": 6812454.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.620301389694214,
|
|
"epoch": 3.28664142779881,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004353677369004635,
|
|
"loss": 5.1281,
|
|
"mean_token_accuracy": 0.20747572928667068,
|
|
"num_tokens": 6823950.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.635411548614502,
|
|
"epoch": 3.292049756625203,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00043506659222557426,
|
|
"loss": 5.3024,
|
|
"mean_token_accuracy": 0.19502145648002625,
|
|
"num_tokens": 6835028.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.581868124008179,
|
|
"epoch": 3.2974580854515954,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000434764865868319,
|
|
"loss": 5.1831,
|
|
"mean_token_accuracy": 0.19999626278877258,
|
|
"num_tokens": 6846316.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.527811479568482,
|
|
"epoch": 3.302866414277988,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000434462558925172,
|
|
"loss": 5.1175,
|
|
"mean_token_accuracy": 0.20996025204658508,
|
|
"num_tokens": 6856463.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.623858213424683,
|
|
"epoch": 3.3082747431043806,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00043415967249471734,
|
|
"loss": 5.1497,
|
|
"mean_token_accuracy": 0.20746323466300964,
|
|
"num_tokens": 6867869.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.582069349288941,
|
|
"epoch": 3.3136830719307735,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000433856207677645,
|
|
"loss": 5.1261,
|
|
"mean_token_accuracy": 0.20743919014930726,
|
|
"num_tokens": 6878545.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.489497947692871,
|
|
"epoch": 3.319091400757166,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000433552165576747,
|
|
"loss": 5.1641,
|
|
"mean_token_accuracy": 0.20064468085765838,
|
|
"num_tokens": 6889399.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.624862194061279,
|
|
"epoch": 3.3244997295835588,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00043324754729691275,
|
|
"loss": 5.1596,
|
|
"mean_token_accuracy": 0.2113773137331009,
|
|
"num_tokens": 6900428.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.69494662284851,
|
|
"epoch": 3.329908058409951,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00043294235394512594,
|
|
"loss": 5.2702,
|
|
"mean_token_accuracy": 0.20045885890722276,
|
|
"num_tokens": 6911112.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.521151447296143,
|
|
"epoch": 3.335316387236344,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004326365866304599,
|
|
"loss": 5.0236,
|
|
"mean_token_accuracy": 0.21755116283893586,
|
|
"num_tokens": 6922383.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.547823572158814,
|
|
"epoch": 3.3407247160627365,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004323302464640738,
|
|
"loss": 5.222,
|
|
"mean_token_accuracy": 0.19735278338193893,
|
|
"num_tokens": 6935153.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.670290184020996,
|
|
"epoch": 3.3461330448891293,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00043202333455920843,
|
|
"loss": 5.2639,
|
|
"mean_token_accuracy": 0.20184647887945176,
|
|
"num_tokens": 6946814.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.546437168121338,
|
|
"epoch": 3.3515413737155217,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004317158520311824,
|
|
"loss": 5.1129,
|
|
"mean_token_accuracy": 0.2076606571674347,
|
|
"num_tokens": 6957230.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.44480447769165,
|
|
"epoch": 3.3569497025419146,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004314077999973879,
|
|
"loss": 5.1318,
|
|
"mean_token_accuracy": 0.19855479449033736,
|
|
"num_tokens": 6968060.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.568084287643432,
|
|
"epoch": 3.362358031368307,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004310991795772869,
|
|
"loss": 5.0431,
|
|
"mean_token_accuracy": 0.21078354120254517,
|
|
"num_tokens": 6977782.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.608031511306763,
|
|
"epoch": 3.3677663601947,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004307899918924065,
|
|
"loss": 5.1228,
|
|
"mean_token_accuracy": 0.20318224132061005,
|
|
"num_tokens": 6988252.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.598913288116455,
|
|
"epoch": 3.3731746890210923,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004304802380663356,
|
|
"loss": 5.1324,
|
|
"mean_token_accuracy": 0.2066564679145813,
|
|
"num_tokens": 6998523.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.519530439376831,
|
|
"epoch": 3.378583017847485,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00043016991922472015,
|
|
"loss": 5.1628,
|
|
"mean_token_accuracy": 0.2022525265812874,
|
|
"num_tokens": 7010058.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.593735790252685,
|
|
"epoch": 3.383991346673878,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00042985903649525977,
|
|
"loss": 5.2746,
|
|
"mean_token_accuracy": 0.19244320541620255,
|
|
"num_tokens": 7021289.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.713921213150025,
|
|
"epoch": 3.3893996755002704,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00042954759100770275,
|
|
"loss": 5.3005,
|
|
"mean_token_accuracy": 0.1927209183573723,
|
|
"num_tokens": 7032649.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.635376024246216,
|
|
"epoch": 3.394808004326663,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00042923558389384303,
|
|
"loss": 5.1584,
|
|
"mean_token_accuracy": 0.2046155095100403,
|
|
"num_tokens": 7044120.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.5387166976928714,
|
|
"epoch": 3.4002163331530557,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004289230162875149,
|
|
"loss": 5.1575,
|
|
"mean_token_accuracy": 0.20784637182950974,
|
|
"num_tokens": 7054526.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.431732416152954,
|
|
"epoch": 3.4056246619794486,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00042860988932458987,
|
|
"loss": 5.0371,
|
|
"mean_token_accuracy": 0.21920679807662963,
|
|
"num_tokens": 7064744.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.573424243927002,
|
|
"epoch": 3.411032990805841,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004282962041429721,
|
|
"loss": 5.1564,
|
|
"mean_token_accuracy": 0.20835559219121932,
|
|
"num_tokens": 7076360.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.627830553054809,
|
|
"epoch": 3.4164413196322334,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00042798196188259427,
|
|
"loss": 5.1758,
|
|
"mean_token_accuracy": 0.20924926102161406,
|
|
"num_tokens": 7088114.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.613403463363648,
|
|
"epoch": 3.4218496484586263,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004276671636854133,
|
|
"loss": 5.2014,
|
|
"mean_token_accuracy": 0.19810499548912047,
|
|
"num_tokens": 7099125.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.5529709815979,
|
|
"epoch": 3.427257977285019,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00042735181069540677,
|
|
"loss": 5.1391,
|
|
"mean_token_accuracy": 0.21497857421636582,
|
|
"num_tokens": 7109875.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.55393533706665,
|
|
"epoch": 3.4326663061114115,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004270359040585681,
|
|
"loss": 5.1845,
|
|
"mean_token_accuracy": 0.20396130681037902,
|
|
"num_tokens": 7121388.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.5741424560546875,
|
|
"epoch": 3.4380746349378044,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00042671944492290277,
|
|
"loss": 5.21,
|
|
"mean_token_accuracy": 0.19878134727478028,
|
|
"num_tokens": 7134161.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.624898862838745,
|
|
"epoch": 3.443482963764197,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004264024344384239,
|
|
"loss": 5.2275,
|
|
"mean_token_accuracy": 0.1993679642677307,
|
|
"num_tokens": 7146580.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.573002195358276,
|
|
"epoch": 3.4488912925905897,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00042608487375714847,
|
|
"loss": 5.1357,
|
|
"mean_token_accuracy": 0.19975454062223436,
|
|
"num_tokens": 7156969.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.560785627365112,
|
|
"epoch": 3.454299621416982,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00042576676403309265,
|
|
"loss": 5.0973,
|
|
"mean_token_accuracy": 0.21098205000162124,
|
|
"num_tokens": 7168687.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.597518825531006,
|
|
"epoch": 3.459707950243375,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004254481064222679,
|
|
"loss": 5.1821,
|
|
"mean_token_accuracy": 0.20848626345396043,
|
|
"num_tokens": 7179996.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.607846450805664,
|
|
"epoch": 3.4651162790697674,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00042512890208267677,
|
|
"loss": 5.213,
|
|
"mean_token_accuracy": 0.19970757514238358,
|
|
"num_tokens": 7191214.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.581211709976197,
|
|
"epoch": 3.4705246078961602,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00042480915217430853,
|
|
"loss": 5.2306,
|
|
"mean_token_accuracy": 0.20193232595920563,
|
|
"num_tokens": 7203491.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.650781297683716,
|
|
"epoch": 3.4759329367225527,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00042448885785913507,
|
|
"loss": 5.1573,
|
|
"mean_token_accuracy": 0.20115691423416138,
|
|
"num_tokens": 7215018.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.562850618362427,
|
|
"epoch": 3.4813412655489455,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00042416802030110677,
|
|
"loss": 5.1011,
|
|
"mean_token_accuracy": 0.2093229651451111,
|
|
"num_tokens": 7226209.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.552242565155029,
|
|
"epoch": 3.486749594375338,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00042384664066614787,
|
|
"loss": 5.1159,
|
|
"mean_token_accuracy": 0.2053811177611351,
|
|
"num_tokens": 7236505.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.487067079544067,
|
|
"epoch": 3.492157923201731,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004235247201221528,
|
|
"loss": 5.0375,
|
|
"mean_token_accuracy": 0.21327000260353088,
|
|
"num_tokens": 7247414.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.582530832290649,
|
|
"epoch": 3.497566252028123,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004232022598389816,
|
|
"loss": 5.2253,
|
|
"mean_token_accuracy": 0.2056236296892166,
|
|
"num_tokens": 7258666.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.610252571105957,
|
|
"epoch": 3.502974580854516,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004228792609884557,
|
|
"loss": 5.1518,
|
|
"mean_token_accuracy": 0.20571167320013045,
|
|
"num_tokens": 7270201.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.6072530269622805,
|
|
"epoch": 3.5083829096809085,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00042255572474435364,
|
|
"loss": 5.2132,
|
|
"mean_token_accuracy": 0.20459086447954178,
|
|
"num_tokens": 7282254.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.589420318603516,
|
|
"epoch": 3.5137912385073014,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004222316522824069,
|
|
"loss": 5.2566,
|
|
"mean_token_accuracy": 0.19173479825258255,
|
|
"num_tokens": 7294579.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.562285947799682,
|
|
"epoch": 3.5191995673336938,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004219070447802955,
|
|
"loss": 5.2104,
|
|
"mean_token_accuracy": 0.19856292456388475,
|
|
"num_tokens": 7305822.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.595698451995849,
|
|
"epoch": 3.5246078961600866,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00042158190341764384,
|
|
"loss": 5.2233,
|
|
"mean_token_accuracy": 0.2017119914293289,
|
|
"num_tokens": 7317543.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.565688610076904,
|
|
"epoch": 3.530016224986479,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00042125622937601645,
|
|
"loss": 5.1714,
|
|
"mean_token_accuracy": 0.20284906327724456,
|
|
"num_tokens": 7327804.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.535323143005371,
|
|
"epoch": 3.535424553812872,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004209300238389135,
|
|
"loss": 5.1222,
|
|
"mean_token_accuracy": 0.20388976186513902,
|
|
"num_tokens": 7337045.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.550784778594971,
|
|
"epoch": 3.5408328826392643,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004206032879917667,
|
|
"loss": 5.2015,
|
|
"mean_token_accuracy": 0.21165899336338043,
|
|
"num_tokens": 7348184.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.5675712585449215,
|
|
"epoch": 3.546241211465657,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004202760230219348,
|
|
"loss": 5.1187,
|
|
"mean_token_accuracy": 0.20928198099136353,
|
|
"num_tokens": 7359571.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.499993753433228,
|
|
"epoch": 3.5516495402920496,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00041994823011869955,
|
|
"loss": 5.1258,
|
|
"mean_token_accuracy": 0.20613025724887848,
|
|
"num_tokens": 7370429.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.59150800704956,
|
|
"epoch": 3.5570578691184425,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00041961991047326116,
|
|
"loss": 5.0892,
|
|
"mean_token_accuracy": 0.2082501992583275,
|
|
"num_tokens": 7381569.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.499939250946045,
|
|
"epoch": 3.5624661979448353,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004192910652787338,
|
|
"loss": 5.0841,
|
|
"mean_token_accuracy": 0.20715413987636566,
|
|
"num_tokens": 7392746.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.549461317062378,
|
|
"epoch": 3.5678745267712277,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004189616957301419,
|
|
"loss": 5.1891,
|
|
"mean_token_accuracy": 0.19182858914136885,
|
|
"num_tokens": 7405520.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.501055097579956,
|
|
"epoch": 3.57328285559762,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004186318030244149,
|
|
"loss": 5.1565,
|
|
"mean_token_accuracy": 0.2044055238366127,
|
|
"num_tokens": 7417788.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.576556062698364,
|
|
"epoch": 3.578691184424013,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00041830138836038383,
|
|
"loss": 5.2256,
|
|
"mean_token_accuracy": 0.20100464969873427,
|
|
"num_tokens": 7428912.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.583335733413696,
|
|
"epoch": 3.584099513250406,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004179704529387763,
|
|
"loss": 5.1013,
|
|
"mean_token_accuracy": 0.21064824014902114,
|
|
"num_tokens": 7439797.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.462785291671753,
|
|
"epoch": 3.5895078420767983,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004176389979622125,
|
|
"loss": 5.0936,
|
|
"mean_token_accuracy": 0.20896225422620773,
|
|
"num_tokens": 7451755.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.588175821304321,
|
|
"epoch": 3.5949161709031907,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004173070246352003,
|
|
"loss": 5.1926,
|
|
"mean_token_accuracy": 0.20159001499414445,
|
|
"num_tokens": 7463393.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.448696327209473,
|
|
"epoch": 3.6003244997295836,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00041697453416413184,
|
|
"loss": 5.1384,
|
|
"mean_token_accuracy": 0.20230467319488527,
|
|
"num_tokens": 7474404.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.559615850448608,
|
|
"epoch": 3.6057328285559764,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000416641527757278,
|
|
"loss": 5.1194,
|
|
"mean_token_accuracy": 0.20770540982484817,
|
|
"num_tokens": 7485495.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.4712920665740965,
|
|
"epoch": 3.611141157382369,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00041630800662478514,
|
|
"loss": 5.1036,
|
|
"mean_token_accuracy": 0.20918697565793992,
|
|
"num_tokens": 7497305.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.491168975830078,
|
|
"epoch": 3.6165494862087613,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00041597397197866957,
|
|
"loss": 5.0288,
|
|
"mean_token_accuracy": 0.21322695314884185,
|
|
"num_tokens": 7509008.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.470600843429565,
|
|
"epoch": 3.621957815035154,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004156394250328141,
|
|
"loss": 5.1054,
|
|
"mean_token_accuracy": 0.2145393580198288,
|
|
"num_tokens": 7520079.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.501084804534912,
|
|
"epoch": 3.627366143861547,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000415304367002963,
|
|
"loss": 5.1446,
|
|
"mean_token_accuracy": 0.2071783125400543,
|
|
"num_tokens": 7531294.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.548788261413574,
|
|
"epoch": 3.6327744726879394,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00041496879910671796,
|
|
"loss": 5.1571,
|
|
"mean_token_accuracy": 0.20829515308141708,
|
|
"num_tokens": 7542508.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.529984998703003,
|
|
"epoch": 3.638182801514332,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004146327225635336,
|
|
"loss": 5.102,
|
|
"mean_token_accuracy": 0.20913522839546203,
|
|
"num_tokens": 7553288.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.604449224472046,
|
|
"epoch": 3.6435911303407247,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004142961385947127,
|
|
"loss": 5.2801,
|
|
"mean_token_accuracy": 0.20308633297681808,
|
|
"num_tokens": 7564532.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.537926387786865,
|
|
"epoch": 3.6489994591671175,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00041395904842340215,
|
|
"loss": 5.1105,
|
|
"mean_token_accuracy": 0.21103381514549255,
|
|
"num_tokens": 7575109.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.472426414489746,
|
|
"epoch": 3.65440778799351,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004136214532745887,
|
|
"loss": 5.1106,
|
|
"mean_token_accuracy": 0.20961469113826753,
|
|
"num_tokens": 7585871.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.582272386550903,
|
|
"epoch": 3.6598161168199024,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004132833543750936,
|
|
"loss": 5.1941,
|
|
"mean_token_accuracy": 0.20975491106510163,
|
|
"num_tokens": 7596821.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.451750040054321,
|
|
"epoch": 3.6652244456462952,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00041294475295356917,
|
|
"loss": 5.1274,
|
|
"mean_token_accuracy": 0.20204561948776245,
|
|
"num_tokens": 7608101.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.545164632797241,
|
|
"epoch": 3.670632774472688,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004126056502404937,
|
|
"loss": 5.1188,
|
|
"mean_token_accuracy": 0.2001086190342903,
|
|
"num_tokens": 7618722.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.506054496765136,
|
|
"epoch": 3.6760411032990805,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004122660474681672,
|
|
"loss": 5.1293,
|
|
"mean_token_accuracy": 0.207930351793766,
|
|
"num_tokens": 7630061.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.536324167251587,
|
|
"epoch": 3.6814494321254734,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00041192594587070706,
|
|
"loss": 5.1173,
|
|
"mean_token_accuracy": 0.20743480771780015,
|
|
"num_tokens": 7641830.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.484718084335327,
|
|
"epoch": 3.686857760951866,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00041158534668404325,
|
|
"loss": 5.1588,
|
|
"mean_token_accuracy": 0.20602032393217087,
|
|
"num_tokens": 7652750.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.5122642993927,
|
|
"epoch": 3.6922660897782587,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00041124425114591386,
|
|
"loss": 5.065,
|
|
"mean_token_accuracy": 0.21273843944072723,
|
|
"num_tokens": 7663405.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.462223911285401,
|
|
"epoch": 3.697674418604651,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00041090266049586107,
|
|
"loss": 5.0796,
|
|
"mean_token_accuracy": 0.19900004267692567,
|
|
"num_tokens": 7673958.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.500379133224487,
|
|
"epoch": 3.703082747431044,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004105605759752258,
|
|
"loss": 5.0774,
|
|
"mean_token_accuracy": 0.2081347331404686,
|
|
"num_tokens": 7685475.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.420930194854736,
|
|
"epoch": 3.7084910762574363,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00041021799882714436,
|
|
"loss": 5.0815,
|
|
"mean_token_accuracy": 0.20927014648914338,
|
|
"num_tokens": 7697296.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.539216470718384,
|
|
"epoch": 3.713899405083829,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004098749302965426,
|
|
"loss": 5.2197,
|
|
"mean_token_accuracy": 0.20556779056787491,
|
|
"num_tokens": 7709414.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.562742471694946,
|
|
"epoch": 3.7193077339102216,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004095313716301324,
|
|
"loss": 5.2042,
|
|
"mean_token_accuracy": 0.1960515037178993,
|
|
"num_tokens": 7720965.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.509733247756958,
|
|
"epoch": 3.7247160627366145,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004091873240764069,
|
|
"loss": 5.1567,
|
|
"mean_token_accuracy": 0.19600157886743547,
|
|
"num_tokens": 7733044.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.624178552627564,
|
|
"epoch": 3.730124391563007,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00040884278888563566,
|
|
"loss": 5.1338,
|
|
"mean_token_accuracy": 0.2047843962907791,
|
|
"num_tokens": 7743878.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.433482074737549,
|
|
"epoch": 3.7355327203893998,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00040849776730986014,
|
|
"loss": 5.1322,
|
|
"mean_token_accuracy": 0.21231569796800615,
|
|
"num_tokens": 7755445.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.584301090240478,
|
|
"epoch": 3.740941049215792,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004081522606028899,
|
|
"loss": 5.1974,
|
|
"mean_token_accuracy": 0.2005298689007759,
|
|
"num_tokens": 7765990.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.445444393157959,
|
|
"epoch": 3.746349378042185,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00040780627002029674,
|
|
"loss": 5.0037,
|
|
"mean_token_accuracy": 0.21728041470050813,
|
|
"num_tokens": 7777064.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.393490314483643,
|
|
"epoch": 3.7517577068685775,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004074597968194116,
|
|
"loss": 5.0414,
|
|
"mean_token_accuracy": 0.2053399369120598,
|
|
"num_tokens": 7788618.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.508790159225464,
|
|
"epoch": 3.7571660356949703,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004071128422593186,
|
|
"loss": 5.1788,
|
|
"mean_token_accuracy": 0.1951671987771988,
|
|
"num_tokens": 7798484.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.535230827331543,
|
|
"epoch": 3.7625743645213627,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004067654076008515,
|
|
"loss": 5.1358,
|
|
"mean_token_accuracy": 0.20426508635282517,
|
|
"num_tokens": 7810270.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.502645969390869,
|
|
"epoch": 3.7679826933477556,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00040641749410658877,
|
|
"loss": 5.2302,
|
|
"mean_token_accuracy": 0.20067734271287918,
|
|
"num_tokens": 7822616.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.5701946258544925,
|
|
"epoch": 3.773391022174148,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004060691030408487,
|
|
"loss": 5.1298,
|
|
"mean_token_accuracy": 0.2053201138973236,
|
|
"num_tokens": 7834933.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.543779182434082,
|
|
"epoch": 3.778799351000541,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00040572023566968536,
|
|
"loss": 5.1516,
|
|
"mean_token_accuracy": 0.20760101526975633,
|
|
"num_tokens": 7847686.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.480415725708008,
|
|
"epoch": 3.7842076798269337,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00040537089326088345,
|
|
"loss": 5.1359,
|
|
"mean_token_accuracy": 0.20168039649724961,
|
|
"num_tokens": 7859087.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 3.7842076798269337,
|
|
"eval_entropy": 5.405016609755429,
|
|
"eval_loss": 5.846310615539551,
|
|
"eval_mean_token_accuracy": 0.17764256713403898,
|
|
"eval_num_tokens": 7859087.0,
|
|
"eval_runtime": 1.934,
|
|
"eval_samples_per_second": 1817.514,
|
|
"eval_steps_per_second": 227.512,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.551335096359253,
|
|
"epoch": 3.789616008653326,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004050210770839541,
|
|
"loss": 5.2017,
|
|
"mean_token_accuracy": 0.20217640548944474,
|
|
"num_tokens": 7870684.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.5254621505737305,
|
|
"epoch": 3.7950243374797186,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004046707884101303,
|
|
"loss": 5.0938,
|
|
"mean_token_accuracy": 0.20737007558345794,
|
|
"num_tokens": 7880017.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.4221290111541744,
|
|
"epoch": 3.8004326663061114,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00040432002851236164,
|
|
"loss": 5.0639,
|
|
"mean_token_accuracy": 0.20686111152172088,
|
|
"num_tokens": 7891751.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.604721164703369,
|
|
"epoch": 3.8058409951325043,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004039687986653106,
|
|
"loss": 5.189,
|
|
"mean_token_accuracy": 0.20483125299215316,
|
|
"num_tokens": 7902602.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.49931149482727,
|
|
"epoch": 3.8112493239588967,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00040361710014534714,
|
|
"loss": 5.1038,
|
|
"mean_token_accuracy": 0.20371359288692475,
|
|
"num_tokens": 7913548.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.553002882003784,
|
|
"epoch": 3.816657652785289,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004032649342305445,
|
|
"loss": 5.1024,
|
|
"mean_token_accuracy": 0.20623021125793456,
|
|
"num_tokens": 7925042.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.573310327529907,
|
|
"epoch": 3.822065981611682,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004029123022006745,
|
|
"loss": 5.1528,
|
|
"mean_token_accuracy": 0.20344366431236266,
|
|
"num_tokens": 7937113.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.508363580703735,
|
|
"epoch": 3.827474310438075,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004025592053372027,
|
|
"loss": 5.1108,
|
|
"mean_token_accuracy": 0.20847974568605424,
|
|
"num_tokens": 7947796.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.569135475158691,
|
|
"epoch": 3.8328826392644673,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004022056449232839,
|
|
"loss": 5.1303,
|
|
"mean_token_accuracy": 0.20409662127494813,
|
|
"num_tokens": 7959642.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.491322946548462,
|
|
"epoch": 3.8382909680908597,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004018516222437575,
|
|
"loss": 5.1154,
|
|
"mean_token_accuracy": 0.20295630991458893,
|
|
"num_tokens": 7969996.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.436260223388672,
|
|
"epoch": 3.8436992969172525,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004014971385851428,
|
|
"loss": 5.0914,
|
|
"mean_token_accuracy": 0.20745493471622467,
|
|
"num_tokens": 7980079.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.539716911315918,
|
|
"epoch": 3.8491076257436454,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000401142195235634,
|
|
"loss": 5.1408,
|
|
"mean_token_accuracy": 0.20869143307209015,
|
|
"num_tokens": 7990453.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.585785007476806,
|
|
"epoch": 3.854515954570038,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004007867934850963,
|
|
"loss": 5.2264,
|
|
"mean_token_accuracy": 0.2021670088171959,
|
|
"num_tokens": 8001810.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.557123470306396,
|
|
"epoch": 3.8599242833964302,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004004309346250603,
|
|
"loss": 5.1271,
|
|
"mean_token_accuracy": 0.21573206782341003,
|
|
"num_tokens": 8011821.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.477389097213745,
|
|
"epoch": 3.865332612222823,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004000746199487181,
|
|
"loss": 5.0808,
|
|
"mean_token_accuracy": 0.2135476812720299,
|
|
"num_tokens": 8024106.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.414725351333618,
|
|
"epoch": 3.870740941049216,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000399717850750918,
|
|
"loss": 5.1146,
|
|
"mean_token_accuracy": 0.21074670404195786,
|
|
"num_tokens": 8035630.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.544428586959839,
|
|
"epoch": 3.8761492698756084,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00039936062832816,
|
|
"loss": 5.1236,
|
|
"mean_token_accuracy": 0.21060896068811416,
|
|
"num_tokens": 8047130.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.50486741065979,
|
|
"epoch": 3.8815575987020012,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003990029539785913,
|
|
"loss": 5.1107,
|
|
"mean_token_accuracy": 0.20224897414445878,
|
|
"num_tokens": 8059484.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.589583253860473,
|
|
"epoch": 3.8869659275283936,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00039864482900200135,
|
|
"loss": 5.1796,
|
|
"mean_token_accuracy": 0.2071842595934868,
|
|
"num_tokens": 8069467.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.582700395584107,
|
|
"epoch": 3.8923742563547865,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003982862546998169,
|
|
"loss": 5.1986,
|
|
"mean_token_accuracy": 0.2042753741145134,
|
|
"num_tokens": 8081390.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.450141668319702,
|
|
"epoch": 3.897782585181179,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003979272323750981,
|
|
"loss": 5.0107,
|
|
"mean_token_accuracy": 0.21870003789663314,
|
|
"num_tokens": 8091202.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.491279125213623,
|
|
"epoch": 3.903190914007572,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003975677633325327,
|
|
"loss": 5.1074,
|
|
"mean_token_accuracy": 0.20239077061414718,
|
|
"num_tokens": 8101900.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.511361217498779,
|
|
"epoch": 3.908599242833964,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003972078488784321,
|
|
"loss": 5.1301,
|
|
"mean_token_accuracy": 0.20085853785276414,
|
|
"num_tokens": 8114500.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.440670728683472,
|
|
"epoch": 3.914007571660357,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003968474903207263,
|
|
"loss": 5.0203,
|
|
"mean_token_accuracy": 0.22119539231061935,
|
|
"num_tokens": 8125319.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.51855697631836,
|
|
"epoch": 3.9194159004867495,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003964866889689592,
|
|
"loss": 5.205,
|
|
"mean_token_accuracy": 0.2095368891954422,
|
|
"num_tokens": 8137077.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.4741452693939205,
|
|
"epoch": 3.9248242293131423,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003961254461342834,
|
|
"loss": 5.0732,
|
|
"mean_token_accuracy": 0.20840024203062057,
|
|
"num_tokens": 8148430.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.437690401077271,
|
|
"epoch": 3.9302325581395348,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003957637631294566,
|
|
"loss": 5.0948,
|
|
"mean_token_accuracy": 0.2089670866727829,
|
|
"num_tokens": 8159250.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.53446397781372,
|
|
"epoch": 3.9356408869659276,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00039540164126883554,
|
|
"loss": 5.1914,
|
|
"mean_token_accuracy": 0.1981954827904701,
|
|
"num_tokens": 8170177.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.558510875701904,
|
|
"epoch": 3.94104921579232,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003950390818683718,
|
|
"loss": 5.1299,
|
|
"mean_token_accuracy": 0.20893062353134156,
|
|
"num_tokens": 8181351.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.437467765808106,
|
|
"epoch": 3.946457544618713,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003946760862456071,
|
|
"loss": 5.1447,
|
|
"mean_token_accuracy": 0.21065412312746049,
|
|
"num_tokens": 8193472.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.538116121292115,
|
|
"epoch": 3.9518658734451053,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003943126557196685,
|
|
"loss": 5.2253,
|
|
"mean_token_accuracy": 0.21236508339643478,
|
|
"num_tokens": 8205449.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.431160593032837,
|
|
"epoch": 3.957274202271498,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00039394879161126333,
|
|
"loss": 4.984,
|
|
"mean_token_accuracy": 0.21492949426174163,
|
|
"num_tokens": 8216520.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.514980363845825,
|
|
"epoch": 3.9626825310978906,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003935844952426745,
|
|
"loss": 5.2061,
|
|
"mean_token_accuracy": 0.1991907238960266,
|
|
"num_tokens": 8228727.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.552783584594726,
|
|
"epoch": 3.9680908599242835,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00039321976793775586,
|
|
"loss": 5.1275,
|
|
"mean_token_accuracy": 0.20329211056232452,
|
|
"num_tokens": 8240423.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.43239917755127,
|
|
"epoch": 3.973499188750676,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003928546110219275,
|
|
"loss": 5.0584,
|
|
"mean_token_accuracy": 0.20820835381746292,
|
|
"num_tokens": 8252757.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.32614221572876,
|
|
"epoch": 3.9789075175770687,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00039248902582217024,
|
|
"loss": 5.0328,
|
|
"mean_token_accuracy": 0.20457035601139067,
|
|
"num_tokens": 8263462.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.485026884078979,
|
|
"epoch": 3.9843158464034616,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00039212301366702153,
|
|
"loss": 5.0722,
|
|
"mean_token_accuracy": 0.20876231044530869,
|
|
"num_tokens": 8274364.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.492988204956054,
|
|
"epoch": 3.989724175229854,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00039175657588657063,
|
|
"loss": 5.1129,
|
|
"mean_token_accuracy": 0.20553327947854996,
|
|
"num_tokens": 8286026.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.548378753662109,
|
|
"epoch": 3.9951325040562464,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003913897138124531,
|
|
"loss": 5.1431,
|
|
"mean_token_accuracy": 0.20458422750234603,
|
|
"num_tokens": 8297990.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.561356332567003,
|
|
"epoch": 4.0,
|
|
"grad_norm": 2.53125,
|
|
"learning_rate": 0.0003910224287778466,
|
|
"loss": 5.2022,
|
|
"mean_token_accuracy": 0.19577878382470873,
|
|
"num_tokens": 8307556.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.516380786895752,
|
|
"epoch": 4.005408328826393,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00039065472211746586,
|
|
"loss": 4.9244,
|
|
"mean_token_accuracy": 0.22025375068187714,
|
|
"num_tokens": 8320805.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.472707653045655,
|
|
"epoch": 4.010816657652786,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003902865951675576,
|
|
"loss": 4.9085,
|
|
"mean_token_accuracy": 0.21889005601406097,
|
|
"num_tokens": 8332390.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.439716720581055,
|
|
"epoch": 4.016224986479178,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00038991804926589616,
|
|
"loss": 4.8137,
|
|
"mean_token_accuracy": 0.2251627489924431,
|
|
"num_tokens": 8342325.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.519870328903198,
|
|
"epoch": 4.0216333153055706,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000389549085751778,
|
|
"loss": 4.8283,
|
|
"mean_token_accuracy": 0.22672139108181,
|
|
"num_tokens": 8352115.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.502739715576172,
|
|
"epoch": 4.027041644131963,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00038917970596601765,
|
|
"loss": 4.8956,
|
|
"mean_token_accuracy": 0.22899759858846663,
|
|
"num_tokens": 8364265.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.582091665267944,
|
|
"epoch": 4.032449972958356,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003888099112509419,
|
|
"loss": 4.8653,
|
|
"mean_token_accuracy": 0.22723956555128097,
|
|
"num_tokens": 8375979.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.493903160095215,
|
|
"epoch": 4.037858301784748,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00038843970295038564,
|
|
"loss": 4.8672,
|
|
"mean_token_accuracy": 0.22776562422513963,
|
|
"num_tokens": 8387794.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.4661518096923825,
|
|
"epoch": 4.043266630611141,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00038806908240968666,
|
|
"loss": 4.7663,
|
|
"mean_token_accuracy": 0.22786850184202195,
|
|
"num_tokens": 8398796.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.422012758255005,
|
|
"epoch": 4.048674959437534,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000387698050975681,
|
|
"loss": 4.868,
|
|
"mean_token_accuracy": 0.23278263956308365,
|
|
"num_tokens": 8408788.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.490291261672974,
|
|
"epoch": 4.054083288263927,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003873266099966976,
|
|
"loss": 4.8855,
|
|
"mean_token_accuracy": 0.22312892526388167,
|
|
"num_tokens": 8420373.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.4261997699737545,
|
|
"epoch": 4.059491617090319,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00038695476082255395,
|
|
"loss": 4.8505,
|
|
"mean_token_accuracy": 0.23360294848680496,
|
|
"num_tokens": 8430399.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.485253810882568,
|
|
"epoch": 4.064899945916712,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003865825048045507,
|
|
"loss": 4.9551,
|
|
"mean_token_accuracy": 0.22004724740982057,
|
|
"num_tokens": 8442649.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.599865484237671,
|
|
"epoch": 4.0703082747431045,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003862098432954672,
|
|
"loss": 4.9584,
|
|
"mean_token_accuracy": 0.21665267944335936,
|
|
"num_tokens": 8453985.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.543493270874023,
|
|
"epoch": 4.075716603569497,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003858367776495561,
|
|
"loss": 4.9667,
|
|
"mean_token_accuracy": 0.22337262481451034,
|
|
"num_tokens": 8465566.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.4220458507537845,
|
|
"epoch": 4.081124932395889,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00038546330922253886,
|
|
"loss": 4.7813,
|
|
"mean_token_accuracy": 0.2385610744357109,
|
|
"num_tokens": 8477426.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.424596786499023,
|
|
"epoch": 4.086533261222282,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003850894393716007,
|
|
"loss": 4.891,
|
|
"mean_token_accuracy": 0.22899001091718674,
|
|
"num_tokens": 8489654.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.378886795043945,
|
|
"epoch": 4.091941590048675,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003847151694553855,
|
|
"loss": 4.8464,
|
|
"mean_token_accuracy": 0.22364416271448134,
|
|
"num_tokens": 8500398.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.38915376663208,
|
|
"epoch": 4.097349918875068,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003843405008339908,
|
|
"loss": 4.7978,
|
|
"mean_token_accuracy": 0.22305359542369843,
|
|
"num_tokens": 8511974.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.463239288330078,
|
|
"epoch": 4.10275824770146,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00038396543486896354,
|
|
"loss": 4.8771,
|
|
"mean_token_accuracy": 0.22338348031044006,
|
|
"num_tokens": 8523140.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.437161207199097,
|
|
"epoch": 4.108166576527853,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00038358997292329406,
|
|
"loss": 4.8475,
|
|
"mean_token_accuracy": 0.23260872215032577,
|
|
"num_tokens": 8534261.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.496507406234741,
|
|
"epoch": 4.113574905354246,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00038321411636141214,
|
|
"loss": 4.9264,
|
|
"mean_token_accuracy": 0.218464657664299,
|
|
"num_tokens": 8544620.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.444603824615479,
|
|
"epoch": 4.1189832341806385,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003828378665491811,
|
|
"loss": 4.8743,
|
|
"mean_token_accuracy": 0.2219197779893875,
|
|
"num_tokens": 8555427.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.430146598815918,
|
|
"epoch": 4.1243915630070305,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000382461224853894,
|
|
"loss": 4.8361,
|
|
"mean_token_accuracy": 0.22686755806207656,
|
|
"num_tokens": 8566956.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.493064975738525,
|
|
"epoch": 4.129799891833423,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003820841926442674,
|
|
"loss": 5.0026,
|
|
"mean_token_accuracy": 0.21972213834524154,
|
|
"num_tokens": 8579373.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.394481182098389,
|
|
"epoch": 4.135208220659816,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003817067712904372,
|
|
"loss": 4.8128,
|
|
"mean_token_accuracy": 0.22822466492652893,
|
|
"num_tokens": 8590979.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.426569604873658,
|
|
"epoch": 4.140616549486209,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00038132896216395355,
|
|
"loss": 4.8639,
|
|
"mean_token_accuracy": 0.21969206333160402,
|
|
"num_tokens": 8602514.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.362453746795654,
|
|
"epoch": 4.146024878312601,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003809507666377758,
|
|
"loss": 4.8171,
|
|
"mean_token_accuracy": 0.23407612293958663,
|
|
"num_tokens": 8613766.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.50042085647583,
|
|
"epoch": 4.151433207138994,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00038057218608626714,
|
|
"loss": 4.9248,
|
|
"mean_token_accuracy": 0.21400074809789657,
|
|
"num_tokens": 8625245.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.4962865829467775,
|
|
"epoch": 4.156841535965387,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003801932218851903,
|
|
"loss": 4.8794,
|
|
"mean_token_accuracy": 0.22843072265386583,
|
|
"num_tokens": 8635653.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.381710243225098,
|
|
"epoch": 4.16224986479178,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000379813875411702,
|
|
"loss": 4.8153,
|
|
"mean_token_accuracy": 0.22896498590707778,
|
|
"num_tokens": 8646671.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.364479875564575,
|
|
"epoch": 4.167658193618172,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00037943414804434826,
|
|
"loss": 4.7778,
|
|
"mean_token_accuracy": 0.22782211005687714,
|
|
"num_tokens": 8656441.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.4185450077056885,
|
|
"epoch": 4.173066522444564,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003790540411630592,
|
|
"loss": 4.8505,
|
|
"mean_token_accuracy": 0.22811439931392669,
|
|
"num_tokens": 8666632.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.423897933959961,
|
|
"epoch": 4.178474851270957,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00037867355614914404,
|
|
"loss": 4.9034,
|
|
"mean_token_accuracy": 0.22682637721300125,
|
|
"num_tokens": 8678269.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.439021587371826,
|
|
"epoch": 4.18388318009735,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003782926943852864,
|
|
"loss": 4.9067,
|
|
"mean_token_accuracy": 0.21999775618314743,
|
|
"num_tokens": 8690004.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.507899904251099,
|
|
"epoch": 4.189291508923742,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00037791145725553853,
|
|
"loss": 4.9178,
|
|
"mean_token_accuracy": 0.218192021548748,
|
|
"num_tokens": 8702306.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.421620464324951,
|
|
"epoch": 4.194699837750135,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003775298461453172,
|
|
"loss": 4.9528,
|
|
"mean_token_accuracy": 0.22453745305538178,
|
|
"num_tokens": 8714070.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.428178834915161,
|
|
"epoch": 4.200108166576528,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000377147862441398,
|
|
"loss": 4.8898,
|
|
"mean_token_accuracy": 0.22364890873432158,
|
|
"num_tokens": 8724433.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.447650480270386,
|
|
"epoch": 4.205516495402921,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00037676550753191075,
|
|
"loss": 4.8255,
|
|
"mean_token_accuracy": 0.23286972492933272,
|
|
"num_tokens": 8735485.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.391544246673584,
|
|
"epoch": 4.210924824229313,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003763827828063339,
|
|
"loss": 4.8589,
|
|
"mean_token_accuracy": 0.21500830799341203,
|
|
"num_tokens": 8746356.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.404104900360108,
|
|
"epoch": 4.2163331530557056,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00037599968965549,
|
|
"loss": 4.8712,
|
|
"mean_token_accuracy": 0.22796602547168732,
|
|
"num_tokens": 8757667.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.412394237518311,
|
|
"epoch": 4.221741481882098,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003756162294715405,
|
|
"loss": 4.8361,
|
|
"mean_token_accuracy": 0.23020044714212418,
|
|
"num_tokens": 8768565.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.36433777809143,
|
|
"epoch": 4.227149810708491,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00037523240364798063,
|
|
"loss": 4.8658,
|
|
"mean_token_accuracy": 0.2293240800499916,
|
|
"num_tokens": 8779608.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.442551517486573,
|
|
"epoch": 4.232558139534884,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00037484821357963423,
|
|
"loss": 4.8974,
|
|
"mean_token_accuracy": 0.23103394657373427,
|
|
"num_tokens": 8791492.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.424734306335449,
|
|
"epoch": 4.237966468361276,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000374463660662649,
|
|
"loss": 4.9164,
|
|
"mean_token_accuracy": 0.2203960731625557,
|
|
"num_tokens": 8803280.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.4306309700012205,
|
|
"epoch": 4.243374797187669,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000374078746294491,
|
|
"loss": 4.9312,
|
|
"mean_token_accuracy": 0.22432847917079926,
|
|
"num_tokens": 8814626.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.395336437225342,
|
|
"epoch": 4.248783126014062,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00037369347187393996,
|
|
"loss": 4.7864,
|
|
"mean_token_accuracy": 0.23629676103591918,
|
|
"num_tokens": 8825520.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.3539650440216064,
|
|
"epoch": 4.254191454840455,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003733078388010842,
|
|
"loss": 4.805,
|
|
"mean_token_accuracy": 0.23205492049455642,
|
|
"num_tokens": 8835980.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.483497524261475,
|
|
"epoch": 4.259599783666847,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003729218484773149,
|
|
"loss": 4.9902,
|
|
"mean_token_accuracy": 0.2130762293934822,
|
|
"num_tokens": 8846714.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.492846965789795,
|
|
"epoch": 4.2650081124932395,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003725355023053221,
|
|
"loss": 4.9065,
|
|
"mean_token_accuracy": 0.22502609044313432,
|
|
"num_tokens": 8858215.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.372627115249633,
|
|
"epoch": 4.270416441319632,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003721488016890884,
|
|
"loss": 4.8197,
|
|
"mean_token_accuracy": 0.23295669555664061,
|
|
"num_tokens": 8870587.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.28197865486145,
|
|
"epoch": 4.275824770146025,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00037176174803388496,
|
|
"loss": 4.8624,
|
|
"mean_token_accuracy": 0.22682570666074753,
|
|
"num_tokens": 8881335.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.396572685241699,
|
|
"epoch": 4.281233098972417,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00037137434274626545,
|
|
"loss": 4.8376,
|
|
"mean_token_accuracy": 0.22881019860506058,
|
|
"num_tokens": 8893294.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.471124839782715,
|
|
"epoch": 4.28664142779881,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003709865872340618,
|
|
"loss": 4.9166,
|
|
"mean_token_accuracy": 0.22167122811079026,
|
|
"num_tokens": 8903914.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.452703475952148,
|
|
"epoch": 4.292049756625203,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003705984829063782,
|
|
"loss": 4.8952,
|
|
"mean_token_accuracy": 0.2186586394906044,
|
|
"num_tokens": 8915591.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.3851700782775875,
|
|
"epoch": 4.297458085451596,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003702100311735869,
|
|
"loss": 4.898,
|
|
"mean_token_accuracy": 0.21510908305644988,
|
|
"num_tokens": 8927031.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.3314032554626465,
|
|
"epoch": 4.302866414277988,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00036982123344732223,
|
|
"loss": 4.8029,
|
|
"mean_token_accuracy": 0.23135318458080292,
|
|
"num_tokens": 8937220.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.383946418762207,
|
|
"epoch": 4.308274743104381,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00036943209114047613,
|
|
"loss": 4.8239,
|
|
"mean_token_accuracy": 0.2245507761836052,
|
|
"num_tokens": 8948032.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.343261575698852,
|
|
"epoch": 4.3136830719307735,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003690426056671926,
|
|
"loss": 4.9368,
|
|
"mean_token_accuracy": 0.21605639308691024,
|
|
"num_tokens": 8960735.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.365348625183105,
|
|
"epoch": 4.319091400757166,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003686527784428627,
|
|
"loss": 4.802,
|
|
"mean_token_accuracy": 0.23148000985383987,
|
|
"num_tokens": 8971801.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.445390844345093,
|
|
"epoch": 4.324499729583558,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00036826261088411955,
|
|
"loss": 4.9174,
|
|
"mean_token_accuracy": 0.21819399148225785,
|
|
"num_tokens": 8983041.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 4.324499729583558,
|
|
"eval_entropy": 5.256173555417495,
|
|
"eval_loss": 5.853297233581543,
|
|
"eval_mean_token_accuracy": 0.17812022828412327,
|
|
"eval_num_tokens": 8983041.0,
|
|
"eval_runtime": 1.9375,
|
|
"eval_samples_per_second": 1814.172,
|
|
"eval_steps_per_second": 227.094,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"entropy": 5.395314359664917,
|
|
"epoch": 4.329908058409951,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003678721044088329,
|
|
"loss": 4.8891,
|
|
"mean_token_accuracy": 0.22300076931715013,
|
|
"num_tokens": 8993935.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 5.417083501815796,
|
|
"epoch": 4.335316387236344,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003674812604361043,
|
|
"loss": 4.8451,
|
|
"mean_token_accuracy": 0.22310078740119935,
|
|
"num_tokens": 9003896.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 5.417169666290283,
|
|
"epoch": 4.340724716062737,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003670900803862615,
|
|
"loss": 4.8902,
|
|
"mean_token_accuracy": 0.22965257316827775,
|
|
"num_tokens": 9014323.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 5.377078676223755,
|
|
"epoch": 4.346133044889129,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003666985656808537,
|
|
"loss": 4.9276,
|
|
"mean_token_accuracy": 0.22262843698263168,
|
|
"num_tokens": 9024252.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 5.453081893920898,
|
|
"epoch": 4.351541373715522,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00036630671774264635,
|
|
"loss": 4.9615,
|
|
"mean_token_accuracy": 0.21441965997219087,
|
|
"num_tokens": 9035178.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 5.464565658569336,
|
|
"epoch": 4.356949702541915,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003659145379956158,
|
|
"loss": 4.87,
|
|
"mean_token_accuracy": 0.22782188504934311,
|
|
"num_tokens": 9045807.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 5.50302529335022,
|
|
"epoch": 4.3623580313683075,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00036552202786494393,
|
|
"loss": 4.9132,
|
|
"mean_token_accuracy": 0.22224678099155426,
|
|
"num_tokens": 9056927.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 5.4103028774261475,
|
|
"epoch": 4.367766360194699,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00036512918877701367,
|
|
"loss": 4.9064,
|
|
"mean_token_accuracy": 0.2255491554737091,
|
|
"num_tokens": 9070141.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 5.362295913696289,
|
|
"epoch": 4.373174689021092,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00036473602215940316,
|
|
"loss": 4.8078,
|
|
"mean_token_accuracy": 0.23299480080604554,
|
|
"num_tokens": 9081595.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 5.409234237670899,
|
|
"epoch": 4.378583017847485,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003643425294408807,
|
|
"loss": 4.8946,
|
|
"mean_token_accuracy": 0.22599810659885405,
|
|
"num_tokens": 9092158.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 5.413430023193359,
|
|
"epoch": 4.383991346673878,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00036394871205139984,
|
|
"loss": 4.8958,
|
|
"mean_token_accuracy": 0.22218936383724214,
|
|
"num_tokens": 9102958.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 5.413196992874146,
|
|
"epoch": 4.38939967550027,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003635545714220938,
|
|
"loss": 4.8484,
|
|
"mean_token_accuracy": 0.22495583444833755,
|
|
"num_tokens": 9113793.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 5.470857620239258,
|
|
"epoch": 4.394808004326663,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003631601089852706,
|
|
"loss": 4.9621,
|
|
"mean_token_accuracy": 0.22081742137670518,
|
|
"num_tokens": 9125149.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 5.357998609542847,
|
|
"epoch": 4.400216333153056,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003627653261744075,
|
|
"loss": 4.7697,
|
|
"mean_token_accuracy": 0.23348610997200012,
|
|
"num_tokens": 9136222.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 5.419644355773926,
|
|
"epoch": 4.405624661979449,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00036237022442414636,
|
|
"loss": 4.933,
|
|
"mean_token_accuracy": 0.22234244644641876,
|
|
"num_tokens": 9147933.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 5.423195886611938,
|
|
"epoch": 4.411032990805841,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003619748051702878,
|
|
"loss": 4.9544,
|
|
"mean_token_accuracy": 0.22362978905439376,
|
|
"num_tokens": 9159421.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 5.435204029083252,
|
|
"epoch": 4.416441319632233,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003615790698497862,
|
|
"loss": 4.861,
|
|
"mean_token_accuracy": 0.22939157634973525,
|
|
"num_tokens": 9170676.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 5.428468894958496,
|
|
"epoch": 4.421849648458626,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00036118301990074463,
|
|
"loss": 4.8677,
|
|
"mean_token_accuracy": 0.22782688438892365,
|
|
"num_tokens": 9181267.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 5.32914023399353,
|
|
"epoch": 4.427257977285019,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003607866567624097,
|
|
"loss": 4.9124,
|
|
"mean_token_accuracy": 0.2182913079857826,
|
|
"num_tokens": 9192999.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 5.430573225021362,
|
|
"epoch": 4.432666306111411,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00036038998187516583,
|
|
"loss": 4.8276,
|
|
"mean_token_accuracy": 0.22866062670946122,
|
|
"num_tokens": 9203778.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 5.402996969223023,
|
|
"epoch": 4.438074634937804,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00035999299668053043,
|
|
"loss": 4.8715,
|
|
"mean_token_accuracy": 0.22328359186649321,
|
|
"num_tokens": 9214331.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 5.39242467880249,
|
|
"epoch": 4.443482963764197,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003595957026211487,
|
|
"loss": 4.8946,
|
|
"mean_token_accuracy": 0.22709545642137527,
|
|
"num_tokens": 9224809.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 5.498127031326294,
|
|
"epoch": 4.44889129259059,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00035919810114078815,
|
|
"loss": 5.0167,
|
|
"mean_token_accuracy": 0.21283081471920012,
|
|
"num_tokens": 9235852.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 5.43682074546814,
|
|
"epoch": 4.4542996214169825,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00035880019368433354,
|
|
"loss": 4.8726,
|
|
"mean_token_accuracy": 0.23312048017978668,
|
|
"num_tokens": 9246296.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 5.428623342514038,
|
|
"epoch": 4.4597079502433745,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003584019816977813,
|
|
"loss": 4.8738,
|
|
"mean_token_accuracy": 0.23407673388719558,
|
|
"num_tokens": 9257761.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 5.406604623794555,
|
|
"epoch": 4.465116279069767,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00035800346662823493,
|
|
"loss": 4.9574,
|
|
"mean_token_accuracy": 0.21968206614255906,
|
|
"num_tokens": 9268796.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 5.500720596313476,
|
|
"epoch": 4.47052460789616,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000357604649923899,
|
|
"loss": 4.9712,
|
|
"mean_token_accuracy": 0.21589330285787584,
|
|
"num_tokens": 9280625.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 5.380571031570435,
|
|
"epoch": 4.475932936722553,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003572055330340744,
|
|
"loss": 4.8079,
|
|
"mean_token_accuracy": 0.22965731024742125,
|
|
"num_tokens": 9291096.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 5.401041507720947,
|
|
"epoch": 4.481341265548945,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00035680611740915264,
|
|
"loss": 4.845,
|
|
"mean_token_accuracy": 0.23129767328500747,
|
|
"num_tokens": 9301664.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 5.282374048233033,
|
|
"epoch": 4.486749594375338,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003564064045006113,
|
|
"loss": 4.8545,
|
|
"mean_token_accuracy": 0.22348852455615997,
|
|
"num_tokens": 9312907.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 5.4265923500061035,
|
|
"epoch": 4.492157923201731,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00035600639576100786,
|
|
"loss": 4.9584,
|
|
"mean_token_accuracy": 0.21911619156599044,
|
|
"num_tokens": 9323806.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 5.509041357040405,
|
|
"epoch": 4.497566252028124,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00035560609264397497,
|
|
"loss": 4.9637,
|
|
"mean_token_accuracy": 0.21484742313623428,
|
|
"num_tokens": 9336394.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 5.394302320480347,
|
|
"epoch": 4.502974580854516,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00035520549660421515,
|
|
"loss": 4.8674,
|
|
"mean_token_accuracy": 0.23138714134693145,
|
|
"num_tokens": 9349177.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 5.412147045135498,
|
|
"epoch": 4.5083829096809085,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003548046090974954,
|
|
"loss": 4.9425,
|
|
"mean_token_accuracy": 0.22144696861505508,
|
|
"num_tokens": 9360033.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 5.404139518737793,
|
|
"epoch": 4.513791238507301,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00035440343158064176,
|
|
"loss": 4.9076,
|
|
"mean_token_accuracy": 0.21946440041065216,
|
|
"num_tokens": 9371300.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 5.397027015686035,
|
|
"epoch": 4.519199567333694,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003540019655115341,
|
|
"loss": 4.9228,
|
|
"mean_token_accuracy": 0.22527797520160675,
|
|
"num_tokens": 9383112.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 5.3305552959442135,
|
|
"epoch": 4.524607896160086,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00035360021234910133,
|
|
"loss": 4.8047,
|
|
"mean_token_accuracy": 0.23085560500621796,
|
|
"num_tokens": 9394062.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 5.448763990402222,
|
|
"epoch": 4.530016224986479,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00035319817355331537,
|
|
"loss": 4.9479,
|
|
"mean_token_accuracy": 0.2256651446223259,
|
|
"num_tokens": 9406016.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 5.475788640975952,
|
|
"epoch": 4.535424553812872,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003527958505851858,
|
|
"loss": 4.9628,
|
|
"mean_token_accuracy": 0.22180442810058593,
|
|
"num_tokens": 9417286.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 5.349158048629761,
|
|
"epoch": 4.540832882639265,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00035239324490675546,
|
|
"loss": 4.8271,
|
|
"mean_token_accuracy": 0.22391604632139206,
|
|
"num_tokens": 9428794.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 5.442984342575073,
|
|
"epoch": 4.546241211465657,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003519903579810943,
|
|
"loss": 4.9665,
|
|
"mean_token_accuracy": 0.21939528733491898,
|
|
"num_tokens": 9440524.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 5.373453378677368,
|
|
"epoch": 4.55164954029205,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003515871912722944,
|
|
"loss": 4.8329,
|
|
"mean_token_accuracy": 0.2274889200925827,
|
|
"num_tokens": 9450796.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 5.428749704360962,
|
|
"epoch": 4.5570578691184425,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003511837462454643,
|
|
"loss": 4.9664,
|
|
"mean_token_accuracy": 0.22118359804153442,
|
|
"num_tokens": 9462498.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 5.488862752914429,
|
|
"epoch": 4.562466197944835,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00035078002436672443,
|
|
"loss": 4.8941,
|
|
"mean_token_accuracy": 0.22850249111652374,
|
|
"num_tokens": 9473395.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 5.4306478023529055,
|
|
"epoch": 4.567874526771227,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003503760271032007,
|
|
"loss": 4.9295,
|
|
"mean_token_accuracy": 0.2237750545144081,
|
|
"num_tokens": 9485099.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 5.446264028549194,
|
|
"epoch": 4.57328285559762,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003499717559230205,
|
|
"loss": 4.9456,
|
|
"mean_token_accuracy": 0.21358946412801744,
|
|
"num_tokens": 9495129.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 5.425550889968872,
|
|
"epoch": 4.578691184424013,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00034956721229530586,
|
|
"loss": 4.9736,
|
|
"mean_token_accuracy": 0.22066534012556077,
|
|
"num_tokens": 9506933.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 5.438918590545654,
|
|
"epoch": 4.584099513250406,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003491623976901695,
|
|
"loss": 4.8375,
|
|
"mean_token_accuracy": 0.22926700562238694,
|
|
"num_tokens": 9518262.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 5.479493904113769,
|
|
"epoch": 4.589507842076799,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003487573135787085,
|
|
"loss": 4.8944,
|
|
"mean_token_accuracy": 0.22381129413843154,
|
|
"num_tokens": 9530829.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 5.288048219680786,
|
|
"epoch": 4.594916170903191,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00034835196143299956,
|
|
"loss": 4.7721,
|
|
"mean_token_accuracy": 0.23331840932369233,
|
|
"num_tokens": 9541762.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 5.352424430847168,
|
|
"epoch": 4.600324499729584,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00034794634272609313,
|
|
"loss": 4.8674,
|
|
"mean_token_accuracy": 0.22781521081924438,
|
|
"num_tokens": 9552344.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 5.297861337661743,
|
|
"epoch": 4.605732828555976,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003475404589320089,
|
|
"loss": 4.7967,
|
|
"mean_token_accuracy": 0.2222983106970787,
|
|
"num_tokens": 9563806.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 5.428743457794189,
|
|
"epoch": 4.611141157382368,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003471343115257291,
|
|
"loss": 4.9458,
|
|
"mean_token_accuracy": 0.21898980140686036,
|
|
"num_tokens": 9575305.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 5.516720008850098,
|
|
"epoch": 4.616549486208761,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00034672790198319453,
|
|
"loss": 4.9777,
|
|
"mean_token_accuracy": 0.22088178843259812,
|
|
"num_tokens": 9586155.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 5.4321221828460695,
|
|
"epoch": 4.621957815035154,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003463212317812985,
|
|
"loss": 4.9553,
|
|
"mean_token_accuracy": 0.2235315263271332,
|
|
"num_tokens": 9597731.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 5.528243541717529,
|
|
"epoch": 4.627366143861547,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003459143023978813,
|
|
"loss": 4.9336,
|
|
"mean_token_accuracy": 0.22308813631534577,
|
|
"num_tokens": 9609438.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 5.425971794128418,
|
|
"epoch": 4.63277447268794,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003455071153117252,
|
|
"loss": 4.9122,
|
|
"mean_token_accuracy": 0.2167592778801918,
|
|
"num_tokens": 9620665.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 5.431094121932984,
|
|
"epoch": 4.638182801514332,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003450996720025493,
|
|
"loss": 4.9398,
|
|
"mean_token_accuracy": 0.22015288919210435,
|
|
"num_tokens": 9631868.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 5.3476465225219725,
|
|
"epoch": 4.643591130340725,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00034469197395100323,
|
|
"loss": 4.8077,
|
|
"mean_token_accuracy": 0.22980467230081558,
|
|
"num_tokens": 9643879.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 5.423450756072998,
|
|
"epoch": 4.6489994591671175,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00034428402263866285,
|
|
"loss": 4.8751,
|
|
"mean_token_accuracy": 0.23372543305158616,
|
|
"num_tokens": 9656277.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 5.401059675216675,
|
|
"epoch": 4.6544077879935095,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00034387581954802406,
|
|
"loss": 4.9536,
|
|
"mean_token_accuracy": 0.21922286599874496,
|
|
"num_tokens": 9666590.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 5.478869819641114,
|
|
"epoch": 4.659816116819902,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00034346736616249786,
|
|
"loss": 4.9065,
|
|
"mean_token_accuracy": 0.22272006422281265,
|
|
"num_tokens": 9677248.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 5.360098934173584,
|
|
"epoch": 4.665224445646295,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00034305866396640473,
|
|
"loss": 4.8628,
|
|
"mean_token_accuracy": 0.2242682993412018,
|
|
"num_tokens": 9689617.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 5.371276187896728,
|
|
"epoch": 4.670632774472688,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00034264971444496954,
|
|
"loss": 4.9742,
|
|
"mean_token_accuracy": 0.2143051564693451,
|
|
"num_tokens": 9702663.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 5.420400857925415,
|
|
"epoch": 4.676041103299081,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00034224051908431574,
|
|
"loss": 4.976,
|
|
"mean_token_accuracy": 0.2184092655777931,
|
|
"num_tokens": 9714005.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 5.470933246612549,
|
|
"epoch": 4.681449432125473,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003418310793714602,
|
|
"loss": 4.8525,
|
|
"mean_token_accuracy": 0.23021799325942993,
|
|
"num_tokens": 9724931.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 5.415314674377441,
|
|
"epoch": 4.686857760951866,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003414213967943077,
|
|
"loss": 4.881,
|
|
"mean_token_accuracy": 0.22342510968446733,
|
|
"num_tokens": 9734948.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 5.320093631744385,
|
|
"epoch": 4.692266089778259,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00034101147284164583,
|
|
"loss": 4.827,
|
|
"mean_token_accuracy": 0.22371249943971633,
|
|
"num_tokens": 9746424.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 5.4916672706604,
|
|
"epoch": 4.6976744186046515,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00034060130900313905,
|
|
"loss": 4.9784,
|
|
"mean_token_accuracy": 0.2240543022751808,
|
|
"num_tokens": 9758434.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 5.365119361877442,
|
|
"epoch": 4.7030827474310435,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00034019090676932383,
|
|
"loss": 4.8687,
|
|
"mean_token_accuracy": 0.22543973326683045,
|
|
"num_tokens": 9768869.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 5.446120548248291,
|
|
"epoch": 4.708491076257436,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003397802676316027,
|
|
"loss": 5.0063,
|
|
"mean_token_accuracy": 0.21798664927482606,
|
|
"num_tokens": 9780001.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 5.4792985916137695,
|
|
"epoch": 4.713899405083829,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00033936939308223935,
|
|
"loss": 4.961,
|
|
"mean_token_accuracy": 0.21688656359910966,
|
|
"num_tokens": 9790609.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 5.404905891418457,
|
|
"epoch": 4.719307733910222,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00033895828461435275,
|
|
"loss": 4.8758,
|
|
"mean_token_accuracy": 0.23324273377656937,
|
|
"num_tokens": 9802064.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 5.359910249710083,
|
|
"epoch": 4.724716062736614,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003385469437219121,
|
|
"loss": 4.9172,
|
|
"mean_token_accuracy": 0.22497787773609162,
|
|
"num_tokens": 9812124.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 5.420432138442993,
|
|
"epoch": 4.730124391563007,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00033813537189973105,
|
|
"loss": 4.7939,
|
|
"mean_token_accuracy": 0.23899793475866318,
|
|
"num_tokens": 9824371.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 5.372838926315308,
|
|
"epoch": 4.7355327203894,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00033772357064346266,
|
|
"loss": 4.8401,
|
|
"mean_token_accuracy": 0.2293306663632393,
|
|
"num_tokens": 9834919.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 5.344186544418335,
|
|
"epoch": 4.740941049215793,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003373115414495935,
|
|
"loss": 4.8977,
|
|
"mean_token_accuracy": 0.2206204667687416,
|
|
"num_tokens": 9845435.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 5.42653112411499,
|
|
"epoch": 4.746349378042185,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003368992858154389,
|
|
"loss": 4.8848,
|
|
"mean_token_accuracy": 0.22370723485946656,
|
|
"num_tokens": 9856091.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 5.340574312210083,
|
|
"epoch": 4.7517577068685775,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00033648680523913643,
|
|
"loss": 4.7422,
|
|
"mean_token_accuracy": 0.23152807205915452,
|
|
"num_tokens": 9866861.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 5.344461679458618,
|
|
"epoch": 4.75716603569497,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003360741012196417,
|
|
"loss": 4.8515,
|
|
"mean_token_accuracy": 0.22048368006944657,
|
|
"num_tokens": 9877852.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 5.406573343276977,
|
|
"epoch": 4.762574364521363,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000335661175256722,
|
|
"loss": 4.7993,
|
|
"mean_token_accuracy": 0.23371399194002151,
|
|
"num_tokens": 9888938.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 5.418169927597046,
|
|
"epoch": 4.767982693347756,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003352480288509513,
|
|
"loss": 4.8704,
|
|
"mean_token_accuracy": 0.2362971380352974,
|
|
"num_tokens": 9901782.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 5.381399631500244,
|
|
"epoch": 4.773391022174148,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00033483466350370445,
|
|
"loss": 4.9857,
|
|
"mean_token_accuracy": 0.21602330952882767,
|
|
"num_tokens": 9913254.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 5.40879077911377,
|
|
"epoch": 4.778799351000541,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00033442108071715215,
|
|
"loss": 4.8902,
|
|
"mean_token_accuracy": 0.23002710789442063,
|
|
"num_tokens": 9923170.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 5.480787515640259,
|
|
"epoch": 4.784207679826934,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00033400728199425525,
|
|
"loss": 4.8945,
|
|
"mean_token_accuracy": 0.22826845496892928,
|
|
"num_tokens": 9934087.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 5.352041912078858,
|
|
"epoch": 4.789616008653326,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00033359326883875917,
|
|
"loss": 4.976,
|
|
"mean_token_accuracy": 0.21525231152772903,
|
|
"num_tokens": 9945463.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 5.373518085479736,
|
|
"epoch": 4.795024337479719,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00033317904275518873,
|
|
"loss": 4.8342,
|
|
"mean_token_accuracy": 0.2260815382003784,
|
|
"num_tokens": 9955602.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 5.393237686157226,
|
|
"epoch": 4.800432666306111,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003327646052488425,
|
|
"loss": 4.8622,
|
|
"mean_token_accuracy": 0.22815865874290467,
|
|
"num_tokens": 9965899.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 5.45719199180603,
|
|
"epoch": 4.805840995132504,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003323499578257873,
|
|
"loss": 5.0196,
|
|
"mean_token_accuracy": 0.21645006388425828,
|
|
"num_tokens": 9978047.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 5.538795709609985,
|
|
"epoch": 4.811249323958897,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00033193510199285286,
|
|
"loss": 5.0631,
|
|
"mean_token_accuracy": 0.21536046266555786,
|
|
"num_tokens": 9991760.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 5.454887819290161,
|
|
"epoch": 4.816657652785289,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00033152003925762615,
|
|
"loss": 4.9,
|
|
"mean_token_accuracy": 0.22307134568691253,
|
|
"num_tokens": 10002540.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 5.440888166427612,
|
|
"epoch": 4.822065981611682,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003311047711284462,
|
|
"loss": 4.9163,
|
|
"mean_token_accuracy": 0.21778731495141984,
|
|
"num_tokens": 10013955.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 5.395072555541992,
|
|
"epoch": 4.827474310438075,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003306892991143983,
|
|
"loss": 4.9017,
|
|
"mean_token_accuracy": 0.22153952419757844,
|
|
"num_tokens": 10024845.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 5.40009183883667,
|
|
"epoch": 4.832882639264467,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003302736247253087,
|
|
"loss": 4.9417,
|
|
"mean_token_accuracy": 0.2241146445274353,
|
|
"num_tokens": 10036440.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 5.4532633304595945,
|
|
"epoch": 4.83829096809086,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003298577494717391,
|
|
"loss": 4.937,
|
|
"mean_token_accuracy": 0.223230342566967,
|
|
"num_tokens": 10049891.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 5.36614499092102,
|
|
"epoch": 4.8436992969172525,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003294416748649812,
|
|
"loss": 4.8228,
|
|
"mean_token_accuracy": 0.23392285108566285,
|
|
"num_tokens": 10061105.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 5.392772626876831,
|
|
"epoch": 4.849107625743645,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00032902540241705106,
|
|
"loss": 4.8764,
|
|
"mean_token_accuracy": 0.22940007746219634,
|
|
"num_tokens": 10072721.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 5.358159923553467,
|
|
"epoch": 4.854515954570038,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003286089336406837,
|
|
"loss": 4.8746,
|
|
"mean_token_accuracy": 0.22514583319425582,
|
|
"num_tokens": 10083288.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 5.437113046646118,
|
|
"epoch": 4.85992428339643,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00032819227004932745,
|
|
"loss": 4.9371,
|
|
"mean_token_accuracy": 0.22263206988573075,
|
|
"num_tokens": 10093879.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 5.411635780334473,
|
|
"epoch": 4.865332612222823,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000327775413157139,
|
|
"loss": 4.8701,
|
|
"mean_token_accuracy": 0.2231753095984459,
|
|
"num_tokens": 10104861.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"epoch": 4.865332612222823,
|
|
"eval_entropy": 5.258417320251465,
|
|
"eval_loss": 5.827299118041992,
|
|
"eval_mean_token_accuracy": 0.18074417342855173,
|
|
"eval_num_tokens": 10104861.0,
|
|
"eval_runtime": 1.9396,
|
|
"eval_samples_per_second": 1812.27,
|
|
"eval_steps_per_second": 226.856,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"entropy": 5.364871978759766,
|
|
"epoch": 4.870740941049216,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003273583644789772,
|
|
"loss": 4.8505,
|
|
"mean_token_accuracy": 0.22941972315311432,
|
|
"num_tokens": 10116420.0,
|
|
"step": 4505
|
|
},
|
|
{
|
|
"entropy": 5.339483261108398,
|
|
"epoch": 4.876149269875609,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00032694112553039795,
|
|
"loss": 4.8365,
|
|
"mean_token_accuracy": 0.22257211804389954,
|
|
"num_tokens": 10127289.0,
|
|
"step": 4510
|
|
},
|
|
{
|
|
"entropy": 5.477032852172852,
|
|
"epoch": 4.881557598702001,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003265236978276484,
|
|
"loss": 5.0121,
|
|
"mean_token_accuracy": 0.2121005669236183,
|
|
"num_tokens": 10139216.0,
|
|
"step": 4515
|
|
},
|
|
{
|
|
"entropy": 5.458974981307984,
|
|
"epoch": 4.886965927528394,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00032610608288766185,
|
|
"loss": 4.8601,
|
|
"mean_token_accuracy": 0.23256896883249284,
|
|
"num_tokens": 10150010.0,
|
|
"step": 4520
|
|
},
|
|
{
|
|
"entropy": 5.356452131271363,
|
|
"epoch": 4.8923742563547865,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00032568828222805216,
|
|
"loss": 4.8596,
|
|
"mean_token_accuracy": 0.22746411710977554,
|
|
"num_tokens": 10161317.0,
|
|
"step": 4525
|
|
},
|
|
{
|
|
"entropy": 5.411697196960449,
|
|
"epoch": 4.897782585181179,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003252702973671075,
|
|
"loss": 4.8798,
|
|
"mean_token_accuracy": 0.23586968183517457,
|
|
"num_tokens": 10172681.0,
|
|
"step": 4530
|
|
},
|
|
{
|
|
"entropy": 5.426865911483764,
|
|
"epoch": 4.903190914007571,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00032485212982378615,
|
|
"loss": 4.8979,
|
|
"mean_token_accuracy": 0.22188460528850557,
|
|
"num_tokens": 10184755.0,
|
|
"step": 4535
|
|
},
|
|
{
|
|
"entropy": 5.367524147033691,
|
|
"epoch": 4.908599242833964,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003244337811177098,
|
|
"loss": 4.8555,
|
|
"mean_token_accuracy": 0.23568345606327057,
|
|
"num_tokens": 10195252.0,
|
|
"step": 4540
|
|
},
|
|
{
|
|
"entropy": 5.4170879364013675,
|
|
"epoch": 4.914007571660357,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003240152527691586,
|
|
"loss": 4.9813,
|
|
"mean_token_accuracy": 0.20997077375650405,
|
|
"num_tokens": 10207660.0,
|
|
"step": 4545
|
|
},
|
|
{
|
|
"entropy": 5.3983988761901855,
|
|
"epoch": 4.91941590048675,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00032359654629906555,
|
|
"loss": 4.9016,
|
|
"mean_token_accuracy": 0.225002783536911,
|
|
"num_tokens": 10218154.0,
|
|
"step": 4550
|
|
},
|
|
{
|
|
"entropy": 5.38257360458374,
|
|
"epoch": 4.924824229313142,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00032317766322901106,
|
|
"loss": 4.8467,
|
|
"mean_token_accuracy": 0.22159490287303923,
|
|
"num_tokens": 10229624.0,
|
|
"step": 4555
|
|
},
|
|
{
|
|
"entropy": 5.397515869140625,
|
|
"epoch": 4.930232558139535,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000322758605081217,
|
|
"loss": 4.8347,
|
|
"mean_token_accuracy": 0.23437041491270066,
|
|
"num_tokens": 10240338.0,
|
|
"step": 4560
|
|
},
|
|
{
|
|
"entropy": 5.347846746444702,
|
|
"epoch": 4.935640886965928,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00032233937337854167,
|
|
"loss": 4.8834,
|
|
"mean_token_accuracy": 0.22012484669685364,
|
|
"num_tokens": 10250721.0,
|
|
"step": 4565
|
|
},
|
|
{
|
|
"entropy": 5.378142976760865,
|
|
"epoch": 4.9410492157923205,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003219199696444741,
|
|
"loss": 4.9395,
|
|
"mean_token_accuracy": 0.21675911843776702,
|
|
"num_tokens": 10262596.0,
|
|
"step": 4570
|
|
},
|
|
{
|
|
"entropy": 5.517351198196411,
|
|
"epoch": 4.9464575446187125,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00032150039540312843,
|
|
"loss": 5.0315,
|
|
"mean_token_accuracy": 0.2136908695101738,
|
|
"num_tokens": 10275043.0,
|
|
"step": 4575
|
|
},
|
|
{
|
|
"entropy": 5.434215068817139,
|
|
"epoch": 4.951865873445105,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00032108065217923836,
|
|
"loss": 4.9846,
|
|
"mean_token_accuracy": 0.22139339596033097,
|
|
"num_tokens": 10286593.0,
|
|
"step": 4580
|
|
},
|
|
{
|
|
"entropy": 5.599158239364624,
|
|
"epoch": 4.957274202271498,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00032066074149815184,
|
|
"loss": 5.0125,
|
|
"mean_token_accuracy": 0.22067911177873611,
|
|
"num_tokens": 10297459.0,
|
|
"step": 4585
|
|
},
|
|
{
|
|
"entropy": 5.327614498138428,
|
|
"epoch": 4.962682531097891,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00032024066488582517,
|
|
"loss": 4.7265,
|
|
"mean_token_accuracy": 0.2337689995765686,
|
|
"num_tokens": 10308397.0,
|
|
"step": 4590
|
|
},
|
|
{
|
|
"entropy": 5.448577737808227,
|
|
"epoch": 4.968090859924283,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003198204238688178,
|
|
"loss": 5.0121,
|
|
"mean_token_accuracy": 0.21802543550729753,
|
|
"num_tokens": 10319648.0,
|
|
"step": 4595
|
|
},
|
|
{
|
|
"entropy": 5.476305913925171,
|
|
"epoch": 4.973499188750676,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00031940001997428645,
|
|
"loss": 4.9716,
|
|
"mean_token_accuracy": 0.21534976363182068,
|
|
"num_tokens": 10330922.0,
|
|
"step": 4600
|
|
},
|
|
{
|
|
"entropy": 5.411015129089355,
|
|
"epoch": 4.978907517577069,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00031897945472997996,
|
|
"loss": 4.8681,
|
|
"mean_token_accuracy": 0.2207595258951187,
|
|
"num_tokens": 10341959.0,
|
|
"step": 4605
|
|
},
|
|
{
|
|
"entropy": 5.428595876693725,
|
|
"epoch": 4.984315846403462,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003185587296642333,
|
|
"loss": 4.9517,
|
|
"mean_token_accuracy": 0.22491201907396316,
|
|
"num_tokens": 10353125.0,
|
|
"step": 4610
|
|
},
|
|
{
|
|
"entropy": 5.445161485671997,
|
|
"epoch": 4.9897241752298545,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003181378463059625,
|
|
"loss": 4.9732,
|
|
"mean_token_accuracy": 0.22063891738653182,
|
|
"num_tokens": 10364528.0,
|
|
"step": 4615
|
|
},
|
|
{
|
|
"entropy": 5.383210897445679,
|
|
"epoch": 4.995132504056246,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003177168061846584,
|
|
"loss": 4.8386,
|
|
"mean_token_accuracy": 0.23045475482940675,
|
|
"num_tokens": 10375252.0,
|
|
"step": 4620
|
|
},
|
|
{
|
|
"entropy": 5.344102011786567,
|
|
"epoch": 5.0,
|
|
"grad_norm": 2.421875,
|
|
"learning_rate": 0.0003172956108303819,
|
|
"loss": 4.8171,
|
|
"mean_token_accuracy": 0.23124853438801235,
|
|
"num_tokens": 10384445.0,
|
|
"step": 4625
|
|
},
|
|
{
|
|
"entropy": 5.4094737529754635,
|
|
"epoch": 5.005408328826393,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003168742617737579,
|
|
"loss": 4.7504,
|
|
"mean_token_accuracy": 0.2368357867002487,
|
|
"num_tokens": 10396122.0,
|
|
"step": 4630
|
|
},
|
|
{
|
|
"entropy": 5.456797742843628,
|
|
"epoch": 5.010816657652786,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00031645276054597006,
|
|
"loss": 4.7054,
|
|
"mean_token_accuracy": 0.24138387739658357,
|
|
"num_tokens": 10409135.0,
|
|
"step": 4635
|
|
},
|
|
{
|
|
"entropy": 5.443136548995971,
|
|
"epoch": 5.016224986479178,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00031603110867875435,
|
|
"loss": 4.681,
|
|
"mean_token_accuracy": 0.24719539880752564,
|
|
"num_tokens": 10421068.0,
|
|
"step": 4640
|
|
},
|
|
{
|
|
"entropy": 5.370496559143066,
|
|
"epoch": 5.0216333153055706,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000315609307704395,
|
|
"loss": 4.6282,
|
|
"mean_token_accuracy": 0.247990919649601,
|
|
"num_tokens": 10432117.0,
|
|
"step": 4645
|
|
},
|
|
{
|
|
"entropy": 5.422119855880737,
|
|
"epoch": 5.027041644131963,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003151873591557176,
|
|
"loss": 4.6745,
|
|
"mean_token_accuracy": 0.2398219257593155,
|
|
"num_tokens": 10443547.0,
|
|
"step": 4650
|
|
},
|
|
{
|
|
"entropy": 5.401692152023315,
|
|
"epoch": 5.032449972958356,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00031476526456608415,
|
|
"loss": 4.662,
|
|
"mean_token_accuracy": 0.2470522865653038,
|
|
"num_tokens": 10455038.0,
|
|
"step": 4655
|
|
},
|
|
{
|
|
"entropy": 5.418014287948608,
|
|
"epoch": 5.037858301784748,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003143430254693873,
|
|
"loss": 4.7398,
|
|
"mean_token_accuracy": 0.23160385191440583,
|
|
"num_tokens": 10466682.0,
|
|
"step": 4660
|
|
},
|
|
{
|
|
"entropy": 5.3870460987091064,
|
|
"epoch": 5.043266630611141,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003139206434000451,
|
|
"loss": 4.7029,
|
|
"mean_token_accuracy": 0.23973186761140824,
|
|
"num_tokens": 10478575.0,
|
|
"step": 4665
|
|
},
|
|
{
|
|
"entropy": 5.398724603652954,
|
|
"epoch": 5.048674959437534,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003134981198929948,
|
|
"loss": 4.6911,
|
|
"mean_token_accuracy": 0.24589354246854783,
|
|
"num_tokens": 10490088.0,
|
|
"step": 4670
|
|
},
|
|
{
|
|
"entropy": 5.361037969589233,
|
|
"epoch": 5.054083288263927,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00031307545648368777,
|
|
"loss": 4.5515,
|
|
"mean_token_accuracy": 0.2569524824619293,
|
|
"num_tokens": 10499877.0,
|
|
"step": 4675
|
|
},
|
|
{
|
|
"entropy": 5.290037393569946,
|
|
"epoch": 5.059491617090319,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00031265265470808393,
|
|
"loss": 4.5856,
|
|
"mean_token_accuracy": 0.2514470607042313,
|
|
"num_tokens": 10509830.0,
|
|
"step": 4680
|
|
},
|
|
{
|
|
"entropy": 5.323128175735474,
|
|
"epoch": 5.064899945916712,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00031222971610264595,
|
|
"loss": 4.5863,
|
|
"mean_token_accuracy": 0.24955299645662307,
|
|
"num_tokens": 10520072.0,
|
|
"step": 4685
|
|
},
|
|
{
|
|
"entropy": 5.4172218322753904,
|
|
"epoch": 5.0703082747431045,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00031180664220433347,
|
|
"loss": 4.7385,
|
|
"mean_token_accuracy": 0.2422848954796791,
|
|
"num_tokens": 10531260.0,
|
|
"step": 4690
|
|
},
|
|
{
|
|
"entropy": 5.32621283531189,
|
|
"epoch": 5.075716603569497,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00031138343455059823,
|
|
"loss": 4.6243,
|
|
"mean_token_accuracy": 0.25065034329891206,
|
|
"num_tokens": 10542971.0,
|
|
"step": 4695
|
|
},
|
|
{
|
|
"entropy": 5.309153366088867,
|
|
"epoch": 5.081124932395889,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003109600946793778,
|
|
"loss": 4.603,
|
|
"mean_token_accuracy": 0.24724968522787094,
|
|
"num_tokens": 10554101.0,
|
|
"step": 4700
|
|
},
|
|
{
|
|
"entropy": 5.489874315261841,
|
|
"epoch": 5.086533261222282,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00031053662412909024,
|
|
"loss": 4.766,
|
|
"mean_token_accuracy": 0.2362542137503624,
|
|
"num_tokens": 10565866.0,
|
|
"step": 4705
|
|
},
|
|
{
|
|
"entropy": 5.270193719863892,
|
|
"epoch": 5.091941590048675,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003101130244386283,
|
|
"loss": 4.6018,
|
|
"mean_token_accuracy": 0.2521262288093567,
|
|
"num_tokens": 10577276.0,
|
|
"step": 4710
|
|
},
|
|
{
|
|
"entropy": 5.451295471191406,
|
|
"epoch": 5.097349918875068,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00030968929714735456,
|
|
"loss": 4.7675,
|
|
"mean_token_accuracy": 0.23866018354892732,
|
|
"num_tokens": 10588625.0,
|
|
"step": 4715
|
|
},
|
|
{
|
|
"entropy": 5.276379251480103,
|
|
"epoch": 5.10275824770146,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003092654437950947,
|
|
"loss": 4.667,
|
|
"mean_token_accuracy": 0.24477856755256652,
|
|
"num_tokens": 10599833.0,
|
|
"step": 4720
|
|
},
|
|
{
|
|
"entropy": 5.342433643341065,
|
|
"epoch": 5.108166576527853,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003088414659221331,
|
|
"loss": 4.6658,
|
|
"mean_token_accuracy": 0.24113594442605973,
|
|
"num_tokens": 10610156.0,
|
|
"step": 4725
|
|
},
|
|
{
|
|
"entropy": 5.338064956665039,
|
|
"epoch": 5.113574905354246,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000308417365069206,
|
|
"loss": 4.6675,
|
|
"mean_token_accuracy": 0.2470059275627136,
|
|
"num_tokens": 10619839.0,
|
|
"step": 4730
|
|
},
|
|
{
|
|
"entropy": 5.325254440307617,
|
|
"epoch": 5.1189832341806385,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00030799314277749697,
|
|
"loss": 4.6645,
|
|
"mean_token_accuracy": 0.24437775313854218,
|
|
"num_tokens": 10631671.0,
|
|
"step": 4735
|
|
},
|
|
{
|
|
"entropy": 5.362559413909912,
|
|
"epoch": 5.1243915630070305,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00030756880058863085,
|
|
"loss": 4.6686,
|
|
"mean_token_accuracy": 0.24862882494926453,
|
|
"num_tokens": 10642597.0,
|
|
"step": 4740
|
|
},
|
|
{
|
|
"entropy": 5.388282442092896,
|
|
"epoch": 5.129799891833423,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00030714434004466823,
|
|
"loss": 4.7312,
|
|
"mean_token_accuracy": 0.23776959478855134,
|
|
"num_tokens": 10653323.0,
|
|
"step": 4745
|
|
},
|
|
{
|
|
"entropy": 5.456527280807495,
|
|
"epoch": 5.135208220659816,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00030671976268809947,
|
|
"loss": 4.7483,
|
|
"mean_token_accuracy": 0.23958356380462648,
|
|
"num_tokens": 10665572.0,
|
|
"step": 4750
|
|
},
|
|
{
|
|
"entropy": 5.326587152481079,
|
|
"epoch": 5.140616549486209,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00030629507006183995,
|
|
"loss": 4.5972,
|
|
"mean_token_accuracy": 0.24627719074487686,
|
|
"num_tokens": 10676686.0,
|
|
"step": 4755
|
|
},
|
|
{
|
|
"entropy": 5.327364301681518,
|
|
"epoch": 5.146024878312601,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0003058702637092234,
|
|
"loss": 4.7426,
|
|
"mean_token_accuracy": 0.23948701322078705,
|
|
"num_tokens": 10687188.0,
|
|
"step": 4760
|
|
},
|
|
{
|
|
"entropy": 5.375910949707031,
|
|
"epoch": 5.151433207138994,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00030544534517399724,
|
|
"loss": 4.6788,
|
|
"mean_token_accuracy": 0.23960106521844865,
|
|
"num_tokens": 10698343.0,
|
|
"step": 4765
|
|
},
|
|
{
|
|
"entropy": 5.360807180404663,
|
|
"epoch": 5.156841535965387,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003050203160003163,
|
|
"loss": 4.6531,
|
|
"mean_token_accuracy": 0.2448326051235199,
|
|
"num_tokens": 10710732.0,
|
|
"step": 4770
|
|
},
|
|
{
|
|
"entropy": 5.3397845268249515,
|
|
"epoch": 5.16224986479178,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003045951777327377,
|
|
"loss": 4.7305,
|
|
"mean_token_accuracy": 0.24151260852813722,
|
|
"num_tokens": 10722397.0,
|
|
"step": 4775
|
|
},
|
|
{
|
|
"entropy": 5.234175968170166,
|
|
"epoch": 5.167658193618172,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003041699319162148,
|
|
"loss": 4.6156,
|
|
"mean_token_accuracy": 0.24147363305091857,
|
|
"num_tokens": 10733714.0,
|
|
"step": 4780
|
|
},
|
|
{
|
|
"entropy": 5.377866172790528,
|
|
"epoch": 5.173066522444564,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00030374458009609193,
|
|
"loss": 4.6854,
|
|
"mean_token_accuracy": 0.24338002800941466,
|
|
"num_tokens": 10745748.0,
|
|
"step": 4785
|
|
},
|
|
{
|
|
"entropy": 5.49366569519043,
|
|
"epoch": 5.178474851270957,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003033191238180985,
|
|
"loss": 4.7272,
|
|
"mean_token_accuracy": 0.24319095760583878,
|
|
"num_tokens": 10756094.0,
|
|
"step": 4790
|
|
},
|
|
{
|
|
"entropy": 5.39813814163208,
|
|
"epoch": 5.18388318009735,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003028935646283438,
|
|
"loss": 4.7833,
|
|
"mean_token_accuracy": 0.23881425112485885,
|
|
"num_tokens": 10768068.0,
|
|
"step": 4795
|
|
},
|
|
{
|
|
"entropy": 5.294648456573486,
|
|
"epoch": 5.189291508923742,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00030246790407331063,
|
|
"loss": 4.6257,
|
|
"mean_token_accuracy": 0.24490901380777358,
|
|
"num_tokens": 10778917.0,
|
|
"step": 4800
|
|
},
|
|
{
|
|
"entropy": 5.37889347076416,
|
|
"epoch": 5.194699837750135,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00030204214369985066,
|
|
"loss": 4.6544,
|
|
"mean_token_accuracy": 0.2472248300909996,
|
|
"num_tokens": 10790552.0,
|
|
"step": 4805
|
|
},
|
|
{
|
|
"entropy": 5.349570274353027,
|
|
"epoch": 5.200108166576528,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00030161628505517797,
|
|
"loss": 4.7069,
|
|
"mean_token_accuracy": 0.2412051886320114,
|
|
"num_tokens": 10803199.0,
|
|
"step": 4810
|
|
},
|
|
{
|
|
"entropy": 5.27286286354065,
|
|
"epoch": 5.205516495402921,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00030119032968686397,
|
|
"loss": 4.6165,
|
|
"mean_token_accuracy": 0.2528571054339409,
|
|
"num_tokens": 10814339.0,
|
|
"step": 4815
|
|
},
|
|
{
|
|
"entropy": 5.341352224349976,
|
|
"epoch": 5.210924824229313,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00030076427914283127,
|
|
"loss": 4.7042,
|
|
"mean_token_accuracy": 0.23464954644441605,
|
|
"num_tokens": 10825776.0,
|
|
"step": 4820
|
|
},
|
|
{
|
|
"entropy": 5.4154784202575685,
|
|
"epoch": 5.2163331530557056,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00030033813497134864,
|
|
"loss": 4.6491,
|
|
"mean_token_accuracy": 0.24242058545351028,
|
|
"num_tokens": 10837179.0,
|
|
"step": 4825
|
|
},
|
|
{
|
|
"entropy": 5.299468564987182,
|
|
"epoch": 5.221741481882098,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000299911898721025,
|
|
"loss": 4.6861,
|
|
"mean_token_accuracy": 0.24601320922374725,
|
|
"num_tokens": 10847205.0,
|
|
"step": 4830
|
|
},
|
|
{
|
|
"entropy": 5.338000965118408,
|
|
"epoch": 5.227149810708491,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00029948557194080385,
|
|
"loss": 4.6733,
|
|
"mean_token_accuracy": 0.23981924802064897,
|
|
"num_tokens": 10859895.0,
|
|
"step": 4835
|
|
},
|
|
{
|
|
"entropy": 5.293436193466187,
|
|
"epoch": 5.232558139534884,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0002990591561799576,
|
|
"loss": 4.7111,
|
|
"mean_token_accuracy": 0.2329314947128296,
|
|
"num_tokens": 10871286.0,
|
|
"step": 4840
|
|
},
|
|
{
|
|
"entropy": 5.294149112701416,
|
|
"epoch": 5.237966468361276,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002986326529880823,
|
|
"loss": 4.555,
|
|
"mean_token_accuracy": 0.254327729344368,
|
|
"num_tokens": 10883479.0,
|
|
"step": 4845
|
|
},
|
|
{
|
|
"entropy": 5.403821563720703,
|
|
"epoch": 5.243374797187669,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0002982060639150913,
|
|
"loss": 4.7191,
|
|
"mean_token_accuracy": 0.24863158464431762,
|
|
"num_tokens": 10895432.0,
|
|
"step": 4850
|
|
},
|
|
{
|
|
"entropy": 5.332467746734619,
|
|
"epoch": 5.248783126014062,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002977793905112106,
|
|
"loss": 4.7308,
|
|
"mean_token_accuracy": 0.23886541575193404,
|
|
"num_tokens": 10906891.0,
|
|
"step": 4855
|
|
},
|
|
{
|
|
"entropy": 5.34052267074585,
|
|
"epoch": 5.254191454840455,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00029735263432697183,
|
|
"loss": 4.7436,
|
|
"mean_token_accuracy": 0.23779460191726684,
|
|
"num_tokens": 10917706.0,
|
|
"step": 4860
|
|
},
|
|
{
|
|
"entropy": 5.352297401428222,
|
|
"epoch": 5.259599783666847,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002969257969132085,
|
|
"loss": 4.7439,
|
|
"mean_token_accuracy": 0.2351348504424095,
|
|
"num_tokens": 10929325.0,
|
|
"step": 4865
|
|
},
|
|
{
|
|
"entropy": 5.390991687774658,
|
|
"epoch": 5.2650081124932395,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0002964988798210485,
|
|
"loss": 4.7294,
|
|
"mean_token_accuracy": 0.2302283838391304,
|
|
"num_tokens": 10939899.0,
|
|
"step": 4870
|
|
},
|
|
{
|
|
"entropy": 5.330938768386841,
|
|
"epoch": 5.270416441319632,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002960718846019095,
|
|
"loss": 4.6803,
|
|
"mean_token_accuracy": 0.24241009652614592,
|
|
"num_tokens": 10949801.0,
|
|
"step": 4875
|
|
},
|
|
{
|
|
"entropy": 5.20372633934021,
|
|
"epoch": 5.275824770146025,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002956448128074931,
|
|
"loss": 4.5608,
|
|
"mean_token_accuracy": 0.2531586140394211,
|
|
"num_tokens": 10960191.0,
|
|
"step": 4880
|
|
},
|
|
{
|
|
"entropy": 5.452007865905761,
|
|
"epoch": 5.281233098972417,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002952176659897794,
|
|
"loss": 4.8143,
|
|
"mean_token_accuracy": 0.23516862839460373,
|
|
"num_tokens": 10972818.0,
|
|
"step": 4885
|
|
},
|
|
{
|
|
"entropy": 5.402248859405518,
|
|
"epoch": 5.28664142779881,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002947904457010207,
|
|
"loss": 4.7184,
|
|
"mean_token_accuracy": 0.2314800277352333,
|
|
"num_tokens": 10984320.0,
|
|
"step": 4890
|
|
},
|
|
{
|
|
"entropy": 5.31989278793335,
|
|
"epoch": 5.292049756625203,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002943631534937366,
|
|
"loss": 4.6974,
|
|
"mean_token_accuracy": 0.2443007066845894,
|
|
"num_tokens": 10995262.0,
|
|
"step": 4895
|
|
},
|
|
{
|
|
"entropy": 5.251620531082153,
|
|
"epoch": 5.297458085451596,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002939357909207079,
|
|
"loss": 4.6905,
|
|
"mean_token_accuracy": 0.2374425411224365,
|
|
"num_tokens": 11005788.0,
|
|
"step": 4900
|
|
},
|
|
{
|
|
"entropy": 5.369648170471192,
|
|
"epoch": 5.302866414277988,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002935083595349713,
|
|
"loss": 4.6844,
|
|
"mean_token_accuracy": 0.24747762084007263,
|
|
"num_tokens": 11016772.0,
|
|
"step": 4905
|
|
},
|
|
{
|
|
"entropy": 5.378639268875122,
|
|
"epoch": 5.308274743104381,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00029308086088981335,
|
|
"loss": 4.745,
|
|
"mean_token_accuracy": 0.23901977092027665,
|
|
"num_tokens": 11028917.0,
|
|
"step": 4910
|
|
},
|
|
{
|
|
"entropy": 5.355211925506592,
|
|
"epoch": 5.3136830719307735,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00029265329653876515,
|
|
"loss": 4.7429,
|
|
"mean_token_accuracy": 0.2346576139330864,
|
|
"num_tokens": 11040500.0,
|
|
"step": 4915
|
|
},
|
|
{
|
|
"entropy": 5.336022567749024,
|
|
"epoch": 5.319091400757166,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002922256680355965,
|
|
"loss": 4.6745,
|
|
"mean_token_accuracy": 0.24615731537342073,
|
|
"num_tokens": 11052173.0,
|
|
"step": 4920
|
|
},
|
|
{
|
|
"entropy": 5.399751138687134,
|
|
"epoch": 5.324499729583558,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00029179797693431046,
|
|
"loss": 4.6824,
|
|
"mean_token_accuracy": 0.24473892748355866,
|
|
"num_tokens": 11061917.0,
|
|
"step": 4925
|
|
},
|
|
{
|
|
"entropy": 5.335155010223389,
|
|
"epoch": 5.329908058409951,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002913702247891374,
|
|
"loss": 4.6623,
|
|
"mean_token_accuracy": 0.25152056068181994,
|
|
"num_tokens": 11072820.0,
|
|
"step": 4930
|
|
},
|
|
{
|
|
"entropy": 5.391182994842529,
|
|
"epoch": 5.335316387236344,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002909424131545296,
|
|
"loss": 4.7412,
|
|
"mean_token_accuracy": 0.24323034435510635,
|
|
"num_tokens": 11083140.0,
|
|
"step": 4935
|
|
},
|
|
{
|
|
"entropy": 5.39636926651001,
|
|
"epoch": 5.340724716062737,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00029051454358515553,
|
|
"loss": 4.7158,
|
|
"mean_token_accuracy": 0.23843424171209335,
|
|
"num_tokens": 11094314.0,
|
|
"step": 4940
|
|
},
|
|
{
|
|
"entropy": 5.373786926269531,
|
|
"epoch": 5.346133044889129,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00029008661763589436,
|
|
"loss": 4.7586,
|
|
"mean_token_accuracy": 0.24009737521409988,
|
|
"num_tokens": 11106748.0,
|
|
"step": 4945
|
|
},
|
|
{
|
|
"entropy": 5.384488534927368,
|
|
"epoch": 5.351541373715522,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002896586368618297,
|
|
"loss": 4.7431,
|
|
"mean_token_accuracy": 0.23956231474876405,
|
|
"num_tokens": 11118702.0,
|
|
"step": 4950
|
|
},
|
|
{
|
|
"entropy": 5.388779211044311,
|
|
"epoch": 5.356949702541915,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00028923060281824483,
|
|
"loss": 4.7312,
|
|
"mean_token_accuracy": 0.23479487150907516,
|
|
"num_tokens": 11130782.0,
|
|
"step": 4955
|
|
},
|
|
{
|
|
"entropy": 5.32146053314209,
|
|
"epoch": 5.3623580313683075,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002888025170606165,
|
|
"loss": 4.6939,
|
|
"mean_token_accuracy": 0.24479596614837645,
|
|
"num_tokens": 11142536.0,
|
|
"step": 4960
|
|
},
|
|
{
|
|
"entropy": 5.330479097366333,
|
|
"epoch": 5.367766360194699,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00028837438114460924,
|
|
"loss": 4.6974,
|
|
"mean_token_accuracy": 0.24041297882795334,
|
|
"num_tokens": 11154027.0,
|
|
"step": 4965
|
|
},
|
|
{
|
|
"entropy": 5.352686643600464,
|
|
"epoch": 5.373174689021092,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002879461966260699,
|
|
"loss": 4.7124,
|
|
"mean_token_accuracy": 0.23900691270828248,
|
|
"num_tokens": 11164759.0,
|
|
"step": 4970
|
|
},
|
|
{
|
|
"entropy": 5.406398963928223,
|
|
"epoch": 5.378583017847485,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00028751796506102224,
|
|
"loss": 4.7285,
|
|
"mean_token_accuracy": 0.23848156481981278,
|
|
"num_tokens": 11176341.0,
|
|
"step": 4975
|
|
},
|
|
{
|
|
"entropy": 5.336361265182495,
|
|
"epoch": 5.383991346673878,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00028708968800566056,
|
|
"loss": 4.6952,
|
|
"mean_token_accuracy": 0.24185363054275513,
|
|
"num_tokens": 11188958.0,
|
|
"step": 4980
|
|
},
|
|
{
|
|
"entropy": 5.4095543384552,
|
|
"epoch": 5.38939967550027,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00028666136701634474,
|
|
"loss": 4.7852,
|
|
"mean_token_accuracy": 0.23133932203054428,
|
|
"num_tokens": 11200895.0,
|
|
"step": 4985
|
|
},
|
|
{
|
|
"entropy": 5.414268732070923,
|
|
"epoch": 5.394808004326663,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002862330036495942,
|
|
"loss": 4.7009,
|
|
"mean_token_accuracy": 0.24816498756408692,
|
|
"num_tokens": 11211563.0,
|
|
"step": 4990
|
|
},
|
|
{
|
|
"entropy": 5.454798412322998,
|
|
"epoch": 5.400216333153056,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002858045994620825,
|
|
"loss": 4.7856,
|
|
"mean_token_accuracy": 0.23839604258537292,
|
|
"num_tokens": 11223323.0,
|
|
"step": 4995
|
|
},
|
|
{
|
|
"entropy": 5.369642114639282,
|
|
"epoch": 5.405624661979449,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00028537615601063133,
|
|
"loss": 4.6974,
|
|
"mean_token_accuracy": 0.23828154504299165,
|
|
"num_tokens": 11235991.0,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"epoch": 5.405624661979449,
|
|
"eval_entropy": 5.2523489930412985,
|
|
"eval_loss": 5.8562445640563965,
|
|
"eval_mean_token_accuracy": 0.17933705442331055,
|
|
"eval_num_tokens": 11235991.0,
|
|
"eval_runtime": 1.9396,
|
|
"eval_samples_per_second": 1812.27,
|
|
"eval_steps_per_second": 226.856,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"entropy": 5.367483282089234,
|
|
"epoch": 5.411032990805841,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002849476748522052,
|
|
"loss": 4.6792,
|
|
"mean_token_accuracy": 0.23938343673944473,
|
|
"num_tokens": 11247406.0,
|
|
"step": 5005
|
|
},
|
|
{
|
|
"entropy": 5.199149751663208,
|
|
"epoch": 5.416441319632233,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002845191575439057,
|
|
"loss": 4.6466,
|
|
"mean_token_accuracy": 0.23921885341405869,
|
|
"num_tokens": 11259031.0,
|
|
"step": 5010
|
|
},
|
|
{
|
|
"entropy": 5.307330083847046,
|
|
"epoch": 5.421849648458626,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002840906056429658,
|
|
"loss": 4.7212,
|
|
"mean_token_accuracy": 0.2314278081059456,
|
|
"num_tokens": 11270321.0,
|
|
"step": 5015
|
|
},
|
|
{
|
|
"entropy": 5.377045106887818,
|
|
"epoch": 5.427257977285019,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002836620207067439,
|
|
"loss": 4.8392,
|
|
"mean_token_accuracy": 0.22439004927873613,
|
|
"num_tokens": 11282209.0,
|
|
"step": 5020
|
|
},
|
|
{
|
|
"entropy": 5.4350920677185055,
|
|
"epoch": 5.432666306111411,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00028323340429271875,
|
|
"loss": 4.7418,
|
|
"mean_token_accuracy": 0.23842664211988449,
|
|
"num_tokens": 11292987.0,
|
|
"step": 5025
|
|
},
|
|
{
|
|
"entropy": 5.342994022369385,
|
|
"epoch": 5.438074634937804,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002828047579584833,
|
|
"loss": 4.7,
|
|
"mean_token_accuracy": 0.23450141847133638,
|
|
"num_tokens": 11303197.0,
|
|
"step": 5030
|
|
},
|
|
{
|
|
"entropy": 5.267442750930786,
|
|
"epoch": 5.443482963764197,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00028237608326173944,
|
|
"loss": 4.6237,
|
|
"mean_token_accuracy": 0.24997031986713408,
|
|
"num_tokens": 11315229.0,
|
|
"step": 5035
|
|
},
|
|
{
|
|
"entropy": 5.2953126430511475,
|
|
"epoch": 5.44889129259059,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00028194738176029206,
|
|
"loss": 4.72,
|
|
"mean_token_accuracy": 0.23903416991233825,
|
|
"num_tokens": 11326033.0,
|
|
"step": 5040
|
|
},
|
|
{
|
|
"entropy": 5.308441019058227,
|
|
"epoch": 5.4542996214169825,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00028151865501204316,
|
|
"loss": 4.7014,
|
|
"mean_token_accuracy": 0.2340387225151062,
|
|
"num_tokens": 11335670.0,
|
|
"step": 5045
|
|
},
|
|
{
|
|
"entropy": 5.329196262359619,
|
|
"epoch": 5.4597079502433745,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00028108990457498696,
|
|
"loss": 4.6606,
|
|
"mean_token_accuracy": 0.24951518326997757,
|
|
"num_tokens": 11346882.0,
|
|
"step": 5050
|
|
},
|
|
{
|
|
"entropy": 5.300083351135254,
|
|
"epoch": 5.465116279069767,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00028066113200720343,
|
|
"loss": 4.6495,
|
|
"mean_token_accuracy": 0.2417772516608238,
|
|
"num_tokens": 11358141.0,
|
|
"step": 5055
|
|
},
|
|
{
|
|
"entropy": 5.2658356666564945,
|
|
"epoch": 5.47052460789616,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00028023233886685314,
|
|
"loss": 4.6803,
|
|
"mean_token_accuracy": 0.2394333228468895,
|
|
"num_tokens": 11368981.0,
|
|
"step": 5060
|
|
},
|
|
{
|
|
"entropy": 5.365193653106689,
|
|
"epoch": 5.475932936722553,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002798035267121713,
|
|
"loss": 4.7998,
|
|
"mean_token_accuracy": 0.23782767951488495,
|
|
"num_tokens": 11381613.0,
|
|
"step": 5065
|
|
},
|
|
{
|
|
"entropy": 5.390643835067749,
|
|
"epoch": 5.481341265548945,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002793746971014623,
|
|
"loss": 4.7052,
|
|
"mean_token_accuracy": 0.24154542833566667,
|
|
"num_tokens": 11392321.0,
|
|
"step": 5070
|
|
},
|
|
{
|
|
"entropy": 5.305334281921387,
|
|
"epoch": 5.486749594375338,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002789458515930939,
|
|
"loss": 4.7411,
|
|
"mean_token_accuracy": 0.23760878443717956,
|
|
"num_tokens": 11403546.0,
|
|
"step": 5075
|
|
},
|
|
{
|
|
"entropy": 5.34538836479187,
|
|
"epoch": 5.492157923201731,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00027851699174549164,
|
|
"loss": 4.6527,
|
|
"mean_token_accuracy": 0.24089958220720292,
|
|
"num_tokens": 11415167.0,
|
|
"step": 5080
|
|
},
|
|
{
|
|
"entropy": 5.34430422782898,
|
|
"epoch": 5.497566252028124,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002780881191171333,
|
|
"loss": 4.6696,
|
|
"mean_token_accuracy": 0.242888243496418,
|
|
"num_tokens": 11426715.0,
|
|
"step": 5085
|
|
},
|
|
{
|
|
"entropy": 5.323763990402222,
|
|
"epoch": 5.502974580854516,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00027765923526654304,
|
|
"loss": 4.6456,
|
|
"mean_token_accuracy": 0.24371725469827651,
|
|
"num_tokens": 11437760.0,
|
|
"step": 5090
|
|
},
|
|
{
|
|
"entropy": 5.33249797821045,
|
|
"epoch": 5.5083829096809085,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00027723034175228554,
|
|
"loss": 4.7267,
|
|
"mean_token_accuracy": 0.23850819319486619,
|
|
"num_tokens": 11448329.0,
|
|
"step": 5095
|
|
},
|
|
{
|
|
"entropy": 5.266516923904419,
|
|
"epoch": 5.513791238507301,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000276801440132961,
|
|
"loss": 4.5491,
|
|
"mean_token_accuracy": 0.24946196526288986,
|
|
"num_tokens": 11459173.0,
|
|
"step": 5100
|
|
},
|
|
{
|
|
"entropy": 5.326828336715698,
|
|
"epoch": 5.519199567333694,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002763725319671989,
|
|
"loss": 4.6495,
|
|
"mean_token_accuracy": 0.2440410077571869,
|
|
"num_tokens": 11470232.0,
|
|
"step": 5105
|
|
},
|
|
{
|
|
"entropy": 5.2256217956542965,
|
|
"epoch": 5.524607896160086,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002759436188136525,
|
|
"loss": 4.6625,
|
|
"mean_token_accuracy": 0.24137548059225084,
|
|
"num_tokens": 11481181.0,
|
|
"step": 5110
|
|
},
|
|
{
|
|
"entropy": 5.317529535293579,
|
|
"epoch": 5.530016224986479,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00027551470223099306,
|
|
"loss": 4.7077,
|
|
"mean_token_accuracy": 0.2416066661477089,
|
|
"num_tokens": 11491772.0,
|
|
"step": 5115
|
|
},
|
|
{
|
|
"entropy": 5.376226139068604,
|
|
"epoch": 5.535424553812872,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002750857837779045,
|
|
"loss": 4.6671,
|
|
"mean_token_accuracy": 0.24373465329408645,
|
|
"num_tokens": 11503895.0,
|
|
"step": 5120
|
|
},
|
|
{
|
|
"entropy": 5.338111639022827,
|
|
"epoch": 5.540832882639265,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002746568650130775,
|
|
"loss": 4.67,
|
|
"mean_token_accuracy": 0.24686562716960908,
|
|
"num_tokens": 11514189.0,
|
|
"step": 5125
|
|
},
|
|
{
|
|
"entropy": 5.218132066726684,
|
|
"epoch": 5.546241211465657,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002742279474952039,
|
|
"loss": 4.7018,
|
|
"mean_token_accuracy": 0.23857082277536393,
|
|
"num_tokens": 11526142.0,
|
|
"step": 5130
|
|
},
|
|
{
|
|
"entropy": 5.378658866882324,
|
|
"epoch": 5.55164954029205,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00027379903278297086,
|
|
"loss": 4.7284,
|
|
"mean_token_accuracy": 0.23357025235891343,
|
|
"num_tokens": 11537061.0,
|
|
"step": 5135
|
|
},
|
|
{
|
|
"entropy": 5.376234340667724,
|
|
"epoch": 5.5570578691184425,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00027337012243505536,
|
|
"loss": 4.8086,
|
|
"mean_token_accuracy": 0.22693837732076644,
|
|
"num_tokens": 11547319.0,
|
|
"step": 5140
|
|
},
|
|
{
|
|
"entropy": 5.349218940734863,
|
|
"epoch": 5.562466197944835,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00027294121801011887,
|
|
"loss": 4.7279,
|
|
"mean_token_accuracy": 0.24360288232564925,
|
|
"num_tokens": 11558766.0,
|
|
"step": 5145
|
|
},
|
|
{
|
|
"entropy": 5.287031078338623,
|
|
"epoch": 5.567874526771227,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00027251232106680084,
|
|
"loss": 4.6559,
|
|
"mean_token_accuracy": 0.24310977905988693,
|
|
"num_tokens": 11570039.0,
|
|
"step": 5150
|
|
},
|
|
{
|
|
"entropy": 5.250191783905029,
|
|
"epoch": 5.57328285559762,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00027208343316371383,
|
|
"loss": 4.6814,
|
|
"mean_token_accuracy": 0.24727738797664642,
|
|
"num_tokens": 11581664.0,
|
|
"step": 5155
|
|
},
|
|
{
|
|
"entropy": 5.3602530002594,
|
|
"epoch": 5.578691184424013,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002716545558594373,
|
|
"loss": 4.7007,
|
|
"mean_token_accuracy": 0.2360749915242195,
|
|
"num_tokens": 11592689.0,
|
|
"step": 5160
|
|
},
|
|
{
|
|
"entropy": 5.361602115631103,
|
|
"epoch": 5.584099513250406,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002712256907125126,
|
|
"loss": 4.6214,
|
|
"mean_token_accuracy": 0.2448100984096527,
|
|
"num_tokens": 11604315.0,
|
|
"step": 5165
|
|
},
|
|
{
|
|
"entropy": 5.373456716537476,
|
|
"epoch": 5.589507842076799,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002707968392814366,
|
|
"loss": 4.758,
|
|
"mean_token_accuracy": 0.2406460613012314,
|
|
"num_tokens": 11614510.0,
|
|
"step": 5170
|
|
},
|
|
{
|
|
"entropy": 5.3373541831970215,
|
|
"epoch": 5.594916170903191,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00027036800312465635,
|
|
"loss": 4.6573,
|
|
"mean_token_accuracy": 0.24531605690717698,
|
|
"num_tokens": 11625501.0,
|
|
"step": 5175
|
|
},
|
|
{
|
|
"entropy": 5.411145305633545,
|
|
"epoch": 5.600324499729584,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002699391838005634,
|
|
"loss": 4.7306,
|
|
"mean_token_accuracy": 0.24030441790819168,
|
|
"num_tokens": 11636841.0,
|
|
"step": 5180
|
|
},
|
|
{
|
|
"entropy": 5.278686761856079,
|
|
"epoch": 5.605732828555976,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00026951038286748824,
|
|
"loss": 4.6274,
|
|
"mean_token_accuracy": 0.25050508975982666,
|
|
"num_tokens": 11647475.0,
|
|
"step": 5185
|
|
},
|
|
{
|
|
"entropy": 5.364908742904663,
|
|
"epoch": 5.611141157382368,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002690816018836945,
|
|
"loss": 4.7311,
|
|
"mean_token_accuracy": 0.2368357762694359,
|
|
"num_tokens": 11658887.0,
|
|
"step": 5190
|
|
},
|
|
{
|
|
"entropy": 5.371151351928711,
|
|
"epoch": 5.616549486208761,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002686528424073731,
|
|
"loss": 4.795,
|
|
"mean_token_accuracy": 0.23770710229873657,
|
|
"num_tokens": 11669658.0,
|
|
"step": 5195
|
|
},
|
|
{
|
|
"entropy": 5.347035932540893,
|
|
"epoch": 5.621957815035154,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00026822410599663704,
|
|
"loss": 4.7511,
|
|
"mean_token_accuracy": 0.23053957521915436,
|
|
"num_tokens": 11681792.0,
|
|
"step": 5200
|
|
},
|
|
{
|
|
"entropy": 5.356085634231567,
|
|
"epoch": 5.627366143861547,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00026779539420951554,
|
|
"loss": 4.7209,
|
|
"mean_token_accuracy": 0.23890049904584884,
|
|
"num_tokens": 11691903.0,
|
|
"step": 5205
|
|
},
|
|
{
|
|
"entropy": 5.387093019485474,
|
|
"epoch": 5.63277447268794,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002673667086039481,
|
|
"loss": 4.732,
|
|
"mean_token_accuracy": 0.2470971092581749,
|
|
"num_tokens": 11703383.0,
|
|
"step": 5210
|
|
},
|
|
{
|
|
"entropy": 5.3501953125,
|
|
"epoch": 5.638182801514332,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002669380507377792,
|
|
"loss": 4.7245,
|
|
"mean_token_accuracy": 0.23862203508615493,
|
|
"num_tokens": 11714721.0,
|
|
"step": 5215
|
|
},
|
|
{
|
|
"entropy": 5.3102460384368895,
|
|
"epoch": 5.643591130340725,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002665094221687525,
|
|
"loss": 4.739,
|
|
"mean_token_accuracy": 0.22990864813327788,
|
|
"num_tokens": 11726016.0,
|
|
"step": 5220
|
|
},
|
|
{
|
|
"entropy": 5.316795015335083,
|
|
"epoch": 5.6489994591671175,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002660808244545054,
|
|
"loss": 4.5611,
|
|
"mean_token_accuracy": 0.2529520198702812,
|
|
"num_tokens": 11737589.0,
|
|
"step": 5225
|
|
},
|
|
{
|
|
"entropy": 5.278080129623413,
|
|
"epoch": 5.6544077879935095,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00026565225915256276,
|
|
"loss": 4.6423,
|
|
"mean_token_accuracy": 0.25084590166807175,
|
|
"num_tokens": 11747969.0,
|
|
"step": 5230
|
|
},
|
|
{
|
|
"entropy": 5.246661424636841,
|
|
"epoch": 5.659816116819902,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000265223727820332,
|
|
"loss": 4.6324,
|
|
"mean_token_accuracy": 0.23852156102657318,
|
|
"num_tokens": 11759535.0,
|
|
"step": 5235
|
|
},
|
|
{
|
|
"entropy": 5.350465726852417,
|
|
"epoch": 5.665224445646295,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002647952320150969,
|
|
"loss": 4.7811,
|
|
"mean_token_accuracy": 0.23218382745981217,
|
|
"num_tokens": 11770863.0,
|
|
"step": 5240
|
|
},
|
|
{
|
|
"entropy": 5.358258485794067,
|
|
"epoch": 5.670632774472688,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002643667732940122,
|
|
"loss": 4.6847,
|
|
"mean_token_accuracy": 0.23707840740680694,
|
|
"num_tokens": 11782643.0,
|
|
"step": 5245
|
|
},
|
|
{
|
|
"entropy": 5.385804748535156,
|
|
"epoch": 5.676041103299081,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00026393835321409796,
|
|
"loss": 4.7206,
|
|
"mean_token_accuracy": 0.2470710515975952,
|
|
"num_tokens": 11794468.0,
|
|
"step": 5250
|
|
},
|
|
{
|
|
"entropy": 5.344185447692871,
|
|
"epoch": 5.681449432125473,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00026350997333223375,
|
|
"loss": 4.665,
|
|
"mean_token_accuracy": 0.2475633889436722,
|
|
"num_tokens": 11807051.0,
|
|
"step": 5255
|
|
},
|
|
{
|
|
"entropy": 5.248823547363282,
|
|
"epoch": 5.686857760951866,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00026308163520515293,
|
|
"loss": 4.682,
|
|
"mean_token_accuracy": 0.24120833575725556,
|
|
"num_tokens": 11817067.0,
|
|
"step": 5260
|
|
},
|
|
{
|
|
"entropy": 5.313168907165528,
|
|
"epoch": 5.692266089778259,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00026265334038943735,
|
|
"loss": 4.7285,
|
|
"mean_token_accuracy": 0.24198154956102372,
|
|
"num_tokens": 11828111.0,
|
|
"step": 5265
|
|
},
|
|
{
|
|
"entropy": 5.337206602096558,
|
|
"epoch": 5.6976744186046515,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002622250904415113,
|
|
"loss": 4.7066,
|
|
"mean_token_accuracy": 0.24249098151922227,
|
|
"num_tokens": 11838536.0,
|
|
"step": 5270
|
|
},
|
|
{
|
|
"entropy": 5.363395118713379,
|
|
"epoch": 5.7030827474310435,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002617968869176362,
|
|
"loss": 4.7115,
|
|
"mean_token_accuracy": 0.24164711385965348,
|
|
"num_tokens": 11848899.0,
|
|
"step": 5275
|
|
},
|
|
{
|
|
"entropy": 5.319475507736206,
|
|
"epoch": 5.708491076257436,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002613687313739044,
|
|
"loss": 4.7365,
|
|
"mean_token_accuracy": 0.24336863607168197,
|
|
"num_tokens": 11861023.0,
|
|
"step": 5280
|
|
},
|
|
{
|
|
"entropy": 5.294900560379029,
|
|
"epoch": 5.713899405083829,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002609406253662344,
|
|
"loss": 4.7261,
|
|
"mean_token_accuracy": 0.23934586048126222,
|
|
"num_tokens": 11871953.0,
|
|
"step": 5285
|
|
},
|
|
{
|
|
"entropy": 5.36879415512085,
|
|
"epoch": 5.719307733910222,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002605125704503643,
|
|
"loss": 4.7401,
|
|
"mean_token_accuracy": 0.23740456253290176,
|
|
"num_tokens": 11882874.0,
|
|
"step": 5290
|
|
},
|
|
{
|
|
"entropy": 5.375743198394775,
|
|
"epoch": 5.724716062736614,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002600845681818467,
|
|
"loss": 4.645,
|
|
"mean_token_accuracy": 0.2470102548599243,
|
|
"num_tokens": 11893431.0,
|
|
"step": 5295
|
|
},
|
|
{
|
|
"entropy": 5.304393625259399,
|
|
"epoch": 5.730124391563007,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002596566201160427,
|
|
"loss": 4.7046,
|
|
"mean_token_accuracy": 0.23565791994333268,
|
|
"num_tokens": 11905311.0,
|
|
"step": 5300
|
|
},
|
|
{
|
|
"entropy": 5.278746271133423,
|
|
"epoch": 5.7355327203894,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00025922872780811664,
|
|
"loss": 4.7544,
|
|
"mean_token_accuracy": 0.23648327142000197,
|
|
"num_tokens": 11916342.0,
|
|
"step": 5305
|
|
},
|
|
{
|
|
"entropy": 5.37562665939331,
|
|
"epoch": 5.740941049215793,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00025880089281303027,
|
|
"loss": 4.7333,
|
|
"mean_token_accuracy": 0.23712480962276458,
|
|
"num_tokens": 11927485.0,
|
|
"step": 5310
|
|
},
|
|
{
|
|
"entropy": 5.331275892257691,
|
|
"epoch": 5.746349378042185,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00025837311668553683,
|
|
"loss": 4.6203,
|
|
"mean_token_accuracy": 0.24342997819185258,
|
|
"num_tokens": 11937968.0,
|
|
"step": 5315
|
|
},
|
|
{
|
|
"entropy": 5.25931224822998,
|
|
"epoch": 5.7517577068685775,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002579454009801757,
|
|
"loss": 4.5841,
|
|
"mean_token_accuracy": 0.253123314678669,
|
|
"num_tokens": 11947510.0,
|
|
"step": 5320
|
|
},
|
|
{
|
|
"entropy": 5.339625215530395,
|
|
"epoch": 5.75716603569497,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002575177472512669,
|
|
"loss": 4.7537,
|
|
"mean_token_accuracy": 0.2361442491412163,
|
|
"num_tokens": 11960002.0,
|
|
"step": 5325
|
|
},
|
|
{
|
|
"entropy": 5.2995460510253904,
|
|
"epoch": 5.762574364521363,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000257090157052905,
|
|
"loss": 4.6639,
|
|
"mean_token_accuracy": 0.2371893495321274,
|
|
"num_tokens": 11970895.0,
|
|
"step": 5330
|
|
},
|
|
{
|
|
"entropy": 5.319638729095459,
|
|
"epoch": 5.767982693347756,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00025666263193895375,
|
|
"loss": 4.6829,
|
|
"mean_token_accuracy": 0.23675003349781037,
|
|
"num_tokens": 11982910.0,
|
|
"step": 5335
|
|
},
|
|
{
|
|
"entropy": 5.361830759048462,
|
|
"epoch": 5.773391022174148,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00025623517346304027,
|
|
"loss": 4.7541,
|
|
"mean_token_accuracy": 0.23789821267127992,
|
|
"num_tokens": 11994066.0,
|
|
"step": 5340
|
|
},
|
|
{
|
|
"entropy": 5.321666479110718,
|
|
"epoch": 5.778799351000541,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00025580778317855,
|
|
"loss": 4.6563,
|
|
"mean_token_accuracy": 0.24208394140005113,
|
|
"num_tokens": 12004821.0,
|
|
"step": 5345
|
|
},
|
|
{
|
|
"entropy": 5.382619571685791,
|
|
"epoch": 5.784207679826934,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002553804626386199,
|
|
"loss": 4.7432,
|
|
"mean_token_accuracy": 0.237531116604805,
|
|
"num_tokens": 12014973.0,
|
|
"step": 5350
|
|
},
|
|
{
|
|
"entropy": 5.375884485244751,
|
|
"epoch": 5.789616008653326,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002549532133961339,
|
|
"loss": 4.8122,
|
|
"mean_token_accuracy": 0.22716958075761795,
|
|
"num_tokens": 12026528.0,
|
|
"step": 5355
|
|
},
|
|
{
|
|
"entropy": 5.443640184402466,
|
|
"epoch": 5.795024337479719,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00025452603700371675,
|
|
"loss": 4.8453,
|
|
"mean_token_accuracy": 0.22859499156475066,
|
|
"num_tokens": 12038733.0,
|
|
"step": 5360
|
|
},
|
|
{
|
|
"entropy": 5.403985834121704,
|
|
"epoch": 5.800432666306111,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002540989350137285,
|
|
"loss": 4.7745,
|
|
"mean_token_accuracy": 0.23548087626695632,
|
|
"num_tokens": 12049821.0,
|
|
"step": 5365
|
|
},
|
|
{
|
|
"entropy": 5.403467178344727,
|
|
"epoch": 5.805840995132504,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00025367190897825873,
|
|
"loss": 4.7679,
|
|
"mean_token_accuracy": 0.23344067931175233,
|
|
"num_tokens": 12060982.0,
|
|
"step": 5370
|
|
},
|
|
{
|
|
"entropy": 5.326777935028076,
|
|
"epoch": 5.811249323958897,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002532449604491209,
|
|
"loss": 4.6344,
|
|
"mean_token_accuracy": 0.24498462826013565,
|
|
"num_tokens": 12070842.0,
|
|
"step": 5375
|
|
},
|
|
{
|
|
"entropy": 5.2458696365356445,
|
|
"epoch": 5.816657652785289,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002528180909778471,
|
|
"loss": 4.628,
|
|
"mean_token_accuracy": 0.2526636555790901,
|
|
"num_tokens": 12080992.0,
|
|
"step": 5380
|
|
},
|
|
{
|
|
"entropy": 5.296866321563721,
|
|
"epoch": 5.822065981611682,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00025239130211568206,
|
|
"loss": 4.7124,
|
|
"mean_token_accuracy": 0.24308098405599593,
|
|
"num_tokens": 12091905.0,
|
|
"step": 5385
|
|
},
|
|
{
|
|
"entropy": 5.38506851196289,
|
|
"epoch": 5.827474310438075,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00025196459541357725,
|
|
"loss": 4.5971,
|
|
"mean_token_accuracy": 0.24719685912132264,
|
|
"num_tokens": 12101577.0,
|
|
"step": 5390
|
|
},
|
|
{
|
|
"entropy": 5.441853380203247,
|
|
"epoch": 5.832882639264467,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00025153797242218604,
|
|
"loss": 4.8111,
|
|
"mean_token_accuracy": 0.23143941164016724,
|
|
"num_tokens": 12112593.0,
|
|
"step": 5395
|
|
},
|
|
{
|
|
"entropy": 5.272561359405517,
|
|
"epoch": 5.83829096809086,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00025111143469185715,
|
|
"loss": 4.6561,
|
|
"mean_token_accuracy": 0.24364881068468094,
|
|
"num_tokens": 12122947.0,
|
|
"step": 5400
|
|
},
|
|
{
|
|
"entropy": 5.227482652664184,
|
|
"epoch": 5.8436992969172525,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002506849837726299,
|
|
"loss": 4.5813,
|
|
"mean_token_accuracy": 0.2607313424348831,
|
|
"num_tokens": 12133557.0,
|
|
"step": 5405
|
|
},
|
|
{
|
|
"entropy": 5.387114715576172,
|
|
"epoch": 5.849107625743645,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002502586212142278,
|
|
"loss": 4.7723,
|
|
"mean_token_accuracy": 0.2342202603816986,
|
|
"num_tokens": 12144735.0,
|
|
"step": 5410
|
|
},
|
|
{
|
|
"entropy": 5.304342794418335,
|
|
"epoch": 5.854515954570038,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00024983234856605337,
|
|
"loss": 4.7002,
|
|
"mean_token_accuracy": 0.24287605434656143,
|
|
"num_tokens": 12155786.0,
|
|
"step": 5415
|
|
},
|
|
{
|
|
"entropy": 5.334001541137695,
|
|
"epoch": 5.85992428339643,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00024940616737718235,
|
|
"loss": 4.7876,
|
|
"mean_token_accuracy": 0.23260244429111482,
|
|
"num_tokens": 12168846.0,
|
|
"step": 5420
|
|
},
|
|
{
|
|
"entropy": 5.2925127983093265,
|
|
"epoch": 5.865332612222823,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002489800791963583,
|
|
"loss": 4.6371,
|
|
"mean_token_accuracy": 0.23738714456558227,
|
|
"num_tokens": 12179743.0,
|
|
"step": 5425
|
|
},
|
|
{
|
|
"entropy": 5.39697380065918,
|
|
"epoch": 5.870740941049216,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00024855408557198657,
|
|
"loss": 4.7311,
|
|
"mean_token_accuracy": 0.24559662640094757,
|
|
"num_tokens": 12190754.0,
|
|
"step": 5430
|
|
},
|
|
{
|
|
"entropy": 5.398286008834839,
|
|
"epoch": 5.876149269875609,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000248128188052129,
|
|
"loss": 4.7163,
|
|
"mean_token_accuracy": 0.2345484212040901,
|
|
"num_tokens": 12201293.0,
|
|
"step": 5435
|
|
},
|
|
{
|
|
"entropy": 5.366581201553345,
|
|
"epoch": 5.881557598702001,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00024770238818449804,
|
|
"loss": 4.7808,
|
|
"mean_token_accuracy": 0.236073300242424,
|
|
"num_tokens": 12214361.0,
|
|
"step": 5440
|
|
},
|
|
{
|
|
"entropy": 5.426185035705567,
|
|
"epoch": 5.886965927528394,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00024727668751645164,
|
|
"loss": 4.8565,
|
|
"mean_token_accuracy": 0.2289385050535202,
|
|
"num_tokens": 12226766.0,
|
|
"step": 5445
|
|
},
|
|
{
|
|
"entropy": 5.371487760543824,
|
|
"epoch": 5.8923742563547865,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0002468510875949868,
|
|
"loss": 4.7624,
|
|
"mean_token_accuracy": 0.23546064645051956,
|
|
"num_tokens": 12237339.0,
|
|
"step": 5450
|
|
},
|
|
{
|
|
"entropy": 5.306449317932129,
|
|
"epoch": 5.897782585181179,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00024642558996673474,
|
|
"loss": 4.6658,
|
|
"mean_token_accuracy": 0.2567827194929123,
|
|
"num_tokens": 12248567.0,
|
|
"step": 5455
|
|
},
|
|
{
|
|
"entropy": 5.3372447967529295,
|
|
"epoch": 5.903190914007571,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00024600019617795465,
|
|
"loss": 4.677,
|
|
"mean_token_accuracy": 0.25030782371759414,
|
|
"num_tokens": 12258968.0,
|
|
"step": 5460
|
|
},
|
|
{
|
|
"entropy": 5.418646192550659,
|
|
"epoch": 5.908599242833964,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002455749077745288,
|
|
"loss": 4.7297,
|
|
"mean_token_accuracy": 0.24659058153629304,
|
|
"num_tokens": 12271233.0,
|
|
"step": 5465
|
|
},
|
|
{
|
|
"entropy": 5.327474737167359,
|
|
"epoch": 5.914007571660357,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00024514972630195603,
|
|
"loss": 4.6951,
|
|
"mean_token_accuracy": 0.231184484064579,
|
|
"num_tokens": 12280634.0,
|
|
"step": 5470
|
|
},
|
|
{
|
|
"entropy": 5.262243461608887,
|
|
"epoch": 5.91941590048675,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002447246533053469,
|
|
"loss": 4.7305,
|
|
"mean_token_accuracy": 0.24198452681303023,
|
|
"num_tokens": 12292282.0,
|
|
"step": 5475
|
|
},
|
|
{
|
|
"entropy": 5.362786149978637,
|
|
"epoch": 5.924824229313142,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00024429969032941756,
|
|
"loss": 4.7529,
|
|
"mean_token_accuracy": 0.239649298787117,
|
|
"num_tokens": 12303940.0,
|
|
"step": 5480
|
|
},
|
|
{
|
|
"entropy": 5.368451642990112,
|
|
"epoch": 5.930232558139535,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00024387483891848456,
|
|
"loss": 4.7389,
|
|
"mean_token_accuracy": 0.23966737538576127,
|
|
"num_tokens": 12315400.0,
|
|
"step": 5485
|
|
},
|
|
{
|
|
"entropy": 5.39257779121399,
|
|
"epoch": 5.935640886965928,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00024345010061645889,
|
|
"loss": 4.755,
|
|
"mean_token_accuracy": 0.23356909155845643,
|
|
"num_tokens": 12327462.0,
|
|
"step": 5490
|
|
},
|
|
{
|
|
"entropy": 5.318007278442383,
|
|
"epoch": 5.9410492157923205,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00024302547696684047,
|
|
"loss": 4.7158,
|
|
"mean_token_accuracy": 0.24026378095149994,
|
|
"num_tokens": 12338966.0,
|
|
"step": 5495
|
|
},
|
|
{
|
|
"entropy": 5.219871282577515,
|
|
"epoch": 5.9464575446187125,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00024260096951271253,
|
|
"loss": 4.5916,
|
|
"mean_token_accuracy": 0.2521056979894638,
|
|
"num_tokens": 12349663.0,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"epoch": 5.9464575446187125,
|
|
"eval_entropy": 5.210777058384635,
|
|
"eval_loss": 5.826916217803955,
|
|
"eval_mean_token_accuracy": 0.1816450426693667,
|
|
"eval_num_tokens": 12349663.0,
|
|
"eval_runtime": 1.9329,
|
|
"eval_samples_per_second": 1818.47,
|
|
"eval_steps_per_second": 227.632,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"entropy": 5.30617847442627,
|
|
"epoch": 5.951865873445105,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00024217657979673634,
|
|
"loss": 4.6653,
|
|
"mean_token_accuracy": 0.23816222101449966,
|
|
"num_tokens": 12361304.0,
|
|
"step": 5505
|
|
},
|
|
{
|
|
"entropy": 5.2627264022827145,
|
|
"epoch": 5.957274202271498,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00024175230936114507,
|
|
"loss": 4.6127,
|
|
"mean_token_accuracy": 0.25264633595943453,
|
|
"num_tokens": 12372162.0,
|
|
"step": 5510
|
|
},
|
|
{
|
|
"entropy": 5.315018129348755,
|
|
"epoch": 5.962682531097891,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002413281597477384,
|
|
"loss": 4.6519,
|
|
"mean_token_accuracy": 0.2445184111595154,
|
|
"num_tokens": 12383916.0,
|
|
"step": 5515
|
|
},
|
|
{
|
|
"entropy": 5.352186298370361,
|
|
"epoch": 5.968090859924283,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00024090413249787696,
|
|
"loss": 4.7641,
|
|
"mean_token_accuracy": 0.24259584099054338,
|
|
"num_tokens": 12394787.0,
|
|
"step": 5520
|
|
},
|
|
{
|
|
"entropy": 5.3336670875549315,
|
|
"epoch": 5.973499188750676,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002404802291524768,
|
|
"loss": 4.6652,
|
|
"mean_token_accuracy": 0.25006339251995086,
|
|
"num_tokens": 12405905.0,
|
|
"step": 5525
|
|
},
|
|
{
|
|
"entropy": 5.30520224571228,
|
|
"epoch": 5.978907517577069,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00024005645125200375,
|
|
"loss": 4.7283,
|
|
"mean_token_accuracy": 0.23745063543319703,
|
|
"num_tokens": 12416785.0,
|
|
"step": 5530
|
|
},
|
|
{
|
|
"entropy": 5.291027069091797,
|
|
"epoch": 5.984315846403462,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002396328003364675,
|
|
"loss": 4.6483,
|
|
"mean_token_accuracy": 0.2419503167271614,
|
|
"num_tokens": 12427900.0,
|
|
"step": 5535
|
|
},
|
|
{
|
|
"entropy": 5.411478710174561,
|
|
"epoch": 5.9897241752298545,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00023920927794541665,
|
|
"loss": 4.7761,
|
|
"mean_token_accuracy": 0.2441563457250595,
|
|
"num_tokens": 12439823.0,
|
|
"step": 5540
|
|
},
|
|
{
|
|
"entropy": 5.315765380859375,
|
|
"epoch": 5.995132504056246,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002387858856179324,
|
|
"loss": 4.6982,
|
|
"mean_token_accuracy": 0.23900126516819,
|
|
"num_tokens": 12450988.0,
|
|
"step": 5545
|
|
},
|
|
{
|
|
"entropy": 5.429989443884955,
|
|
"epoch": 6.0,
|
|
"grad_norm": 1.9609375,
|
|
"learning_rate": 0.00023836262489262361,
|
|
"loss": 4.7636,
|
|
"mean_token_accuracy": 0.24497108161449432,
|
|
"num_tokens": 12461334.0,
|
|
"step": 5550
|
|
},
|
|
{
|
|
"entropy": 5.292334842681885,
|
|
"epoch": 6.005408328826393,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00023793949730762049,
|
|
"loss": 4.4971,
|
|
"mean_token_accuracy": 0.2586694613099098,
|
|
"num_tokens": 12473206.0,
|
|
"step": 5555
|
|
},
|
|
{
|
|
"entropy": 5.322540378570556,
|
|
"epoch": 6.010816657652786,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00023751650440056982,
|
|
"loss": 4.5997,
|
|
"mean_token_accuracy": 0.25555664151906965,
|
|
"num_tokens": 12484247.0,
|
|
"step": 5560
|
|
},
|
|
{
|
|
"entropy": 5.3309849262237545,
|
|
"epoch": 6.016224986479178,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00023709364770862885,
|
|
"loss": 4.4988,
|
|
"mean_token_accuracy": 0.26663260012865064,
|
|
"num_tokens": 12495234.0,
|
|
"step": 5565
|
|
},
|
|
{
|
|
"entropy": 5.298547649383545,
|
|
"epoch": 6.0216333153055706,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00023667092876845976,
|
|
"loss": 4.6107,
|
|
"mean_token_accuracy": 0.2537641391158104,
|
|
"num_tokens": 12506516.0,
|
|
"step": 5570
|
|
},
|
|
{
|
|
"entropy": 5.314134550094605,
|
|
"epoch": 6.027041644131963,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002362483491162239,
|
|
"loss": 4.5192,
|
|
"mean_token_accuracy": 0.2655138850212097,
|
|
"num_tokens": 12517238.0,
|
|
"step": 5575
|
|
},
|
|
{
|
|
"entropy": 5.352644443511963,
|
|
"epoch": 6.032449972958356,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00023582591028757706,
|
|
"loss": 4.5337,
|
|
"mean_token_accuracy": 0.2598115220665932,
|
|
"num_tokens": 12527212.0,
|
|
"step": 5580
|
|
},
|
|
{
|
|
"entropy": 5.313656187057495,
|
|
"epoch": 6.037858301784748,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00023540361381766274,
|
|
"loss": 4.5239,
|
|
"mean_token_accuracy": 0.25893490463495256,
|
|
"num_tokens": 12537939.0,
|
|
"step": 5585
|
|
},
|
|
{
|
|
"entropy": 5.276715421676636,
|
|
"epoch": 6.043266630611141,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002349814612411073,
|
|
"loss": 4.5382,
|
|
"mean_token_accuracy": 0.2535753384232521,
|
|
"num_tokens": 12547987.0,
|
|
"step": 5590
|
|
},
|
|
{
|
|
"entropy": 5.299291133880615,
|
|
"epoch": 6.048674959437534,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00023455945409201411,
|
|
"loss": 4.539,
|
|
"mean_token_accuracy": 0.2553084596991539,
|
|
"num_tokens": 12558505.0,
|
|
"step": 5595
|
|
},
|
|
{
|
|
"entropy": 5.31742753982544,
|
|
"epoch": 6.054083288263927,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00023413759390395828,
|
|
"loss": 4.4841,
|
|
"mean_token_accuracy": 0.258325582742691,
|
|
"num_tokens": 12569748.0,
|
|
"step": 5600
|
|
},
|
|
{
|
|
"entropy": 5.2306907176971436,
|
|
"epoch": 6.059491617090319,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002337158822099807,
|
|
"loss": 4.4695,
|
|
"mean_token_accuracy": 0.26026298850774765,
|
|
"num_tokens": 12581420.0,
|
|
"step": 5605
|
|
},
|
|
{
|
|
"entropy": 5.328782939910889,
|
|
"epoch": 6.064899945916712,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002332943205425825,
|
|
"loss": 4.5561,
|
|
"mean_token_accuracy": 0.25838291347026826,
|
|
"num_tokens": 12593781.0,
|
|
"step": 5610
|
|
},
|
|
{
|
|
"entropy": 5.270041227340698,
|
|
"epoch": 6.0703082747431045,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00023287291043371978,
|
|
"loss": 4.5151,
|
|
"mean_token_accuracy": 0.25252519398927686,
|
|
"num_tokens": 12606264.0,
|
|
"step": 5615
|
|
},
|
|
{
|
|
"entropy": 5.420348644256592,
|
|
"epoch": 6.075716603569497,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00023245165341479796,
|
|
"loss": 4.5686,
|
|
"mean_token_accuracy": 0.26410178989171984,
|
|
"num_tokens": 12617419.0,
|
|
"step": 5620
|
|
},
|
|
{
|
|
"entropy": 5.313140344619751,
|
|
"epoch": 6.081124932395889,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00023203055101666593,
|
|
"loss": 4.5032,
|
|
"mean_token_accuracy": 0.25365361124277114,
|
|
"num_tokens": 12628628.0,
|
|
"step": 5625
|
|
},
|
|
{
|
|
"entropy": 5.243945503234864,
|
|
"epoch": 6.086533261222282,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0002316096047696108,
|
|
"loss": 4.4663,
|
|
"mean_token_accuracy": 0.255739326775074,
|
|
"num_tokens": 12638375.0,
|
|
"step": 5630
|
|
},
|
|
{
|
|
"entropy": 5.221099758148194,
|
|
"epoch": 6.091941590048675,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002311888162033522,
|
|
"loss": 4.515,
|
|
"mean_token_accuracy": 0.2507918506860733,
|
|
"num_tokens": 12649174.0,
|
|
"step": 5635
|
|
},
|
|
{
|
|
"entropy": 5.267313003540039,
|
|
"epoch": 6.097349918875068,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00023076818684703682,
|
|
"loss": 4.4768,
|
|
"mean_token_accuracy": 0.2619173377752304,
|
|
"num_tokens": 12659454.0,
|
|
"step": 5640
|
|
},
|
|
{
|
|
"entropy": 5.278536653518676,
|
|
"epoch": 6.10275824770146,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00023034771822923268,
|
|
"loss": 4.5147,
|
|
"mean_token_accuracy": 0.25881255865097047,
|
|
"num_tokens": 12672093.0,
|
|
"step": 5645
|
|
},
|
|
{
|
|
"entropy": 5.326001310348511,
|
|
"epoch": 6.108166576527853,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002299274118779236,
|
|
"loss": 4.548,
|
|
"mean_token_accuracy": 0.2538618862628937,
|
|
"num_tokens": 12683628.0,
|
|
"step": 5650
|
|
},
|
|
{
|
|
"entropy": 5.279880094528198,
|
|
"epoch": 6.113574905354246,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00022950726932050386,
|
|
"loss": 4.5155,
|
|
"mean_token_accuracy": 0.25998761504888535,
|
|
"num_tokens": 12694726.0,
|
|
"step": 5655
|
|
},
|
|
{
|
|
"entropy": 5.332649898529053,
|
|
"epoch": 6.1189832341806385,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00022908729208377267,
|
|
"loss": 4.566,
|
|
"mean_token_accuracy": 0.25722457766532897,
|
|
"num_tokens": 12705292.0,
|
|
"step": 5660
|
|
},
|
|
{
|
|
"entropy": 5.35866117477417,
|
|
"epoch": 6.1243915630070305,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00022866748169392815,
|
|
"loss": 4.5629,
|
|
"mean_token_accuracy": 0.25580236315727234,
|
|
"num_tokens": 12717721.0,
|
|
"step": 5665
|
|
},
|
|
{
|
|
"entropy": 5.294124698638916,
|
|
"epoch": 6.129799891833423,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00022824783967656208,
|
|
"loss": 4.555,
|
|
"mean_token_accuracy": 0.257673142850399,
|
|
"num_tokens": 12728360.0,
|
|
"step": 5670
|
|
},
|
|
{
|
|
"entropy": 5.295912647247315,
|
|
"epoch": 6.135208220659816,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00022782836755665475,
|
|
"loss": 4.5768,
|
|
"mean_token_accuracy": 0.25268289148807527,
|
|
"num_tokens": 12739771.0,
|
|
"step": 5675
|
|
},
|
|
{
|
|
"entropy": 5.318285894393921,
|
|
"epoch": 6.140616549486209,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00022740906685856876,
|
|
"loss": 4.5381,
|
|
"mean_token_accuracy": 0.2523332178592682,
|
|
"num_tokens": 12750024.0,
|
|
"step": 5680
|
|
},
|
|
{
|
|
"entropy": 5.2828937530517575,
|
|
"epoch": 6.146024878312601,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00022698993910604383,
|
|
"loss": 4.5017,
|
|
"mean_token_accuracy": 0.2568276524543762,
|
|
"num_tokens": 12761160.0,
|
|
"step": 5685
|
|
},
|
|
{
|
|
"entropy": 5.337252140045166,
|
|
"epoch": 6.151433207138994,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000226570985822191,
|
|
"loss": 4.5069,
|
|
"mean_token_accuracy": 0.26051444858312606,
|
|
"num_tokens": 12773043.0,
|
|
"step": 5690
|
|
},
|
|
{
|
|
"entropy": 5.213260841369629,
|
|
"epoch": 6.156841535965387,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00022615220852948765,
|
|
"loss": 4.5126,
|
|
"mean_token_accuracy": 0.25155836939811704,
|
|
"num_tokens": 12783479.0,
|
|
"step": 5695
|
|
},
|
|
{
|
|
"entropy": 5.3765974044799805,
|
|
"epoch": 6.16224986479178,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00022573360874977145,
|
|
"loss": 4.616,
|
|
"mean_token_accuracy": 0.25408181399106977,
|
|
"num_tokens": 12794889.0,
|
|
"step": 5700
|
|
},
|
|
{
|
|
"entropy": 5.314326620101928,
|
|
"epoch": 6.167658193618172,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00022531518800423484,
|
|
"loss": 4.5688,
|
|
"mean_token_accuracy": 0.25127440243959426,
|
|
"num_tokens": 12806161.0,
|
|
"step": 5705
|
|
},
|
|
{
|
|
"entropy": 5.348446846008301,
|
|
"epoch": 6.173066522444564,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002248969478134199,
|
|
"loss": 4.6028,
|
|
"mean_token_accuracy": 0.247745880484581,
|
|
"num_tokens": 12817829.0,
|
|
"step": 5710
|
|
},
|
|
{
|
|
"entropy": 5.3233452320098875,
|
|
"epoch": 6.178474851270957,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00022447888969721232,
|
|
"loss": 4.6208,
|
|
"mean_token_accuracy": 0.24070498198270798,
|
|
"num_tokens": 12828212.0,
|
|
"step": 5715
|
|
},
|
|
{
|
|
"entropy": 5.377896308898926,
|
|
"epoch": 6.18388318009735,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00022406101517483636,
|
|
"loss": 4.5649,
|
|
"mean_token_accuracy": 0.2563469722867012,
|
|
"num_tokens": 12840058.0,
|
|
"step": 5720
|
|
},
|
|
{
|
|
"entropy": 5.282167196273804,
|
|
"epoch": 6.189291508923742,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00022364332576484897,
|
|
"loss": 4.5678,
|
|
"mean_token_accuracy": 0.2506422162055969,
|
|
"num_tokens": 12852833.0,
|
|
"step": 5725
|
|
},
|
|
{
|
|
"entropy": 5.332680130004883,
|
|
"epoch": 6.194699837750135,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002232258229851345,
|
|
"loss": 4.5506,
|
|
"mean_token_accuracy": 0.2593381077051163,
|
|
"num_tokens": 12863487.0,
|
|
"step": 5730
|
|
},
|
|
{
|
|
"entropy": 5.253411293029785,
|
|
"epoch": 6.200108166576528,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00022280850835289885,
|
|
"loss": 4.4592,
|
|
"mean_token_accuracy": 0.2645656645298004,
|
|
"num_tokens": 12873952.0,
|
|
"step": 5735
|
|
},
|
|
{
|
|
"entropy": 5.246587324142456,
|
|
"epoch": 6.205516495402921,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00022239138338466452,
|
|
"loss": 4.5619,
|
|
"mean_token_accuracy": 0.24997915476560592,
|
|
"num_tokens": 12884462.0,
|
|
"step": 5740
|
|
},
|
|
{
|
|
"entropy": 5.28809084892273,
|
|
"epoch": 6.210924824229313,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00022197444959626451,
|
|
"loss": 4.5116,
|
|
"mean_token_accuracy": 0.2621185913681984,
|
|
"num_tokens": 12894996.0,
|
|
"step": 5745
|
|
},
|
|
{
|
|
"entropy": 5.266299915313721,
|
|
"epoch": 6.2163331530557056,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00022155770850283723,
|
|
"loss": 4.4885,
|
|
"mean_token_accuracy": 0.26162062436342237,
|
|
"num_tokens": 12906256.0,
|
|
"step": 5750
|
|
},
|
|
{
|
|
"entropy": 5.291370534896851,
|
|
"epoch": 6.221741481882098,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00022114116161882065,
|
|
"loss": 4.6133,
|
|
"mean_token_accuracy": 0.2538672149181366,
|
|
"num_tokens": 12918738.0,
|
|
"step": 5755
|
|
},
|
|
{
|
|
"entropy": 5.297574472427368,
|
|
"epoch": 6.227149810708491,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00022072481045794733,
|
|
"loss": 4.5306,
|
|
"mean_token_accuracy": 0.2529829189181328,
|
|
"num_tokens": 12929918.0,
|
|
"step": 5760
|
|
},
|
|
{
|
|
"entropy": 5.341555976867676,
|
|
"epoch": 6.232558139534884,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00022030865653323823,
|
|
"loss": 4.5834,
|
|
"mean_token_accuracy": 0.2557613417506218,
|
|
"num_tokens": 12941427.0,
|
|
"step": 5765
|
|
},
|
|
{
|
|
"entropy": 5.27587251663208,
|
|
"epoch": 6.237966468361276,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002198927013569977,
|
|
"loss": 4.5224,
|
|
"mean_token_accuracy": 0.26029486358165743,
|
|
"num_tokens": 12952889.0,
|
|
"step": 5770
|
|
},
|
|
{
|
|
"entropy": 5.293698120117187,
|
|
"epoch": 6.243374797187669,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002194769464408076,
|
|
"loss": 4.5405,
|
|
"mean_token_accuracy": 0.25903954952955244,
|
|
"num_tokens": 12962928.0,
|
|
"step": 5775
|
|
},
|
|
{
|
|
"entropy": 5.292485284805298,
|
|
"epoch": 6.248783126014062,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0002190613932955226,
|
|
"loss": 4.5041,
|
|
"mean_token_accuracy": 0.26182852387428285,
|
|
"num_tokens": 12972901.0,
|
|
"step": 5780
|
|
},
|
|
{
|
|
"entropy": 5.3199540138244625,
|
|
"epoch": 6.254191454840455,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00021864604343126365,
|
|
"loss": 4.5971,
|
|
"mean_token_accuracy": 0.2563637360930443,
|
|
"num_tokens": 12985296.0,
|
|
"step": 5785
|
|
},
|
|
{
|
|
"entropy": 5.388424825668335,
|
|
"epoch": 6.259599783666847,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00021823089835741312,
|
|
"loss": 4.6626,
|
|
"mean_token_accuracy": 0.24707048535346984,
|
|
"num_tokens": 12997069.0,
|
|
"step": 5790
|
|
},
|
|
{
|
|
"entropy": 5.309629964828491,
|
|
"epoch": 6.2650081124932395,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000217815959582609,
|
|
"loss": 4.5382,
|
|
"mean_token_accuracy": 0.26082683354616165,
|
|
"num_tokens": 13007147.0,
|
|
"step": 5795
|
|
},
|
|
{
|
|
"entropy": 5.2757481098175045,
|
|
"epoch": 6.270416441319632,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00021740122861473992,
|
|
"loss": 4.565,
|
|
"mean_token_accuracy": 0.25873169898986814,
|
|
"num_tokens": 13019084.0,
|
|
"step": 5800
|
|
},
|
|
{
|
|
"entropy": 5.296625423431396,
|
|
"epoch": 6.275824770146025,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00021698670696093932,
|
|
"loss": 4.5718,
|
|
"mean_token_accuracy": 0.2559361010789871,
|
|
"num_tokens": 13030751.0,
|
|
"step": 5805
|
|
},
|
|
{
|
|
"entropy": 5.308929347991944,
|
|
"epoch": 6.281233098972417,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00021657239612757963,
|
|
"loss": 4.5264,
|
|
"mean_token_accuracy": 0.25520056784152984,
|
|
"num_tokens": 13042731.0,
|
|
"step": 5810
|
|
},
|
|
{
|
|
"entropy": 5.287246942520142,
|
|
"epoch": 6.28664142779881,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00021615829762026734,
|
|
"loss": 4.5062,
|
|
"mean_token_accuracy": 0.25439499169588087,
|
|
"num_tokens": 13053442.0,
|
|
"step": 5815
|
|
},
|
|
{
|
|
"entropy": 5.32212347984314,
|
|
"epoch": 6.292049756625203,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00021574441294383752,
|
|
"loss": 4.5618,
|
|
"mean_token_accuracy": 0.2544273599982262,
|
|
"num_tokens": 13063911.0,
|
|
"step": 5820
|
|
},
|
|
{
|
|
"entropy": 5.356846141815185,
|
|
"epoch": 6.297458085451596,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00021533074360234806,
|
|
"loss": 4.6193,
|
|
"mean_token_accuracy": 0.24670790135860443,
|
|
"num_tokens": 13075887.0,
|
|
"step": 5825
|
|
},
|
|
{
|
|
"entropy": 5.2510631561279295,
|
|
"epoch": 6.302866414277988,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00021491729109907414,
|
|
"loss": 4.4989,
|
|
"mean_token_accuracy": 0.26197862476110456,
|
|
"num_tokens": 13087320.0,
|
|
"step": 5830
|
|
},
|
|
{
|
|
"entropy": 5.280094766616822,
|
|
"epoch": 6.308274743104381,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00021450405693650333,
|
|
"loss": 4.523,
|
|
"mean_token_accuracy": 0.2632941335439682,
|
|
"num_tokens": 13099078.0,
|
|
"step": 5835
|
|
},
|
|
{
|
|
"entropy": 5.339206075668335,
|
|
"epoch": 6.3136830719307735,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00021409104261632932,
|
|
"loss": 4.6057,
|
|
"mean_token_accuracy": 0.25635338574647903,
|
|
"num_tokens": 13111909.0,
|
|
"step": 5840
|
|
},
|
|
{
|
|
"entropy": 5.233299303054809,
|
|
"epoch": 6.319091400757166,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00021367824963944727,
|
|
"loss": 4.5003,
|
|
"mean_token_accuracy": 0.25619914680719375,
|
|
"num_tokens": 13123297.0,
|
|
"step": 5845
|
|
},
|
|
{
|
|
"entropy": 5.2498924255371096,
|
|
"epoch": 6.324499729583558,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00021326567950594766,
|
|
"loss": 4.5748,
|
|
"mean_token_accuracy": 0.25029214322566984,
|
|
"num_tokens": 13133959.0,
|
|
"step": 5850
|
|
},
|
|
{
|
|
"entropy": 5.231543350219726,
|
|
"epoch": 6.329908058409951,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0002128533337151114,
|
|
"loss": 4.431,
|
|
"mean_token_accuracy": 0.2638119250535965,
|
|
"num_tokens": 13144858.0,
|
|
"step": 5855
|
|
},
|
|
{
|
|
"entropy": 5.267545080184936,
|
|
"epoch": 6.335316387236344,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002124412137654041,
|
|
"loss": 4.5728,
|
|
"mean_token_accuracy": 0.2533300340175629,
|
|
"num_tokens": 13156437.0,
|
|
"step": 5860
|
|
},
|
|
{
|
|
"entropy": 5.2572516918182375,
|
|
"epoch": 6.340724716062737,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00021202932115447056,
|
|
"loss": 4.5132,
|
|
"mean_token_accuracy": 0.26316333562135696,
|
|
"num_tokens": 13166860.0,
|
|
"step": 5865
|
|
},
|
|
{
|
|
"entropy": 5.306639671325684,
|
|
"epoch": 6.346133044889129,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00021161765737912937,
|
|
"loss": 4.604,
|
|
"mean_token_accuracy": 0.2563185691833496,
|
|
"num_tokens": 13178563.0,
|
|
"step": 5870
|
|
},
|
|
{
|
|
"entropy": 5.320454168319702,
|
|
"epoch": 6.351541373715522,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00021120622393536775,
|
|
"loss": 4.5228,
|
|
"mean_token_accuracy": 0.2562135517597198,
|
|
"num_tokens": 13189915.0,
|
|
"step": 5875
|
|
},
|
|
{
|
|
"entropy": 5.350077199935913,
|
|
"epoch": 6.356949702541915,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00021079502231833585,
|
|
"loss": 4.5468,
|
|
"mean_token_accuracy": 0.2602106213569641,
|
|
"num_tokens": 13201767.0,
|
|
"step": 5880
|
|
},
|
|
{
|
|
"entropy": 5.277065277099609,
|
|
"epoch": 6.3623580313683075,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00021038405402234116,
|
|
"loss": 4.5516,
|
|
"mean_token_accuracy": 0.2614173337817192,
|
|
"num_tokens": 13212265.0,
|
|
"step": 5885
|
|
},
|
|
{
|
|
"entropy": 5.252430582046509,
|
|
"epoch": 6.367766360194699,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002099733205408435,
|
|
"loss": 4.577,
|
|
"mean_token_accuracy": 0.2522157892584801,
|
|
"num_tokens": 13223503.0,
|
|
"step": 5890
|
|
},
|
|
{
|
|
"entropy": 5.31737642288208,
|
|
"epoch": 6.373174689021092,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002095628233664492,
|
|
"loss": 4.5574,
|
|
"mean_token_accuracy": 0.2532134160399437,
|
|
"num_tokens": 13234797.0,
|
|
"step": 5895
|
|
},
|
|
{
|
|
"entropy": 5.284051609039307,
|
|
"epoch": 6.378583017847485,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002091525639909061,
|
|
"loss": 4.5262,
|
|
"mean_token_accuracy": 0.2593584477901459,
|
|
"num_tokens": 13246421.0,
|
|
"step": 5900
|
|
},
|
|
{
|
|
"entropy": 5.280451393127441,
|
|
"epoch": 6.383991346673878,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00020874254390509757,
|
|
"loss": 4.5255,
|
|
"mean_token_accuracy": 0.2569135382771492,
|
|
"num_tokens": 13256747.0,
|
|
"step": 5905
|
|
},
|
|
{
|
|
"entropy": 5.309819602966309,
|
|
"epoch": 6.38939967550027,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00020833276459903766,
|
|
"loss": 4.5199,
|
|
"mean_token_accuracy": 0.2580658808350563,
|
|
"num_tokens": 13267620.0,
|
|
"step": 5910
|
|
},
|
|
{
|
|
"entropy": 5.353440952301026,
|
|
"epoch": 6.394808004326663,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002079232275618651,
|
|
"loss": 4.5008,
|
|
"mean_token_accuracy": 0.2587909549474716,
|
|
"num_tokens": 13277818.0,
|
|
"step": 5915
|
|
},
|
|
{
|
|
"entropy": 5.161814117431641,
|
|
"epoch": 6.400216333153056,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002075139342818388,
|
|
"loss": 4.4867,
|
|
"mean_token_accuracy": 0.25960992723703386,
|
|
"num_tokens": 13289905.0,
|
|
"step": 5920
|
|
},
|
|
{
|
|
"entropy": 5.262392234802246,
|
|
"epoch": 6.405624661979449,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00020710488624633118,
|
|
"loss": 4.4738,
|
|
"mean_token_accuracy": 0.26027931123971937,
|
|
"num_tokens": 13302465.0,
|
|
"step": 5925
|
|
},
|
|
{
|
|
"entropy": 5.314022016525269,
|
|
"epoch": 6.411032990805841,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00020669608494182384,
|
|
"loss": 4.5826,
|
|
"mean_token_accuracy": 0.24711670726537704,
|
|
"num_tokens": 13314366.0,
|
|
"step": 5930
|
|
},
|
|
{
|
|
"entropy": 5.17999415397644,
|
|
"epoch": 6.416441319632233,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00020628753185390164,
|
|
"loss": 4.4051,
|
|
"mean_token_accuracy": 0.2698850393295288,
|
|
"num_tokens": 13324981.0,
|
|
"step": 5935
|
|
},
|
|
{
|
|
"entropy": 5.257151412963867,
|
|
"epoch": 6.421849648458626,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002058792284672475,
|
|
"loss": 4.5399,
|
|
"mean_token_accuracy": 0.25577238500118255,
|
|
"num_tokens": 13335705.0,
|
|
"step": 5940
|
|
},
|
|
{
|
|
"entropy": 5.142347049713135,
|
|
"epoch": 6.427257977285019,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002054711762656369,
|
|
"loss": 4.4773,
|
|
"mean_token_accuracy": 0.25621184408664704,
|
|
"num_tokens": 13346515.0,
|
|
"step": 5945
|
|
},
|
|
{
|
|
"entropy": 5.226314496994019,
|
|
"epoch": 6.432666306111411,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00020506337673193236,
|
|
"loss": 4.5162,
|
|
"mean_token_accuracy": 0.25663149654865264,
|
|
"num_tokens": 13357780.0,
|
|
"step": 5950
|
|
},
|
|
{
|
|
"entropy": 5.264117336273193,
|
|
"epoch": 6.438074634937804,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00020465583134807836,
|
|
"loss": 4.5072,
|
|
"mean_token_accuracy": 0.2576764702796936,
|
|
"num_tokens": 13369549.0,
|
|
"step": 5955
|
|
},
|
|
{
|
|
"entropy": 5.358155202865601,
|
|
"epoch": 6.443482963764197,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00020424854159509587,
|
|
"loss": 4.5782,
|
|
"mean_token_accuracy": 0.2542428493499756,
|
|
"num_tokens": 13380259.0,
|
|
"step": 5960
|
|
},
|
|
{
|
|
"entropy": 5.34224648475647,
|
|
"epoch": 6.44889129259059,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0002038415089530767,
|
|
"loss": 4.6161,
|
|
"mean_token_accuracy": 0.2523256987333298,
|
|
"num_tokens": 13392049.0,
|
|
"step": 5965
|
|
},
|
|
{
|
|
"entropy": 5.279792070388794,
|
|
"epoch": 6.4542996214169825,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00020343473490117846,
|
|
"loss": 4.5851,
|
|
"mean_token_accuracy": 0.24941320419311525,
|
|
"num_tokens": 13403874.0,
|
|
"step": 5970
|
|
},
|
|
{
|
|
"entropy": 5.266143655776977,
|
|
"epoch": 6.4597079502433745,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00020302822091761884,
|
|
"loss": 4.5394,
|
|
"mean_token_accuracy": 0.264926378428936,
|
|
"num_tokens": 13415194.0,
|
|
"step": 5975
|
|
},
|
|
{
|
|
"entropy": 5.287329530715942,
|
|
"epoch": 6.465116279069767,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00020262196847967091,
|
|
"loss": 4.5716,
|
|
"mean_token_accuracy": 0.2551450803875923,
|
|
"num_tokens": 13425159.0,
|
|
"step": 5980
|
|
},
|
|
{
|
|
"entropy": 5.321452713012695,
|
|
"epoch": 6.47052460789616,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00020221597906365667,
|
|
"loss": 4.5586,
|
|
"mean_token_accuracy": 0.2508067712187767,
|
|
"num_tokens": 13437279.0,
|
|
"step": 5985
|
|
},
|
|
{
|
|
"entropy": 5.327033519744873,
|
|
"epoch": 6.475932936722553,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00020181025414494285,
|
|
"loss": 4.5559,
|
|
"mean_token_accuracy": 0.252538600564003,
|
|
"num_tokens": 13449170.0,
|
|
"step": 5990
|
|
},
|
|
{
|
|
"entropy": 5.341722440719605,
|
|
"epoch": 6.481341265548945,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002014047951979344,
|
|
"loss": 4.613,
|
|
"mean_token_accuracy": 0.24629263430833817,
|
|
"num_tokens": 13460216.0,
|
|
"step": 5995
|
|
},
|
|
{
|
|
"entropy": 5.306678056716919,
|
|
"epoch": 6.486749594375338,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002009996036960704,
|
|
"loss": 4.586,
|
|
"mean_token_accuracy": 0.25587851107120513,
|
|
"num_tokens": 13470468.0,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"epoch": 6.486749594375338,
|
|
"eval_entropy": 5.205435578389602,
|
|
"eval_loss": 5.85944938659668,
|
|
"eval_mean_token_accuracy": 0.18033043237572366,
|
|
"eval_num_tokens": 13470468.0,
|
|
"eval_runtime": 1.9431,
|
|
"eval_samples_per_second": 1808.924,
|
|
"eval_steps_per_second": 226.437,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"entropy": 5.347350549697876,
|
|
"epoch": 6.492157923201731,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00020059468111181757,
|
|
"loss": 4.5934,
|
|
"mean_token_accuracy": 0.25713061094284057,
|
|
"num_tokens": 13482690.0,
|
|
"step": 6005
|
|
},
|
|
{
|
|
"entropy": 5.201942014694214,
|
|
"epoch": 6.497566252028124,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00020019002891666555,
|
|
"loss": 4.4928,
|
|
"mean_token_accuracy": 0.2603228658437729,
|
|
"num_tokens": 13494806.0,
|
|
"step": 6010
|
|
},
|
|
{
|
|
"entropy": 5.189780139923096,
|
|
"epoch": 6.502974580854516,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00019978564858112125,
|
|
"loss": 4.476,
|
|
"mean_token_accuracy": 0.2650508269667625,
|
|
"num_tokens": 13506968.0,
|
|
"step": 6015
|
|
},
|
|
{
|
|
"entropy": 5.325718593597412,
|
|
"epoch": 6.5083829096809085,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00019938154157470378,
|
|
"loss": 4.5991,
|
|
"mean_token_accuracy": 0.25761554688215255,
|
|
"num_tokens": 13518817.0,
|
|
"step": 6020
|
|
},
|
|
{
|
|
"entropy": 5.3316652297973635,
|
|
"epoch": 6.513791238507301,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00019897770936593912,
|
|
"loss": 4.5236,
|
|
"mean_token_accuracy": 0.2606540858745575,
|
|
"num_tokens": 13529054.0,
|
|
"step": 6025
|
|
},
|
|
{
|
|
"entropy": 5.272928905487061,
|
|
"epoch": 6.519199567333694,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001985741534223543,
|
|
"loss": 4.5391,
|
|
"mean_token_accuracy": 0.2551047787070274,
|
|
"num_tokens": 13540462.0,
|
|
"step": 6030
|
|
},
|
|
{
|
|
"entropy": 5.366613721847534,
|
|
"epoch": 6.524607896160086,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00019817087521047244,
|
|
"loss": 4.6492,
|
|
"mean_token_accuracy": 0.2518571466207504,
|
|
"num_tokens": 13551949.0,
|
|
"step": 6035
|
|
},
|
|
{
|
|
"entropy": 5.349287366867065,
|
|
"epoch": 6.530016224986479,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00019776787619580767,
|
|
"loss": 4.5774,
|
|
"mean_token_accuracy": 0.2593910336494446,
|
|
"num_tokens": 13562567.0,
|
|
"step": 6040
|
|
},
|
|
{
|
|
"entropy": 5.260582685470581,
|
|
"epoch": 6.535424553812872,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0001973651578428593,
|
|
"loss": 4.5409,
|
|
"mean_token_accuracy": 0.2613977700471878,
|
|
"num_tokens": 13573696.0,
|
|
"step": 6045
|
|
},
|
|
{
|
|
"entropy": 5.356221771240234,
|
|
"epoch": 6.540832882639265,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00019696272161510663,
|
|
"loss": 4.6144,
|
|
"mean_token_accuracy": 0.24852794408798218,
|
|
"num_tokens": 13585941.0,
|
|
"step": 6050
|
|
},
|
|
{
|
|
"entropy": 5.213946151733398,
|
|
"epoch": 6.546241211465657,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00019656056897500408,
|
|
"loss": 4.4148,
|
|
"mean_token_accuracy": 0.2675089165568352,
|
|
"num_tokens": 13597665.0,
|
|
"step": 6055
|
|
},
|
|
{
|
|
"entropy": 5.204039430618286,
|
|
"epoch": 6.55164954029205,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00019615870138397506,
|
|
"loss": 4.5419,
|
|
"mean_token_accuracy": 0.2532116249203682,
|
|
"num_tokens": 13608732.0,
|
|
"step": 6060
|
|
},
|
|
{
|
|
"entropy": 5.368945455551147,
|
|
"epoch": 6.5570578691184425,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001957571203024075,
|
|
"loss": 4.6435,
|
|
"mean_token_accuracy": 0.25293977558612823,
|
|
"num_tokens": 13620710.0,
|
|
"step": 6065
|
|
},
|
|
{
|
|
"entropy": 5.219055843353272,
|
|
"epoch": 6.562466197944835,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00019535582718964773,
|
|
"loss": 4.4922,
|
|
"mean_token_accuracy": 0.2548223987221718,
|
|
"num_tokens": 13632157.0,
|
|
"step": 6070
|
|
},
|
|
{
|
|
"entropy": 5.2971395492553714,
|
|
"epoch": 6.567874526771227,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00019495482350399612,
|
|
"loss": 4.4912,
|
|
"mean_token_accuracy": 0.2708602637052536,
|
|
"num_tokens": 13642833.0,
|
|
"step": 6075
|
|
},
|
|
{
|
|
"entropy": 5.3008780002594,
|
|
"epoch": 6.57328285559762,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001945541107027009,
|
|
"loss": 4.606,
|
|
"mean_token_accuracy": 0.2507108375430107,
|
|
"num_tokens": 13654985.0,
|
|
"step": 6080
|
|
},
|
|
{
|
|
"entropy": 5.252521514892578,
|
|
"epoch": 6.578691184424013,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001941536902419533,
|
|
"loss": 4.5604,
|
|
"mean_token_accuracy": 0.25265324413776397,
|
|
"num_tokens": 13666180.0,
|
|
"step": 6085
|
|
},
|
|
{
|
|
"entropy": 5.262233924865723,
|
|
"epoch": 6.584099513250406,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00019375356357688206,
|
|
"loss": 4.5218,
|
|
"mean_token_accuracy": 0.2639009147882462,
|
|
"num_tokens": 13677993.0,
|
|
"step": 6090
|
|
},
|
|
{
|
|
"entropy": 5.373490905761718,
|
|
"epoch": 6.589507842076799,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00019335373216154862,
|
|
"loss": 4.6691,
|
|
"mean_token_accuracy": 0.24974068254232407,
|
|
"num_tokens": 13687427.0,
|
|
"step": 6095
|
|
},
|
|
{
|
|
"entropy": 5.222711420059204,
|
|
"epoch": 6.594916170903191,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00019295419744894116,
|
|
"loss": 4.4707,
|
|
"mean_token_accuracy": 0.26479039937257765,
|
|
"num_tokens": 13698255.0,
|
|
"step": 6100
|
|
},
|
|
{
|
|
"entropy": 5.32066855430603,
|
|
"epoch": 6.600324499729584,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00019255496089096985,
|
|
"loss": 4.5819,
|
|
"mean_token_accuracy": 0.2568788334727287,
|
|
"num_tokens": 13710471.0,
|
|
"step": 6105
|
|
},
|
|
{
|
|
"entropy": 5.3814596176147464,
|
|
"epoch": 6.605732828555976,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00019215602393846104,
|
|
"loss": 4.6806,
|
|
"mean_token_accuracy": 0.24636783450841904,
|
|
"num_tokens": 13721605.0,
|
|
"step": 6110
|
|
},
|
|
{
|
|
"entropy": 5.305346965789795,
|
|
"epoch": 6.611141157382368,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00019175738804115279,
|
|
"loss": 4.571,
|
|
"mean_token_accuracy": 0.254509425163269,
|
|
"num_tokens": 13732639.0,
|
|
"step": 6115
|
|
},
|
|
{
|
|
"entropy": 5.285580730438232,
|
|
"epoch": 6.616549486208761,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00019135905464768882,
|
|
"loss": 4.5422,
|
|
"mean_token_accuracy": 0.25357137620449066,
|
|
"num_tokens": 13743644.0,
|
|
"step": 6120
|
|
},
|
|
{
|
|
"entropy": 5.262369537353516,
|
|
"epoch": 6.621957815035154,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00019096102520561344,
|
|
"loss": 4.4737,
|
|
"mean_token_accuracy": 0.2676727697253227,
|
|
"num_tokens": 13754887.0,
|
|
"step": 6125
|
|
},
|
|
{
|
|
"entropy": 5.255896806716919,
|
|
"epoch": 6.627366143861547,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00019056330116136666,
|
|
"loss": 4.6468,
|
|
"mean_token_accuracy": 0.25256345719099044,
|
|
"num_tokens": 13766267.0,
|
|
"step": 6130
|
|
},
|
|
{
|
|
"entropy": 5.2915407657623295,
|
|
"epoch": 6.63277447268794,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00019016588396027862,
|
|
"loss": 4.5626,
|
|
"mean_token_accuracy": 0.2577078938484192,
|
|
"num_tokens": 13776532.0,
|
|
"step": 6135
|
|
},
|
|
{
|
|
"entropy": 5.275694704055786,
|
|
"epoch": 6.638182801514332,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00018976877504656438,
|
|
"loss": 4.4941,
|
|
"mean_token_accuracy": 0.256391341984272,
|
|
"num_tokens": 13787511.0,
|
|
"step": 6140
|
|
},
|
|
{
|
|
"entropy": 5.269114828109741,
|
|
"epoch": 6.643591130340725,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001893719758633185,
|
|
"loss": 4.5345,
|
|
"mean_token_accuracy": 0.2575741857290268,
|
|
"num_tokens": 13799264.0,
|
|
"step": 6145
|
|
},
|
|
{
|
|
"entropy": 5.336614894866943,
|
|
"epoch": 6.6489994591671175,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00018897548785251018,
|
|
"loss": 4.6162,
|
|
"mean_token_accuracy": 0.2510768577456474,
|
|
"num_tokens": 13810271.0,
|
|
"step": 6150
|
|
},
|
|
{
|
|
"entropy": 5.267123031616211,
|
|
"epoch": 6.6544077879935095,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0001885793124549779,
|
|
"loss": 4.5449,
|
|
"mean_token_accuracy": 0.26072639375925066,
|
|
"num_tokens": 13820980.0,
|
|
"step": 6155
|
|
},
|
|
{
|
|
"entropy": 5.36530237197876,
|
|
"epoch": 6.659816116819902,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00018818345111042383,
|
|
"loss": 4.7083,
|
|
"mean_token_accuracy": 0.24492036551237106,
|
|
"num_tokens": 13832522.0,
|
|
"step": 6160
|
|
},
|
|
{
|
|
"entropy": 5.270572233200073,
|
|
"epoch": 6.665224445646295,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00018778790525740907,
|
|
"loss": 4.4893,
|
|
"mean_token_accuracy": 0.25949892550706866,
|
|
"num_tokens": 13844453.0,
|
|
"step": 6165
|
|
},
|
|
{
|
|
"entropy": 5.346529150009156,
|
|
"epoch": 6.670632774472688,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00018739267633334806,
|
|
"loss": 4.6312,
|
|
"mean_token_accuracy": 0.2432834178209305,
|
|
"num_tokens": 13856776.0,
|
|
"step": 6170
|
|
},
|
|
{
|
|
"entropy": 5.280401086807251,
|
|
"epoch": 6.676041103299081,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00018699776577450378,
|
|
"loss": 4.5389,
|
|
"mean_token_accuracy": 0.2519763261079788,
|
|
"num_tokens": 13868050.0,
|
|
"step": 6175
|
|
},
|
|
{
|
|
"entropy": 5.316688442230225,
|
|
"epoch": 6.681449432125473,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00018660317501598212,
|
|
"loss": 4.6368,
|
|
"mean_token_accuracy": 0.25246660113334657,
|
|
"num_tokens": 13879866.0,
|
|
"step": 6180
|
|
},
|
|
{
|
|
"entropy": 5.313631153106689,
|
|
"epoch": 6.686857760951866,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0001862089054917266,
|
|
"loss": 4.5359,
|
|
"mean_token_accuracy": 0.25359778553247453,
|
|
"num_tokens": 13891782.0,
|
|
"step": 6185
|
|
},
|
|
{
|
|
"entropy": 5.301898622512818,
|
|
"epoch": 6.692266089778259,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001858149586345137,
|
|
"loss": 4.61,
|
|
"mean_token_accuracy": 0.24771793484687804,
|
|
"num_tokens": 13903587.0,
|
|
"step": 6190
|
|
},
|
|
{
|
|
"entropy": 5.275982904434204,
|
|
"epoch": 6.6976744186046515,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00018542133587594735,
|
|
"loss": 4.5328,
|
|
"mean_token_accuracy": 0.2582537904381752,
|
|
"num_tokens": 13914921.0,
|
|
"step": 6195
|
|
},
|
|
{
|
|
"entropy": 5.345368671417236,
|
|
"epoch": 6.7030827474310435,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00018502803864645334,
|
|
"loss": 4.5087,
|
|
"mean_token_accuracy": 0.2624422237277031,
|
|
"num_tokens": 13925412.0,
|
|
"step": 6200
|
|
},
|
|
{
|
|
"entropy": 5.314154863357544,
|
|
"epoch": 6.708491076257436,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00018463506837527484,
|
|
"loss": 4.5352,
|
|
"mean_token_accuracy": 0.25585331171751025,
|
|
"num_tokens": 13935783.0,
|
|
"step": 6205
|
|
},
|
|
{
|
|
"entropy": 5.223713541030884,
|
|
"epoch": 6.713899405083829,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00018424242649046663,
|
|
"loss": 4.5488,
|
|
"mean_token_accuracy": 0.2652630880475044,
|
|
"num_tokens": 13945233.0,
|
|
"step": 6210
|
|
},
|
|
{
|
|
"entropy": 5.287110948562622,
|
|
"epoch": 6.719307733910222,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00018385011441889037,
|
|
"loss": 4.534,
|
|
"mean_token_accuracy": 0.2620464950799942,
|
|
"num_tokens": 13955411.0,
|
|
"step": 6215
|
|
},
|
|
{
|
|
"entropy": 5.372185945510864,
|
|
"epoch": 6.724716062736614,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00018345813358620894,
|
|
"loss": 4.6738,
|
|
"mean_token_accuracy": 0.24594725221395491,
|
|
"num_tokens": 13966607.0,
|
|
"step": 6220
|
|
},
|
|
{
|
|
"entropy": 5.41127290725708,
|
|
"epoch": 6.730124391563007,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0001830664854168816,
|
|
"loss": 4.6541,
|
|
"mean_token_accuracy": 0.24991250187158584,
|
|
"num_tokens": 13978122.0,
|
|
"step": 6225
|
|
},
|
|
{
|
|
"entropy": 5.294037580490112,
|
|
"epoch": 6.7355327203894,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00018267517133415862,
|
|
"loss": 4.5625,
|
|
"mean_token_accuracy": 0.25710591971874236,
|
|
"num_tokens": 13989301.0,
|
|
"step": 6230
|
|
},
|
|
{
|
|
"entropy": 5.420369052886963,
|
|
"epoch": 6.740941049215793,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00018228419276007646,
|
|
"loss": 4.6588,
|
|
"mean_token_accuracy": 0.2500617653131485,
|
|
"num_tokens": 14000674.0,
|
|
"step": 6235
|
|
},
|
|
{
|
|
"entropy": 5.287535095214844,
|
|
"epoch": 6.746349378042185,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00018189355111545208,
|
|
"loss": 4.5298,
|
|
"mean_token_accuracy": 0.2545910432934761,
|
|
"num_tokens": 14012884.0,
|
|
"step": 6240
|
|
},
|
|
{
|
|
"entropy": 5.258394336700439,
|
|
"epoch": 6.7517577068685775,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001815032478198781,
|
|
"loss": 4.5505,
|
|
"mean_token_accuracy": 0.2501890644431114,
|
|
"num_tokens": 14023624.0,
|
|
"step": 6245
|
|
},
|
|
{
|
|
"entropy": 5.363884782791137,
|
|
"epoch": 6.75716603569497,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00018111328429171747,
|
|
"loss": 4.6186,
|
|
"mean_token_accuracy": 0.24845457822084427,
|
|
"num_tokens": 14034873.0,
|
|
"step": 6250
|
|
},
|
|
{
|
|
"entropy": 5.322827386856079,
|
|
"epoch": 6.762574364521363,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00018072366194809886,
|
|
"loss": 4.5171,
|
|
"mean_token_accuracy": 0.2651121810078621,
|
|
"num_tokens": 14045707.0,
|
|
"step": 6255
|
|
},
|
|
{
|
|
"entropy": 5.300434160232544,
|
|
"epoch": 6.767982693347756,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00018033438220491055,
|
|
"loss": 4.578,
|
|
"mean_token_accuracy": 0.25883566588163376,
|
|
"num_tokens": 14056774.0,
|
|
"step": 6260
|
|
},
|
|
{
|
|
"entropy": 5.233197164535523,
|
|
"epoch": 6.773391022174148,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.000179945446476796,
|
|
"loss": 4.4646,
|
|
"mean_token_accuracy": 0.2604628339409828,
|
|
"num_tokens": 14067358.0,
|
|
"step": 6265
|
|
},
|
|
{
|
|
"entropy": 5.26256046295166,
|
|
"epoch": 6.778799351000541,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00017955685617714867,
|
|
"loss": 4.5451,
|
|
"mean_token_accuracy": 0.2557960584759712,
|
|
"num_tokens": 14078336.0,
|
|
"step": 6270
|
|
},
|
|
{
|
|
"entropy": 5.358508682250976,
|
|
"epoch": 6.784207679826934,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001791686127181066,
|
|
"loss": 4.6984,
|
|
"mean_token_accuracy": 0.24683898389339448,
|
|
"num_tokens": 14091700.0,
|
|
"step": 6275
|
|
},
|
|
{
|
|
"entropy": 5.334011697769165,
|
|
"epoch": 6.789616008653326,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00017878071751054736,
|
|
"loss": 4.5772,
|
|
"mean_token_accuracy": 0.2534466966986656,
|
|
"num_tokens": 14103322.0,
|
|
"step": 6280
|
|
},
|
|
{
|
|
"entropy": 5.38698000907898,
|
|
"epoch": 6.795024337479719,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00017839317196408305,
|
|
"loss": 4.6312,
|
|
"mean_token_accuracy": 0.2574057936668396,
|
|
"num_tokens": 14114758.0,
|
|
"step": 6285
|
|
},
|
|
{
|
|
"entropy": 5.1754943370819095,
|
|
"epoch": 6.800432666306111,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00017800597748705494,
|
|
"loss": 4.4573,
|
|
"mean_token_accuracy": 0.26355628967285155,
|
|
"num_tokens": 14124935.0,
|
|
"step": 6290
|
|
},
|
|
{
|
|
"entropy": 5.256164407730102,
|
|
"epoch": 6.805840995132504,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001776191354865289,
|
|
"loss": 4.5197,
|
|
"mean_token_accuracy": 0.26415102928876877,
|
|
"num_tokens": 14136102.0,
|
|
"step": 6295
|
|
},
|
|
{
|
|
"entropy": 5.313513135910034,
|
|
"epoch": 6.811249323958897,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00017723264736828952,
|
|
"loss": 4.5294,
|
|
"mean_token_accuracy": 0.26630495935678483,
|
|
"num_tokens": 14147042.0,
|
|
"step": 6300
|
|
},
|
|
{
|
|
"entropy": 5.21271915435791,
|
|
"epoch": 6.816657652785289,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00017684651453683554,
|
|
"loss": 4.4673,
|
|
"mean_token_accuracy": 0.26117067188024523,
|
|
"num_tokens": 14158045.0,
|
|
"step": 6305
|
|
},
|
|
{
|
|
"entropy": 5.261816072463989,
|
|
"epoch": 6.822065981611682,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00017646073839537458,
|
|
"loss": 4.5802,
|
|
"mean_token_accuracy": 0.2506229683756828,
|
|
"num_tokens": 14169815.0,
|
|
"step": 6310
|
|
},
|
|
{
|
|
"entropy": 5.23839545249939,
|
|
"epoch": 6.827474310438075,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00017607532034581797,
|
|
"loss": 4.5359,
|
|
"mean_token_accuracy": 0.252497473359108,
|
|
"num_tokens": 14179965.0,
|
|
"step": 6315
|
|
},
|
|
{
|
|
"entropy": 5.35866527557373,
|
|
"epoch": 6.832882639264467,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00017569026178877584,
|
|
"loss": 4.6189,
|
|
"mean_token_accuracy": 0.25480786710977554,
|
|
"num_tokens": 14191142.0,
|
|
"step": 6320
|
|
},
|
|
{
|
|
"entropy": 5.324539232254028,
|
|
"epoch": 6.83829096809086,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001753055641235519,
|
|
"loss": 4.5238,
|
|
"mean_token_accuracy": 0.2647758945822716,
|
|
"num_tokens": 14201765.0,
|
|
"step": 6325
|
|
},
|
|
{
|
|
"entropy": 5.287074995040894,
|
|
"epoch": 6.8436992969172525,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00017492122874813824,
|
|
"loss": 4.5552,
|
|
"mean_token_accuracy": 0.25649518221616746,
|
|
"num_tokens": 14211995.0,
|
|
"step": 6330
|
|
},
|
|
{
|
|
"entropy": 5.3656528949737545,
|
|
"epoch": 6.849107625743645,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001745372570592107,
|
|
"loss": 4.7022,
|
|
"mean_token_accuracy": 0.237104232609272,
|
|
"num_tokens": 14225488.0,
|
|
"step": 6335
|
|
},
|
|
{
|
|
"entropy": 5.30509843826294,
|
|
"epoch": 6.854515954570038,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00017415365045212312,
|
|
"loss": 4.5316,
|
|
"mean_token_accuracy": 0.25440850108861923,
|
|
"num_tokens": 14237459.0,
|
|
"step": 6340
|
|
},
|
|
{
|
|
"entropy": 5.3194056987762455,
|
|
"epoch": 6.85992428339643,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00017377041032090278,
|
|
"loss": 4.5925,
|
|
"mean_token_accuracy": 0.24982351958751678,
|
|
"num_tokens": 14248136.0,
|
|
"step": 6345
|
|
},
|
|
{
|
|
"entropy": 5.2875903129577635,
|
|
"epoch": 6.865332612222823,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00017338753805824536,
|
|
"loss": 4.6165,
|
|
"mean_token_accuracy": 0.24947860985994338,
|
|
"num_tokens": 14260239.0,
|
|
"step": 6350
|
|
},
|
|
{
|
|
"entropy": 5.242704629898071,
|
|
"epoch": 6.870740941049216,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00017300503505550939,
|
|
"loss": 4.5193,
|
|
"mean_token_accuracy": 0.25569299757480624,
|
|
"num_tokens": 14271825.0,
|
|
"step": 6355
|
|
},
|
|
{
|
|
"entropy": 5.218107748031616,
|
|
"epoch": 6.876149269875609,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00017262290270271175,
|
|
"loss": 4.437,
|
|
"mean_token_accuracy": 0.27283566296100614,
|
|
"num_tokens": 14282517.0,
|
|
"step": 6360
|
|
},
|
|
{
|
|
"entropy": 5.362441921234131,
|
|
"epoch": 6.881557598702001,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00017224114238852234,
|
|
"loss": 4.6174,
|
|
"mean_token_accuracy": 0.25439168959856034,
|
|
"num_tokens": 14293548.0,
|
|
"step": 6365
|
|
},
|
|
{
|
|
"entropy": 5.310395526885986,
|
|
"epoch": 6.886965927528394,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001718597555002588,
|
|
"loss": 4.6001,
|
|
"mean_token_accuracy": 0.24933263957500457,
|
|
"num_tokens": 14305782.0,
|
|
"step": 6370
|
|
},
|
|
{
|
|
"entropy": 5.251075267791748,
|
|
"epoch": 6.8923742563547865,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00017147874342388227,
|
|
"loss": 4.5173,
|
|
"mean_token_accuracy": 0.2571871355175972,
|
|
"num_tokens": 14315968.0,
|
|
"step": 6375
|
|
},
|
|
{
|
|
"entropy": 5.303053379058838,
|
|
"epoch": 6.897782585181179,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0001710981075439913,
|
|
"loss": 4.55,
|
|
"mean_token_accuracy": 0.25021299421787263,
|
|
"num_tokens": 14327579.0,
|
|
"step": 6380
|
|
},
|
|
{
|
|
"entropy": 5.3887430191040036,
|
|
"epoch": 6.903190914007571,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00017071784924381766,
|
|
"loss": 4.6147,
|
|
"mean_token_accuracy": 0.25370681285858154,
|
|
"num_tokens": 14338620.0,
|
|
"step": 6385
|
|
},
|
|
{
|
|
"entropy": 5.288820552825928,
|
|
"epoch": 6.908599242833964,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00017033796990522081,
|
|
"loss": 4.5288,
|
|
"mean_token_accuracy": 0.2533402979373932,
|
|
"num_tokens": 14348638.0,
|
|
"step": 6390
|
|
},
|
|
{
|
|
"entropy": 5.264960098266601,
|
|
"epoch": 6.914007571660357,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00016995847090868322,
|
|
"loss": 4.571,
|
|
"mean_token_accuracy": 0.2598321259021759,
|
|
"num_tokens": 14359253.0,
|
|
"step": 6395
|
|
},
|
|
{
|
|
"entropy": 5.2884820938110355,
|
|
"epoch": 6.91941590048675,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00016957935363330506,
|
|
"loss": 4.5481,
|
|
"mean_token_accuracy": 0.25481689870357516,
|
|
"num_tokens": 14370659.0,
|
|
"step": 6400
|
|
},
|
|
{
|
|
"entropy": 5.303628444671631,
|
|
"epoch": 6.924824229313142,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00016920061945679951,
|
|
"loss": 4.593,
|
|
"mean_token_accuracy": 0.2431500717997551,
|
|
"num_tokens": 14381136.0,
|
|
"step": 6405
|
|
},
|
|
{
|
|
"entropy": 5.313886547088623,
|
|
"epoch": 6.930232558139535,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001688222697554872,
|
|
"loss": 4.5992,
|
|
"mean_token_accuracy": 0.24208810776472092,
|
|
"num_tokens": 14392789.0,
|
|
"step": 6410
|
|
},
|
|
{
|
|
"entropy": 5.29232931137085,
|
|
"epoch": 6.935640886965928,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000168444305904292,
|
|
"loss": 4.5152,
|
|
"mean_token_accuracy": 0.2591824933886528,
|
|
"num_tokens": 14402808.0,
|
|
"step": 6415
|
|
},
|
|
{
|
|
"entropy": 5.329613304138183,
|
|
"epoch": 6.9410492157923205,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00016806672927673552,
|
|
"loss": 4.6154,
|
|
"mean_token_accuracy": 0.2500679537653923,
|
|
"num_tokens": 14413961.0,
|
|
"step": 6420
|
|
},
|
|
{
|
|
"entropy": 5.270015859603882,
|
|
"epoch": 6.9464575446187125,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00016768954124493188,
|
|
"loss": 4.5184,
|
|
"mean_token_accuracy": 0.2611670717597008,
|
|
"num_tokens": 14424416.0,
|
|
"step": 6425
|
|
},
|
|
{
|
|
"entropy": 5.359933090209961,
|
|
"epoch": 6.951865873445105,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00016731274317958333,
|
|
"loss": 4.6579,
|
|
"mean_token_accuracy": 0.25377337336540223,
|
|
"num_tokens": 14435771.0,
|
|
"step": 6430
|
|
},
|
|
{
|
|
"entropy": 5.260810852050781,
|
|
"epoch": 6.957274202271498,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0001669363364499749,
|
|
"loss": 4.516,
|
|
"mean_token_accuracy": 0.2577952012419701,
|
|
"num_tokens": 14447106.0,
|
|
"step": 6435
|
|
},
|
|
{
|
|
"entropy": 5.2826904296875,
|
|
"epoch": 6.962682531097891,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00016656032242396958,
|
|
"loss": 4.576,
|
|
"mean_token_accuracy": 0.25429315865039825,
|
|
"num_tokens": 14458698.0,
|
|
"step": 6440
|
|
},
|
|
{
|
|
"entropy": 5.245286178588867,
|
|
"epoch": 6.968090859924283,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00016618470246800322,
|
|
"loss": 4.5301,
|
|
"mean_token_accuracy": 0.25542558133602145,
|
|
"num_tokens": 14469440.0,
|
|
"step": 6445
|
|
},
|
|
{
|
|
"entropy": 5.3228531837463375,
|
|
"epoch": 6.973499188750676,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00016580947794707937,
|
|
"loss": 4.5714,
|
|
"mean_token_accuracy": 0.26038903146982195,
|
|
"num_tokens": 14480256.0,
|
|
"step": 6450
|
|
},
|
|
{
|
|
"entropy": 5.294363594055175,
|
|
"epoch": 6.978907517577069,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00016543465022476512,
|
|
"loss": 4.6286,
|
|
"mean_token_accuracy": 0.24763010442256927,
|
|
"num_tokens": 14492347.0,
|
|
"step": 6455
|
|
},
|
|
{
|
|
"entropy": 5.279555368423462,
|
|
"epoch": 6.984315846403462,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0001650602206631852,
|
|
"loss": 4.5608,
|
|
"mean_token_accuracy": 0.25490829944610593,
|
|
"num_tokens": 14503492.0,
|
|
"step": 6460
|
|
},
|
|
{
|
|
"entropy": 5.356185722351074,
|
|
"epoch": 6.9897241752298545,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00016468619062301739,
|
|
"loss": 4.6848,
|
|
"mean_token_accuracy": 0.24446437805891036,
|
|
"num_tokens": 14515843.0,
|
|
"step": 6465
|
|
},
|
|
{
|
|
"entropy": 5.2494855403900145,
|
|
"epoch": 6.995132504056246,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001643125614634877,
|
|
"loss": 4.5414,
|
|
"mean_token_accuracy": 0.25641597509384156,
|
|
"num_tokens": 14526733.0,
|
|
"step": 6470
|
|
},
|
|
{
|
|
"entropy": 5.309608353508843,
|
|
"epoch": 7.0,
|
|
"grad_norm": 2.734375,
|
|
"learning_rate": 0.0001639393345423656,
|
|
"loss": 4.6709,
|
|
"mean_token_accuracy": 0.24487148059739006,
|
|
"num_tokens": 14538223.0,
|
|
"step": 6475
|
|
},
|
|
{
|
|
"entropy": 5.2985520362854,
|
|
"epoch": 7.005408328826393,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00016356651121595832,
|
|
"loss": 4.4382,
|
|
"mean_token_accuracy": 0.2609593763947487,
|
|
"num_tokens": 14549388.0,
|
|
"step": 6480
|
|
},
|
|
{
|
|
"entropy": 5.318699789047241,
|
|
"epoch": 7.010816657652786,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001631940928391068,
|
|
"loss": 4.4704,
|
|
"mean_token_accuracy": 0.267617829144001,
|
|
"num_tokens": 14561300.0,
|
|
"step": 6485
|
|
},
|
|
{
|
|
"entropy": 5.2496805667877195,
|
|
"epoch": 7.016224986479178,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00016282208076518032,
|
|
"loss": 4.3971,
|
|
"mean_token_accuracy": 0.2688437163829803,
|
|
"num_tokens": 14573169.0,
|
|
"step": 6490
|
|
},
|
|
{
|
|
"entropy": 5.273196411132813,
|
|
"epoch": 7.0216333153055706,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00016245047634607155,
|
|
"loss": 4.4932,
|
|
"mean_token_accuracy": 0.25926129072904586,
|
|
"num_tokens": 14585035.0,
|
|
"step": 6495
|
|
},
|
|
{
|
|
"entropy": 5.187214756011963,
|
|
"epoch": 7.027041644131963,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0001620792809321919,
|
|
"loss": 4.4095,
|
|
"mean_token_accuracy": 0.2736878976225853,
|
|
"num_tokens": 14596184.0,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"epoch": 7.027041644131963,
|
|
"eval_entropy": 5.189655908671292,
|
|
"eval_loss": 5.865362167358398,
|
|
"eval_mean_token_accuracy": 0.18087120874022897,
|
|
"eval_num_tokens": 14596184.0,
|
|
"eval_runtime": 1.9346,
|
|
"eval_samples_per_second": 1816.872,
|
|
"eval_steps_per_second": 227.432,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"entropy": 5.266715431213379,
|
|
"epoch": 7.032449972958356,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00016170849587246623,
|
|
"loss": 4.4106,
|
|
"mean_token_accuracy": 0.27497953474521636,
|
|
"num_tokens": 14606987.0,
|
|
"step": 6505
|
|
},
|
|
{
|
|
"entropy": 5.217801427841186,
|
|
"epoch": 7.037858301784748,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00016133812251432826,
|
|
"loss": 4.4092,
|
|
"mean_token_accuracy": 0.2655544489622116,
|
|
"num_tokens": 14617600.0,
|
|
"step": 6510
|
|
},
|
|
{
|
|
"entropy": 5.391884279251099,
|
|
"epoch": 7.043266630611141,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0001609681622037159,
|
|
"loss": 4.6012,
|
|
"mean_token_accuracy": 0.2641146957874298,
|
|
"num_tokens": 14629551.0,
|
|
"step": 6515
|
|
},
|
|
{
|
|
"entropy": 5.3228614807128904,
|
|
"epoch": 7.048674959437534,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00016059861628506557,
|
|
"loss": 4.5102,
|
|
"mean_token_accuracy": 0.26080356538295746,
|
|
"num_tokens": 14641322.0,
|
|
"step": 6520
|
|
},
|
|
{
|
|
"entropy": 5.260049104690552,
|
|
"epoch": 7.054083288263927,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00016022948610130804,
|
|
"loss": 4.4303,
|
|
"mean_token_accuracy": 0.27646683156490326,
|
|
"num_tokens": 14653380.0,
|
|
"step": 6525
|
|
},
|
|
{
|
|
"entropy": 5.274545335769654,
|
|
"epoch": 7.059491617090319,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0001598607729938632,
|
|
"loss": 4.4216,
|
|
"mean_token_accuracy": 0.2758495047688484,
|
|
"num_tokens": 14664043.0,
|
|
"step": 6530
|
|
},
|
|
{
|
|
"entropy": 5.177192497253418,
|
|
"epoch": 7.064899945916712,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00015949247830263538,
|
|
"loss": 4.3567,
|
|
"mean_token_accuracy": 0.2732002645730972,
|
|
"num_tokens": 14675117.0,
|
|
"step": 6535
|
|
},
|
|
{
|
|
"entropy": 5.221408557891846,
|
|
"epoch": 7.0703082747431045,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00015912460336600838,
|
|
"loss": 4.4076,
|
|
"mean_token_accuracy": 0.2736793726682663,
|
|
"num_tokens": 14686908.0,
|
|
"step": 6540
|
|
},
|
|
{
|
|
"entropy": 5.242983293533325,
|
|
"epoch": 7.075716603569497,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00015875714952084054,
|
|
"loss": 4.408,
|
|
"mean_token_accuracy": 0.2641566827893257,
|
|
"num_tokens": 14698997.0,
|
|
"step": 6545
|
|
},
|
|
{
|
|
"entropy": 5.297539854049683,
|
|
"epoch": 7.081124932395889,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00015839011810245984,
|
|
"loss": 4.5191,
|
|
"mean_token_accuracy": 0.26138575077056886,
|
|
"num_tokens": 14709586.0,
|
|
"step": 6550
|
|
},
|
|
{
|
|
"entropy": 5.402618503570556,
|
|
"epoch": 7.086533261222282,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001580235104446596,
|
|
"loss": 4.538,
|
|
"mean_token_accuracy": 0.2686152383685112,
|
|
"num_tokens": 14720821.0,
|
|
"step": 6555
|
|
},
|
|
{
|
|
"entropy": 5.35269136428833,
|
|
"epoch": 7.091941590048675,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00015765732787969267,
|
|
"loss": 4.5534,
|
|
"mean_token_accuracy": 0.25714511424303055,
|
|
"num_tokens": 14732218.0,
|
|
"step": 6560
|
|
},
|
|
{
|
|
"entropy": 5.317057752609253,
|
|
"epoch": 7.097349918875068,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00015729157173826724,
|
|
"loss": 4.5026,
|
|
"mean_token_accuracy": 0.2648747518658638,
|
|
"num_tokens": 14742306.0,
|
|
"step": 6565
|
|
},
|
|
{
|
|
"entropy": 5.269532346725464,
|
|
"epoch": 7.10275824770146,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00015692624334954236,
|
|
"loss": 4.4238,
|
|
"mean_token_accuracy": 0.27280645817518234,
|
|
"num_tokens": 14752924.0,
|
|
"step": 6570
|
|
},
|
|
{
|
|
"entropy": 5.333494710922241,
|
|
"epoch": 7.108166576527853,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001565613440411219,
|
|
"loss": 4.4649,
|
|
"mean_token_accuracy": 0.26433314830064775,
|
|
"num_tokens": 14764161.0,
|
|
"step": 6575
|
|
},
|
|
{
|
|
"entropy": 5.259645795822143,
|
|
"epoch": 7.113574905354246,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0001561968751390509,
|
|
"loss": 4.4317,
|
|
"mean_token_accuracy": 0.2752742484211922,
|
|
"num_tokens": 14774520.0,
|
|
"step": 6580
|
|
},
|
|
{
|
|
"entropy": 5.257570123672485,
|
|
"epoch": 7.1189832341806385,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00015583283796781022,
|
|
"loss": 4.4377,
|
|
"mean_token_accuracy": 0.2647978216409683,
|
|
"num_tokens": 14785243.0,
|
|
"step": 6585
|
|
},
|
|
{
|
|
"entropy": 5.320237159729004,
|
|
"epoch": 7.1243915630070305,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00015546923385031169,
|
|
"loss": 4.4816,
|
|
"mean_token_accuracy": 0.2674393802881241,
|
|
"num_tokens": 14796152.0,
|
|
"step": 6590
|
|
},
|
|
{
|
|
"entropy": 5.2931108474731445,
|
|
"epoch": 7.129799891833423,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00015510606410789368,
|
|
"loss": 4.435,
|
|
"mean_token_accuracy": 0.2677324801683426,
|
|
"num_tokens": 14807849.0,
|
|
"step": 6595
|
|
},
|
|
{
|
|
"entropy": 5.250351858139038,
|
|
"epoch": 7.135208220659816,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001547433300603157,
|
|
"loss": 4.4858,
|
|
"mean_token_accuracy": 0.2633852496743202,
|
|
"num_tokens": 14819402.0,
|
|
"step": 6600
|
|
},
|
|
{
|
|
"entropy": 5.3230983257293705,
|
|
"epoch": 7.140616549486209,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00015438103302575413,
|
|
"loss": 4.5342,
|
|
"mean_token_accuracy": 0.25321471095085146,
|
|
"num_tokens": 14831801.0,
|
|
"step": 6605
|
|
},
|
|
{
|
|
"entropy": 5.268402242660523,
|
|
"epoch": 7.146024878312601,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001540191743207974,
|
|
"loss": 4.4263,
|
|
"mean_token_accuracy": 0.2719721645116806,
|
|
"num_tokens": 14842758.0,
|
|
"step": 6610
|
|
},
|
|
{
|
|
"entropy": 5.24076476097107,
|
|
"epoch": 7.151433207138994,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0001536577552604407,
|
|
"loss": 4.4311,
|
|
"mean_token_accuracy": 0.26891272217035295,
|
|
"num_tokens": 14853493.0,
|
|
"step": 6615
|
|
},
|
|
{
|
|
"entropy": 5.282127714157104,
|
|
"epoch": 7.156841535965387,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00015329677715808176,
|
|
"loss": 4.4302,
|
|
"mean_token_accuracy": 0.26485222280025483,
|
|
"num_tokens": 14863561.0,
|
|
"step": 6620
|
|
},
|
|
{
|
|
"entropy": 5.249979448318482,
|
|
"epoch": 7.16224986479178,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00015293624132551591,
|
|
"loss": 4.4502,
|
|
"mean_token_accuracy": 0.26533009558916093,
|
|
"num_tokens": 14874501.0,
|
|
"step": 6625
|
|
},
|
|
{
|
|
"entropy": 5.193007755279541,
|
|
"epoch": 7.167658193618172,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0001525761490729312,
|
|
"loss": 4.3466,
|
|
"mean_token_accuracy": 0.26985053420066835,
|
|
"num_tokens": 14885457.0,
|
|
"step": 6630
|
|
},
|
|
{
|
|
"entropy": 5.179992246627807,
|
|
"epoch": 7.173066522444564,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0001522165017089036,
|
|
"loss": 4.3717,
|
|
"mean_token_accuracy": 0.27363738119602204,
|
|
"num_tokens": 14895812.0,
|
|
"step": 6635
|
|
},
|
|
{
|
|
"entropy": 5.267845964431762,
|
|
"epoch": 7.178474851270957,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00015185730054039266,
|
|
"loss": 4.3855,
|
|
"mean_token_accuracy": 0.2817357927560806,
|
|
"num_tokens": 14907367.0,
|
|
"step": 6640
|
|
},
|
|
{
|
|
"entropy": 5.195575523376465,
|
|
"epoch": 7.18388318009735,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00015149854687273594,
|
|
"loss": 4.4024,
|
|
"mean_token_accuracy": 0.2697136580944061,
|
|
"num_tokens": 14918931.0,
|
|
"step": 6645
|
|
},
|
|
{
|
|
"entropy": 5.25063533782959,
|
|
"epoch": 7.189291508923742,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.0001511402420096455,
|
|
"loss": 4.4453,
|
|
"mean_token_accuracy": 0.27210221588611605,
|
|
"num_tokens": 14929518.0,
|
|
"step": 6650
|
|
},
|
|
{
|
|
"entropy": 5.208005428314209,
|
|
"epoch": 7.194699837750135,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001507823872532017,
|
|
"loss": 4.3853,
|
|
"mean_token_accuracy": 0.27794582694768905,
|
|
"num_tokens": 14940249.0,
|
|
"step": 6655
|
|
},
|
|
{
|
|
"entropy": 5.311082792282105,
|
|
"epoch": 7.200108166576528,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00015042498390384966,
|
|
"loss": 4.4062,
|
|
"mean_token_accuracy": 0.2760100021958351,
|
|
"num_tokens": 14950645.0,
|
|
"step": 6660
|
|
},
|
|
{
|
|
"entropy": 5.311803197860717,
|
|
"epoch": 7.205516495402921,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00015006803326039403,
|
|
"loss": 4.4312,
|
|
"mean_token_accuracy": 0.27072658240795133,
|
|
"num_tokens": 14961687.0,
|
|
"step": 6665
|
|
},
|
|
{
|
|
"entropy": 5.2980845928192135,
|
|
"epoch": 7.210924824229313,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00014971153661999425,
|
|
"loss": 4.4641,
|
|
"mean_token_accuracy": 0.26890019327402115,
|
|
"num_tokens": 14973119.0,
|
|
"step": 6670
|
|
},
|
|
{
|
|
"entropy": 5.285019636154175,
|
|
"epoch": 7.2163331530557056,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0001493554952781599,
|
|
"loss": 4.48,
|
|
"mean_token_accuracy": 0.26675955057144163,
|
|
"num_tokens": 14984826.0,
|
|
"step": 6675
|
|
},
|
|
{
|
|
"entropy": 5.241651487350464,
|
|
"epoch": 7.221741481882098,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00014899991052874618,
|
|
"loss": 4.3867,
|
|
"mean_token_accuracy": 0.2717937245965004,
|
|
"num_tokens": 14996853.0,
|
|
"step": 6680
|
|
},
|
|
{
|
|
"entropy": 5.3143280982971195,
|
|
"epoch": 7.227149810708491,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00014864478366394853,
|
|
"loss": 4.4662,
|
|
"mean_token_accuracy": 0.27232622653245925,
|
|
"num_tokens": 15006941.0,
|
|
"step": 6685
|
|
},
|
|
{
|
|
"entropy": 5.2965874671936035,
|
|
"epoch": 7.232558139534884,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001482901159742992,
|
|
"loss": 4.5063,
|
|
"mean_token_accuracy": 0.26382807791233065,
|
|
"num_tokens": 15018477.0,
|
|
"step": 6690
|
|
},
|
|
{
|
|
"entropy": 5.317858171463013,
|
|
"epoch": 7.237966468361276,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001479359087486613,
|
|
"loss": 4.479,
|
|
"mean_token_accuracy": 0.26958292424678804,
|
|
"num_tokens": 15030454.0,
|
|
"step": 6695
|
|
},
|
|
{
|
|
"entropy": 5.260033035278321,
|
|
"epoch": 7.243374797187669,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00014758216327422457,
|
|
"loss": 4.4216,
|
|
"mean_token_accuracy": 0.27488421499729154,
|
|
"num_tokens": 15040662.0,
|
|
"step": 6700
|
|
},
|
|
{
|
|
"entropy": 5.250268363952637,
|
|
"epoch": 7.248783126014062,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00014722888083650094,
|
|
"loss": 4.4934,
|
|
"mean_token_accuracy": 0.26202374547719953,
|
|
"num_tokens": 15051414.0,
|
|
"step": 6705
|
|
},
|
|
{
|
|
"entropy": 5.233264303207397,
|
|
"epoch": 7.254191454840455,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00014687606271931961,
|
|
"loss": 4.4366,
|
|
"mean_token_accuracy": 0.26409810185432436,
|
|
"num_tokens": 15062563.0,
|
|
"step": 6710
|
|
},
|
|
{
|
|
"entropy": 5.273464727401733,
|
|
"epoch": 7.259599783666847,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00014652371020482246,
|
|
"loss": 4.5254,
|
|
"mean_token_accuracy": 0.26107025146484375,
|
|
"num_tokens": 15074443.0,
|
|
"step": 6715
|
|
},
|
|
{
|
|
"entropy": 5.2914787292480465,
|
|
"epoch": 7.2650081124932395,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0001461718245734594,
|
|
"loss": 4.4673,
|
|
"mean_token_accuracy": 0.26807126253843305,
|
|
"num_tokens": 15085920.0,
|
|
"step": 6720
|
|
},
|
|
{
|
|
"entropy": 5.265882682800293,
|
|
"epoch": 7.270416441319632,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0001458204071039834,
|
|
"loss": 4.3968,
|
|
"mean_token_accuracy": 0.27347314804792405,
|
|
"num_tokens": 15096851.0,
|
|
"step": 6725
|
|
},
|
|
{
|
|
"entropy": 5.250567054748535,
|
|
"epoch": 7.275824770146025,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0001454694590734466,
|
|
"loss": 4.3424,
|
|
"mean_token_accuracy": 0.28372384905815123,
|
|
"num_tokens": 15107217.0,
|
|
"step": 6730
|
|
},
|
|
{
|
|
"entropy": 5.233177900314331,
|
|
"epoch": 7.281233098972417,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0001451189817571949,
|
|
"loss": 4.4114,
|
|
"mean_token_accuracy": 0.27561776638031005,
|
|
"num_tokens": 15117464.0,
|
|
"step": 6735
|
|
},
|
|
{
|
|
"entropy": 5.2102500915527346,
|
|
"epoch": 7.28664142779881,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00014476897642886355,
|
|
"loss": 4.4105,
|
|
"mean_token_accuracy": 0.26500155925750735,
|
|
"num_tokens": 15127801.0,
|
|
"step": 6740
|
|
},
|
|
{
|
|
"entropy": 5.353866100311279,
|
|
"epoch": 7.292049756625203,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00014441944436037287,
|
|
"loss": 4.5991,
|
|
"mean_token_accuracy": 0.2600067913532257,
|
|
"num_tokens": 15139117.0,
|
|
"step": 6745
|
|
},
|
|
{
|
|
"entropy": 5.248448514938355,
|
|
"epoch": 7.297458085451596,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00014407038682192312,
|
|
"loss": 4.4641,
|
|
"mean_token_accuracy": 0.25678576678037646,
|
|
"num_tokens": 15150205.0,
|
|
"step": 6750
|
|
},
|
|
{
|
|
"entropy": 5.331437826156616,
|
|
"epoch": 7.302866414277988,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0001437218050819902,
|
|
"loss": 4.4861,
|
|
"mean_token_accuracy": 0.26236632466316223,
|
|
"num_tokens": 15160813.0,
|
|
"step": 6755
|
|
},
|
|
{
|
|
"entropy": 5.222689580917359,
|
|
"epoch": 7.308274743104381,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.000143373700407321,
|
|
"loss": 4.453,
|
|
"mean_token_accuracy": 0.26178296506404874,
|
|
"num_tokens": 15171609.0,
|
|
"step": 6760
|
|
},
|
|
{
|
|
"entropy": 5.256084823608399,
|
|
"epoch": 7.3136830719307735,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00014302607406292873,
|
|
"loss": 4.4065,
|
|
"mean_token_accuracy": 0.2691895917057991,
|
|
"num_tokens": 15183030.0,
|
|
"step": 6765
|
|
},
|
|
{
|
|
"entropy": 5.292045545578003,
|
|
"epoch": 7.319091400757166,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.00014267892731208826,
|
|
"loss": 4.4934,
|
|
"mean_token_accuracy": 0.2661156073212624,
|
|
"num_tokens": 15193365.0,
|
|
"step": 6770
|
|
},
|
|
{
|
|
"entropy": 5.222332572937011,
|
|
"epoch": 7.324499729583558,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0001423322614163318,
|
|
"loss": 4.4423,
|
|
"mean_token_accuracy": 0.26444998383522034,
|
|
"num_tokens": 15205075.0,
|
|
"step": 6775
|
|
},
|
|
{
|
|
"entropy": 5.279045438766479,
|
|
"epoch": 7.329908058409951,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00014198607763544392,
|
|
"loss": 4.4855,
|
|
"mean_token_accuracy": 0.26089412420988084,
|
|
"num_tokens": 15216655.0,
|
|
"step": 6780
|
|
},
|
|
{
|
|
"entropy": 5.238213157653808,
|
|
"epoch": 7.335316387236344,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00014164037722745715,
|
|
"loss": 4.4565,
|
|
"mean_token_accuracy": 0.2755573257803917,
|
|
"num_tokens": 15228130.0,
|
|
"step": 6785
|
|
},
|
|
{
|
|
"entropy": 5.246974849700928,
|
|
"epoch": 7.340724716062737,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0001412951614486479,
|
|
"loss": 4.3874,
|
|
"mean_token_accuracy": 0.2681003510951996,
|
|
"num_tokens": 15239019.0,
|
|
"step": 6790
|
|
},
|
|
{
|
|
"entropy": 5.313682508468628,
|
|
"epoch": 7.346133044889129,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00014095043155353085,
|
|
"loss": 4.4707,
|
|
"mean_token_accuracy": 0.2685552045702934,
|
|
"num_tokens": 15250325.0,
|
|
"step": 6795
|
|
},
|
|
{
|
|
"entropy": 5.273488330841064,
|
|
"epoch": 7.351541373715522,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.0001406061887948553,
|
|
"loss": 4.5299,
|
|
"mean_token_accuracy": 0.25533734560012816,
|
|
"num_tokens": 15261670.0,
|
|
"step": 6800
|
|
},
|
|
{
|
|
"entropy": 5.350579786300659,
|
|
"epoch": 7.356949702541915,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00014026243442360025,
|
|
"loss": 4.5157,
|
|
"mean_token_accuracy": 0.2633839249610901,
|
|
"num_tokens": 15273660.0,
|
|
"step": 6805
|
|
},
|
|
{
|
|
"entropy": 5.255425548553466,
|
|
"epoch": 7.3623580313683075,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001399191696889699,
|
|
"loss": 4.3872,
|
|
"mean_token_accuracy": 0.2716802015900612,
|
|
"num_tokens": 15284593.0,
|
|
"step": 6810
|
|
},
|
|
{
|
|
"entropy": 5.2395524978637695,
|
|
"epoch": 7.367766360194699,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0001395763958383891,
|
|
"loss": 4.4985,
|
|
"mean_token_accuracy": 0.25602276176214217,
|
|
"num_tokens": 15295860.0,
|
|
"step": 6815
|
|
},
|
|
{
|
|
"entropy": 5.2247922897338865,
|
|
"epoch": 7.373174689021092,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0001392341141174987,
|
|
"loss": 4.4366,
|
|
"mean_token_accuracy": 0.2658582925796509,
|
|
"num_tokens": 15306718.0,
|
|
"step": 6820
|
|
},
|
|
{
|
|
"entropy": 5.287804651260376,
|
|
"epoch": 7.378583017847485,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00013889232577015125,
|
|
"loss": 4.4061,
|
|
"mean_token_accuracy": 0.27729910612106323,
|
|
"num_tokens": 15318213.0,
|
|
"step": 6825
|
|
},
|
|
{
|
|
"entropy": 5.253314352035522,
|
|
"epoch": 7.383991346673878,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.00013855103203840673,
|
|
"loss": 4.4581,
|
|
"mean_token_accuracy": 0.26829472482204436,
|
|
"num_tokens": 15328932.0,
|
|
"step": 6830
|
|
},
|
|
{
|
|
"entropy": 5.265571594238281,
|
|
"epoch": 7.38939967550027,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00013821023416252693,
|
|
"loss": 4.4683,
|
|
"mean_token_accuracy": 0.27067202925682066,
|
|
"num_tokens": 15340370.0,
|
|
"step": 6835
|
|
},
|
|
{
|
|
"entropy": 5.236275100708008,
|
|
"epoch": 7.394808004326663,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00013786993338097236,
|
|
"loss": 4.4934,
|
|
"mean_token_accuracy": 0.26339503973722456,
|
|
"num_tokens": 15351567.0,
|
|
"step": 6840
|
|
},
|
|
{
|
|
"entropy": 5.251045656204224,
|
|
"epoch": 7.400216333153056,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00013753013093039673,
|
|
"loss": 4.4299,
|
|
"mean_token_accuracy": 0.27526747435331345,
|
|
"num_tokens": 15361945.0,
|
|
"step": 6845
|
|
},
|
|
{
|
|
"entropy": 5.1956963539123535,
|
|
"epoch": 7.405624661979449,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000137190828045643,
|
|
"loss": 4.3612,
|
|
"mean_token_accuracy": 0.2704841122031212,
|
|
"num_tokens": 15374267.0,
|
|
"step": 6850
|
|
},
|
|
{
|
|
"entropy": 5.204835987091064,
|
|
"epoch": 7.411032990805841,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00013685202595973861,
|
|
"loss": 4.3898,
|
|
"mean_token_accuracy": 0.2724447131156921,
|
|
"num_tokens": 15384323.0,
|
|
"step": 6855
|
|
},
|
|
{
|
|
"entropy": 5.2820392608642575,
|
|
"epoch": 7.416441319632233,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00013651372590389122,
|
|
"loss": 4.4356,
|
|
"mean_token_accuracy": 0.2677796557545662,
|
|
"num_tokens": 15395034.0,
|
|
"step": 6860
|
|
},
|
|
{
|
|
"entropy": 5.321906566619873,
|
|
"epoch": 7.421849648458626,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00013617592910748378,
|
|
"loss": 4.5115,
|
|
"mean_token_accuracy": 0.26134859919548037,
|
|
"num_tokens": 15405948.0,
|
|
"step": 6865
|
|
},
|
|
{
|
|
"entropy": 5.237935876846313,
|
|
"epoch": 7.427257977285019,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00013583863679807092,
|
|
"loss": 4.4847,
|
|
"mean_token_accuracy": 0.26738051474094393,
|
|
"num_tokens": 15417200.0,
|
|
"step": 6870
|
|
},
|
|
{
|
|
"entropy": 5.183591508865357,
|
|
"epoch": 7.432666306111411,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00013550185020137344,
|
|
"loss": 4.4145,
|
|
"mean_token_accuracy": 0.2659712180495262,
|
|
"num_tokens": 15428525.0,
|
|
"step": 6875
|
|
},
|
|
{
|
|
"entropy": 5.257800436019897,
|
|
"epoch": 7.438074634937804,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.0001351655705412747,
|
|
"loss": 4.484,
|
|
"mean_token_accuracy": 0.26218918710947037,
|
|
"num_tokens": 15440464.0,
|
|
"step": 6880
|
|
},
|
|
{
|
|
"entropy": 5.255518770217895,
|
|
"epoch": 7.443482963764197,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0001348297990398158,
|
|
"loss": 4.4214,
|
|
"mean_token_accuracy": 0.271384996175766,
|
|
"num_tokens": 15452141.0,
|
|
"step": 6885
|
|
},
|
|
{
|
|
"entropy": 5.234218406677246,
|
|
"epoch": 7.44889129259059,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 0.00013449453691719108,
|
|
"loss": 4.3876,
|
|
"mean_token_accuracy": 0.2696350783109665,
|
|
"num_tokens": 15463550.0,
|
|
"step": 6890
|
|
},
|
|
{
|
|
"entropy": 5.271007299423218,
|
|
"epoch": 7.4542996214169825,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0001341597853917439,
|
|
"loss": 4.4957,
|
|
"mean_token_accuracy": 0.2652488797903061,
|
|
"num_tokens": 15475021.0,
|
|
"step": 6895
|
|
},
|
|
{
|
|
"entropy": 5.270128393173218,
|
|
"epoch": 7.4597079502433745,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00013382554567996214,
|
|
"loss": 4.4424,
|
|
"mean_token_accuracy": 0.26731889694929123,
|
|
"num_tokens": 15486567.0,
|
|
"step": 6900
|
|
},
|
|
{
|
|
"entropy": 5.272231483459473,
|
|
"epoch": 7.465116279069767,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00013349181899647346,
|
|
"loss": 4.4755,
|
|
"mean_token_accuracy": 0.2659808754920959,
|
|
"num_tokens": 15498133.0,
|
|
"step": 6905
|
|
},
|
|
{
|
|
"entropy": 5.275656795501709,
|
|
"epoch": 7.47052460789616,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00013315860655404167,
|
|
"loss": 4.534,
|
|
"mean_token_accuracy": 0.25830481201410294,
|
|
"num_tokens": 15509723.0,
|
|
"step": 6910
|
|
},
|
|
{
|
|
"entropy": 5.289685010910034,
|
|
"epoch": 7.475932936722553,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001328259095635615,
|
|
"loss": 4.4754,
|
|
"mean_token_accuracy": 0.25790139734745027,
|
|
"num_tokens": 15519590.0,
|
|
"step": 6915
|
|
},
|
|
{
|
|
"entropy": 5.314891719818116,
|
|
"epoch": 7.481341265548945,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0001324937292340545,
|
|
"loss": 4.4822,
|
|
"mean_token_accuracy": 0.26420515179634096,
|
|
"num_tokens": 15531112.0,
|
|
"step": 6920
|
|
},
|
|
{
|
|
"entropy": 5.202852964401245,
|
|
"epoch": 7.486749594375338,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00013216206677266475,
|
|
"loss": 4.3969,
|
|
"mean_token_accuracy": 0.27031752169132234,
|
|
"num_tokens": 15541341.0,
|
|
"step": 6925
|
|
},
|
|
{
|
|
"entropy": 5.2771759033203125,
|
|
"epoch": 7.492157923201731,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0001318309233846544,
|
|
"loss": 4.5074,
|
|
"mean_token_accuracy": 0.26005752086639405,
|
|
"num_tokens": 15553025.0,
|
|
"step": 6930
|
|
},
|
|
{
|
|
"entropy": 5.355898332595825,
|
|
"epoch": 7.497566252028124,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00013150030027339935,
|
|
"loss": 4.5317,
|
|
"mean_token_accuracy": 0.25956417471170423,
|
|
"num_tokens": 15563824.0,
|
|
"step": 6935
|
|
},
|
|
{
|
|
"entropy": 5.304265356063842,
|
|
"epoch": 7.502974580854516,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00013117019864038483,
|
|
"loss": 4.4809,
|
|
"mean_token_accuracy": 0.26720154881477354,
|
|
"num_tokens": 15575129.0,
|
|
"step": 6940
|
|
},
|
|
{
|
|
"entropy": 5.210094833374024,
|
|
"epoch": 7.5083829096809085,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00013084061968520063,
|
|
"loss": 4.4199,
|
|
"mean_token_accuracy": 0.26588934659957886,
|
|
"num_tokens": 15586018.0,
|
|
"step": 6945
|
|
},
|
|
{
|
|
"entropy": 5.286025619506836,
|
|
"epoch": 7.513791238507301,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00013051156460553776,
|
|
"loss": 4.5548,
|
|
"mean_token_accuracy": 0.2573856830596924,
|
|
"num_tokens": 15597062.0,
|
|
"step": 6950
|
|
},
|
|
{
|
|
"entropy": 5.321948909759522,
|
|
"epoch": 7.519199567333694,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00013018303459718305,
|
|
"loss": 4.6132,
|
|
"mean_token_accuracy": 0.25284685492515563,
|
|
"num_tokens": 15608666.0,
|
|
"step": 6955
|
|
},
|
|
{
|
|
"entropy": 5.313227939605713,
|
|
"epoch": 7.524607896160086,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00012985503085401523,
|
|
"loss": 4.4752,
|
|
"mean_token_accuracy": 0.26611890345811845,
|
|
"num_tokens": 15619764.0,
|
|
"step": 6960
|
|
},
|
|
{
|
|
"entropy": 5.32073860168457,
|
|
"epoch": 7.530016224986479,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00012952755456800067,
|
|
"loss": 4.5238,
|
|
"mean_token_accuracy": 0.25668325573205947,
|
|
"num_tokens": 15631611.0,
|
|
"step": 6965
|
|
},
|
|
{
|
|
"entropy": 5.304954671859742,
|
|
"epoch": 7.535424553812872,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.000129200606929189,
|
|
"loss": 4.4748,
|
|
"mean_token_accuracy": 0.2711686208844185,
|
|
"num_tokens": 15642392.0,
|
|
"step": 6970
|
|
},
|
|
{
|
|
"entropy": 5.279082250595093,
|
|
"epoch": 7.540832882639265,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0001288741891257087,
|
|
"loss": 4.4776,
|
|
"mean_token_accuracy": 0.2665621921420097,
|
|
"num_tokens": 15655304.0,
|
|
"step": 6975
|
|
},
|
|
{
|
|
"entropy": 5.23156328201294,
|
|
"epoch": 7.546241211465657,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00012854830234376277,
|
|
"loss": 4.4494,
|
|
"mean_token_accuracy": 0.26530476957559584,
|
|
"num_tokens": 15666055.0,
|
|
"step": 6980
|
|
},
|
|
{
|
|
"entropy": 5.2366766929626465,
|
|
"epoch": 7.55164954029205,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0001282229477676246,
|
|
"loss": 4.468,
|
|
"mean_token_accuracy": 0.2705321416258812,
|
|
"num_tokens": 15676287.0,
|
|
"step": 6985
|
|
},
|
|
{
|
|
"entropy": 5.34303846359253,
|
|
"epoch": 7.5570578691184425,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0001278981265796334,
|
|
"loss": 4.5326,
|
|
"mean_token_accuracy": 0.2570274189114571,
|
|
"num_tokens": 15687448.0,
|
|
"step": 6990
|
|
},
|
|
{
|
|
"entropy": 5.214194250106812,
|
|
"epoch": 7.562466197944835,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00012757383996019024,
|
|
"loss": 4.4278,
|
|
"mean_token_accuracy": 0.2692372933030128,
|
|
"num_tokens": 15698303.0,
|
|
"step": 6995
|
|
},
|
|
{
|
|
"entropy": 5.2056872844696045,
|
|
"epoch": 7.567874526771227,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0001272500890877533,
|
|
"loss": 4.3527,
|
|
"mean_token_accuracy": 0.2759406238794327,
|
|
"num_tokens": 15708905.0,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"epoch": 7.567874526771227,
|
|
"eval_entropy": 5.156692019375888,
|
|
"eval_loss": 5.879018306732178,
|
|
"eval_mean_token_accuracy": 0.18067381798543714,
|
|
"eval_num_tokens": 15708905.0,
|
|
"eval_runtime": 1.9384,
|
|
"eval_samples_per_second": 1813.35,
|
|
"eval_steps_per_second": 226.991,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"entropy": 5.186620092391967,
|
|
"epoch": 7.57328285559762,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00012692687513883393,
|
|
"loss": 4.4178,
|
|
"mean_token_accuracy": 0.2676037520170212,
|
|
"num_tokens": 15719394.0,
|
|
"step": 7005
|
|
},
|
|
{
|
|
"entropy": 5.315431070327759,
|
|
"epoch": 7.578691184424013,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00012660419928799265,
|
|
"loss": 4.481,
|
|
"mean_token_accuracy": 0.2667367056012154,
|
|
"num_tokens": 15729754.0,
|
|
"step": 7010
|
|
},
|
|
{
|
|
"entropy": 5.276031351089477,
|
|
"epoch": 7.584099513250406,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00012628206270783404,
|
|
"loss": 4.4759,
|
|
"mean_token_accuracy": 0.2635058224201202,
|
|
"num_tokens": 15741194.0,
|
|
"step": 7015
|
|
},
|
|
{
|
|
"entropy": 5.250506258010864,
|
|
"epoch": 7.589507842076799,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00012596046656900323,
|
|
"loss": 4.4431,
|
|
"mean_token_accuracy": 0.2722063735127449,
|
|
"num_tokens": 15752594.0,
|
|
"step": 7020
|
|
},
|
|
{
|
|
"entropy": 5.272899198532104,
|
|
"epoch": 7.594916170903191,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001256394120401813,
|
|
"loss": 4.4738,
|
|
"mean_token_accuracy": 0.26401168256998064,
|
|
"num_tokens": 15764121.0,
|
|
"step": 7025
|
|
},
|
|
{
|
|
"entropy": 5.279602527618408,
|
|
"epoch": 7.600324499729584,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00012531890028808122,
|
|
"loss": 4.4193,
|
|
"mean_token_accuracy": 0.26759782135486604,
|
|
"num_tokens": 15775842.0,
|
|
"step": 7030
|
|
},
|
|
{
|
|
"entropy": 5.2658774852752686,
|
|
"epoch": 7.605732828555976,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.0001249989324774434,
|
|
"loss": 4.3924,
|
|
"mean_token_accuracy": 0.2718045458197594,
|
|
"num_tokens": 15787663.0,
|
|
"step": 7035
|
|
},
|
|
{
|
|
"entropy": 5.274012041091919,
|
|
"epoch": 7.611141157382368,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00012467950977103158,
|
|
"loss": 4.4314,
|
|
"mean_token_accuracy": 0.26743387132883073,
|
|
"num_tokens": 15799675.0,
|
|
"step": 7040
|
|
},
|
|
{
|
|
"entropy": 5.235326862335205,
|
|
"epoch": 7.616549486208761,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00012436063332962852,
|
|
"loss": 4.4894,
|
|
"mean_token_accuracy": 0.26909134835004805,
|
|
"num_tokens": 15810772.0,
|
|
"step": 7045
|
|
},
|
|
{
|
|
"entropy": 5.2840930938720705,
|
|
"epoch": 7.621957815035154,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00012404230431203227,
|
|
"loss": 4.5276,
|
|
"mean_token_accuracy": 0.26369849890470504,
|
|
"num_tokens": 15823711.0,
|
|
"step": 7050
|
|
},
|
|
{
|
|
"entropy": 5.225018787384033,
|
|
"epoch": 7.627366143861547,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00012372452387505087,
|
|
"loss": 4.4004,
|
|
"mean_token_accuracy": 0.2768779143691063,
|
|
"num_tokens": 15834603.0,
|
|
"step": 7055
|
|
},
|
|
{
|
|
"entropy": 5.2185302734375,
|
|
"epoch": 7.63277447268794,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001234072931734993,
|
|
"loss": 4.3904,
|
|
"mean_token_accuracy": 0.2656655788421631,
|
|
"num_tokens": 15846238.0,
|
|
"step": 7060
|
|
},
|
|
{
|
|
"entropy": 5.3304859638214115,
|
|
"epoch": 7.638182801514332,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00012309061336019465,
|
|
"loss": 4.5526,
|
|
"mean_token_accuracy": 0.2568135380744934,
|
|
"num_tokens": 15858560.0,
|
|
"step": 7065
|
|
},
|
|
{
|
|
"entropy": 5.2238997459411625,
|
|
"epoch": 7.643591130340725,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00012277448558595212,
|
|
"loss": 4.4055,
|
|
"mean_token_accuracy": 0.26717972606420515,
|
|
"num_tokens": 15870624.0,
|
|
"step": 7070
|
|
},
|
|
{
|
|
"entropy": 5.243378162384033,
|
|
"epoch": 7.6489994591671175,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00012245891099958074,
|
|
"loss": 4.4254,
|
|
"mean_token_accuracy": 0.26449286937713623,
|
|
"num_tokens": 15881199.0,
|
|
"step": 7075
|
|
},
|
|
{
|
|
"entropy": 5.291801691055298,
|
|
"epoch": 7.6544077879935095,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001221438907478793,
|
|
"loss": 4.5295,
|
|
"mean_token_accuracy": 0.2559365391731262,
|
|
"num_tokens": 15893270.0,
|
|
"step": 7080
|
|
},
|
|
{
|
|
"entropy": 5.207455348968506,
|
|
"epoch": 7.659816116819902,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00012182942597563213,
|
|
"loss": 4.3871,
|
|
"mean_token_accuracy": 0.26937498450279235,
|
|
"num_tokens": 15904509.0,
|
|
"step": 7085
|
|
},
|
|
{
|
|
"entropy": 5.2135786533355715,
|
|
"epoch": 7.665224445646295,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00012151551782560499,
|
|
"loss": 4.3325,
|
|
"mean_token_accuracy": 0.28553567826747894,
|
|
"num_tokens": 15914913.0,
|
|
"step": 7090
|
|
},
|
|
{
|
|
"entropy": 5.291334247589111,
|
|
"epoch": 7.670632774472688,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00012120216743854068,
|
|
"loss": 4.4578,
|
|
"mean_token_accuracy": 0.2734370157122612,
|
|
"num_tokens": 15925742.0,
|
|
"step": 7095
|
|
},
|
|
{
|
|
"entropy": 5.19296703338623,
|
|
"epoch": 7.676041103299081,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00012088937595315525,
|
|
"loss": 4.4442,
|
|
"mean_token_accuracy": 0.27159425616264343,
|
|
"num_tokens": 15936165.0,
|
|
"step": 7100
|
|
},
|
|
{
|
|
"entropy": 5.216121244430542,
|
|
"epoch": 7.681449432125473,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0001205771445061339,
|
|
"loss": 4.5146,
|
|
"mean_token_accuracy": 0.26815264523029325,
|
|
"num_tokens": 15948570.0,
|
|
"step": 7105
|
|
},
|
|
{
|
|
"entropy": 5.298116827011109,
|
|
"epoch": 7.686857760951866,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001202654742321262,
|
|
"loss": 4.4862,
|
|
"mean_token_accuracy": 0.26067458987236025,
|
|
"num_tokens": 15959890.0,
|
|
"step": 7110
|
|
},
|
|
{
|
|
"entropy": 5.269825124740601,
|
|
"epoch": 7.692266089778259,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00011995436626374272,
|
|
"loss": 4.414,
|
|
"mean_token_accuracy": 0.27869476974010465,
|
|
"num_tokens": 15970632.0,
|
|
"step": 7115
|
|
},
|
|
{
|
|
"entropy": 5.250730371475219,
|
|
"epoch": 7.6976744186046515,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 0.00011964382173155053,
|
|
"loss": 4.4332,
|
|
"mean_token_accuracy": 0.26520663797855376,
|
|
"num_tokens": 15981352.0,
|
|
"step": 7120
|
|
},
|
|
{
|
|
"entropy": 5.239840412139893,
|
|
"epoch": 7.7030827474310435,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 0.00011933384176406915,
|
|
"loss": 4.4191,
|
|
"mean_token_accuracy": 0.2603464901447296,
|
|
"num_tokens": 15991619.0,
|
|
"step": 7125
|
|
},
|
|
{
|
|
"entropy": 5.219773054122925,
|
|
"epoch": 7.708491076257436,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0001190244274877666,
|
|
"loss": 4.4355,
|
|
"mean_token_accuracy": 0.2651257306337357,
|
|
"num_tokens": 16001588.0,
|
|
"step": 7130
|
|
},
|
|
{
|
|
"entropy": 5.265380239486694,
|
|
"epoch": 7.713899405083829,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00011871558002705501,
|
|
"loss": 4.4742,
|
|
"mean_token_accuracy": 0.26553222090005874,
|
|
"num_tokens": 16013230.0,
|
|
"step": 7135
|
|
},
|
|
{
|
|
"entropy": 5.27869553565979,
|
|
"epoch": 7.719307733910222,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00011840730050428666,
|
|
"loss": 4.4705,
|
|
"mean_token_accuracy": 0.2703775465488434,
|
|
"num_tokens": 16024310.0,
|
|
"step": 7140
|
|
},
|
|
{
|
|
"entropy": 5.34291639328003,
|
|
"epoch": 7.724716062736614,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0001180995900397503,
|
|
"loss": 4.5202,
|
|
"mean_token_accuracy": 0.2699713662266731,
|
|
"num_tokens": 16035188.0,
|
|
"step": 7145
|
|
},
|
|
{
|
|
"entropy": 5.301549196243286,
|
|
"epoch": 7.730124391563007,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00011779244975166623,
|
|
"loss": 4.4556,
|
|
"mean_token_accuracy": 0.26819572746753695,
|
|
"num_tokens": 16046354.0,
|
|
"step": 7150
|
|
},
|
|
{
|
|
"entropy": 5.244382905960083,
|
|
"epoch": 7.7355327203894,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00011748588075618305,
|
|
"loss": 4.4902,
|
|
"mean_token_accuracy": 0.2710270330309868,
|
|
"num_tokens": 16056809.0,
|
|
"step": 7155
|
|
},
|
|
{
|
|
"entropy": 5.2672436237335205,
|
|
"epoch": 7.740941049215793,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.00011717988416737322,
|
|
"loss": 4.4842,
|
|
"mean_token_accuracy": 0.26911704391241076,
|
|
"num_tokens": 16067440.0,
|
|
"step": 7160
|
|
},
|
|
{
|
|
"entropy": 5.225475645065307,
|
|
"epoch": 7.746349378042185,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00011687446109722904,
|
|
"loss": 4.4134,
|
|
"mean_token_accuracy": 0.26953894197940825,
|
|
"num_tokens": 16078126.0,
|
|
"step": 7165
|
|
},
|
|
{
|
|
"entropy": 5.271527671813965,
|
|
"epoch": 7.7517577068685775,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00011656961265565864,
|
|
"loss": 4.4973,
|
|
"mean_token_accuracy": 0.2660005271434784,
|
|
"num_tokens": 16090823.0,
|
|
"step": 7170
|
|
},
|
|
{
|
|
"entropy": 5.244742012023925,
|
|
"epoch": 7.75716603569497,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00011626533995048204,
|
|
"loss": 4.4135,
|
|
"mean_token_accuracy": 0.27221386432647704,
|
|
"num_tokens": 16102544.0,
|
|
"step": 7175
|
|
},
|
|
{
|
|
"entropy": 5.329130983352661,
|
|
"epoch": 7.762574364521363,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00011596164408742675,
|
|
"loss": 4.5403,
|
|
"mean_token_accuracy": 0.2668018117547035,
|
|
"num_tokens": 16113573.0,
|
|
"step": 7180
|
|
},
|
|
{
|
|
"entropy": 5.175236225128174,
|
|
"epoch": 7.767982693347756,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00011565852617012452,
|
|
"loss": 4.405,
|
|
"mean_token_accuracy": 0.2706847369670868,
|
|
"num_tokens": 16124999.0,
|
|
"step": 7185
|
|
},
|
|
{
|
|
"entropy": 5.1959755420684814,
|
|
"epoch": 7.773391022174148,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00011535598730010633,
|
|
"loss": 4.4167,
|
|
"mean_token_accuracy": 0.26864219158887864,
|
|
"num_tokens": 16134941.0,
|
|
"step": 7190
|
|
},
|
|
{
|
|
"entropy": 5.2813338279724125,
|
|
"epoch": 7.778799351000541,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00011505402857679921,
|
|
"loss": 4.4453,
|
|
"mean_token_accuracy": 0.27222750931978223,
|
|
"num_tokens": 16146336.0,
|
|
"step": 7195
|
|
},
|
|
{
|
|
"entropy": 5.324097204208374,
|
|
"epoch": 7.784207679826934,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00011475265109752187,
|
|
"loss": 4.5239,
|
|
"mean_token_accuracy": 0.2651411861181259,
|
|
"num_tokens": 16158539.0,
|
|
"step": 7200
|
|
},
|
|
{
|
|
"entropy": 5.291914606094361,
|
|
"epoch": 7.789616008653326,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00011445185595748074,
|
|
"loss": 4.5578,
|
|
"mean_token_accuracy": 0.2536396816372871,
|
|
"num_tokens": 16170878.0,
|
|
"step": 7205
|
|
},
|
|
{
|
|
"entropy": 5.184276103973389,
|
|
"epoch": 7.795024337479719,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0001141516442497661,
|
|
"loss": 4.3938,
|
|
"mean_token_accuracy": 0.2640189856290817,
|
|
"num_tokens": 16182076.0,
|
|
"step": 7210
|
|
},
|
|
{
|
|
"entropy": 5.1988012313842775,
|
|
"epoch": 7.800432666306111,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00011385201706534803,
|
|
"loss": 4.4374,
|
|
"mean_token_accuracy": 0.26373240649700164,
|
|
"num_tokens": 16192850.0,
|
|
"step": 7215
|
|
},
|
|
{
|
|
"entropy": 5.181417512893677,
|
|
"epoch": 7.805840995132504,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001135529754930722,
|
|
"loss": 4.4055,
|
|
"mean_token_accuracy": 0.2642504543066025,
|
|
"num_tokens": 16204118.0,
|
|
"step": 7220
|
|
},
|
|
{
|
|
"entropy": 5.340326118469238,
|
|
"epoch": 7.811249323958897,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00011325452061965662,
|
|
"loss": 4.4875,
|
|
"mean_token_accuracy": 0.2545256793498993,
|
|
"num_tokens": 16215513.0,
|
|
"step": 7225
|
|
},
|
|
{
|
|
"entropy": 5.272321319580078,
|
|
"epoch": 7.816657652785289,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00011295665352968692,
|
|
"loss": 4.4997,
|
|
"mean_token_accuracy": 0.26032399386167526,
|
|
"num_tokens": 16228169.0,
|
|
"step": 7230
|
|
},
|
|
{
|
|
"entropy": 5.321831369400025,
|
|
"epoch": 7.822065981611682,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.00011265937530561271,
|
|
"loss": 4.4363,
|
|
"mean_token_accuracy": 0.26349303424358367,
|
|
"num_tokens": 16239200.0,
|
|
"step": 7235
|
|
},
|
|
{
|
|
"entropy": 5.329130411148071,
|
|
"epoch": 7.827474310438075,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00011236268702774377,
|
|
"loss": 4.5294,
|
|
"mean_token_accuracy": 0.25417055785655973,
|
|
"num_tokens": 16250979.0,
|
|
"step": 7240
|
|
},
|
|
{
|
|
"entropy": 5.24636173248291,
|
|
"epoch": 7.832882639264467,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00011206658977424605,
|
|
"loss": 4.449,
|
|
"mean_token_accuracy": 0.2664190322160721,
|
|
"num_tokens": 16262103.0,
|
|
"step": 7245
|
|
},
|
|
{
|
|
"entropy": 5.209012651443482,
|
|
"epoch": 7.83829096809086,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00011177108462113761,
|
|
"loss": 4.461,
|
|
"mean_token_accuracy": 0.26312619298696516,
|
|
"num_tokens": 16272948.0,
|
|
"step": 7250
|
|
},
|
|
{
|
|
"entropy": 5.213464069366455,
|
|
"epoch": 7.8436992969172525,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00011147617264228491,
|
|
"loss": 4.3709,
|
|
"mean_token_accuracy": 0.27457325309515,
|
|
"num_tokens": 16283180.0,
|
|
"step": 7255
|
|
},
|
|
{
|
|
"entropy": 5.2294981479644775,
|
|
"epoch": 7.849107625743645,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00011118185490939861,
|
|
"loss": 4.394,
|
|
"mean_token_accuracy": 0.2742414534091949,
|
|
"num_tokens": 16294886.0,
|
|
"step": 7260
|
|
},
|
|
{
|
|
"entropy": 5.237207269668579,
|
|
"epoch": 7.854515954570038,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00011088813249203019,
|
|
"loss": 4.4166,
|
|
"mean_token_accuracy": 0.27937147617340086,
|
|
"num_tokens": 16305979.0,
|
|
"step": 7265
|
|
},
|
|
{
|
|
"entropy": 5.233218860626221,
|
|
"epoch": 7.85992428339643,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00011059500645756762,
|
|
"loss": 4.4258,
|
|
"mean_token_accuracy": 0.2637356072664261,
|
|
"num_tokens": 16318645.0,
|
|
"step": 7270
|
|
},
|
|
{
|
|
"entropy": 5.21151762008667,
|
|
"epoch": 7.865332612222823,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00011030247787123147,
|
|
"loss": 4.4026,
|
|
"mean_token_accuracy": 0.2698768854141235,
|
|
"num_tokens": 16329907.0,
|
|
"step": 7275
|
|
},
|
|
{
|
|
"entropy": 5.275745964050293,
|
|
"epoch": 7.870740941049216,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00011001054779607121,
|
|
"loss": 4.5388,
|
|
"mean_token_accuracy": 0.2556803718209267,
|
|
"num_tokens": 16342531.0,
|
|
"step": 7280
|
|
},
|
|
{
|
|
"entropy": 5.258225727081299,
|
|
"epoch": 7.876149269875609,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00010971921729296172,
|
|
"loss": 4.374,
|
|
"mean_token_accuracy": 0.2737454742193222,
|
|
"num_tokens": 16355394.0,
|
|
"step": 7285
|
|
},
|
|
{
|
|
"entropy": 5.162232971191406,
|
|
"epoch": 7.881557598702001,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00010942848742059845,
|
|
"loss": 4.2829,
|
|
"mean_token_accuracy": 0.28719756901264193,
|
|
"num_tokens": 16366049.0,
|
|
"step": 7290
|
|
},
|
|
{
|
|
"entropy": 5.245200109481812,
|
|
"epoch": 7.886965927528394,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00010913835923549453,
|
|
"loss": 4.4269,
|
|
"mean_token_accuracy": 0.2689394310116768,
|
|
"num_tokens": 16378597.0,
|
|
"step": 7295
|
|
},
|
|
{
|
|
"entropy": 5.306468772888183,
|
|
"epoch": 7.8923742563547865,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00010884883379197643,
|
|
"loss": 4.5634,
|
|
"mean_token_accuracy": 0.26228204369544983,
|
|
"num_tokens": 16391417.0,
|
|
"step": 7300
|
|
},
|
|
{
|
|
"entropy": 5.2524865627288815,
|
|
"epoch": 7.897782585181179,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00010855991214218028,
|
|
"loss": 4.4373,
|
|
"mean_token_accuracy": 0.27614626288414,
|
|
"num_tokens": 16402188.0,
|
|
"step": 7305
|
|
},
|
|
{
|
|
"entropy": 5.285575008392334,
|
|
"epoch": 7.903190914007571,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00010827159533604814,
|
|
"loss": 4.4777,
|
|
"mean_token_accuracy": 0.26792855858802794,
|
|
"num_tokens": 16415124.0,
|
|
"step": 7310
|
|
},
|
|
{
|
|
"entropy": 5.259737920761109,
|
|
"epoch": 7.908599242833964,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00010798388442132375,
|
|
"loss": 4.4697,
|
|
"mean_token_accuracy": 0.27016521692276,
|
|
"num_tokens": 16426536.0,
|
|
"step": 7315
|
|
},
|
|
{
|
|
"entropy": 5.259300470352173,
|
|
"epoch": 7.914007571660357,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00010769678044354927,
|
|
"loss": 4.441,
|
|
"mean_token_accuracy": 0.27272363752126694,
|
|
"num_tokens": 16438417.0,
|
|
"step": 7320
|
|
},
|
|
{
|
|
"entropy": 5.246110248565674,
|
|
"epoch": 7.91941590048675,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00010741028444606141,
|
|
"loss": 4.4572,
|
|
"mean_token_accuracy": 0.2611702874302864,
|
|
"num_tokens": 16449271.0,
|
|
"step": 7325
|
|
},
|
|
{
|
|
"entropy": 5.326108026504516,
|
|
"epoch": 7.924824229313142,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00010712439746998711,
|
|
"loss": 4.5412,
|
|
"mean_token_accuracy": 0.26421731561422346,
|
|
"num_tokens": 16460387.0,
|
|
"step": 7330
|
|
},
|
|
{
|
|
"entropy": 5.246243190765381,
|
|
"epoch": 7.930232558139535,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00010683912055424036,
|
|
"loss": 4.4207,
|
|
"mean_token_accuracy": 0.2688411593437195,
|
|
"num_tokens": 16471262.0,
|
|
"step": 7335
|
|
},
|
|
{
|
|
"entropy": 5.272463607788086,
|
|
"epoch": 7.935640886965928,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00010655445473551809,
|
|
"loss": 4.4714,
|
|
"mean_token_accuracy": 0.26707434803247454,
|
|
"num_tokens": 16482605.0,
|
|
"step": 7340
|
|
},
|
|
{
|
|
"entropy": 5.317854928970337,
|
|
"epoch": 7.9410492157923205,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00010627040104829658,
|
|
"loss": 4.5334,
|
|
"mean_token_accuracy": 0.261533884704113,
|
|
"num_tokens": 16493951.0,
|
|
"step": 7345
|
|
},
|
|
{
|
|
"entropy": 5.278520345687866,
|
|
"epoch": 7.9464575446187125,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00010598696052482759,
|
|
"loss": 4.4314,
|
|
"mean_token_accuracy": 0.27727963626384733,
|
|
"num_tokens": 16506236.0,
|
|
"step": 7350
|
|
},
|
|
{
|
|
"entropy": 5.181040239334107,
|
|
"epoch": 7.951865873445105,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0001057041341951347,
|
|
"loss": 4.3385,
|
|
"mean_token_accuracy": 0.28301757425069807,
|
|
"num_tokens": 16518339.0,
|
|
"step": 7355
|
|
},
|
|
{
|
|
"entropy": 5.281914949417114,
|
|
"epoch": 7.957274202271498,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00010542192308700925,
|
|
"loss": 4.4263,
|
|
"mean_token_accuracy": 0.2641948163509369,
|
|
"num_tokens": 16528733.0,
|
|
"step": 7360
|
|
},
|
|
{
|
|
"entropy": 5.254223918914795,
|
|
"epoch": 7.962682531097891,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00010514032822600733,
|
|
"loss": 4.4892,
|
|
"mean_token_accuracy": 0.26294231563806536,
|
|
"num_tokens": 16540776.0,
|
|
"step": 7365
|
|
},
|
|
{
|
|
"entropy": 5.23925404548645,
|
|
"epoch": 7.968090859924283,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00010485935063544507,
|
|
"loss": 4.4394,
|
|
"mean_token_accuracy": 0.27023291140794753,
|
|
"num_tokens": 16550818.0,
|
|
"step": 7370
|
|
},
|
|
{
|
|
"entropy": 5.266943311691284,
|
|
"epoch": 7.973499188750676,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00010457899133639579,
|
|
"loss": 4.4524,
|
|
"mean_token_accuracy": 0.2584550857543945,
|
|
"num_tokens": 16561791.0,
|
|
"step": 7375
|
|
},
|
|
{
|
|
"entropy": 5.277328729629517,
|
|
"epoch": 7.978907517577069,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00010429925134768584,
|
|
"loss": 4.4563,
|
|
"mean_token_accuracy": 0.269451779127121,
|
|
"num_tokens": 16573360.0,
|
|
"step": 7380
|
|
},
|
|
{
|
|
"entropy": 5.257878255844116,
|
|
"epoch": 7.984315846403462,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00010402013168589092,
|
|
"loss": 4.4255,
|
|
"mean_token_accuracy": 0.27087407410144804,
|
|
"num_tokens": 16584061.0,
|
|
"step": 7385
|
|
},
|
|
{
|
|
"entropy": 5.276725196838379,
|
|
"epoch": 7.9897241752298545,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00010374163336533261,
|
|
"loss": 4.4514,
|
|
"mean_token_accuracy": 0.269998237490654,
|
|
"num_tokens": 16594178.0,
|
|
"step": 7390
|
|
},
|
|
{
|
|
"entropy": 5.282387208938599,
|
|
"epoch": 7.995132504056246,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00010346375739807448,
|
|
"loss": 4.5348,
|
|
"mean_token_accuracy": 0.2592212691903114,
|
|
"num_tokens": 16605577.0,
|
|
"step": 7395
|
|
},
|
|
{
|
|
"entropy": 5.259804566701253,
|
|
"epoch": 8.0,
|
|
"grad_norm": 2.21875,
|
|
"learning_rate": 0.00010318650479391825,
|
|
"loss": 4.6801,
|
|
"mean_token_accuracy": 0.2569085856278737,
|
|
"num_tokens": 16615112.0,
|
|
"step": 7400
|
|
},
|
|
{
|
|
"entropy": 5.267020797729492,
|
|
"epoch": 8.005408328826393,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00010290987656040076,
|
|
"loss": 4.3394,
|
|
"mean_token_accuracy": 0.2833008199930191,
|
|
"num_tokens": 16625050.0,
|
|
"step": 7405
|
|
},
|
|
{
|
|
"entropy": 5.248191928863525,
|
|
"epoch": 8.010816657652786,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00010263387370278948,
|
|
"loss": 4.4361,
|
|
"mean_token_accuracy": 0.2671979993581772,
|
|
"num_tokens": 16636536.0,
|
|
"step": 7410
|
|
},
|
|
{
|
|
"entropy": 5.325454187393189,
|
|
"epoch": 8.016224986479179,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001023584972240795,
|
|
"loss": 4.4196,
|
|
"mean_token_accuracy": 0.27599603682756424,
|
|
"num_tokens": 16647439.0,
|
|
"step": 7415
|
|
},
|
|
{
|
|
"entropy": 5.291596221923828,
|
|
"epoch": 8.021633315305571,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00010208374812498955,
|
|
"loss": 4.3839,
|
|
"mean_token_accuracy": 0.26871761232614516,
|
|
"num_tokens": 16659778.0,
|
|
"step": 7420
|
|
},
|
|
{
|
|
"entropy": 5.2614256858825685,
|
|
"epoch": 8.027041644131963,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00010180962740395853,
|
|
"loss": 4.3929,
|
|
"mean_token_accuracy": 0.27438968122005464,
|
|
"num_tokens": 16671095.0,
|
|
"step": 7425
|
|
},
|
|
{
|
|
"entropy": 5.224966812133789,
|
|
"epoch": 8.032449972958355,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.00010153613605714173,
|
|
"loss": 4.3383,
|
|
"mean_token_accuracy": 0.2776650130748749,
|
|
"num_tokens": 16682401.0,
|
|
"step": 7430
|
|
},
|
|
{
|
|
"entropy": 5.285452461242675,
|
|
"epoch": 8.037858301784748,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00010126327507840744,
|
|
"loss": 4.4128,
|
|
"mean_token_accuracy": 0.2766126379370689,
|
|
"num_tokens": 16692606.0,
|
|
"step": 7435
|
|
},
|
|
{
|
|
"entropy": 5.181665992736816,
|
|
"epoch": 8.043266630611141,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00010099104545933288,
|
|
"loss": 4.3111,
|
|
"mean_token_accuracy": 0.29082414507865906,
|
|
"num_tokens": 16704771.0,
|
|
"step": 7440
|
|
},
|
|
{
|
|
"entropy": 5.194856023788452,
|
|
"epoch": 8.048674959437534,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00010071944818920127,
|
|
"loss": 4.4004,
|
|
"mean_token_accuracy": 0.27537615448236463,
|
|
"num_tokens": 16717512.0,
|
|
"step": 7445
|
|
},
|
|
{
|
|
"entropy": 5.216123151779175,
|
|
"epoch": 8.054083288263927,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001004484842549978,
|
|
"loss": 4.382,
|
|
"mean_token_accuracy": 0.2776669323444366,
|
|
"num_tokens": 16729457.0,
|
|
"step": 7450
|
|
},
|
|
{
|
|
"entropy": 5.254812717437744,
|
|
"epoch": 8.05949161709032,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00010017815464140583,
|
|
"loss": 4.3632,
|
|
"mean_token_accuracy": 0.2793877571821213,
|
|
"num_tokens": 16741195.0,
|
|
"step": 7455
|
|
},
|
|
{
|
|
"entropy": 5.210014200210571,
|
|
"epoch": 8.064899945916713,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 9.990846033080389e-05,
|
|
"loss": 4.352,
|
|
"mean_token_accuracy": 0.2738141268491745,
|
|
"num_tokens": 16752261.0,
|
|
"step": 7460
|
|
},
|
|
{
|
|
"entropy": 5.240028667449951,
|
|
"epoch": 8.070308274743104,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 9.963940230326176e-05,
|
|
"loss": 4.3783,
|
|
"mean_token_accuracy": 0.28560387194156645,
|
|
"num_tokens": 16763492.0,
|
|
"step": 7465
|
|
},
|
|
{
|
|
"entropy": 5.184400320053101,
|
|
"epoch": 8.075716603569496,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 9.937098153653687e-05,
|
|
"loss": 4.332,
|
|
"mean_token_accuracy": 0.2865113377571106,
|
|
"num_tokens": 16775218.0,
|
|
"step": 7470
|
|
},
|
|
{
|
|
"entropy": 5.167940378189087,
|
|
"epoch": 8.08112493239589,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 9.910319900607104e-05,
|
|
"loss": 4.29,
|
|
"mean_token_accuracy": 0.27895103991031645,
|
|
"num_tokens": 16786589.0,
|
|
"step": 7475
|
|
},
|
|
{
|
|
"entropy": 5.232013368606568,
|
|
"epoch": 8.086533261222282,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 9.883605568498635e-05,
|
|
"loss": 4.3351,
|
|
"mean_token_accuracy": 0.2818018883466721,
|
|
"num_tokens": 16798308.0,
|
|
"step": 7480
|
|
},
|
|
{
|
|
"entropy": 5.307837438583374,
|
|
"epoch": 8.091941590048675,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 9.856955254408248e-05,
|
|
"loss": 4.4263,
|
|
"mean_token_accuracy": 0.26896942853927613,
|
|
"num_tokens": 16810039.0,
|
|
"step": 7485
|
|
},
|
|
{
|
|
"entropy": 5.216094350814819,
|
|
"epoch": 8.097349918875068,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 9.830369055183244e-05,
|
|
"loss": 4.3857,
|
|
"mean_token_accuracy": 0.2688171058893204,
|
|
"num_tokens": 16820571.0,
|
|
"step": 7490
|
|
},
|
|
{
|
|
"entropy": 5.219715881347656,
|
|
"epoch": 8.10275824770146,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 9.80384706743792e-05,
|
|
"loss": 4.3534,
|
|
"mean_token_accuracy": 0.27653409242630006,
|
|
"num_tokens": 16831165.0,
|
|
"step": 7495
|
|
},
|
|
{
|
|
"entropy": 5.312400674819946,
|
|
"epoch": 8.108166576527854,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 9.777389387553243e-05,
|
|
"loss": 4.4824,
|
|
"mean_token_accuracy": 0.2587782502174377,
|
|
"num_tokens": 16842599.0,
|
|
"step": 7500
|
|
},
|
|
{
|
|
"epoch": 8.108166576527854,
|
|
"eval_entropy": 5.149628858132796,
|
|
"eval_loss": 5.887925624847412,
|
|
"eval_mean_token_accuracy": 0.18019113271412524,
|
|
"eval_num_tokens": 16842599.0,
|
|
"eval_runtime": 1.9306,
|
|
"eval_samples_per_second": 1820.704,
|
|
"eval_steps_per_second": 227.912,
|
|
"step": 7500
|
|
},
|
|
{
|
|
"entropy": 5.299224042892456,
|
|
"epoch": 8.113574905354245,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 9.750996111676497e-05,
|
|
"loss": 4.4655,
|
|
"mean_token_accuracy": 0.2678083196282387,
|
|
"num_tokens": 16854408.0,
|
|
"step": 7505
|
|
},
|
|
{
|
|
"entropy": 5.327978992462159,
|
|
"epoch": 8.118983234180638,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 9.724667335720894e-05,
|
|
"loss": 4.4564,
|
|
"mean_token_accuracy": 0.26968016624450686,
|
|
"num_tokens": 16865662.0,
|
|
"step": 7510
|
|
},
|
|
{
|
|
"entropy": 5.302506017684936,
|
|
"epoch": 8.12439156300703,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 9.698403155365268e-05,
|
|
"loss": 4.4165,
|
|
"mean_token_accuracy": 0.27912467122077944,
|
|
"num_tokens": 16876621.0,
|
|
"step": 7515
|
|
},
|
|
{
|
|
"entropy": 5.233722400665283,
|
|
"epoch": 8.129799891833423,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 9.672203666053714e-05,
|
|
"loss": 4.3839,
|
|
"mean_token_accuracy": 0.2759858310222626,
|
|
"num_tokens": 16888989.0,
|
|
"step": 7520
|
|
},
|
|
{
|
|
"entropy": 5.304172611236572,
|
|
"epoch": 8.135208220659816,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 9.646068962995235e-05,
|
|
"loss": 4.4659,
|
|
"mean_token_accuracy": 0.26399523615837095,
|
|
"num_tokens": 16900153.0,
|
|
"step": 7525
|
|
},
|
|
{
|
|
"entropy": 5.206618595123291,
|
|
"epoch": 8.140616549486209,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 9.619999141163415e-05,
|
|
"loss": 4.4625,
|
|
"mean_token_accuracy": 0.26713075041770934,
|
|
"num_tokens": 16913304.0,
|
|
"step": 7530
|
|
},
|
|
{
|
|
"entropy": 5.268564462661743,
|
|
"epoch": 8.146024878312602,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 9.593994295296024e-05,
|
|
"loss": 4.4307,
|
|
"mean_token_accuracy": 0.27510613650083543,
|
|
"num_tokens": 16925050.0,
|
|
"step": 7535
|
|
},
|
|
{
|
|
"entropy": 5.170204019546508,
|
|
"epoch": 8.151433207138995,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 9.56805451989474e-05,
|
|
"loss": 4.403,
|
|
"mean_token_accuracy": 0.27537869215011596,
|
|
"num_tokens": 16935940.0,
|
|
"step": 7540
|
|
},
|
|
{
|
|
"entropy": 5.2763372421264645,
|
|
"epoch": 8.156841535965386,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 9.54217990922478e-05,
|
|
"loss": 4.4205,
|
|
"mean_token_accuracy": 0.27138821333646773,
|
|
"num_tokens": 16947285.0,
|
|
"step": 7545
|
|
},
|
|
{
|
|
"entropy": 5.2447432518005375,
|
|
"epoch": 8.162249864791779,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 9.516370557314525e-05,
|
|
"loss": 4.3281,
|
|
"mean_token_accuracy": 0.27640354484319685,
|
|
"num_tokens": 16958781.0,
|
|
"step": 7550
|
|
},
|
|
{
|
|
"entropy": 5.2224359035491945,
|
|
"epoch": 8.167658193618172,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 9.490626557955221e-05,
|
|
"loss": 4.3826,
|
|
"mean_token_accuracy": 0.28325906246900556,
|
|
"num_tokens": 16970277.0,
|
|
"step": 7555
|
|
},
|
|
{
|
|
"entropy": 5.278955030441284,
|
|
"epoch": 8.173066522444564,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 9.464948004700626e-05,
|
|
"loss": 4.4181,
|
|
"mean_token_accuracy": 0.2788705050945282,
|
|
"num_tokens": 16983250.0,
|
|
"step": 7560
|
|
},
|
|
{
|
|
"entropy": 5.257851219177246,
|
|
"epoch": 8.178474851270957,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 9.439334990866658e-05,
|
|
"loss": 4.4096,
|
|
"mean_token_accuracy": 0.27978136837482454,
|
|
"num_tokens": 16994531.0,
|
|
"step": 7565
|
|
},
|
|
{
|
|
"entropy": 5.286068439483643,
|
|
"epoch": 8.18388318009735,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 9.413787609531072e-05,
|
|
"loss": 4.3978,
|
|
"mean_token_accuracy": 0.27611191272735597,
|
|
"num_tokens": 17006107.0,
|
|
"step": 7570
|
|
},
|
|
{
|
|
"entropy": 5.2852174758911135,
|
|
"epoch": 8.189291508923743,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 9.38830595353311e-05,
|
|
"loss": 4.4555,
|
|
"mean_token_accuracy": 0.2655415117740631,
|
|
"num_tokens": 17017243.0,
|
|
"step": 7575
|
|
},
|
|
{
|
|
"entropy": 5.251709032058716,
|
|
"epoch": 8.194699837750136,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 9.362890115473161e-05,
|
|
"loss": 4.3735,
|
|
"mean_token_accuracy": 0.2740049809217453,
|
|
"num_tokens": 17026915.0,
|
|
"step": 7580
|
|
},
|
|
{
|
|
"entropy": 5.237537527084351,
|
|
"epoch": 8.200108166576527,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 9.337540187712453e-05,
|
|
"loss": 4.3818,
|
|
"mean_token_accuracy": 0.2733478993177414,
|
|
"num_tokens": 17036135.0,
|
|
"step": 7585
|
|
},
|
|
{
|
|
"entropy": 5.244228315353394,
|
|
"epoch": 8.20551649540292,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 9.312256262372674e-05,
|
|
"loss": 4.3687,
|
|
"mean_token_accuracy": 0.2807093560695648,
|
|
"num_tokens": 17046660.0,
|
|
"step": 7590
|
|
},
|
|
{
|
|
"entropy": 5.196211671829223,
|
|
"epoch": 8.210924824229313,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 9.28703843133566e-05,
|
|
"loss": 4.3203,
|
|
"mean_token_accuracy": 0.28123302161693575,
|
|
"num_tokens": 17057923.0,
|
|
"step": 7595
|
|
},
|
|
{
|
|
"entropy": 5.255496454238892,
|
|
"epoch": 8.216333153055706,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 9.261886786243087e-05,
|
|
"loss": 4.4208,
|
|
"mean_token_accuracy": 0.2741557121276855,
|
|
"num_tokens": 17070053.0,
|
|
"step": 7600
|
|
},
|
|
{
|
|
"entropy": 5.29184365272522,
|
|
"epoch": 8.221741481882098,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 9.236801418496071e-05,
|
|
"loss": 4.4098,
|
|
"mean_token_accuracy": 0.27229327857494356,
|
|
"num_tokens": 17082311.0,
|
|
"step": 7605
|
|
},
|
|
{
|
|
"entropy": 5.271917533874512,
|
|
"epoch": 8.227149810708491,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 9.21178241925491e-05,
|
|
"loss": 4.4886,
|
|
"mean_token_accuracy": 0.2599450185894966,
|
|
"num_tokens": 17093073.0,
|
|
"step": 7610
|
|
},
|
|
{
|
|
"entropy": 5.247313594818115,
|
|
"epoch": 8.232558139534884,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 9.186829879438696e-05,
|
|
"loss": 4.3597,
|
|
"mean_token_accuracy": 0.27071188390254974,
|
|
"num_tokens": 17102753.0,
|
|
"step": 7615
|
|
},
|
|
{
|
|
"entropy": 5.2359003067016605,
|
|
"epoch": 8.237966468361277,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 9.161943889725021e-05,
|
|
"loss": 4.3914,
|
|
"mean_token_accuracy": 0.27498864829540254,
|
|
"num_tokens": 17113519.0,
|
|
"step": 7620
|
|
},
|
|
{
|
|
"entropy": 5.269698810577393,
|
|
"epoch": 8.24337479718767,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 9.137124540549638e-05,
|
|
"loss": 4.3864,
|
|
"mean_token_accuracy": 0.27518289238214494,
|
|
"num_tokens": 17125634.0,
|
|
"step": 7625
|
|
},
|
|
{
|
|
"entropy": 5.2179258346557615,
|
|
"epoch": 8.248783126014061,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 9.1123719221061e-05,
|
|
"loss": 4.3294,
|
|
"mean_token_accuracy": 0.2796433538198471,
|
|
"num_tokens": 17136982.0,
|
|
"step": 7630
|
|
},
|
|
{
|
|
"entropy": 5.213311672210693,
|
|
"epoch": 8.254191454840454,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 9.087686124345486e-05,
|
|
"loss": 4.4017,
|
|
"mean_token_accuracy": 0.265230293571949,
|
|
"num_tokens": 17148299.0,
|
|
"step": 7635
|
|
},
|
|
{
|
|
"entropy": 5.219297790527344,
|
|
"epoch": 8.259599783666847,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 9.063067236976055e-05,
|
|
"loss": 4.4151,
|
|
"mean_token_accuracy": 0.2716340392827988,
|
|
"num_tokens": 17161224.0,
|
|
"step": 7640
|
|
},
|
|
{
|
|
"entropy": 5.1686044216156,
|
|
"epoch": 8.26500811249324,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 9.038515349462887e-05,
|
|
"loss": 4.3326,
|
|
"mean_token_accuracy": 0.2796026110649109,
|
|
"num_tokens": 17172406.0,
|
|
"step": 7645
|
|
},
|
|
{
|
|
"entropy": 5.292226552963257,
|
|
"epoch": 8.270416441319632,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 9.014030551027601e-05,
|
|
"loss": 4.4107,
|
|
"mean_token_accuracy": 0.2771429270505905,
|
|
"num_tokens": 17182626.0,
|
|
"step": 7650
|
|
},
|
|
{
|
|
"entropy": 5.210595512390137,
|
|
"epoch": 8.275824770146025,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 8.989612930648014e-05,
|
|
"loss": 4.3838,
|
|
"mean_token_accuracy": 0.2739571645855904,
|
|
"num_tokens": 17195651.0,
|
|
"step": 7655
|
|
},
|
|
{
|
|
"entropy": 5.1909442901611325,
|
|
"epoch": 8.281233098972418,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 8.965262577057811e-05,
|
|
"loss": 4.4115,
|
|
"mean_token_accuracy": 0.27030020356178286,
|
|
"num_tokens": 17207321.0,
|
|
"step": 7660
|
|
},
|
|
{
|
|
"entropy": 5.209118270874024,
|
|
"epoch": 8.286641427798811,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 8.940979578746236e-05,
|
|
"loss": 4.3779,
|
|
"mean_token_accuracy": 0.2773276060819626,
|
|
"num_tokens": 17218212.0,
|
|
"step": 7665
|
|
},
|
|
{
|
|
"entropy": 5.273705339431762,
|
|
"epoch": 8.292049756625202,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 8.916764023957766e-05,
|
|
"loss": 4.4076,
|
|
"mean_token_accuracy": 0.2829126238822937,
|
|
"num_tokens": 17230095.0,
|
|
"step": 7670
|
|
},
|
|
{
|
|
"entropy": 5.31418194770813,
|
|
"epoch": 8.297458085451595,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 8.892616000691766e-05,
|
|
"loss": 4.4994,
|
|
"mean_token_accuracy": 0.26071294099092485,
|
|
"num_tokens": 17241380.0,
|
|
"step": 7675
|
|
},
|
|
{
|
|
"entropy": 5.232287263870239,
|
|
"epoch": 8.302866414277988,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 8.868535596702232e-05,
|
|
"loss": 4.3691,
|
|
"mean_token_accuracy": 0.27473134547472,
|
|
"num_tokens": 17252136.0,
|
|
"step": 7680
|
|
},
|
|
{
|
|
"entropy": 5.228588962554932,
|
|
"epoch": 8.30827474310438,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 8.844522899497387e-05,
|
|
"loss": 4.391,
|
|
"mean_token_accuracy": 0.27318202555179594,
|
|
"num_tokens": 17263150.0,
|
|
"step": 7685
|
|
},
|
|
{
|
|
"entropy": 5.156239414215088,
|
|
"epoch": 8.313683071930773,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 8.82057799633944e-05,
|
|
"loss": 4.356,
|
|
"mean_token_accuracy": 0.2779333233833313,
|
|
"num_tokens": 17274993.0,
|
|
"step": 7690
|
|
},
|
|
{
|
|
"entropy": 5.236821556091309,
|
|
"epoch": 8.319091400757166,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 8.796700974244218e-05,
|
|
"loss": 4.4003,
|
|
"mean_token_accuracy": 0.2698005959391594,
|
|
"num_tokens": 17286559.0,
|
|
"step": 7695
|
|
},
|
|
{
|
|
"entropy": 5.256543445587158,
|
|
"epoch": 8.32449972958356,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 8.772891919980881e-05,
|
|
"loss": 4.3621,
|
|
"mean_token_accuracy": 0.2786628529429436,
|
|
"num_tokens": 17298034.0,
|
|
"step": 7700
|
|
},
|
|
{
|
|
"entropy": 5.243308067321777,
|
|
"epoch": 8.329908058409952,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 8.74915092007158e-05,
|
|
"loss": 4.2905,
|
|
"mean_token_accuracy": 0.28707629144191743,
|
|
"num_tokens": 17309311.0,
|
|
"step": 7705
|
|
},
|
|
{
|
|
"entropy": 5.210727405548096,
|
|
"epoch": 8.335316387236343,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 8.725478060791175e-05,
|
|
"loss": 4.3595,
|
|
"mean_token_accuracy": 0.27717672735452653,
|
|
"num_tokens": 17320414.0,
|
|
"step": 7710
|
|
},
|
|
{
|
|
"entropy": 5.259465026855469,
|
|
"epoch": 8.340724716062736,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 8.701873428166874e-05,
|
|
"loss": 4.4445,
|
|
"mean_token_accuracy": 0.27204022705554964,
|
|
"num_tokens": 17330743.0,
|
|
"step": 7715
|
|
},
|
|
{
|
|
"entropy": 5.203076314926148,
|
|
"epoch": 8.346133044889129,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 8.678337107977976e-05,
|
|
"loss": 4.4294,
|
|
"mean_token_accuracy": 0.2719213977456093,
|
|
"num_tokens": 17341636.0,
|
|
"step": 7720
|
|
},
|
|
{
|
|
"entropy": 5.196039819717408,
|
|
"epoch": 8.351541373715522,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 8.654869185755532e-05,
|
|
"loss": 4.4055,
|
|
"mean_token_accuracy": 0.2695676162838936,
|
|
"num_tokens": 17352979.0,
|
|
"step": 7725
|
|
},
|
|
{
|
|
"entropy": 5.224801349639892,
|
|
"epoch": 8.356949702541915,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 8.631469746782011e-05,
|
|
"loss": 4.4183,
|
|
"mean_token_accuracy": 0.2684234485030174,
|
|
"num_tokens": 17364723.0,
|
|
"step": 7730
|
|
},
|
|
{
|
|
"entropy": 5.265634155273437,
|
|
"epoch": 8.362358031368307,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 8.608138876091036e-05,
|
|
"loss": 4.4545,
|
|
"mean_token_accuracy": 0.26871850788593293,
|
|
"num_tokens": 17375078.0,
|
|
"step": 7735
|
|
},
|
|
{
|
|
"entropy": 5.270746803283691,
|
|
"epoch": 8.3677663601947,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 8.584876658467041e-05,
|
|
"loss": 4.4699,
|
|
"mean_token_accuracy": 0.2669924110174179,
|
|
"num_tokens": 17386414.0,
|
|
"step": 7740
|
|
},
|
|
{
|
|
"entropy": 5.247923278808594,
|
|
"epoch": 8.373174689021093,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 8.561683178444979e-05,
|
|
"loss": 4.3901,
|
|
"mean_token_accuracy": 0.27611470967531204,
|
|
"num_tokens": 17397756.0,
|
|
"step": 7745
|
|
},
|
|
{
|
|
"entropy": 5.215674495697021,
|
|
"epoch": 8.378583017847484,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 8.53855852031002e-05,
|
|
"loss": 4.3475,
|
|
"mean_token_accuracy": 0.2810898020863533,
|
|
"num_tokens": 17408425.0,
|
|
"step": 7750
|
|
},
|
|
{
|
|
"entropy": 5.230283403396607,
|
|
"epoch": 8.383991346673877,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 8.515502768097204e-05,
|
|
"loss": 4.3938,
|
|
"mean_token_accuracy": 0.2764684855937958,
|
|
"num_tokens": 17420746.0,
|
|
"step": 7755
|
|
},
|
|
{
|
|
"entropy": 5.265897607803344,
|
|
"epoch": 8.38939967550027,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 8.492516005591204e-05,
|
|
"loss": 4.3357,
|
|
"mean_token_accuracy": 0.2852681130170822,
|
|
"num_tokens": 17431634.0,
|
|
"step": 7760
|
|
},
|
|
{
|
|
"entropy": 5.249261999130249,
|
|
"epoch": 8.394808004326663,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 8.469598316325971e-05,
|
|
"loss": 4.3549,
|
|
"mean_token_accuracy": 0.2787042915821075,
|
|
"num_tokens": 17442069.0,
|
|
"step": 7765
|
|
},
|
|
{
|
|
"entropy": 5.329693603515625,
|
|
"epoch": 8.400216333153056,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 8.446749783584432e-05,
|
|
"loss": 4.4711,
|
|
"mean_token_accuracy": 0.2622985377907753,
|
|
"num_tokens": 17451921.0,
|
|
"step": 7770
|
|
},
|
|
{
|
|
"entropy": 5.244504451751709,
|
|
"epoch": 8.405624661979449,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 8.423970490398206e-05,
|
|
"loss": 4.378,
|
|
"mean_token_accuracy": 0.2794926315546036,
|
|
"num_tokens": 17463804.0,
|
|
"step": 7775
|
|
},
|
|
{
|
|
"entropy": 5.225708723068237,
|
|
"epoch": 8.411032990805841,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 8.401260519547305e-05,
|
|
"loss": 4.3669,
|
|
"mean_token_accuracy": 0.2802198722958565,
|
|
"num_tokens": 17474682.0,
|
|
"step": 7780
|
|
},
|
|
{
|
|
"entropy": 5.186916875839233,
|
|
"epoch": 8.416441319632234,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 8.378619953559807e-05,
|
|
"loss": 4.3807,
|
|
"mean_token_accuracy": 0.2746036395430565,
|
|
"num_tokens": 17486272.0,
|
|
"step": 7785
|
|
},
|
|
{
|
|
"entropy": 5.294495677947998,
|
|
"epoch": 8.421849648458625,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 8.356048874711589e-05,
|
|
"loss": 4.5028,
|
|
"mean_token_accuracy": 0.25705422163009645,
|
|
"num_tokens": 17498422.0,
|
|
"step": 7790
|
|
},
|
|
{
|
|
"entropy": 5.238483953475952,
|
|
"epoch": 8.427257977285018,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 8.333547365026e-05,
|
|
"loss": 4.4496,
|
|
"mean_token_accuracy": 0.26462914794683456,
|
|
"num_tokens": 17509923.0,
|
|
"step": 7795
|
|
},
|
|
{
|
|
"entropy": 5.225796890258789,
|
|
"epoch": 8.432666306111411,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 8.311115506273579e-05,
|
|
"loss": 4.3597,
|
|
"mean_token_accuracy": 0.28231186866760255,
|
|
"num_tokens": 17520962.0,
|
|
"step": 7800
|
|
},
|
|
{
|
|
"entropy": 5.230301570892334,
|
|
"epoch": 8.438074634937804,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 8.288753379971756e-05,
|
|
"loss": 4.3682,
|
|
"mean_token_accuracy": 0.2796242982149124,
|
|
"num_tokens": 17532336.0,
|
|
"step": 7805
|
|
},
|
|
{
|
|
"entropy": 5.23316068649292,
|
|
"epoch": 8.443482963764197,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 8.26646106738454e-05,
|
|
"loss": 4.3407,
|
|
"mean_token_accuracy": 0.2812255650758743,
|
|
"num_tokens": 17541684.0,
|
|
"step": 7810
|
|
},
|
|
{
|
|
"entropy": 5.240253973007202,
|
|
"epoch": 8.44889129259059,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 8.244238649522249e-05,
|
|
"loss": 4.2813,
|
|
"mean_token_accuracy": 0.28954372406005857,
|
|
"num_tokens": 17551788.0,
|
|
"step": 7815
|
|
},
|
|
{
|
|
"entropy": 5.234849977493286,
|
|
"epoch": 8.454299621416983,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 8.222086207141214e-05,
|
|
"loss": 4.4591,
|
|
"mean_token_accuracy": 0.2664370432496071,
|
|
"num_tokens": 17563292.0,
|
|
"step": 7820
|
|
},
|
|
{
|
|
"entropy": 5.256240034103394,
|
|
"epoch": 8.459707950243375,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 8.200003820743449e-05,
|
|
"loss": 4.3798,
|
|
"mean_token_accuracy": 0.28862873911857606,
|
|
"num_tokens": 17574305.0,
|
|
"step": 7825
|
|
},
|
|
{
|
|
"entropy": 5.2228233337402346,
|
|
"epoch": 8.465116279069768,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 8.177991570576396e-05,
|
|
"loss": 4.3526,
|
|
"mean_token_accuracy": 0.2800112724304199,
|
|
"num_tokens": 17584448.0,
|
|
"step": 7830
|
|
},
|
|
{
|
|
"entropy": 5.236324834823608,
|
|
"epoch": 8.47052460789616,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 8.156049536632629e-05,
|
|
"loss": 4.3752,
|
|
"mean_token_accuracy": 0.27668683528900145,
|
|
"num_tokens": 17595877.0,
|
|
"step": 7835
|
|
},
|
|
{
|
|
"entropy": 5.221021127700806,
|
|
"epoch": 8.475932936722552,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 8.13417779864955e-05,
|
|
"loss": 4.4036,
|
|
"mean_token_accuracy": 0.2745181769132614,
|
|
"num_tokens": 17608164.0,
|
|
"step": 7840
|
|
},
|
|
{
|
|
"entropy": 5.207059288024903,
|
|
"epoch": 8.481341265548945,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 8.112376436109107e-05,
|
|
"loss": 4.3531,
|
|
"mean_token_accuracy": 0.2772742807865143,
|
|
"num_tokens": 17619098.0,
|
|
"step": 7845
|
|
},
|
|
{
|
|
"entropy": 5.185477685928345,
|
|
"epoch": 8.486749594375338,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 8.090645528237493e-05,
|
|
"loss": 4.3554,
|
|
"mean_token_accuracy": 0.2703353762626648,
|
|
"num_tokens": 17630032.0,
|
|
"step": 7850
|
|
},
|
|
{
|
|
"entropy": 5.251085567474365,
|
|
"epoch": 8.49215792320173,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 8.068985154004879e-05,
|
|
"loss": 4.3792,
|
|
"mean_token_accuracy": 0.2746999129652977,
|
|
"num_tokens": 17641127.0,
|
|
"step": 7855
|
|
},
|
|
{
|
|
"entropy": 5.271446752548218,
|
|
"epoch": 8.497566252028124,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 8.047395392125126e-05,
|
|
"loss": 4.4197,
|
|
"mean_token_accuracy": 0.27077027410268784,
|
|
"num_tokens": 17652944.0,
|
|
"step": 7860
|
|
},
|
|
{
|
|
"entropy": 5.171000289916992,
|
|
"epoch": 8.502974580854517,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 8.02587632105547e-05,
|
|
"loss": 4.3095,
|
|
"mean_token_accuracy": 0.27184768170118334,
|
|
"num_tokens": 17664345.0,
|
|
"step": 7865
|
|
},
|
|
{
|
|
"entropy": 5.289643955230713,
|
|
"epoch": 8.50838290968091,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 8.004428018996264e-05,
|
|
"loss": 4.3734,
|
|
"mean_token_accuracy": 0.27643611282110214,
|
|
"num_tokens": 17674314.0,
|
|
"step": 7870
|
|
},
|
|
{
|
|
"entropy": 5.190473842620849,
|
|
"epoch": 8.5137912385073,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 7.983050563890692e-05,
|
|
"loss": 4.3995,
|
|
"mean_token_accuracy": 0.2732491731643677,
|
|
"num_tokens": 17685338.0,
|
|
"step": 7875
|
|
},
|
|
{
|
|
"entropy": 5.2136908054351805,
|
|
"epoch": 8.519199567333693,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 7.961744033424473e-05,
|
|
"loss": 4.4123,
|
|
"mean_token_accuracy": 0.2737598612904549,
|
|
"num_tokens": 17695954.0,
|
|
"step": 7880
|
|
},
|
|
{
|
|
"entropy": 5.213007545471191,
|
|
"epoch": 8.524607896160086,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 7.940508505025586e-05,
|
|
"loss": 4.3936,
|
|
"mean_token_accuracy": 0.26886777579784393,
|
|
"num_tokens": 17706627.0,
|
|
"step": 7885
|
|
},
|
|
{
|
|
"entropy": 5.235505199432373,
|
|
"epoch": 8.530016224986479,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 7.919344055863997e-05,
|
|
"loss": 4.4246,
|
|
"mean_token_accuracy": 0.27497789859771726,
|
|
"num_tokens": 17716810.0,
|
|
"step": 7890
|
|
},
|
|
{
|
|
"entropy": 5.291230583190918,
|
|
"epoch": 8.535424553812872,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 7.89825076285135e-05,
|
|
"loss": 4.4683,
|
|
"mean_token_accuracy": 0.27097458243370054,
|
|
"num_tokens": 17727429.0,
|
|
"step": 7895
|
|
},
|
|
{
|
|
"entropy": 5.24913010597229,
|
|
"epoch": 8.540832882639265,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 7.877228702640739e-05,
|
|
"loss": 4.3647,
|
|
"mean_token_accuracy": 0.2827416032552719,
|
|
"num_tokens": 17737198.0,
|
|
"step": 7900
|
|
},
|
|
{
|
|
"entropy": 5.22300124168396,
|
|
"epoch": 8.546241211465658,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 7.856277951626363e-05,
|
|
"loss": 4.3906,
|
|
"mean_token_accuracy": 0.2812996804714203,
|
|
"num_tokens": 17747288.0,
|
|
"step": 7905
|
|
},
|
|
{
|
|
"entropy": 5.237582778930664,
|
|
"epoch": 8.55164954029205,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 7.835398585943307e-05,
|
|
"loss": 4.465,
|
|
"mean_token_accuracy": 0.26551628261804583,
|
|
"num_tokens": 17759238.0,
|
|
"step": 7910
|
|
},
|
|
{
|
|
"entropy": 5.2212389469146725,
|
|
"epoch": 8.557057869118442,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 7.814590681467235e-05,
|
|
"loss": 4.3578,
|
|
"mean_token_accuracy": 0.27837354838848116,
|
|
"num_tokens": 17769306.0,
|
|
"step": 7915
|
|
},
|
|
{
|
|
"entropy": 5.20025601387024,
|
|
"epoch": 8.562466197944834,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 7.793854313814114e-05,
|
|
"loss": 4.3995,
|
|
"mean_token_accuracy": 0.26891224533319474,
|
|
"num_tokens": 17780446.0,
|
|
"step": 7920
|
|
},
|
|
{
|
|
"entropy": 5.236232328414917,
|
|
"epoch": 8.567874526771227,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 7.773189558339956e-05,
|
|
"loss": 4.3461,
|
|
"mean_token_accuracy": 0.2760110378265381,
|
|
"num_tokens": 17791217.0,
|
|
"step": 7925
|
|
},
|
|
{
|
|
"entropy": 5.197807884216308,
|
|
"epoch": 8.57328285559762,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 7.752596490140537e-05,
|
|
"loss": 4.2808,
|
|
"mean_token_accuracy": 0.28877498507499694,
|
|
"num_tokens": 17802755.0,
|
|
"step": 7930
|
|
},
|
|
{
|
|
"entropy": 5.190744781494141,
|
|
"epoch": 8.578691184424013,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 7.732075184051098e-05,
|
|
"loss": 4.3644,
|
|
"mean_token_accuracy": 0.2841712892055511,
|
|
"num_tokens": 17814466.0,
|
|
"step": 7935
|
|
},
|
|
{
|
|
"entropy": 5.3480353355407715,
|
|
"epoch": 8.584099513250406,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 7.711625714646127e-05,
|
|
"loss": 4.5253,
|
|
"mean_token_accuracy": 0.26552720963954923,
|
|
"num_tokens": 17826561.0,
|
|
"step": 7940
|
|
},
|
|
{
|
|
"entropy": 5.275240659713745,
|
|
"epoch": 8.589507842076799,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 7.691248156239041e-05,
|
|
"loss": 4.4284,
|
|
"mean_token_accuracy": 0.2738826394081116,
|
|
"num_tokens": 17837219.0,
|
|
"step": 7945
|
|
},
|
|
{
|
|
"entropy": 5.233570575714111,
|
|
"epoch": 8.594916170903192,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 7.670942582881931e-05,
|
|
"loss": 4.3673,
|
|
"mean_token_accuracy": 0.2775705769658089,
|
|
"num_tokens": 17848173.0,
|
|
"step": 7950
|
|
},
|
|
{
|
|
"entropy": 5.178972005844116,
|
|
"epoch": 8.600324499729584,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 7.650709068365295e-05,
|
|
"loss": 4.3269,
|
|
"mean_token_accuracy": 0.2801755964756012,
|
|
"num_tokens": 17858073.0,
|
|
"step": 7955
|
|
},
|
|
{
|
|
"entropy": 5.278064632415772,
|
|
"epoch": 8.605732828555976,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 7.630547686217774e-05,
|
|
"loss": 4.4647,
|
|
"mean_token_accuracy": 0.27346938848495483,
|
|
"num_tokens": 17871842.0,
|
|
"step": 7960
|
|
},
|
|
{
|
|
"entropy": 5.181892538070679,
|
|
"epoch": 8.611141157382368,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 7.61045850970588e-05,
|
|
"loss": 4.3625,
|
|
"mean_token_accuracy": 0.2778499245643616,
|
|
"num_tokens": 17883217.0,
|
|
"step": 7965
|
|
},
|
|
{
|
|
"entropy": 5.227734661102295,
|
|
"epoch": 8.616549486208761,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 7.59044161183373e-05,
|
|
"loss": 4.3835,
|
|
"mean_token_accuracy": 0.2741628125309944,
|
|
"num_tokens": 17895606.0,
|
|
"step": 7970
|
|
},
|
|
{
|
|
"entropy": 5.2346655368804935,
|
|
"epoch": 8.621957815035154,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 7.570497065342762e-05,
|
|
"loss": 4.3299,
|
|
"mean_token_accuracy": 0.27707968950271605,
|
|
"num_tokens": 17906444.0,
|
|
"step": 7975
|
|
},
|
|
{
|
|
"entropy": 5.281631231307983,
|
|
"epoch": 8.627366143861547,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 7.550624942711517e-05,
|
|
"loss": 4.4333,
|
|
"mean_token_accuracy": 0.27240731120109557,
|
|
"num_tokens": 17917019.0,
|
|
"step": 7980
|
|
},
|
|
{
|
|
"entropy": 5.181613445281982,
|
|
"epoch": 8.63277447268794,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 7.530825316155337e-05,
|
|
"loss": 4.3119,
|
|
"mean_token_accuracy": 0.28016529977321625,
|
|
"num_tokens": 17927930.0,
|
|
"step": 7985
|
|
},
|
|
{
|
|
"entropy": 5.215329837799072,
|
|
"epoch": 8.638182801514333,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 7.511098257626101e-05,
|
|
"loss": 4.3579,
|
|
"mean_token_accuracy": 0.27332532703876494,
|
|
"num_tokens": 17938820.0,
|
|
"step": 7990
|
|
},
|
|
{
|
|
"entropy": 5.256243944168091,
|
|
"epoch": 8.643591130340724,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 7.491443838811991e-05,
|
|
"loss": 4.4215,
|
|
"mean_token_accuracy": 0.27570990175008775,
|
|
"num_tokens": 17949334.0,
|
|
"step": 7995
|
|
},
|
|
{
|
|
"entropy": 5.212791919708252,
|
|
"epoch": 8.648999459167117,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 7.47186213113721e-05,
|
|
"loss": 4.3941,
|
|
"mean_token_accuracy": 0.2721568986773491,
|
|
"num_tokens": 17961454.0,
|
|
"step": 8000
|
|
},
|
|
{
|
|
"epoch": 8.648999459167117,
|
|
"eval_entropy": 5.161833815141158,
|
|
"eval_loss": 5.890415191650391,
|
|
"eval_mean_token_accuracy": 0.18021582644432782,
|
|
"eval_num_tokens": 17961454.0,
|
|
"eval_runtime": 1.9459,
|
|
"eval_samples_per_second": 1806.393,
|
|
"eval_steps_per_second": 226.12,
|
|
"step": 8000
|
|
},
|
|
{
|
|
"entropy": 5.2061888694763185,
|
|
"epoch": 8.65440778799351,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 7.452353205761725e-05,
|
|
"loss": 4.3837,
|
|
"mean_token_accuracy": 0.27418379336595533,
|
|
"num_tokens": 17972430.0,
|
|
"step": 8005
|
|
},
|
|
{
|
|
"entropy": 5.224462270736694,
|
|
"epoch": 8.659816116819902,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 7.432917133581019e-05,
|
|
"loss": 4.4003,
|
|
"mean_token_accuracy": 0.26904485523700716,
|
|
"num_tokens": 17983849.0,
|
|
"step": 8010
|
|
},
|
|
{
|
|
"entropy": 5.302052593231201,
|
|
"epoch": 8.665224445646295,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 7.413553985225824e-05,
|
|
"loss": 4.4456,
|
|
"mean_token_accuracy": 0.2723198503255844,
|
|
"num_tokens": 17994791.0,
|
|
"step": 8015
|
|
},
|
|
{
|
|
"entropy": 5.201394414901733,
|
|
"epoch": 8.670632774472688,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 7.394263831061865e-05,
|
|
"loss": 4.327,
|
|
"mean_token_accuracy": 0.2867824658751488,
|
|
"num_tokens": 18005068.0,
|
|
"step": 8020
|
|
},
|
|
{
|
|
"entropy": 5.292807102203369,
|
|
"epoch": 8.676041103299081,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 7.375046741189611e-05,
|
|
"loss": 4.4743,
|
|
"mean_token_accuracy": 0.2594904348254204,
|
|
"num_tokens": 18016610.0,
|
|
"step": 8025
|
|
},
|
|
{
|
|
"entropy": 5.2265081882476805,
|
|
"epoch": 8.681449432125474,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 7.355902785444002e-05,
|
|
"loss": 4.385,
|
|
"mean_token_accuracy": 0.27682053595781325,
|
|
"num_tokens": 18028854.0,
|
|
"step": 8030
|
|
},
|
|
{
|
|
"entropy": 5.2796917915344235,
|
|
"epoch": 8.686857760951867,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 7.336832033394217e-05,
|
|
"loss": 4.4374,
|
|
"mean_token_accuracy": 0.2706138134002686,
|
|
"num_tokens": 18039766.0,
|
|
"step": 8035
|
|
},
|
|
{
|
|
"entropy": 5.26754789352417,
|
|
"epoch": 8.692266089778258,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 7.317834554343429e-05,
|
|
"loss": 4.3649,
|
|
"mean_token_accuracy": 0.2785991430282593,
|
|
"num_tokens": 18049358.0,
|
|
"step": 8040
|
|
},
|
|
{
|
|
"entropy": 5.220230627059936,
|
|
"epoch": 8.69767441860465,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 7.298910417328503e-05,
|
|
"loss": 4.3396,
|
|
"mean_token_accuracy": 0.2797338664531708,
|
|
"num_tokens": 18060274.0,
|
|
"step": 8045
|
|
},
|
|
{
|
|
"entropy": 5.202096652984619,
|
|
"epoch": 8.703082747431043,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 7.280059691119806e-05,
|
|
"loss": 4.3441,
|
|
"mean_token_accuracy": 0.2757891371846199,
|
|
"num_tokens": 18070688.0,
|
|
"step": 8050
|
|
},
|
|
{
|
|
"entropy": 5.263065767288208,
|
|
"epoch": 8.708491076257436,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 7.261282444220917e-05,
|
|
"loss": 4.417,
|
|
"mean_token_accuracy": 0.2734151378273964,
|
|
"num_tokens": 18081739.0,
|
|
"step": 8055
|
|
},
|
|
{
|
|
"entropy": 5.27297739982605,
|
|
"epoch": 8.71389940508383,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 7.242578744868395e-05,
|
|
"loss": 4.4945,
|
|
"mean_token_accuracy": 0.2637551099061966,
|
|
"num_tokens": 18092978.0,
|
|
"step": 8060
|
|
},
|
|
{
|
|
"entropy": 5.229227018356323,
|
|
"epoch": 8.719307733910222,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 7.223948661031534e-05,
|
|
"loss": 4.4218,
|
|
"mean_token_accuracy": 0.2686552360653877,
|
|
"num_tokens": 18104649.0,
|
|
"step": 8065
|
|
},
|
|
{
|
|
"entropy": 5.2275176525115965,
|
|
"epoch": 8.724716062736615,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 7.205392260412088e-05,
|
|
"loss": 4.374,
|
|
"mean_token_accuracy": 0.27921224683523177,
|
|
"num_tokens": 18115640.0,
|
|
"step": 8070
|
|
},
|
|
{
|
|
"entropy": 5.18868203163147,
|
|
"epoch": 8.730124391563008,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.186909610444065e-05,
|
|
"loss": 4.4006,
|
|
"mean_token_accuracy": 0.2708303615450859,
|
|
"num_tokens": 18126599.0,
|
|
"step": 8075
|
|
},
|
|
{
|
|
"entropy": 5.2919519424438475,
|
|
"epoch": 8.735532720389399,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 7.168500778293466e-05,
|
|
"loss": 4.4438,
|
|
"mean_token_accuracy": 0.27533996999263766,
|
|
"num_tokens": 18137961.0,
|
|
"step": 8080
|
|
},
|
|
{
|
|
"entropy": 5.392880249023437,
|
|
"epoch": 8.740941049215792,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 7.15016583085802e-05,
|
|
"loss": 4.5198,
|
|
"mean_token_accuracy": 0.269387012720108,
|
|
"num_tokens": 18150167.0,
|
|
"step": 8085
|
|
},
|
|
{
|
|
"entropy": 5.369137287139893,
|
|
"epoch": 8.746349378042185,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 7.131904834766973e-05,
|
|
"loss": 4.4539,
|
|
"mean_token_accuracy": 0.2753928184509277,
|
|
"num_tokens": 18162320.0,
|
|
"step": 8090
|
|
},
|
|
{
|
|
"entropy": 5.244376277923584,
|
|
"epoch": 8.751757706868577,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 7.113717856380824e-05,
|
|
"loss": 4.3922,
|
|
"mean_token_accuracy": 0.2785827353596687,
|
|
"num_tokens": 18174213.0,
|
|
"step": 8095
|
|
},
|
|
{
|
|
"entropy": 5.234084749221802,
|
|
"epoch": 8.75716603569497,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 7.095604961791095e-05,
|
|
"loss": 4.3771,
|
|
"mean_token_accuracy": 0.276390939950943,
|
|
"num_tokens": 18184071.0,
|
|
"step": 8100
|
|
},
|
|
{
|
|
"entropy": 5.384398412704468,
|
|
"epoch": 8.762574364521363,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 7.077566216820089e-05,
|
|
"loss": 4.563,
|
|
"mean_token_accuracy": 0.2615682289004326,
|
|
"num_tokens": 18197479.0,
|
|
"step": 8105
|
|
},
|
|
{
|
|
"entropy": 5.248125171661377,
|
|
"epoch": 8.767982693347756,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 7.059601687020647e-05,
|
|
"loss": 4.4967,
|
|
"mean_token_accuracy": 0.25774045586586,
|
|
"num_tokens": 18209824.0,
|
|
"step": 8110
|
|
},
|
|
{
|
|
"entropy": 5.236955165863037,
|
|
"epoch": 8.773391022174149,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 7.041711437675908e-05,
|
|
"loss": 4.3996,
|
|
"mean_token_accuracy": 0.28496327698230745,
|
|
"num_tokens": 18221788.0,
|
|
"step": 8115
|
|
},
|
|
{
|
|
"entropy": 5.211531400680542,
|
|
"epoch": 8.77879935100054,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 7.023895533799087e-05,
|
|
"loss": 4.3849,
|
|
"mean_token_accuracy": 0.2695922046899796,
|
|
"num_tokens": 18233755.0,
|
|
"step": 8120
|
|
},
|
|
{
|
|
"entropy": 5.305582761764526,
|
|
"epoch": 8.784207679826933,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 7.006154040133211e-05,
|
|
"loss": 4.4503,
|
|
"mean_token_accuracy": 0.2699734464287758,
|
|
"num_tokens": 18244591.0,
|
|
"step": 8125
|
|
},
|
|
{
|
|
"entropy": 5.272140264511108,
|
|
"epoch": 8.789616008653326,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 6.98848702115091e-05,
|
|
"loss": 4.3541,
|
|
"mean_token_accuracy": 0.27224867343902587,
|
|
"num_tokens": 18255928.0,
|
|
"step": 8130
|
|
},
|
|
{
|
|
"entropy": 5.252304410934448,
|
|
"epoch": 8.795024337479719,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 6.970894541054182e-05,
|
|
"loss": 4.377,
|
|
"mean_token_accuracy": 0.2829632952809334,
|
|
"num_tokens": 18267288.0,
|
|
"step": 8135
|
|
},
|
|
{
|
|
"entropy": 5.288474798202515,
|
|
"epoch": 8.800432666306111,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 6.953376663774129e-05,
|
|
"loss": 4.5028,
|
|
"mean_token_accuracy": 0.27201053202152253,
|
|
"num_tokens": 18279045.0,
|
|
"step": 8140
|
|
},
|
|
{
|
|
"entropy": 5.22249059677124,
|
|
"epoch": 8.805840995132504,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 6.935933452970762e-05,
|
|
"loss": 4.4312,
|
|
"mean_token_accuracy": 0.2801878720521927,
|
|
"num_tokens": 18291378.0,
|
|
"step": 8145
|
|
},
|
|
{
|
|
"entropy": 5.226069927215576,
|
|
"epoch": 8.811249323958897,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 6.918564972032745e-05,
|
|
"loss": 4.3536,
|
|
"mean_token_accuracy": 0.27620834559202195,
|
|
"num_tokens": 18303089.0,
|
|
"step": 8150
|
|
},
|
|
{
|
|
"entropy": 5.288994550704956,
|
|
"epoch": 8.81665765278529,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 6.901271284077188e-05,
|
|
"loss": 4.3972,
|
|
"mean_token_accuracy": 0.27214264422655104,
|
|
"num_tokens": 18313656.0,
|
|
"step": 8155
|
|
},
|
|
{
|
|
"entropy": 5.217668581008911,
|
|
"epoch": 8.822065981611683,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 6.884052451949382e-05,
|
|
"loss": 4.3788,
|
|
"mean_token_accuracy": 0.28044068813323975,
|
|
"num_tokens": 18324176.0,
|
|
"step": 8160
|
|
},
|
|
{
|
|
"entropy": 5.250646591186523,
|
|
"epoch": 8.827474310438074,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 6.866908538222611e-05,
|
|
"loss": 4.485,
|
|
"mean_token_accuracy": 0.2660000428557396,
|
|
"num_tokens": 18336704.0,
|
|
"step": 8165
|
|
},
|
|
{
|
|
"entropy": 5.2640523433685305,
|
|
"epoch": 8.832882639264467,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 6.84983960519789e-05,
|
|
"loss": 4.3957,
|
|
"mean_token_accuracy": 0.2739565372467041,
|
|
"num_tokens": 18346992.0,
|
|
"step": 8170
|
|
},
|
|
{
|
|
"entropy": 5.224236154556275,
|
|
"epoch": 8.83829096809086,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 6.83284571490377e-05,
|
|
"loss": 4.3353,
|
|
"mean_token_accuracy": 0.28455475717782974,
|
|
"num_tokens": 18358244.0,
|
|
"step": 8175
|
|
},
|
|
{
|
|
"entropy": 5.228202390670776,
|
|
"epoch": 8.843699296917253,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 6.815926929096077e-05,
|
|
"loss": 4.3997,
|
|
"mean_token_accuracy": 0.2765922173857689,
|
|
"num_tokens": 18368951.0,
|
|
"step": 8180
|
|
},
|
|
{
|
|
"entropy": 5.256389808654785,
|
|
"epoch": 8.849107625743645,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 6.799083309257726e-05,
|
|
"loss": 4.3945,
|
|
"mean_token_accuracy": 0.26765153408050535,
|
|
"num_tokens": 18379927.0,
|
|
"step": 8185
|
|
},
|
|
{
|
|
"entropy": 5.180625581741333,
|
|
"epoch": 8.854515954570038,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 6.782314916598465e-05,
|
|
"loss": 4.3075,
|
|
"mean_token_accuracy": 0.2807507276535034,
|
|
"num_tokens": 18390396.0,
|
|
"step": 8190
|
|
},
|
|
{
|
|
"entropy": 5.199189710617065,
|
|
"epoch": 8.859924283396431,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 6.765621812054675e-05,
|
|
"loss": 4.3894,
|
|
"mean_token_accuracy": 0.27569980919361115,
|
|
"num_tokens": 18401640.0,
|
|
"step": 8195
|
|
},
|
|
{
|
|
"entropy": 5.250468635559082,
|
|
"epoch": 8.865332612222822,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 6.749004056289132e-05,
|
|
"loss": 4.433,
|
|
"mean_token_accuracy": 0.27182183861732484,
|
|
"num_tokens": 18413032.0,
|
|
"step": 8200
|
|
},
|
|
{
|
|
"entropy": 5.298527956008911,
|
|
"epoch": 8.870740941049215,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 6.732461709690811e-05,
|
|
"loss": 4.4499,
|
|
"mean_token_accuracy": 0.2689834743738174,
|
|
"num_tokens": 18424249.0,
|
|
"step": 8205
|
|
},
|
|
{
|
|
"entropy": 5.2517828941345215,
|
|
"epoch": 8.876149269875608,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 6.715994832374619e-05,
|
|
"loss": 4.4369,
|
|
"mean_token_accuracy": 0.2671634808182716,
|
|
"num_tokens": 18435532.0,
|
|
"step": 8210
|
|
},
|
|
{
|
|
"entropy": 5.226087808609009,
|
|
"epoch": 8.881557598702,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 6.699603484181245e-05,
|
|
"loss": 4.423,
|
|
"mean_token_accuracy": 0.279066264629364,
|
|
"num_tokens": 18447433.0,
|
|
"step": 8215
|
|
},
|
|
{
|
|
"entropy": 5.291144227981567,
|
|
"epoch": 8.886965927528394,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 6.683287724676874e-05,
|
|
"loss": 4.4541,
|
|
"mean_token_accuracy": 0.2668018788099289,
|
|
"num_tokens": 18459348.0,
|
|
"step": 8220
|
|
},
|
|
{
|
|
"entropy": 5.229798650741577,
|
|
"epoch": 8.892374256354787,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 6.667047613153018e-05,
|
|
"loss": 4.3605,
|
|
"mean_token_accuracy": 0.27678524851799013,
|
|
"num_tokens": 18469626.0,
|
|
"step": 8225
|
|
},
|
|
{
|
|
"entropy": 5.171689558029175,
|
|
"epoch": 8.89778258518118,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 6.650883208626275e-05,
|
|
"loss": 4.3535,
|
|
"mean_token_accuracy": 0.2774387449026108,
|
|
"num_tokens": 18480280.0,
|
|
"step": 8230
|
|
},
|
|
{
|
|
"entropy": 5.27275390625,
|
|
"epoch": 8.903190914007572,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 6.634794569838136e-05,
|
|
"loss": 4.4487,
|
|
"mean_token_accuracy": 0.27441560178995134,
|
|
"num_tokens": 18491169.0,
|
|
"step": 8235
|
|
},
|
|
{
|
|
"entropy": 5.262706327438354,
|
|
"epoch": 8.908599242833965,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.618781755254749e-05,
|
|
"loss": 4.4074,
|
|
"mean_token_accuracy": 0.2747210055589676,
|
|
"num_tokens": 18502987.0,
|
|
"step": 8240
|
|
},
|
|
{
|
|
"entropy": 5.192014837265015,
|
|
"epoch": 8.914007571660356,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 6.60284482306672e-05,
|
|
"loss": 4.3652,
|
|
"mean_token_accuracy": 0.2755765184760094,
|
|
"num_tokens": 18514111.0,
|
|
"step": 8245
|
|
},
|
|
{
|
|
"entropy": 5.259258031845093,
|
|
"epoch": 8.919415900486749,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 6.586983831188888e-05,
|
|
"loss": 4.4011,
|
|
"mean_token_accuracy": 0.2639654025435448,
|
|
"num_tokens": 18524802.0,
|
|
"step": 8250
|
|
},
|
|
{
|
|
"entropy": 5.18294997215271,
|
|
"epoch": 8.924824229313142,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 6.571198837260148e-05,
|
|
"loss": 4.279,
|
|
"mean_token_accuracy": 0.28173806965351106,
|
|
"num_tokens": 18535033.0,
|
|
"step": 8255
|
|
},
|
|
{
|
|
"entropy": 5.2232567310333256,
|
|
"epoch": 8.930232558139535,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 6.555489898643201e-05,
|
|
"loss": 4.3513,
|
|
"mean_token_accuracy": 0.27498531639575957,
|
|
"num_tokens": 18545992.0,
|
|
"step": 8260
|
|
},
|
|
{
|
|
"entropy": 5.265142440795898,
|
|
"epoch": 8.935640886965928,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 6.539857072424357e-05,
|
|
"loss": 4.3872,
|
|
"mean_token_accuracy": 0.2737391531467438,
|
|
"num_tokens": 18557034.0,
|
|
"step": 8265
|
|
},
|
|
{
|
|
"entropy": 5.261191749572754,
|
|
"epoch": 8.94104921579232,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 6.524300415413354e-05,
|
|
"loss": 4.4721,
|
|
"mean_token_accuracy": 0.2628531590104103,
|
|
"num_tokens": 18569772.0,
|
|
"step": 8270
|
|
},
|
|
{
|
|
"entropy": 5.219944715499878,
|
|
"epoch": 8.946457544618713,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 6.508819984143111e-05,
|
|
"loss": 4.3358,
|
|
"mean_token_accuracy": 0.27514045834541323,
|
|
"num_tokens": 18581366.0,
|
|
"step": 8275
|
|
},
|
|
{
|
|
"entropy": 5.183649635314941,
|
|
"epoch": 8.951865873445106,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 6.493415834869555e-05,
|
|
"loss": 4.3413,
|
|
"mean_token_accuracy": 0.27045978903770446,
|
|
"num_tokens": 18592538.0,
|
|
"step": 8280
|
|
},
|
|
{
|
|
"entropy": 5.2639350414276125,
|
|
"epoch": 8.957274202271497,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 6.478088023571403e-05,
|
|
"loss": 4.4997,
|
|
"mean_token_accuracy": 0.26455006897449496,
|
|
"num_tokens": 18604060.0,
|
|
"step": 8285
|
|
},
|
|
{
|
|
"entropy": 5.304011821746826,
|
|
"epoch": 8.96268253109789,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 6.462836605949945e-05,
|
|
"loss": 4.4405,
|
|
"mean_token_accuracy": 0.26662241369485856,
|
|
"num_tokens": 18614038.0,
|
|
"step": 8290
|
|
},
|
|
{
|
|
"entropy": 5.3179931640625,
|
|
"epoch": 8.968090859924283,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 6.447661637428882e-05,
|
|
"loss": 4.4287,
|
|
"mean_token_accuracy": 0.278962804377079,
|
|
"num_tokens": 18624428.0,
|
|
"step": 8295
|
|
},
|
|
{
|
|
"entropy": 5.283885908126831,
|
|
"epoch": 8.973499188750676,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 6.432563173154082e-05,
|
|
"loss": 4.4191,
|
|
"mean_token_accuracy": 0.2719813883304596,
|
|
"num_tokens": 18636738.0,
|
|
"step": 8300
|
|
},
|
|
{
|
|
"entropy": 5.263170909881592,
|
|
"epoch": 8.978907517577069,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 6.417541267993391e-05,
|
|
"loss": 4.3916,
|
|
"mean_token_accuracy": 0.26805350184440613,
|
|
"num_tokens": 18648318.0,
|
|
"step": 8305
|
|
},
|
|
{
|
|
"entropy": 5.248971366882325,
|
|
"epoch": 8.984315846403462,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 6.402595976536451e-05,
|
|
"loss": 4.3916,
|
|
"mean_token_accuracy": 0.2771768897771835,
|
|
"num_tokens": 18659616.0,
|
|
"step": 8310
|
|
},
|
|
{
|
|
"entropy": 5.274503898620606,
|
|
"epoch": 8.989724175229854,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 6.387727353094492e-05,
|
|
"loss": 4.4587,
|
|
"mean_token_accuracy": 0.2656686007976532,
|
|
"num_tokens": 18671787.0,
|
|
"step": 8315
|
|
},
|
|
{
|
|
"entropy": 5.30846529006958,
|
|
"epoch": 8.995132504056247,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 6.372935451700119e-05,
|
|
"loss": 4.4389,
|
|
"mean_token_accuracy": 0.2739927902817726,
|
|
"num_tokens": 18683668.0,
|
|
"step": 8320
|
|
},
|
|
{
|
|
"entropy": 5.242831812964545,
|
|
"epoch": 9.0,
|
|
"grad_norm": 3.171875,
|
|
"learning_rate": 6.358220326107135e-05,
|
|
"loss": 4.4657,
|
|
"mean_token_accuracy": 0.2683419966035419,
|
|
"num_tokens": 18692001.0,
|
|
"step": 8325
|
|
},
|
|
{
|
|
"entropy": 5.253059720993042,
|
|
"epoch": 9.005408328826393,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 6.343582029790345e-05,
|
|
"loss": 4.2977,
|
|
"mean_token_accuracy": 0.28464472442865374,
|
|
"num_tokens": 18703498.0,
|
|
"step": 8330
|
|
},
|
|
{
|
|
"entropy": 5.264838695526123,
|
|
"epoch": 9.010816657652786,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 6.32902061594535e-05,
|
|
"loss": 4.3891,
|
|
"mean_token_accuracy": 0.27829418033361436,
|
|
"num_tokens": 18714493.0,
|
|
"step": 8335
|
|
},
|
|
{
|
|
"entropy": 5.285182905197144,
|
|
"epoch": 9.016224986479179,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 6.314536137488364e-05,
|
|
"loss": 4.4132,
|
|
"mean_token_accuracy": 0.2689006432890892,
|
|
"num_tokens": 18727116.0,
|
|
"step": 8340
|
|
},
|
|
{
|
|
"entropy": 5.177760028839112,
|
|
"epoch": 9.021633315305571,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 6.300128647056007e-05,
|
|
"loss": 4.3211,
|
|
"mean_token_accuracy": 0.27741474360227586,
|
|
"num_tokens": 18738043.0,
|
|
"step": 8345
|
|
},
|
|
{
|
|
"entropy": 5.1945250034332275,
|
|
"epoch": 9.027041644131963,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 6.285798197005136e-05,
|
|
"loss": 4.3659,
|
|
"mean_token_accuracy": 0.2751522779464722,
|
|
"num_tokens": 18749744.0,
|
|
"step": 8350
|
|
},
|
|
{
|
|
"entropy": 5.287675523757935,
|
|
"epoch": 9.032449972958355,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 6.271544839412647e-05,
|
|
"loss": 4.4185,
|
|
"mean_token_accuracy": 0.26764392852783203,
|
|
"num_tokens": 18760644.0,
|
|
"step": 8355
|
|
},
|
|
{
|
|
"entropy": 5.225261878967285,
|
|
"epoch": 9.037858301784748,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 6.257368626075266e-05,
|
|
"loss": 4.3265,
|
|
"mean_token_accuracy": 0.2787309855222702,
|
|
"num_tokens": 18771554.0,
|
|
"step": 8360
|
|
},
|
|
{
|
|
"entropy": 5.215203475952149,
|
|
"epoch": 9.043266630611141,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 6.243269608509388e-05,
|
|
"loss": 4.3615,
|
|
"mean_token_accuracy": 0.27616046369075775,
|
|
"num_tokens": 18782652.0,
|
|
"step": 8365
|
|
},
|
|
{
|
|
"entropy": 5.248007583618164,
|
|
"epoch": 9.048674959437534,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 6.229247837950878e-05,
|
|
"loss": 4.3527,
|
|
"mean_token_accuracy": 0.2813325494527817,
|
|
"num_tokens": 18795266.0,
|
|
"step": 8370
|
|
},
|
|
{
|
|
"entropy": 5.219410991668701,
|
|
"epoch": 9.054083288263927,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 6.215303365354875e-05,
|
|
"loss": 4.3936,
|
|
"mean_token_accuracy": 0.27252517342567445,
|
|
"num_tokens": 18806572.0,
|
|
"step": 8375
|
|
},
|
|
{
|
|
"entropy": 5.190199184417724,
|
|
"epoch": 9.05949161709032,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 6.201436241395634e-05,
|
|
"loss": 4.3858,
|
|
"mean_token_accuracy": 0.27309153527021407,
|
|
"num_tokens": 18818036.0,
|
|
"step": 8380
|
|
},
|
|
{
|
|
"entropy": 5.229004430770874,
|
|
"epoch": 9.064899945916713,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 6.187646516466311e-05,
|
|
"loss": 4.3641,
|
|
"mean_token_accuracy": 0.27973378598690035,
|
|
"num_tokens": 18828245.0,
|
|
"step": 8385
|
|
},
|
|
{
|
|
"entropy": 5.210979080200195,
|
|
"epoch": 9.070308274743104,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 6.173934240678794e-05,
|
|
"loss": 4.364,
|
|
"mean_token_accuracy": 0.28140947222709656,
|
|
"num_tokens": 18839538.0,
|
|
"step": 8390
|
|
},
|
|
{
|
|
"entropy": 5.21007719039917,
|
|
"epoch": 9.075716603569496,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 6.160299463863535e-05,
|
|
"loss": 4.275,
|
|
"mean_token_accuracy": 0.2862601548433304,
|
|
"num_tokens": 18850294.0,
|
|
"step": 8395
|
|
},
|
|
{
|
|
"entropy": 5.230219745635987,
|
|
"epoch": 9.08112493239589,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 6.146742235569336e-05,
|
|
"loss": 4.3487,
|
|
"mean_token_accuracy": 0.2801396310329437,
|
|
"num_tokens": 18861394.0,
|
|
"step": 8400
|
|
},
|
|
{
|
|
"entropy": 5.286949205398559,
|
|
"epoch": 9.086533261222282,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 6.133262605063199e-05,
|
|
"loss": 4.4006,
|
|
"mean_token_accuracy": 0.27951018810272216,
|
|
"num_tokens": 18873436.0,
|
|
"step": 8405
|
|
},
|
|
{
|
|
"entropy": 5.196886825561523,
|
|
"epoch": 9.091941590048675,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 6.119860621330131e-05,
|
|
"loss": 4.2976,
|
|
"mean_token_accuracy": 0.27733597457408904,
|
|
"num_tokens": 18884174.0,
|
|
"step": 8410
|
|
},
|
|
{
|
|
"entropy": 5.276731538772583,
|
|
"epoch": 9.097349918875068,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 6.106536333072973e-05,
|
|
"loss": 4.3985,
|
|
"mean_token_accuracy": 0.2840696394443512,
|
|
"num_tokens": 18895524.0,
|
|
"step": 8415
|
|
},
|
|
{
|
|
"entropy": 5.13250617980957,
|
|
"epoch": 9.10275824770146,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 6.093289788712218e-05,
|
|
"loss": 4.3146,
|
|
"mean_token_accuracy": 0.2773715406656265,
|
|
"num_tokens": 18906412.0,
|
|
"step": 8420
|
|
},
|
|
{
|
|
"entropy": 5.223281335830689,
|
|
"epoch": 9.108166576527854,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 6.080121036385845e-05,
|
|
"loss": 4.4051,
|
|
"mean_token_accuracy": 0.27144336253404616,
|
|
"num_tokens": 18917557.0,
|
|
"step": 8425
|
|
},
|
|
{
|
|
"entropy": 5.357726669311523,
|
|
"epoch": 9.113574905354245,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 6.067030123949119e-05,
|
|
"loss": 4.4702,
|
|
"mean_token_accuracy": 0.26969380080699923,
|
|
"num_tokens": 18928749.0,
|
|
"step": 8430
|
|
},
|
|
{
|
|
"entropy": 5.241640710830689,
|
|
"epoch": 9.118983234180638,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 6.0540170989744566e-05,
|
|
"loss": 4.3953,
|
|
"mean_token_accuracy": 0.27076262384653094,
|
|
"num_tokens": 18938854.0,
|
|
"step": 8435
|
|
},
|
|
{
|
|
"entropy": 5.2210204124450685,
|
|
"epoch": 9.12439156300703,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 6.041082008751215e-05,
|
|
"loss": 4.3834,
|
|
"mean_token_accuracy": 0.28340217620134356,
|
|
"num_tokens": 18949915.0,
|
|
"step": 8440
|
|
},
|
|
{
|
|
"entropy": 5.3062217235565186,
|
|
"epoch": 9.129799891833423,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 6.028224900285545e-05,
|
|
"loss": 4.428,
|
|
"mean_token_accuracy": 0.2801760762929916,
|
|
"num_tokens": 18961828.0,
|
|
"step": 8445
|
|
},
|
|
{
|
|
"entropy": 5.254652070999145,
|
|
"epoch": 9.135208220659816,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 6.015445820300207e-05,
|
|
"loss": 4.4291,
|
|
"mean_token_accuracy": 0.2721879631280899,
|
|
"num_tokens": 18974719.0,
|
|
"step": 8450
|
|
},
|
|
{
|
|
"entropy": 5.231177568435669,
|
|
"epoch": 9.140616549486209,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 6.002744815234405e-05,
|
|
"loss": 4.3446,
|
|
"mean_token_accuracy": 0.28035159409046173,
|
|
"num_tokens": 18985540.0,
|
|
"step": 8455
|
|
},
|
|
{
|
|
"entropy": 5.147705888748169,
|
|
"epoch": 9.146024878312602,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 5.990121931243623e-05,
|
|
"loss": 4.2635,
|
|
"mean_token_accuracy": 0.2873945295810699,
|
|
"num_tokens": 18997235.0,
|
|
"step": 8460
|
|
},
|
|
{
|
|
"entropy": 5.189701318740845,
|
|
"epoch": 9.151433207138995,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.977577214199456e-05,
|
|
"loss": 4.2693,
|
|
"mean_token_accuracy": 0.2875798925757408,
|
|
"num_tokens": 19008375.0,
|
|
"step": 8465
|
|
},
|
|
{
|
|
"entropy": 5.173213052749634,
|
|
"epoch": 9.156841535965386,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 5.9651107096894214e-05,
|
|
"loss": 4.34,
|
|
"mean_token_accuracy": 0.27337823510169984,
|
|
"num_tokens": 19019780.0,
|
|
"step": 8470
|
|
},
|
|
{
|
|
"entropy": 5.280072975158691,
|
|
"epoch": 9.162249864791779,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.952722463016836e-05,
|
|
"loss": 4.4388,
|
|
"mean_token_accuracy": 0.2669890016317368,
|
|
"num_tokens": 19032124.0,
|
|
"step": 8475
|
|
},
|
|
{
|
|
"entropy": 5.229280281066894,
|
|
"epoch": 9.167658193618172,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 5.940412519200617e-05,
|
|
"loss": 4.3928,
|
|
"mean_token_accuracy": 0.27287583947181704,
|
|
"num_tokens": 19042694.0,
|
|
"step": 8480
|
|
},
|
|
{
|
|
"entropy": 5.230866432189941,
|
|
"epoch": 9.173066522444564,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 5.928180922975124e-05,
|
|
"loss": 4.4005,
|
|
"mean_token_accuracy": 0.2765102401375771,
|
|
"num_tokens": 19054342.0,
|
|
"step": 8485
|
|
},
|
|
{
|
|
"entropy": 5.224048089981079,
|
|
"epoch": 9.178474851270957,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 5.916027718790007e-05,
|
|
"loss": 4.3542,
|
|
"mean_token_accuracy": 0.2786171793937683,
|
|
"num_tokens": 19064770.0,
|
|
"step": 8490
|
|
},
|
|
{
|
|
"entropy": 5.190273904800415,
|
|
"epoch": 9.18388318009735,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.903952950810038e-05,
|
|
"loss": 4.3451,
|
|
"mean_token_accuracy": 0.2798835977911949,
|
|
"num_tokens": 19076038.0,
|
|
"step": 8495
|
|
},
|
|
{
|
|
"entropy": 5.225831985473633,
|
|
"epoch": 9.189291508923743,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 5.8919566629149535e-05,
|
|
"loss": 4.381,
|
|
"mean_token_accuracy": 0.26815175712108613,
|
|
"num_tokens": 19088024.0,
|
|
"step": 8500
|
|
},
|
|
{
|
|
"epoch": 9.189291508923743,
|
|
"eval_entropy": 5.169601465355266,
|
|
"eval_loss": 5.894737720489502,
|
|
"eval_mean_token_accuracy": 0.18010994033379987,
|
|
"eval_num_tokens": 19088024.0,
|
|
"eval_runtime": 2.1344,
|
|
"eval_samples_per_second": 1646.85,
|
|
"eval_steps_per_second": 206.149,
|
|
"step": 8500
|
|
},
|
|
{
|
|
"entropy": 5.204227542877197,
|
|
"epoch": 9.194699837750136,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.880038898699291e-05,
|
|
"loss": 4.2897,
|
|
"mean_token_accuracy": 0.2818758457899094,
|
|
"num_tokens": 19099342.0,
|
|
"step": 8505
|
|
},
|
|
{
|
|
"entropy": 5.250473165512085,
|
|
"epoch": 9.200108166576527,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.8681997014722303e-05,
|
|
"loss": 4.3844,
|
|
"mean_token_accuracy": 0.284494549036026,
|
|
"num_tokens": 19111978.0,
|
|
"step": 8510
|
|
},
|
|
{
|
|
"entropy": 5.294572496414185,
|
|
"epoch": 9.20551649540292,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.856439114257446e-05,
|
|
"loss": 4.4544,
|
|
"mean_token_accuracy": 0.26648917943239214,
|
|
"num_tokens": 19123859.0,
|
|
"step": 8515
|
|
},
|
|
{
|
|
"entropy": 5.19370288848877,
|
|
"epoch": 9.210924824229313,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 5.84475717979294e-05,
|
|
"loss": 4.3123,
|
|
"mean_token_accuracy": 0.27751430571079255,
|
|
"num_tokens": 19135058.0,
|
|
"step": 8520
|
|
},
|
|
{
|
|
"entropy": 5.209638786315918,
|
|
"epoch": 9.216333153055706,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 5.833153940530882e-05,
|
|
"loss": 4.3519,
|
|
"mean_token_accuracy": 0.2818104177713394,
|
|
"num_tokens": 19146568.0,
|
|
"step": 8525
|
|
},
|
|
{
|
|
"entropy": 5.195817708969116,
|
|
"epoch": 9.221741481882098,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 5.8216294386374727e-05,
|
|
"loss": 4.3123,
|
|
"mean_token_accuracy": 0.2843390330672264,
|
|
"num_tokens": 19158402.0,
|
|
"step": 8530
|
|
},
|
|
{
|
|
"entropy": 5.210106372833252,
|
|
"epoch": 9.227149810708491,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 5.8101837159927845e-05,
|
|
"loss": 4.393,
|
|
"mean_token_accuracy": 0.2734003245830536,
|
|
"num_tokens": 19170716.0,
|
|
"step": 8535
|
|
},
|
|
{
|
|
"entropy": 5.3270899772644045,
|
|
"epoch": 9.232558139534884,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 5.798816814190595e-05,
|
|
"loss": 4.3842,
|
|
"mean_token_accuracy": 0.27748891413211824,
|
|
"num_tokens": 19184085.0,
|
|
"step": 8540
|
|
},
|
|
{
|
|
"entropy": 5.235583686828614,
|
|
"epoch": 9.237966468361277,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 5.7875287745382504e-05,
|
|
"loss": 4.2693,
|
|
"mean_token_accuracy": 0.2875764399766922,
|
|
"num_tokens": 19194619.0,
|
|
"step": 8545
|
|
},
|
|
{
|
|
"entropy": 5.226613712310791,
|
|
"epoch": 9.24337479718767,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.776319638056516e-05,
|
|
"loss": 4.3954,
|
|
"mean_token_accuracy": 0.26622334867715836,
|
|
"num_tokens": 19206422.0,
|
|
"step": 8550
|
|
},
|
|
{
|
|
"entropy": 5.23198356628418,
|
|
"epoch": 9.248783126014061,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 5.76518944547942e-05,
|
|
"loss": 4.4087,
|
|
"mean_token_accuracy": 0.2703027904033661,
|
|
"num_tokens": 19217723.0,
|
|
"step": 8555
|
|
},
|
|
{
|
|
"entropy": 5.2369402885437015,
|
|
"epoch": 9.254191454840454,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 5.754138237254108e-05,
|
|
"loss": 4.305,
|
|
"mean_token_accuracy": 0.27824212312698365,
|
|
"num_tokens": 19228135.0,
|
|
"step": 8560
|
|
},
|
|
{
|
|
"entropy": 5.232626914978027,
|
|
"epoch": 9.259599783666847,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 5.743166053540692e-05,
|
|
"loss": 4.3303,
|
|
"mean_token_accuracy": 0.28407035768032074,
|
|
"num_tokens": 19238553.0,
|
|
"step": 8565
|
|
},
|
|
{
|
|
"entropy": 5.240845823287964,
|
|
"epoch": 9.26500811249324,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 5.732272934212113e-05,
|
|
"loss": 4.3796,
|
|
"mean_token_accuracy": 0.27226119190454484,
|
|
"num_tokens": 19249152.0,
|
|
"step": 8570
|
|
},
|
|
{
|
|
"entropy": 5.175811147689819,
|
|
"epoch": 9.270416441319632,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.721458918854001e-05,
|
|
"loss": 4.2827,
|
|
"mean_token_accuracy": 0.28636554181575774,
|
|
"num_tokens": 19259790.0,
|
|
"step": 8575
|
|
},
|
|
{
|
|
"entropy": 5.208571815490723,
|
|
"epoch": 9.275824770146025,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 5.7107240467645024e-05,
|
|
"loss": 4.2886,
|
|
"mean_token_accuracy": 0.2828440129756927,
|
|
"num_tokens": 19271654.0,
|
|
"step": 8580
|
|
},
|
|
{
|
|
"entropy": 5.198828458786011,
|
|
"epoch": 9.281233098972418,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 5.700068356954172e-05,
|
|
"loss": 4.3349,
|
|
"mean_token_accuracy": 0.2817742720246315,
|
|
"num_tokens": 19281291.0,
|
|
"step": 8585
|
|
},
|
|
{
|
|
"entropy": 5.182634496688843,
|
|
"epoch": 9.286641427798811,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 5.689491888145811e-05,
|
|
"loss": 4.3134,
|
|
"mean_token_accuracy": 0.27976621091365816,
|
|
"num_tokens": 19291789.0,
|
|
"step": 8590
|
|
},
|
|
{
|
|
"entropy": 5.280173444747925,
|
|
"epoch": 9.292049756625202,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.67899467877433e-05,
|
|
"loss": 4.3669,
|
|
"mean_token_accuracy": 0.27156451195478437,
|
|
"num_tokens": 19301881.0,
|
|
"step": 8595
|
|
},
|
|
{
|
|
"entropy": 5.218665027618409,
|
|
"epoch": 9.297458085451595,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 5.668576766986611e-05,
|
|
"loss": 4.3429,
|
|
"mean_token_accuracy": 0.2805596351623535,
|
|
"num_tokens": 19314800.0,
|
|
"step": 8600
|
|
},
|
|
{
|
|
"entropy": 5.179418754577637,
|
|
"epoch": 9.302866414277988,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 5.658238190641378e-05,
|
|
"loss": 4.3552,
|
|
"mean_token_accuracy": 0.2843093529343605,
|
|
"num_tokens": 19325585.0,
|
|
"step": 8605
|
|
},
|
|
{
|
|
"entropy": 5.231477069854736,
|
|
"epoch": 9.30827474310438,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 5.6479789873090273e-05,
|
|
"loss": 4.3799,
|
|
"mean_token_accuracy": 0.27409094721078875,
|
|
"num_tokens": 19337178.0,
|
|
"step": 8610
|
|
},
|
|
{
|
|
"entropy": 5.1650677680969235,
|
|
"epoch": 9.313683071930773,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.637799194271541e-05,
|
|
"loss": 4.2427,
|
|
"mean_token_accuracy": 0.2865681290626526,
|
|
"num_tokens": 19348196.0,
|
|
"step": 8615
|
|
},
|
|
{
|
|
"entropy": 5.23661470413208,
|
|
"epoch": 9.319091400757166,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 5.6276988485223004e-05,
|
|
"loss": 4.3933,
|
|
"mean_token_accuracy": 0.274845652282238,
|
|
"num_tokens": 19359198.0,
|
|
"step": 8620
|
|
},
|
|
{
|
|
"entropy": 5.227737188339233,
|
|
"epoch": 9.32449972958356,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 5.617677986765987e-05,
|
|
"loss": 4.3127,
|
|
"mean_token_accuracy": 0.2826430663466454,
|
|
"num_tokens": 19369577.0,
|
|
"step": 8625
|
|
},
|
|
{
|
|
"entropy": 5.283739423751831,
|
|
"epoch": 9.329908058409952,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.6077366454184446e-05,
|
|
"loss": 4.451,
|
|
"mean_token_accuracy": 0.2679708421230316,
|
|
"num_tokens": 19381875.0,
|
|
"step": 8630
|
|
},
|
|
{
|
|
"entropy": 5.29633092880249,
|
|
"epoch": 9.335316387236343,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.597874860606521e-05,
|
|
"loss": 4.4328,
|
|
"mean_token_accuracy": 0.2748284608125687,
|
|
"num_tokens": 19393619.0,
|
|
"step": 8635
|
|
},
|
|
{
|
|
"entropy": 5.19608964920044,
|
|
"epoch": 9.340724716062736,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 5.5880926681679695e-05,
|
|
"loss": 4.3422,
|
|
"mean_token_accuracy": 0.2741240128874779,
|
|
"num_tokens": 19404011.0,
|
|
"step": 8640
|
|
},
|
|
{
|
|
"entropy": 5.263300085067749,
|
|
"epoch": 9.346133044889129,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 5.578390103651297e-05,
|
|
"loss": 4.4252,
|
|
"mean_token_accuracy": 0.27124367356300355,
|
|
"num_tokens": 19414713.0,
|
|
"step": 8645
|
|
},
|
|
{
|
|
"entropy": 5.238052225112915,
|
|
"epoch": 9.351541373715522,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.568767202315647e-05,
|
|
"loss": 4.3555,
|
|
"mean_token_accuracy": 0.2829028517007828,
|
|
"num_tokens": 19425400.0,
|
|
"step": 8650
|
|
},
|
|
{
|
|
"entropy": 5.235742521286011,
|
|
"epoch": 9.356949702541915,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 5.559223999130667e-05,
|
|
"loss": 4.3163,
|
|
"mean_token_accuracy": 0.28530324399471285,
|
|
"num_tokens": 19436064.0,
|
|
"step": 8655
|
|
},
|
|
{
|
|
"entropy": 5.133183193206787,
|
|
"epoch": 9.362358031368307,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.549760528776376e-05,
|
|
"loss": 4.2543,
|
|
"mean_token_accuracy": 0.2803094297647476,
|
|
"num_tokens": 19446669.0,
|
|
"step": 8660
|
|
},
|
|
{
|
|
"entropy": 5.285119152069091,
|
|
"epoch": 9.3677663601947,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 5.5403768256430475e-05,
|
|
"loss": 4.4464,
|
|
"mean_token_accuracy": 0.2798794835805893,
|
|
"num_tokens": 19457299.0,
|
|
"step": 8665
|
|
},
|
|
{
|
|
"entropy": 5.3271440982818605,
|
|
"epoch": 9.373174689021093,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.5310729238310854e-05,
|
|
"loss": 4.4925,
|
|
"mean_token_accuracy": 0.2670835092663765,
|
|
"num_tokens": 19469181.0,
|
|
"step": 8670
|
|
},
|
|
{
|
|
"entropy": 5.2894837856292725,
|
|
"epoch": 9.378583017847484,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 5.521848857150884e-05,
|
|
"loss": 4.4182,
|
|
"mean_token_accuracy": 0.27753921300172807,
|
|
"num_tokens": 19480393.0,
|
|
"step": 8675
|
|
},
|
|
{
|
|
"entropy": 5.246707439422607,
|
|
"epoch": 9.383991346673877,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 5.512704659122725e-05,
|
|
"loss": 4.3458,
|
|
"mean_token_accuracy": 0.2840798556804657,
|
|
"num_tokens": 19491412.0,
|
|
"step": 8680
|
|
},
|
|
{
|
|
"entropy": 5.183616399765015,
|
|
"epoch": 9.38939967550027,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.503640362976646e-05,
|
|
"loss": 4.3105,
|
|
"mean_token_accuracy": 0.27510482966899874,
|
|
"num_tokens": 19502601.0,
|
|
"step": 8685
|
|
},
|
|
{
|
|
"entropy": 5.239747524261475,
|
|
"epoch": 9.394808004326663,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 5.4946560016523184e-05,
|
|
"loss": 4.3426,
|
|
"mean_token_accuracy": 0.282394739985466,
|
|
"num_tokens": 19513985.0,
|
|
"step": 8690
|
|
},
|
|
{
|
|
"entropy": 5.270395565032959,
|
|
"epoch": 9.400216333153056,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 5.485751607798937e-05,
|
|
"loss": 4.4115,
|
|
"mean_token_accuracy": 0.26771504133939744,
|
|
"num_tokens": 19525575.0,
|
|
"step": 8695
|
|
},
|
|
{
|
|
"entropy": 5.2745068073272705,
|
|
"epoch": 9.405624661979449,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 5.4769272137750855e-05,
|
|
"loss": 4.4323,
|
|
"mean_token_accuracy": 0.2696091040968895,
|
|
"num_tokens": 19536130.0,
|
|
"step": 8700
|
|
},
|
|
{
|
|
"entropy": 5.225967884063721,
|
|
"epoch": 9.411032990805841,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.468182851648629e-05,
|
|
"loss": 4.3462,
|
|
"mean_token_accuracy": 0.2728260576725006,
|
|
"num_tokens": 19547803.0,
|
|
"step": 8705
|
|
},
|
|
{
|
|
"entropy": 5.226136970520019,
|
|
"epoch": 9.416441319632234,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.4595185531966074e-05,
|
|
"loss": 4.3534,
|
|
"mean_token_accuracy": 0.2794545263051987,
|
|
"num_tokens": 19559622.0,
|
|
"step": 8710
|
|
},
|
|
{
|
|
"entropy": 5.184755754470825,
|
|
"epoch": 9.421849648458625,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.450934349905094e-05,
|
|
"loss": 4.3901,
|
|
"mean_token_accuracy": 0.2785751774907112,
|
|
"num_tokens": 19569914.0,
|
|
"step": 8715
|
|
},
|
|
{
|
|
"entropy": 5.236275959014892,
|
|
"epoch": 9.427257977285018,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 5.44243027296911e-05,
|
|
"loss": 4.3256,
|
|
"mean_token_accuracy": 0.2795374274253845,
|
|
"num_tokens": 19580770.0,
|
|
"step": 8720
|
|
},
|
|
{
|
|
"entropy": 5.241759586334228,
|
|
"epoch": 9.432666306111411,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.434006353292485e-05,
|
|
"loss": 4.3547,
|
|
"mean_token_accuracy": 0.27839161455631256,
|
|
"num_tokens": 19591588.0,
|
|
"step": 8725
|
|
},
|
|
{
|
|
"entropy": 5.266218805313111,
|
|
"epoch": 9.438074634937804,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.425662621487766e-05,
|
|
"loss": 4.3796,
|
|
"mean_token_accuracy": 0.2789332017302513,
|
|
"num_tokens": 19602866.0,
|
|
"step": 8730
|
|
},
|
|
{
|
|
"entropy": 5.2629162788391115,
|
|
"epoch": 9.443482963764197,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 5.417399107876092e-05,
|
|
"loss": 4.3308,
|
|
"mean_token_accuracy": 0.28248959183692934,
|
|
"num_tokens": 19613546.0,
|
|
"step": 8735
|
|
},
|
|
{
|
|
"entropy": 5.300745916366577,
|
|
"epoch": 9.44889129259059,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.4092158424870936e-05,
|
|
"loss": 4.4766,
|
|
"mean_token_accuracy": 0.26634194105863573,
|
|
"num_tokens": 19626221.0,
|
|
"step": 8740
|
|
},
|
|
{
|
|
"entropy": 5.2578410625457765,
|
|
"epoch": 9.454299621416983,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 5.40111285505877e-05,
|
|
"loss": 4.4682,
|
|
"mean_token_accuracy": 0.26660424619913103,
|
|
"num_tokens": 19637166.0,
|
|
"step": 8745
|
|
},
|
|
{
|
|
"entropy": 5.248173809051513,
|
|
"epoch": 9.459707950243375,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.393090175037399e-05,
|
|
"loss": 4.3816,
|
|
"mean_token_accuracy": 0.27967100143432616,
|
|
"num_tokens": 19648448.0,
|
|
"step": 8750
|
|
},
|
|
{
|
|
"entropy": 5.264388561248779,
|
|
"epoch": 9.465116279069768,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.385147831577424e-05,
|
|
"loss": 4.359,
|
|
"mean_token_accuracy": 0.27891606390476226,
|
|
"num_tokens": 19660282.0,
|
|
"step": 8755
|
|
},
|
|
{
|
|
"entropy": 5.2563313961029055,
|
|
"epoch": 9.47052460789616,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 5.377285853541332e-05,
|
|
"loss": 4.4007,
|
|
"mean_token_accuracy": 0.26806887090206144,
|
|
"num_tokens": 19670672.0,
|
|
"step": 8760
|
|
},
|
|
{
|
|
"entropy": 5.250340414047241,
|
|
"epoch": 9.475932936722552,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 5.369504269499574e-05,
|
|
"loss": 4.3829,
|
|
"mean_token_accuracy": 0.27631027698516847,
|
|
"num_tokens": 19682894.0,
|
|
"step": 8765
|
|
},
|
|
{
|
|
"entropy": 5.287089014053345,
|
|
"epoch": 9.481341265548945,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.3618031077304406e-05,
|
|
"loss": 4.3785,
|
|
"mean_token_accuracy": 0.2782739341259003,
|
|
"num_tokens": 19693743.0,
|
|
"step": 8770
|
|
},
|
|
{
|
|
"entropy": 5.228011274337769,
|
|
"epoch": 9.486749594375338,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 5.354182396219974e-05,
|
|
"loss": 4.3878,
|
|
"mean_token_accuracy": 0.284874165058136,
|
|
"num_tokens": 19703891.0,
|
|
"step": 8775
|
|
},
|
|
{
|
|
"entropy": 5.271098661422729,
|
|
"epoch": 9.49215792320173,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 5.3466421626618573e-05,
|
|
"loss": 4.3709,
|
|
"mean_token_accuracy": 0.28179795742034913,
|
|
"num_tokens": 19714163.0,
|
|
"step": 8780
|
|
},
|
|
{
|
|
"entropy": 5.30140495300293,
|
|
"epoch": 9.497566252028124,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.339182434457314e-05,
|
|
"loss": 4.3758,
|
|
"mean_token_accuracy": 0.2765964582562447,
|
|
"num_tokens": 19725750.0,
|
|
"step": 8785
|
|
},
|
|
{
|
|
"entropy": 5.273830223083496,
|
|
"epoch": 9.502974580854517,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 5.3318032387150165e-05,
|
|
"loss": 4.392,
|
|
"mean_token_accuracy": 0.2796762004494667,
|
|
"num_tokens": 19737120.0,
|
|
"step": 8790
|
|
},
|
|
{
|
|
"entropy": 5.2480443000793455,
|
|
"epoch": 9.50838290968091,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 5.324504602250981e-05,
|
|
"loss": 4.386,
|
|
"mean_token_accuracy": 0.2766797184944153,
|
|
"num_tokens": 19748568.0,
|
|
"step": 8795
|
|
},
|
|
{
|
|
"entropy": 5.2622514247894285,
|
|
"epoch": 9.5137912385073,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 5.317286551588464e-05,
|
|
"loss": 4.3654,
|
|
"mean_token_accuracy": 0.2790058806538582,
|
|
"num_tokens": 19760786.0,
|
|
"step": 8800
|
|
},
|
|
{
|
|
"entropy": 5.2104700088500975,
|
|
"epoch": 9.519199567333693,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 5.310149112957882e-05,
|
|
"loss": 4.3638,
|
|
"mean_token_accuracy": 0.28416160196065904,
|
|
"num_tokens": 19771519.0,
|
|
"step": 8805
|
|
},
|
|
{
|
|
"entropy": 5.269663381576538,
|
|
"epoch": 9.524607896160086,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 5.3030923122967036e-05,
|
|
"loss": 4.3258,
|
|
"mean_token_accuracy": 0.28618118166923523,
|
|
"num_tokens": 19782010.0,
|
|
"step": 8810
|
|
},
|
|
{
|
|
"entropy": 5.206872415542603,
|
|
"epoch": 9.530016224986479,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 5.296116175249357e-05,
|
|
"loss": 4.3813,
|
|
"mean_token_accuracy": 0.27461864799261093,
|
|
"num_tokens": 19793436.0,
|
|
"step": 8815
|
|
},
|
|
{
|
|
"entropy": 5.2102879047393795,
|
|
"epoch": 9.535424553812872,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.289220727167139e-05,
|
|
"loss": 4.39,
|
|
"mean_token_accuracy": 0.2763330444693565,
|
|
"num_tokens": 19805005.0,
|
|
"step": 8820
|
|
},
|
|
{
|
|
"entropy": 5.164290904998779,
|
|
"epoch": 9.540832882639265,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 5.282405993108127e-05,
|
|
"loss": 4.3537,
|
|
"mean_token_accuracy": 0.27875770032405855,
|
|
"num_tokens": 19816200.0,
|
|
"step": 8825
|
|
},
|
|
{
|
|
"entropy": 5.254450845718384,
|
|
"epoch": 9.546241211465658,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 5.2756719978370794e-05,
|
|
"loss": 4.3798,
|
|
"mean_token_accuracy": 0.28819774091243744,
|
|
"num_tokens": 19826919.0,
|
|
"step": 8830
|
|
},
|
|
{
|
|
"entropy": 5.273106098175049,
|
|
"epoch": 9.55164954029205,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 5.2690187658253525e-05,
|
|
"loss": 4.4876,
|
|
"mean_token_accuracy": 0.2696785658597946,
|
|
"num_tokens": 19838852.0,
|
|
"step": 8835
|
|
},
|
|
{
|
|
"entropy": 5.309946870803833,
|
|
"epoch": 9.557057869118442,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.262446321250803e-05,
|
|
"loss": 4.3823,
|
|
"mean_token_accuracy": 0.2779736042022705,
|
|
"num_tokens": 19850748.0,
|
|
"step": 8840
|
|
},
|
|
{
|
|
"entropy": 5.25062894821167,
|
|
"epoch": 9.562466197944834,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 5.2559546879977084e-05,
|
|
"loss": 4.4251,
|
|
"mean_token_accuracy": 0.27089960426092147,
|
|
"num_tokens": 19862599.0,
|
|
"step": 8845
|
|
},
|
|
{
|
|
"entropy": 5.204287910461426,
|
|
"epoch": 9.567874526771227,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 5.2495438896566854e-05,
|
|
"loss": 4.3638,
|
|
"mean_token_accuracy": 0.2746440678834915,
|
|
"num_tokens": 19872573.0,
|
|
"step": 8850
|
|
},
|
|
{
|
|
"entropy": 5.2073243141174315,
|
|
"epoch": 9.57328285559762,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 5.2432139495245855e-05,
|
|
"loss": 4.2764,
|
|
"mean_token_accuracy": 0.2863683193922043,
|
|
"num_tokens": 19883859.0,
|
|
"step": 8855
|
|
},
|
|
{
|
|
"entropy": 5.194192028045654,
|
|
"epoch": 9.578691184424013,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 5.236964890604426e-05,
|
|
"loss": 4.3424,
|
|
"mean_token_accuracy": 0.2826130047440529,
|
|
"num_tokens": 19895552.0,
|
|
"step": 8860
|
|
},
|
|
{
|
|
"entropy": 5.1817155361175535,
|
|
"epoch": 9.584099513250406,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 5.230796735605304e-05,
|
|
"loss": 4.3035,
|
|
"mean_token_accuracy": 0.2806200832128525,
|
|
"num_tokens": 19905664.0,
|
|
"step": 8865
|
|
},
|
|
{
|
|
"entropy": 5.229707860946656,
|
|
"epoch": 9.589507842076799,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 5.224709506942305e-05,
|
|
"loss": 4.3978,
|
|
"mean_token_accuracy": 0.2788667678833008,
|
|
"num_tokens": 19916725.0,
|
|
"step": 8870
|
|
},
|
|
{
|
|
"entropy": 5.195999002456665,
|
|
"epoch": 9.594916170903192,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 5.218703226736438e-05,
|
|
"loss": 4.3846,
|
|
"mean_token_accuracy": 0.28048355877399445,
|
|
"num_tokens": 19927584.0,
|
|
"step": 8875
|
|
},
|
|
{
|
|
"entropy": 5.234932374954224,
|
|
"epoch": 9.600324499729584,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 5.212777916814533e-05,
|
|
"loss": 4.3777,
|
|
"mean_token_accuracy": 0.2734958082437515,
|
|
"num_tokens": 19939122.0,
|
|
"step": 8880
|
|
},
|
|
{
|
|
"entropy": 5.274390602111817,
|
|
"epoch": 9.605732828555976,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 5.206933598709179e-05,
|
|
"loss": 4.4356,
|
|
"mean_token_accuracy": 0.27307973355054854,
|
|
"num_tokens": 19950513.0,
|
|
"step": 8885
|
|
},
|
|
{
|
|
"entropy": 5.237869024276733,
|
|
"epoch": 9.611141157382368,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 5.20117029365865e-05,
|
|
"loss": 4.3546,
|
|
"mean_token_accuracy": 0.28013348281383516,
|
|
"num_tokens": 19961960.0,
|
|
"step": 8890
|
|
},
|
|
{
|
|
"entropy": 5.258018970489502,
|
|
"epoch": 9.616549486208761,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 5.195488022606802e-05,
|
|
"loss": 4.3892,
|
|
"mean_token_accuracy": 0.28570207506418227,
|
|
"num_tokens": 19973222.0,
|
|
"step": 8895
|
|
},
|
|
{
|
|
"entropy": 5.18748927116394,
|
|
"epoch": 9.621957815035154,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.189886806203023e-05,
|
|
"loss": 4.3206,
|
|
"mean_token_accuracy": 0.2815970003604889,
|
|
"num_tokens": 19985053.0,
|
|
"step": 8900
|
|
},
|
|
{
|
|
"entropy": 5.2108941078186035,
|
|
"epoch": 9.627366143861547,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 5.184366664802149e-05,
|
|
"loss": 4.3039,
|
|
"mean_token_accuracy": 0.28705120980739596,
|
|
"num_tokens": 19996937.0,
|
|
"step": 8905
|
|
},
|
|
{
|
|
"entropy": 5.248632287979126,
|
|
"epoch": 9.63277447268794,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 5.178927618464384e-05,
|
|
"loss": 4.3207,
|
|
"mean_token_accuracy": 0.2828929841518402,
|
|
"num_tokens": 20008749.0,
|
|
"step": 8910
|
|
},
|
|
{
|
|
"entropy": 5.328709363937378,
|
|
"epoch": 9.638182801514333,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 5.1735696869552364e-05,
|
|
"loss": 4.4551,
|
|
"mean_token_accuracy": 0.2730847284197807,
|
|
"num_tokens": 20020279.0,
|
|
"step": 8915
|
|
},
|
|
{
|
|
"entropy": 5.32074818611145,
|
|
"epoch": 9.643591130340724,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 5.1682928897454416e-05,
|
|
"loss": 4.4716,
|
|
"mean_token_accuracy": 0.2642207533121109,
|
|
"num_tokens": 20032133.0,
|
|
"step": 8920
|
|
},
|
|
{
|
|
"entropy": 5.224866914749145,
|
|
"epoch": 9.648999459167117,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 5.1630972460108904e-05,
|
|
"loss": 4.3004,
|
|
"mean_token_accuracy": 0.28149241805076597,
|
|
"num_tokens": 20044253.0,
|
|
"step": 8925
|
|
},
|
|
{
|
|
"entropy": 5.300191926956177,
|
|
"epoch": 9.65440778799351,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.1579827746325694e-05,
|
|
"loss": 4.4153,
|
|
"mean_token_accuracy": 0.2767783090472221,
|
|
"num_tokens": 20055445.0,
|
|
"step": 8930
|
|
},
|
|
{
|
|
"entropy": 5.330207061767578,
|
|
"epoch": 9.659816116819902,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 5.152949494196472e-05,
|
|
"loss": 4.4145,
|
|
"mean_token_accuracy": 0.2676451072096825,
|
|
"num_tokens": 20067670.0,
|
|
"step": 8935
|
|
},
|
|
{
|
|
"entropy": 5.284798002243042,
|
|
"epoch": 9.665224445646295,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.1479974229935575e-05,
|
|
"loss": 4.4134,
|
|
"mean_token_accuracy": 0.27238773107528685,
|
|
"num_tokens": 20079408.0,
|
|
"step": 8940
|
|
},
|
|
{
|
|
"entropy": 5.168606424331665,
|
|
"epoch": 9.670632774472688,
|
|
"grad_norm": 1.8203125,
|
|
"learning_rate": 5.1431265790196583e-05,
|
|
"loss": 4.3311,
|
|
"mean_token_accuracy": 0.28833879679441454,
|
|
"num_tokens": 20089371.0,
|
|
"step": 8945
|
|
},
|
|
{
|
|
"entropy": 5.259204626083374,
|
|
"epoch": 9.676041103299081,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 5.138336979975439e-05,
|
|
"loss": 4.3544,
|
|
"mean_token_accuracy": 0.2725896716117859,
|
|
"num_tokens": 20099521.0,
|
|
"step": 8950
|
|
},
|
|
{
|
|
"entropy": 5.2556312561035154,
|
|
"epoch": 9.681449432125474,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 5.13362864326631e-05,
|
|
"loss": 4.4079,
|
|
"mean_token_accuracy": 0.2724431216716766,
|
|
"num_tokens": 20111087.0,
|
|
"step": 8955
|
|
},
|
|
{
|
|
"entropy": 5.2816814422607425,
|
|
"epoch": 9.686857760951867,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 5.12900158600238e-05,
|
|
"loss": 4.4158,
|
|
"mean_token_accuracy": 0.2631221890449524,
|
|
"num_tokens": 20122791.0,
|
|
"step": 8960
|
|
},
|
|
{
|
|
"entropy": 5.212072849273682,
|
|
"epoch": 9.692266089778258,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.12445582499838e-05,
|
|
"loss": 4.3076,
|
|
"mean_token_accuracy": 0.2804286301136017,
|
|
"num_tokens": 20133665.0,
|
|
"step": 8965
|
|
},
|
|
{
|
|
"entropy": 5.283759069442749,
|
|
"epoch": 9.69767441860465,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 5.119991376773624e-05,
|
|
"loss": 4.4103,
|
|
"mean_token_accuracy": 0.269738033413887,
|
|
"num_tokens": 20144749.0,
|
|
"step": 8970
|
|
},
|
|
{
|
|
"entropy": 5.196679639816284,
|
|
"epoch": 9.703082747431043,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 5.115608257551923e-05,
|
|
"loss": 4.3065,
|
|
"mean_token_accuracy": 0.2792526543140411,
|
|
"num_tokens": 20154605.0,
|
|
"step": 8975
|
|
},
|
|
{
|
|
"entropy": 5.311254072189331,
|
|
"epoch": 9.708491076257436,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.11130648326154e-05,
|
|
"loss": 4.4272,
|
|
"mean_token_accuracy": 0.2763336971402168,
|
|
"num_tokens": 20165947.0,
|
|
"step": 8980
|
|
},
|
|
{
|
|
"entropy": 5.2654438495635985,
|
|
"epoch": 9.71389940508383,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 5.107086069535136e-05,
|
|
"loss": 4.4041,
|
|
"mean_token_accuracy": 0.2726740688085556,
|
|
"num_tokens": 20176697.0,
|
|
"step": 8985
|
|
},
|
|
{
|
|
"entropy": 5.232499504089356,
|
|
"epoch": 9.719307733910222,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 5.1029470317097044e-05,
|
|
"loss": 4.3613,
|
|
"mean_token_accuracy": 0.27251105606555937,
|
|
"num_tokens": 20189022.0,
|
|
"step": 8990
|
|
},
|
|
{
|
|
"entropy": 5.293680191040039,
|
|
"epoch": 9.724716062736615,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.098889384826519e-05,
|
|
"loss": 4.4537,
|
|
"mean_token_accuracy": 0.27145034968853,
|
|
"num_tokens": 20200603.0,
|
|
"step": 8995
|
|
},
|
|
{
|
|
"entropy": 5.2705987930297855,
|
|
"epoch": 9.730124391563008,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 5.094913143631073e-05,
|
|
"loss": 4.4223,
|
|
"mean_token_accuracy": 0.27229013442993166,
|
|
"num_tokens": 20212258.0,
|
|
"step": 9000
|
|
},
|
|
{
|
|
"epoch": 9.730124391563008,
|
|
"eval_entropy": 5.160298653082414,
|
|
"eval_loss": 5.8957648277282715,
|
|
"eval_mean_token_accuracy": 0.18002823575992477,
|
|
"eval_num_tokens": 20212258.0,
|
|
"eval_runtime": 1.9386,
|
|
"eval_samples_per_second": 1813.122,
|
|
"eval_steps_per_second": 226.963,
|
|
"step": 9000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 9240,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"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": 4686219886510080.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|