7112 lines
193 KiB
JSON
7112 lines
193 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 3.0068758057584875,
|
|
"eval_steps": 500,
|
|
"global_step": 3500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.691965675354004,
|
|
"epoch": 0.004297378599054577,
|
|
"grad_norm": 14.1875,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.8333,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 10107.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.691940116882325,
|
|
"epoch": 0.008594757198109154,
|
|
"grad_norm": 14.0,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7061,
|
|
"mean_token_accuracy": 0.0001254705130122602,
|
|
"num_tokens": 18391.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.690368461608887,
|
|
"epoch": 0.01289213579716373,
|
|
"grad_norm": 10.25,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.4329,
|
|
"mean_token_accuracy": 0.039750428823754194,
|
|
"num_tokens": 27061.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.658441638946533,
|
|
"epoch": 0.017189514396218308,
|
|
"grad_norm": 6.71875,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.1104,
|
|
"mean_token_accuracy": 0.05699652768671513,
|
|
"num_tokens": 36339.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.46299867630005,
|
|
"epoch": 0.021486892995272882,
|
|
"grad_norm": 3.6875,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 9.8252,
|
|
"mean_token_accuracy": 0.05812832042574882,
|
|
"num_tokens": 45770.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.438103103637696,
|
|
"epoch": 0.02578427159432746,
|
|
"grad_norm": 3.109375,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.7045,
|
|
"mean_token_accuracy": 0.05137639828026295,
|
|
"num_tokens": 54575.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.417833423614502,
|
|
"epoch": 0.030081650193382038,
|
|
"grad_norm": 2.640625,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.6957,
|
|
"mean_token_accuracy": 0.05632184371352196,
|
|
"num_tokens": 66403.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.414659309387208,
|
|
"epoch": 0.034379028792436615,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.5706,
|
|
"mean_token_accuracy": 0.05973100401461125,
|
|
"num_tokens": 76510.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.398680400848388,
|
|
"epoch": 0.03867640739149119,
|
|
"grad_norm": 2.453125,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.5207,
|
|
"mean_token_accuracy": 0.05555921196937561,
|
|
"num_tokens": 84836.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.38410243988037,
|
|
"epoch": 0.042973785990545764,
|
|
"grad_norm": 2.328125,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.4948,
|
|
"mean_token_accuracy": 0.05959299132227898,
|
|
"num_tokens": 93197.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.410307884216309,
|
|
"epoch": 0.047271164589600345,
|
|
"grad_norm": 2.921875,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.4126,
|
|
"mean_token_accuracy": 0.07061550468206405,
|
|
"num_tokens": 101546.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.19266414642334,
|
|
"epoch": 0.05156854318865492,
|
|
"grad_norm": 2.078125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.4273,
|
|
"mean_token_accuracy": 0.06822829730808735,
|
|
"num_tokens": 111703.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.230934810638427,
|
|
"epoch": 0.055865921787709494,
|
|
"grad_norm": 2.03125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.2909,
|
|
"mean_token_accuracy": 0.071290984749794,
|
|
"num_tokens": 119894.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.18722620010376,
|
|
"epoch": 0.060163300386764075,
|
|
"grad_norm": 1.8828125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.2147,
|
|
"mean_token_accuracy": 0.07613217085599899,
|
|
"num_tokens": 128885.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.178893566131592,
|
|
"epoch": 0.06446067898581866,
|
|
"grad_norm": 2.03125,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.1672,
|
|
"mean_token_accuracy": 0.07484854124486447,
|
|
"num_tokens": 138106.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.157839488983154,
|
|
"epoch": 0.06875805758487323,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.102,
|
|
"mean_token_accuracy": 0.07415809743106365,
|
|
"num_tokens": 146691.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.076846313476562,
|
|
"epoch": 0.0730554361839278,
|
|
"grad_norm": 2.015625,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.0178,
|
|
"mean_token_accuracy": 0.07684990540146827,
|
|
"num_tokens": 155792.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.131054401397705,
|
|
"epoch": 0.07735281478298238,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.041,
|
|
"mean_token_accuracy": 0.0714216936379671,
|
|
"num_tokens": 166944.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.075201606750488,
|
|
"epoch": 0.08165019338203695,
|
|
"grad_norm": 2.234375,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.8173,
|
|
"mean_token_accuracy": 0.08133668191730976,
|
|
"num_tokens": 175303.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 9.932421398162841,
|
|
"epoch": 0.08594757198109153,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.7966,
|
|
"mean_token_accuracy": 0.0745367206633091,
|
|
"num_tokens": 184708.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.040246391296387,
|
|
"epoch": 0.09024495058014612,
|
|
"grad_norm": 1.859375,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.6871,
|
|
"mean_token_accuracy": 0.07615455649793149,
|
|
"num_tokens": 193835.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 9.88200340270996,
|
|
"epoch": 0.09454232917920069,
|
|
"grad_norm": 1.8984375,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.6105,
|
|
"mean_token_accuracy": 0.08173681348562241,
|
|
"num_tokens": 203344.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 9.8795636177063,
|
|
"epoch": 0.09883970777825526,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.6013,
|
|
"mean_token_accuracy": 0.06998304352164268,
|
|
"num_tokens": 213048.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.673752784729004,
|
|
"epoch": 0.10313708637730984,
|
|
"grad_norm": 2.40625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.4264,
|
|
"mean_token_accuracy": 0.08312096372246743,
|
|
"num_tokens": 221784.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.593516635894776,
|
|
"epoch": 0.10743446497636441,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.3091,
|
|
"mean_token_accuracy": 0.0795404627919197,
|
|
"num_tokens": 230971.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.516329765319824,
|
|
"epoch": 0.11173184357541899,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.2258,
|
|
"mean_token_accuracy": 0.08525455147027969,
|
|
"num_tokens": 240524.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.287061023712159,
|
|
"epoch": 0.11602922217447358,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.1406,
|
|
"mean_token_accuracy": 0.07961341962218285,
|
|
"num_tokens": 249220.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.293281650543213,
|
|
"epoch": 0.12032660077352815,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.1524,
|
|
"mean_token_accuracy": 0.07657627090811729,
|
|
"num_tokens": 258934.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.01927890777588,
|
|
"epoch": 0.12462397937258272,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.0306,
|
|
"mean_token_accuracy": 0.07841632440686226,
|
|
"num_tokens": 267680.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.906111335754394,
|
|
"epoch": 0.1289213579716373,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.9934,
|
|
"mean_token_accuracy": 0.08083904013037682,
|
|
"num_tokens": 276227.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.772429847717286,
|
|
"epoch": 0.1332187365706919,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.8676,
|
|
"mean_token_accuracy": 0.08683294653892518,
|
|
"num_tokens": 286342.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.59535722732544,
|
|
"epoch": 0.13751611516974646,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.875,
|
|
"mean_token_accuracy": 0.0787959948182106,
|
|
"num_tokens": 294994.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.52799425125122,
|
|
"epoch": 0.14181349376880104,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.8586,
|
|
"mean_token_accuracy": 0.083160999417305,
|
|
"num_tokens": 303882.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.462903499603271,
|
|
"epoch": 0.1461108723678556,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.8291,
|
|
"mean_token_accuracy": 0.08582018166780472,
|
|
"num_tokens": 312515.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.407231998443603,
|
|
"epoch": 0.15040825096691018,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.8147,
|
|
"mean_token_accuracy": 0.08485982567071915,
|
|
"num_tokens": 320801.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.276727199554443,
|
|
"epoch": 0.15470562956596476,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.7982,
|
|
"mean_token_accuracy": 0.08752225190401078,
|
|
"num_tokens": 329382.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.329781436920166,
|
|
"epoch": 0.15900300816501933,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.7867,
|
|
"mean_token_accuracy": 0.08166329339146614,
|
|
"num_tokens": 337894.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.223849201202393,
|
|
"epoch": 0.1633003867640739,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.763,
|
|
"mean_token_accuracy": 0.0833988644182682,
|
|
"num_tokens": 346380.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.166347885131836,
|
|
"epoch": 0.16759776536312848,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.8225,
|
|
"mean_token_accuracy": 0.07767418175935745,
|
|
"num_tokens": 356305.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.312477111816406,
|
|
"epoch": 0.17189514396218306,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.7712,
|
|
"mean_token_accuracy": 0.08167372904717922,
|
|
"num_tokens": 364899.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.078741836547852,
|
|
"epoch": 0.17619252256123766,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.7396,
|
|
"mean_token_accuracy": 0.0805903747677803,
|
|
"num_tokens": 373663.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.12859468460083,
|
|
"epoch": 0.18048990116029223,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.7106,
|
|
"mean_token_accuracy": 0.0817422017455101,
|
|
"num_tokens": 382730.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.151737308502197,
|
|
"epoch": 0.1847872797593468,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.7463,
|
|
"mean_token_accuracy": 0.08051776215434074,
|
|
"num_tokens": 392676.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.053222179412842,
|
|
"epoch": 0.18908465835840138,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.7332,
|
|
"mean_token_accuracy": 0.08385580405592918,
|
|
"num_tokens": 401050.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.128051567077637,
|
|
"epoch": 0.19338203695745596,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.6847,
|
|
"mean_token_accuracy": 0.08319979459047318,
|
|
"num_tokens": 410009.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.011648893356323,
|
|
"epoch": 0.19767941555651053,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.6685,
|
|
"mean_token_accuracy": 0.08485263586044312,
|
|
"num_tokens": 419302.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.079462242126464,
|
|
"epoch": 0.2019767941555651,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.5827,
|
|
"mean_token_accuracy": 0.09158513993024826,
|
|
"num_tokens": 427296.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.013971376419068,
|
|
"epoch": 0.20627417275461968,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.7232,
|
|
"mean_token_accuracy": 0.08320990726351737,
|
|
"num_tokens": 436368.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.017411470413208,
|
|
"epoch": 0.21057155135367425,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.6329,
|
|
"mean_token_accuracy": 0.08677152320742607,
|
|
"num_tokens": 445535.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.988703823089599,
|
|
"epoch": 0.21486892995272883,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.6242,
|
|
"mean_token_accuracy": 0.08498941510915756,
|
|
"num_tokens": 454769.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.950494480133057,
|
|
"epoch": 0.2191663085517834,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.6389,
|
|
"mean_token_accuracy": 0.08514668568968772,
|
|
"num_tokens": 463975.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.953547191619873,
|
|
"epoch": 0.22346368715083798,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.6256,
|
|
"mean_token_accuracy": 0.0869191125035286,
|
|
"num_tokens": 472899.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.079424142837524,
|
|
"epoch": 0.22776106574989258,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.6169,
|
|
"mean_token_accuracy": 0.08434697091579438,
|
|
"num_tokens": 481556.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.917314720153809,
|
|
"epoch": 0.23205844434894715,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.5906,
|
|
"mean_token_accuracy": 0.08885264247655869,
|
|
"num_tokens": 490253.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.981470727920533,
|
|
"epoch": 0.23635582294800173,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.6123,
|
|
"mean_token_accuracy": 0.08401632830500602,
|
|
"num_tokens": 498444.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.889223909378051,
|
|
"epoch": 0.2406532015470563,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.5345,
|
|
"mean_token_accuracy": 0.089448182284832,
|
|
"num_tokens": 508330.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.9246303081512455,
|
|
"epoch": 0.24495058014611087,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.6371,
|
|
"mean_token_accuracy": 0.08723574504256248,
|
|
"num_tokens": 517900.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.941996765136719,
|
|
"epoch": 0.24924795874516545,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.6012,
|
|
"mean_token_accuracy": 0.08463213071227074,
|
|
"num_tokens": 527808.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.864819860458374,
|
|
"epoch": 0.25354533734422,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.5607,
|
|
"mean_token_accuracy": 0.08868518695235253,
|
|
"num_tokens": 536931.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 8.03366231918335,
|
|
"epoch": 0.2578427159432746,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.5922,
|
|
"mean_token_accuracy": 0.08514139242470264,
|
|
"num_tokens": 545758.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.859347915649414,
|
|
"epoch": 0.26214009454232917,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.5998,
|
|
"mean_token_accuracy": 0.08533880487084389,
|
|
"num_tokens": 555165.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.95199933052063,
|
|
"epoch": 0.2664374731413838,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.5021,
|
|
"mean_token_accuracy": 0.09082270339131356,
|
|
"num_tokens": 564719.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.8452857494354244,
|
|
"epoch": 0.2707348517404383,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.4768,
|
|
"mean_token_accuracy": 0.0927261844277382,
|
|
"num_tokens": 573572.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.885577821731568,
|
|
"epoch": 0.2750322303394929,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.6561,
|
|
"mean_token_accuracy": 0.08534143045544625,
|
|
"num_tokens": 581497.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.828050327301026,
|
|
"epoch": 0.27932960893854747,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.5095,
|
|
"mean_token_accuracy": 0.08862045854330063,
|
|
"num_tokens": 591107.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.820580291748047,
|
|
"epoch": 0.28362698753760207,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.5394,
|
|
"mean_token_accuracy": 0.08959158807992935,
|
|
"num_tokens": 600241.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.917682647705078,
|
|
"epoch": 0.2879243661366566,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.4679,
|
|
"mean_token_accuracy": 0.087359119951725,
|
|
"num_tokens": 608697.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.84361424446106,
|
|
"epoch": 0.2922217447357112,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.4459,
|
|
"mean_token_accuracy": 0.09482443928718567,
|
|
"num_tokens": 617275.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.767910099029541,
|
|
"epoch": 0.29651912333476577,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.5397,
|
|
"mean_token_accuracy": 0.08964255526661873,
|
|
"num_tokens": 626644.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.85635404586792,
|
|
"epoch": 0.30081650193382037,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.4583,
|
|
"mean_token_accuracy": 0.09355088248848915,
|
|
"num_tokens": 635110.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.811613988876343,
|
|
"epoch": 0.30511388053287497,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.6085,
|
|
"mean_token_accuracy": 0.0853718601167202,
|
|
"num_tokens": 644746.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.915624761581421,
|
|
"epoch": 0.3094112591319295,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.5016,
|
|
"mean_token_accuracy": 0.09187210947275162,
|
|
"num_tokens": 654281.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.767735958099365,
|
|
"epoch": 0.3137086377309841,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.4717,
|
|
"mean_token_accuracy": 0.09581760317087173,
|
|
"num_tokens": 663174.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.733127641677856,
|
|
"epoch": 0.31800601633003867,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.4106,
|
|
"mean_token_accuracy": 0.09181510880589486,
|
|
"num_tokens": 672178.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.764935684204102,
|
|
"epoch": 0.32230339492909327,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.4192,
|
|
"mean_token_accuracy": 0.09482730180025101,
|
|
"num_tokens": 681323.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.748341512680054,
|
|
"epoch": 0.3266007735281478,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.3494,
|
|
"mean_token_accuracy": 0.09762883707880973,
|
|
"num_tokens": 690461.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.726337242126465,
|
|
"epoch": 0.3308981521272024,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.385,
|
|
"mean_token_accuracy": 0.0929147757589817,
|
|
"num_tokens": 699199.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.714869022369385,
|
|
"epoch": 0.33519553072625696,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.3466,
|
|
"mean_token_accuracy": 0.09913949891924859,
|
|
"num_tokens": 707949.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.790979957580566,
|
|
"epoch": 0.33949290932531156,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.4025,
|
|
"mean_token_accuracy": 0.09461145251989364,
|
|
"num_tokens": 715752.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.699273300170899,
|
|
"epoch": 0.3437902879243661,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.3498,
|
|
"mean_token_accuracy": 0.0960754208266735,
|
|
"num_tokens": 724416.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.691787385940552,
|
|
"epoch": 0.3480876665234207,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.3931,
|
|
"mean_token_accuracy": 0.09644346088171005,
|
|
"num_tokens": 733116.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.731271648406983,
|
|
"epoch": 0.3523850451224753,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.3842,
|
|
"mean_token_accuracy": 0.09576574862003326,
|
|
"num_tokens": 742093.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.707010555267334,
|
|
"epoch": 0.35668242372152986,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.3749,
|
|
"mean_token_accuracy": 0.09439021646976471,
|
|
"num_tokens": 750402.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.543523073196411,
|
|
"epoch": 0.36097980232058446,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.3356,
|
|
"mean_token_accuracy": 0.09795518517494202,
|
|
"num_tokens": 760961.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.811382913589478,
|
|
"epoch": 0.365277180919639,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.4027,
|
|
"mean_token_accuracy": 0.09610488563776017,
|
|
"num_tokens": 770554.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.553221845626831,
|
|
"epoch": 0.3695745595186936,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.3782,
|
|
"mean_token_accuracy": 0.09901996925473214,
|
|
"num_tokens": 779172.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.695575714111328,
|
|
"epoch": 0.37387193811774816,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.4409,
|
|
"mean_token_accuracy": 0.09009857177734375,
|
|
"num_tokens": 788040.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.697094821929932,
|
|
"epoch": 0.37816931671680276,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.3336,
|
|
"mean_token_accuracy": 0.0955778069794178,
|
|
"num_tokens": 796786.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.651894569396973,
|
|
"epoch": 0.3824666953158573,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.3424,
|
|
"mean_token_accuracy": 0.0917925551533699,
|
|
"num_tokens": 805520.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.574945497512817,
|
|
"epoch": 0.3867640739149119,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.2618,
|
|
"mean_token_accuracy": 0.09809256345033646,
|
|
"num_tokens": 814939.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.574750566482544,
|
|
"epoch": 0.39106145251396646,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.2789,
|
|
"mean_token_accuracy": 0.09702842682600021,
|
|
"num_tokens": 823862.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.632641363143921,
|
|
"epoch": 0.39535883111302106,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.289,
|
|
"mean_token_accuracy": 0.09691592380404472,
|
|
"num_tokens": 832820.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.676099395751953,
|
|
"epoch": 0.39965620971207566,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.2716,
|
|
"mean_token_accuracy": 0.0997233621776104,
|
|
"num_tokens": 841538.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.588865423202515,
|
|
"epoch": 0.4039535883111302,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.3935,
|
|
"mean_token_accuracy": 0.09301438555121422,
|
|
"num_tokens": 851123.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.607400226593017,
|
|
"epoch": 0.4082509669101848,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.3539,
|
|
"mean_token_accuracy": 0.10057086423039437,
|
|
"num_tokens": 860357.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.581248617172241,
|
|
"epoch": 0.41254834550923936,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.2498,
|
|
"mean_token_accuracy": 0.09947932437062264,
|
|
"num_tokens": 869980.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.5202593326568605,
|
|
"epoch": 0.41684572410829396,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.2247,
|
|
"mean_token_accuracy": 0.10081686452031136,
|
|
"num_tokens": 878250.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.625096321105957,
|
|
"epoch": 0.4211431027073485,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.3273,
|
|
"mean_token_accuracy": 0.09839151874184608,
|
|
"num_tokens": 887624.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.487894821166992,
|
|
"epoch": 0.4254404813064031,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.2795,
|
|
"mean_token_accuracy": 0.10116326585412025,
|
|
"num_tokens": 897120.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.501826333999634,
|
|
"epoch": 0.42973785990545765,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.2414,
|
|
"mean_token_accuracy": 0.1031015545129776,
|
|
"num_tokens": 906215.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.42973785990545765,
|
|
"eval_entropy": 7.422262197142249,
|
|
"eval_loss": 7.298919677734375,
|
|
"eval_mean_token_accuracy": 0.10195983114006284,
|
|
"eval_num_tokens": 906215.0,
|
|
"eval_runtime": 2.6905,
|
|
"eval_samples_per_second": 1319.07,
|
|
"eval_steps_per_second": 165.023,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.528691911697388,
|
|
"epoch": 0.43403523850451226,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.244,
|
|
"mean_token_accuracy": 0.10198947340250016,
|
|
"num_tokens": 915181.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.484815073013306,
|
|
"epoch": 0.4383326171035668,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.2539,
|
|
"mean_token_accuracy": 0.09950242042541504,
|
|
"num_tokens": 924377.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.670762062072754,
|
|
"epoch": 0.4426299957026214,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.3389,
|
|
"mean_token_accuracy": 0.09818454459309578,
|
|
"num_tokens": 933114.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.467870569229126,
|
|
"epoch": 0.44692737430167595,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.2915,
|
|
"mean_token_accuracy": 0.09697328805923462,
|
|
"num_tokens": 943306.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.557570314407348,
|
|
"epoch": 0.45122475290073055,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.1814,
|
|
"mean_token_accuracy": 0.10667803287506103,
|
|
"num_tokens": 951515.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.502040958404541,
|
|
"epoch": 0.45552213149978515,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.3163,
|
|
"mean_token_accuracy": 0.09412212558090687,
|
|
"num_tokens": 962686.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.531167888641358,
|
|
"epoch": 0.4598195100988397,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.3078,
|
|
"mean_token_accuracy": 0.09996521174907684,
|
|
"num_tokens": 972136.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.354623126983642,
|
|
"epoch": 0.4641168886978943,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.2032,
|
|
"mean_token_accuracy": 0.1003618560731411,
|
|
"num_tokens": 981301.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.538769388198853,
|
|
"epoch": 0.46841426729694885,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.2247,
|
|
"mean_token_accuracy": 0.0963795855641365,
|
|
"num_tokens": 990360.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.51032829284668,
|
|
"epoch": 0.47271164589600345,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.2726,
|
|
"mean_token_accuracy": 0.09743484109640121,
|
|
"num_tokens": 999415.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.406736850738525,
|
|
"epoch": 0.477009024495058,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.1982,
|
|
"mean_token_accuracy": 0.10159658044576644,
|
|
"num_tokens": 1008762.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.508565759658813,
|
|
"epoch": 0.4813064030941126,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.1973,
|
|
"mean_token_accuracy": 0.10663462653756142,
|
|
"num_tokens": 1017704.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.388948822021485,
|
|
"epoch": 0.48560378169316715,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.1512,
|
|
"mean_token_accuracy": 0.09958953484892845,
|
|
"num_tokens": 1026251.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.365187549591065,
|
|
"epoch": 0.48990116029222175,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.1484,
|
|
"mean_token_accuracy": 0.10498780906200408,
|
|
"num_tokens": 1036191.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.4464256286621096,
|
|
"epoch": 0.4941985388912763,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.1792,
|
|
"mean_token_accuracy": 0.10612925887107849,
|
|
"num_tokens": 1044936.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.390227842330932,
|
|
"epoch": 0.4984959174903309,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.1896,
|
|
"mean_token_accuracy": 0.10474750399589539,
|
|
"num_tokens": 1053683.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.384358596801758,
|
|
"epoch": 0.5027932960893855,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.1593,
|
|
"mean_token_accuracy": 0.10548826530575753,
|
|
"num_tokens": 1062932.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.423210000991821,
|
|
"epoch": 0.50709067468844,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.1611,
|
|
"mean_token_accuracy": 0.10090402215719223,
|
|
"num_tokens": 1072313.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.389828681945801,
|
|
"epoch": 0.5113880532874946,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.1565,
|
|
"mean_token_accuracy": 0.09966452196240425,
|
|
"num_tokens": 1081675.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.403277826309204,
|
|
"epoch": 0.5156854318865493,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.1649,
|
|
"mean_token_accuracy": 0.1019170843064785,
|
|
"num_tokens": 1091541.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.4162681102752686,
|
|
"epoch": 0.5199828104856038,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.1867,
|
|
"mean_token_accuracy": 0.10488295033574105,
|
|
"num_tokens": 1100724.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.225993299484253,
|
|
"epoch": 0.5242801890846583,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.1452,
|
|
"mean_token_accuracy": 0.10972803980112075,
|
|
"num_tokens": 1108869.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.465368556976318,
|
|
"epoch": 0.5285775676837129,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.1682,
|
|
"mean_token_accuracy": 0.10226655825972557,
|
|
"num_tokens": 1117314.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.255156230926514,
|
|
"epoch": 0.5328749462827675,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.0974,
|
|
"mean_token_accuracy": 0.10857934057712555,
|
|
"num_tokens": 1126786.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.303985261917115,
|
|
"epoch": 0.5371723248818221,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.1011,
|
|
"mean_token_accuracy": 0.10179087147116661,
|
|
"num_tokens": 1136013.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.3519556522369385,
|
|
"epoch": 0.5414697034808766,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.1306,
|
|
"mean_token_accuracy": 0.10694391876459122,
|
|
"num_tokens": 1144970.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.271116876602173,
|
|
"epoch": 0.5457670820799312,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.0926,
|
|
"mean_token_accuracy": 0.11246051043272018,
|
|
"num_tokens": 1153810.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.369027090072632,
|
|
"epoch": 0.5500644606789858,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.0851,
|
|
"mean_token_accuracy": 0.10559632927179337,
|
|
"num_tokens": 1162498.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.294491338729858,
|
|
"epoch": 0.5543618392780404,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.156,
|
|
"mean_token_accuracy": 0.1034268319606781,
|
|
"num_tokens": 1172091.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.400327205657959,
|
|
"epoch": 0.5586592178770949,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.1362,
|
|
"mean_token_accuracy": 0.10459564328193664,
|
|
"num_tokens": 1181400.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.265139150619507,
|
|
"epoch": 0.5629565964761496,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.1364,
|
|
"mean_token_accuracy": 0.10839771702885628,
|
|
"num_tokens": 1189780.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.171574592590332,
|
|
"epoch": 0.5672539750752041,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.0999,
|
|
"mean_token_accuracy": 0.10514928102493286,
|
|
"num_tokens": 1198671.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.378566837310791,
|
|
"epoch": 0.5715513536742587,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.1118,
|
|
"mean_token_accuracy": 0.10252572223544121,
|
|
"num_tokens": 1207173.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.2369647979736325,
|
|
"epoch": 0.5758487322733132,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.0444,
|
|
"mean_token_accuracy": 0.11203232631087304,
|
|
"num_tokens": 1216387.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.302586269378662,
|
|
"epoch": 0.5801461108723679,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.0134,
|
|
"mean_token_accuracy": 0.10791798532009125,
|
|
"num_tokens": 1224461.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.243020248413086,
|
|
"epoch": 0.5844434894714224,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.0899,
|
|
"mean_token_accuracy": 0.11033338010311126,
|
|
"num_tokens": 1233774.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.226488161087036,
|
|
"epoch": 0.588740868070477,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.0447,
|
|
"mean_token_accuracy": 0.11142575591802598,
|
|
"num_tokens": 1242812.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.324519062042237,
|
|
"epoch": 0.5930382466695315,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.1597,
|
|
"mean_token_accuracy": 0.10541425719857216,
|
|
"num_tokens": 1252872.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.069272375106811,
|
|
"epoch": 0.5973356252685862,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.9431,
|
|
"mean_token_accuracy": 0.10554427430033683,
|
|
"num_tokens": 1260852.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.209041118621826,
|
|
"epoch": 0.6016330038676407,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.0431,
|
|
"mean_token_accuracy": 0.11091816946864128,
|
|
"num_tokens": 1268925.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.320328855514527,
|
|
"epoch": 0.6059303824666953,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.099,
|
|
"mean_token_accuracy": 0.10238482132554054,
|
|
"num_tokens": 1278994.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.211482000350952,
|
|
"epoch": 0.6102277610657499,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.0943,
|
|
"mean_token_accuracy": 0.10070538222789764,
|
|
"num_tokens": 1287698.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.211009883880616,
|
|
"epoch": 0.6145251396648045,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.0269,
|
|
"mean_token_accuracy": 0.1104694277048111,
|
|
"num_tokens": 1297475.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.228686857223511,
|
|
"epoch": 0.618822518263859,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.0654,
|
|
"mean_token_accuracy": 0.10902067199349404,
|
|
"num_tokens": 1306836.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.297521018981934,
|
|
"epoch": 0.6231198968629136,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.0899,
|
|
"mean_token_accuracy": 0.10146719440817834,
|
|
"num_tokens": 1315872.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.149651527404785,
|
|
"epoch": 0.6274172754619682,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.0508,
|
|
"mean_token_accuracy": 0.11371816098690032,
|
|
"num_tokens": 1324624.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.250580930709839,
|
|
"epoch": 0.6317146540610228,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.0282,
|
|
"mean_token_accuracy": 0.11503755301237106,
|
|
"num_tokens": 1333058.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.036735582351684,
|
|
"epoch": 0.6360120326600773,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.997,
|
|
"mean_token_accuracy": 0.10714449658989907,
|
|
"num_tokens": 1342376.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.235709524154663,
|
|
"epoch": 0.6403094112591319,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.0323,
|
|
"mean_token_accuracy": 0.10010821968317032,
|
|
"num_tokens": 1351386.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.101161146163941,
|
|
"epoch": 0.6446067898581865,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.9373,
|
|
"mean_token_accuracy": 0.11665342599153519,
|
|
"num_tokens": 1358958.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.305012178421021,
|
|
"epoch": 0.6489041684572411,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.0889,
|
|
"mean_token_accuracy": 0.11094088032841683,
|
|
"num_tokens": 1368599.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.169938325881958,
|
|
"epoch": 0.6532015470562956,
|
|
"grad_norm": 1.8203125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.1182,
|
|
"mean_token_accuracy": 0.10421510338783264,
|
|
"num_tokens": 1378529.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.09354887008667,
|
|
"epoch": 0.6574989256553503,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.8948,
|
|
"mean_token_accuracy": 0.11917778551578521,
|
|
"num_tokens": 1386993.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.101036643981933,
|
|
"epoch": 0.6617963042544048,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.983,
|
|
"mean_token_accuracy": 0.11538739651441574,
|
|
"num_tokens": 1395790.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.051474475860596,
|
|
"epoch": 0.6660936828534594,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.9982,
|
|
"mean_token_accuracy": 0.11313480660319328,
|
|
"num_tokens": 1405587.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.074571561813355,
|
|
"epoch": 0.6703910614525139,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.9625,
|
|
"mean_token_accuracy": 0.11627124771475791,
|
|
"num_tokens": 1414478.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.257329654693604,
|
|
"epoch": 0.6746884400515686,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.0418,
|
|
"mean_token_accuracy": 0.11491334065794945,
|
|
"num_tokens": 1423791.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.128624153137207,
|
|
"epoch": 0.6789858186506231,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.0716,
|
|
"mean_token_accuracy": 0.10399920344352723,
|
|
"num_tokens": 1432955.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.181348657608032,
|
|
"epoch": 0.6832831972496777,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.0327,
|
|
"mean_token_accuracy": 0.10609091967344284,
|
|
"num_tokens": 1441907.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.047503185272217,
|
|
"epoch": 0.6875805758487322,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.9205,
|
|
"mean_token_accuracy": 0.113692706823349,
|
|
"num_tokens": 1451062.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.177039051055909,
|
|
"epoch": 0.6918779544477869,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.995,
|
|
"mean_token_accuracy": 0.10832012668251992,
|
|
"num_tokens": 1460132.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.96923713684082,
|
|
"epoch": 0.6961753330468414,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.8756,
|
|
"mean_token_accuracy": 0.1189111702144146,
|
|
"num_tokens": 1469582.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.084101915359497,
|
|
"epoch": 0.700472711645896,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.9686,
|
|
"mean_token_accuracy": 0.10740544646978378,
|
|
"num_tokens": 1479053.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.053977298736572,
|
|
"epoch": 0.7047700902449506,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.9863,
|
|
"mean_token_accuracy": 0.1092232272028923,
|
|
"num_tokens": 1488189.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.9987997055053714,
|
|
"epoch": 0.7090674688440052,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.8489,
|
|
"mean_token_accuracy": 0.11559001430869102,
|
|
"num_tokens": 1497324.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.033938598632813,
|
|
"epoch": 0.7133648474430597,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.9598,
|
|
"mean_token_accuracy": 0.11667724475264549,
|
|
"num_tokens": 1506543.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.102971458435059,
|
|
"epoch": 0.7176622260421143,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.9417,
|
|
"mean_token_accuracy": 0.1188001312315464,
|
|
"num_tokens": 1516737.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.932536888122558,
|
|
"epoch": 0.7219596046411689,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.8072,
|
|
"mean_token_accuracy": 0.11943197026848792,
|
|
"num_tokens": 1525561.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.0301721572875975,
|
|
"epoch": 0.7262569832402235,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.8388,
|
|
"mean_token_accuracy": 0.12032183110713959,
|
|
"num_tokens": 1533323.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.044216203689575,
|
|
"epoch": 0.730554361839278,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.8906,
|
|
"mean_token_accuracy": 0.1188323475420475,
|
|
"num_tokens": 1542632.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.953540372848511,
|
|
"epoch": 0.7348517404383326,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.8465,
|
|
"mean_token_accuracy": 0.11979897692799568,
|
|
"num_tokens": 1551236.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.919392871856689,
|
|
"epoch": 0.7391491190373872,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.8608,
|
|
"mean_token_accuracy": 0.11899535655975342,
|
|
"num_tokens": 1559674.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.0038087368011475,
|
|
"epoch": 0.7434464976364418,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.9342,
|
|
"mean_token_accuracy": 0.11725401654839515,
|
|
"num_tokens": 1569481.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.0618713855743405,
|
|
"epoch": 0.7477438762354963,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.9678,
|
|
"mean_token_accuracy": 0.11540251150727272,
|
|
"num_tokens": 1578488.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.945058965682984,
|
|
"epoch": 0.752041254834551,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.8741,
|
|
"mean_token_accuracy": 0.11292878687381744,
|
|
"num_tokens": 1586675.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.043763780593872,
|
|
"epoch": 0.7563386334336055,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.9852,
|
|
"mean_token_accuracy": 0.11031034588813782,
|
|
"num_tokens": 1595411.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.071803712844849,
|
|
"epoch": 0.7606360120326601,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.9491,
|
|
"mean_token_accuracy": 0.11442151814699172,
|
|
"num_tokens": 1604046.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.057859277725219,
|
|
"epoch": 0.7649333906317146,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.9508,
|
|
"mean_token_accuracy": 0.10657546371221542,
|
|
"num_tokens": 1613759.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.840588092803955,
|
|
"epoch": 0.7692307692307693,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.922,
|
|
"mean_token_accuracy": 0.11644922271370887,
|
|
"num_tokens": 1623323.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.06750636100769,
|
|
"epoch": 0.7735281478298238,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.8895,
|
|
"mean_token_accuracy": 0.11327373012900352,
|
|
"num_tokens": 1631727.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.9450620174407955,
|
|
"epoch": 0.7778255264288784,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.8165,
|
|
"mean_token_accuracy": 0.1182815119624138,
|
|
"num_tokens": 1639544.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.046970844268799,
|
|
"epoch": 0.7821229050279329,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.928,
|
|
"mean_token_accuracy": 0.11032929420471191,
|
|
"num_tokens": 1648931.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.876349306106567,
|
|
"epoch": 0.7864202836269876,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.8163,
|
|
"mean_token_accuracy": 0.11762315034866333,
|
|
"num_tokens": 1657688.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.0096639633178714,
|
|
"epoch": 0.7907176622260421,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.9127,
|
|
"mean_token_accuracy": 0.11502915844321251,
|
|
"num_tokens": 1666879.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.911726617813111,
|
|
"epoch": 0.7950150408250967,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.8745,
|
|
"mean_token_accuracy": 0.11813249662518502,
|
|
"num_tokens": 1676773.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.955066204071045,
|
|
"epoch": 0.7993124194241513,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.9842,
|
|
"mean_token_accuracy": 0.11074966862797737,
|
|
"num_tokens": 1686144.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.000700998306274,
|
|
"epoch": 0.8036097980232059,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.9536,
|
|
"mean_token_accuracy": 0.10892226025462151,
|
|
"num_tokens": 1695476.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.962090682983399,
|
|
"epoch": 0.8079071766222604,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.852,
|
|
"mean_token_accuracy": 0.11600353494286537,
|
|
"num_tokens": 1704907.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.887026691436768,
|
|
"epoch": 0.812204555221315,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.8937,
|
|
"mean_token_accuracy": 0.11855662986636162,
|
|
"num_tokens": 1714564.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.033007001876831,
|
|
"epoch": 0.8165019338203696,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.9038,
|
|
"mean_token_accuracy": 0.1119144968688488,
|
|
"num_tokens": 1725285.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.916668319702149,
|
|
"epoch": 0.8207993124194242,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.8986,
|
|
"mean_token_accuracy": 0.11464172080159188,
|
|
"num_tokens": 1734331.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.954834079742431,
|
|
"epoch": 0.8250966910184787,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.8419,
|
|
"mean_token_accuracy": 0.11120855137705803,
|
|
"num_tokens": 1742340.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.873659420013428,
|
|
"epoch": 0.8293940696175333,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.8244,
|
|
"mean_token_accuracy": 0.1190465047955513,
|
|
"num_tokens": 1751725.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.828067302703857,
|
|
"epoch": 0.8336914482165879,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.7571,
|
|
"mean_token_accuracy": 0.1218742311000824,
|
|
"num_tokens": 1760294.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.792805624008179,
|
|
"epoch": 0.8379888268156425,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.828,
|
|
"mean_token_accuracy": 0.11585493832826614,
|
|
"num_tokens": 1768912.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.068136978149414,
|
|
"epoch": 0.842286205414697,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 7.0154,
|
|
"mean_token_accuracy": 0.11436814591288566,
|
|
"num_tokens": 1778633.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.926547336578369,
|
|
"epoch": 0.8465835840137517,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.9038,
|
|
"mean_token_accuracy": 0.118341064453125,
|
|
"num_tokens": 1787275.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.972157049179077,
|
|
"epoch": 0.8508809626128062,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.8651,
|
|
"mean_token_accuracy": 0.115602745115757,
|
|
"num_tokens": 1795994.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.753544712066651,
|
|
"epoch": 0.8551783412118608,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.7981,
|
|
"mean_token_accuracy": 0.12302216216921806,
|
|
"num_tokens": 1806379.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.940453767776489,
|
|
"epoch": 0.8594757198109153,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.8262,
|
|
"mean_token_accuracy": 0.12151429355144501,
|
|
"num_tokens": 1816135.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.8594757198109153,
|
|
"eval_entropy": 6.563961205181775,
|
|
"eval_loss": 6.872446537017822,
|
|
"eval_mean_token_accuracy": 0.11898115175767793,
|
|
"eval_num_tokens": 1816135.0,
|
|
"eval_runtime": 2.6328,
|
|
"eval_samples_per_second": 1347.969,
|
|
"eval_steps_per_second": 168.639,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.861291074752808,
|
|
"epoch": 0.86377309840997,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004999998427807679,
|
|
"loss": 6.8278,
|
|
"mean_token_accuracy": 0.11725394353270531,
|
|
"num_tokens": 1824777.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.817271327972412,
|
|
"epoch": 0.8680704770090245,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0004999992040780138,
|
|
"loss": 6.8934,
|
|
"mean_token_accuracy": 0.11908305063843727,
|
|
"num_tokens": 1833807.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.912307357788086,
|
|
"epoch": 0.8723678556080791,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999980740669294,
|
|
"loss": 6.8371,
|
|
"mean_token_accuracy": 0.11982744261622429,
|
|
"num_tokens": 1843375.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.919908285140991,
|
|
"epoch": 0.8766652342071336,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999964527499823,
|
|
"loss": 6.8995,
|
|
"mean_token_accuracy": 0.11144264861941337,
|
|
"num_tokens": 1853036.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.93482608795166,
|
|
"epoch": 0.8809626128061883,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999943401307127,
|
|
"loss": 6.8674,
|
|
"mean_token_accuracy": 0.11142331585288048,
|
|
"num_tokens": 1862041.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.808039283752441,
|
|
"epoch": 0.8852599914052428,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999917362137337,
|
|
"loss": 6.7691,
|
|
"mean_token_accuracy": 0.11688561141490936,
|
|
"num_tokens": 1870707.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.84266152381897,
|
|
"epoch": 0.8895573700042974,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999886410047312,
|
|
"loss": 6.7673,
|
|
"mean_token_accuracy": 0.12210694700479507,
|
|
"num_tokens": 1879787.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.921979093551636,
|
|
"epoch": 0.8938547486033519,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999850545104638,
|
|
"loss": 6.8312,
|
|
"mean_token_accuracy": 0.11770776957273484,
|
|
"num_tokens": 1889413.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.841566896438598,
|
|
"epoch": 0.8981521272024066,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004999809767387633,
|
|
"loss": 6.8172,
|
|
"mean_token_accuracy": 0.1146760605275631,
|
|
"num_tokens": 1898283.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.860340738296509,
|
|
"epoch": 0.9024495058014611,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999764076985337,
|
|
"loss": 6.8437,
|
|
"mean_token_accuracy": 0.11705187186598778,
|
|
"num_tokens": 1907175.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.756522703170776,
|
|
"epoch": 0.9067468844005157,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999713473997519,
|
|
"loss": 6.8871,
|
|
"mean_token_accuracy": 0.11854531466960908,
|
|
"num_tokens": 1918223.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.915897035598755,
|
|
"epoch": 0.9110442629995703,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999657958534677,
|
|
"loss": 6.8237,
|
|
"mean_token_accuracy": 0.11243784725666046,
|
|
"num_tokens": 1928801.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.816089534759522,
|
|
"epoch": 0.9153416415986249,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999597530718034,
|
|
"loss": 6.7938,
|
|
"mean_token_accuracy": 0.12117088660597801,
|
|
"num_tokens": 1937406.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.779704427719116,
|
|
"epoch": 0.9196390201976794,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000499953219067954,
|
|
"loss": 6.7786,
|
|
"mean_token_accuracy": 0.11641940400004387,
|
|
"num_tokens": 1947184.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.911731958389282,
|
|
"epoch": 0.923936398796734,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999461938561873,
|
|
"loss": 6.8079,
|
|
"mean_token_accuracy": 0.11928496584296226,
|
|
"num_tokens": 1956293.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.777858400344849,
|
|
"epoch": 0.9282337773957886,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999386774518432,
|
|
"loss": 6.7726,
|
|
"mean_token_accuracy": 0.11898861080408096,
|
|
"num_tokens": 1964791.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.783628988265991,
|
|
"epoch": 0.9325311559948432,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999306698713349,
|
|
"loss": 6.6966,
|
|
"mean_token_accuracy": 0.12003475353121758,
|
|
"num_tokens": 1973754.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.859310483932495,
|
|
"epoch": 0.9368285345938977,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999221711321477,
|
|
"loss": 6.7784,
|
|
"mean_token_accuracy": 0.12257176861166955,
|
|
"num_tokens": 1983035.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.739272737503052,
|
|
"epoch": 0.9411259131929522,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999131812528393,
|
|
"loss": 6.7912,
|
|
"mean_token_accuracy": 0.12251479029655457,
|
|
"num_tokens": 1992584.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.821235084533692,
|
|
"epoch": 0.9454232917920069,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00049990370025304,
|
|
"loss": 6.8329,
|
|
"mean_token_accuracy": 0.11643655076622964,
|
|
"num_tokens": 2001876.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.851299810409546,
|
|
"epoch": 0.9497206703910615,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998937281534526,
|
|
"loss": 6.7129,
|
|
"mean_token_accuracy": 0.12695161774754524,
|
|
"num_tokens": 2011067.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.812281751632691,
|
|
"epoch": 0.954018048990116,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004998832649758521,
|
|
"loss": 6.8138,
|
|
"mean_token_accuracy": 0.11856605932116508,
|
|
"num_tokens": 2020763.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.753700637817383,
|
|
"epoch": 0.9583154275891707,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004998723107430862,
|
|
"loss": 6.7788,
|
|
"mean_token_accuracy": 0.12092987298965455,
|
|
"num_tokens": 2029534.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.84175591468811,
|
|
"epoch": 0.9626128061882252,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004998608654790741,
|
|
"loss": 6.7432,
|
|
"mean_token_accuracy": 0.11753956079483033,
|
|
"num_tokens": 2039143.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.721464443206787,
|
|
"epoch": 0.9669101847872797,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000499848929208808,
|
|
"loss": 6.6916,
|
|
"mean_token_accuracy": 0.12772009372711182,
|
|
"num_tokens": 2048253.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.705768156051636,
|
|
"epoch": 0.9712075633863343,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004998365019583519,
|
|
"loss": 6.7479,
|
|
"mean_token_accuracy": 0.12803810015320777,
|
|
"num_tokens": 2057234.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.848254632949829,
|
|
"epoch": 0.975504941985389,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004998235837548417,
|
|
"loss": 6.7967,
|
|
"mean_token_accuracy": 0.12506601735949516,
|
|
"num_tokens": 2065431.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.739435911178589,
|
|
"epoch": 0.9798023205844435,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000499810174626486,
|
|
"loss": 6.7851,
|
|
"mean_token_accuracy": 0.12419307976961136,
|
|
"num_tokens": 2074723.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.76078782081604,
|
|
"epoch": 0.984099699183498,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004997962746025646,
|
|
"loss": 6.6718,
|
|
"mean_token_accuracy": 0.1286655180156231,
|
|
"num_tokens": 2084509.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.807924270629883,
|
|
"epoch": 0.9883970777825526,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004997818837134298,
|
|
"loss": 6.8029,
|
|
"mean_token_accuracy": 0.12207063660025597,
|
|
"num_tokens": 2093110.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.6994280338287355,
|
|
"epoch": 0.9926944563816072,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004997670019905057,
|
|
"loss": 6.6601,
|
|
"mean_token_accuracy": 0.12511659264564515,
|
|
"num_tokens": 2102355.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.756126117706299,
|
|
"epoch": 0.9969918349806618,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004997516294662876,
|
|
"loss": 6.7133,
|
|
"mean_token_accuracy": 0.12134261280298234,
|
|
"num_tokens": 2110418.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.7629002465142145,
|
|
"epoch": 1.0008594757198108,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004997357661743433,
|
|
"loss": 6.686,
|
|
"mean_token_accuracy": 0.12661833895577324,
|
|
"num_tokens": 2117866.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.680280160903931,
|
|
"epoch": 1.0051568543188656,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004997194121493118,
|
|
"loss": 6.5054,
|
|
"mean_token_accuracy": 0.12853871434926986,
|
|
"num_tokens": 2126082.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.644075536727906,
|
|
"epoch": 1.0094542329179201,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004997025674269037,
|
|
"loss": 6.4665,
|
|
"mean_token_accuracy": 0.13536003082990647,
|
|
"num_tokens": 2134042.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.691802644729615,
|
|
"epoch": 1.0137516115169747,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004996852320439013,
|
|
"loss": 6.5639,
|
|
"mean_token_accuracy": 0.13069004118442534,
|
|
"num_tokens": 2142570.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.641972494125366,
|
|
"epoch": 1.0180489901160292,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004996674060381578,
|
|
"loss": 6.5072,
|
|
"mean_token_accuracy": 0.12871361523866653,
|
|
"num_tokens": 2151310.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.6996808528900145,
|
|
"epoch": 1.0223463687150838,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004996490894485985,
|
|
"loss": 6.552,
|
|
"mean_token_accuracy": 0.129477708786726,
|
|
"num_tokens": 2160662.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.646242380142212,
|
|
"epoch": 1.0266437473141383,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004996302823152193,
|
|
"loss": 6.5101,
|
|
"mean_token_accuracy": 0.13247828036546708,
|
|
"num_tokens": 2170067.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.618408918380737,
|
|
"epoch": 1.0309411259131929,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004996109846790873,
|
|
"loss": 6.4723,
|
|
"mean_token_accuracy": 0.13233422860503197,
|
|
"num_tokens": 2178850.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.592642974853516,
|
|
"epoch": 1.0352385045122476,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004995911965823412,
|
|
"loss": 6.4956,
|
|
"mean_token_accuracy": 0.13544322028756142,
|
|
"num_tokens": 2188307.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.665491533279419,
|
|
"epoch": 1.0395358831113022,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004995709180681899,
|
|
"loss": 6.4958,
|
|
"mean_token_accuracy": 0.1364510491490364,
|
|
"num_tokens": 2197026.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.558549213409424,
|
|
"epoch": 1.0438332617103567,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000499550149180914,
|
|
"loss": 6.4866,
|
|
"mean_token_accuracy": 0.1290366992354393,
|
|
"num_tokens": 2205537.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.709677457809448,
|
|
"epoch": 1.0481306403094113,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004995288899658641,
|
|
"loss": 6.4893,
|
|
"mean_token_accuracy": 0.1379025362432003,
|
|
"num_tokens": 2214508.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.602940797805786,
|
|
"epoch": 1.0524280189084658,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004995071404694619,
|
|
"loss": 6.6098,
|
|
"mean_token_accuracy": 0.1207184799015522,
|
|
"num_tokens": 2223084.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.706255054473877,
|
|
"epoch": 1.0567253975075204,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004994849007391996,
|
|
"loss": 6.5283,
|
|
"mean_token_accuracy": 0.12673679515719413,
|
|
"num_tokens": 2231406.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.561632061004639,
|
|
"epoch": 1.061022776106575,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004994621708236401,
|
|
"loss": 6.4622,
|
|
"mean_token_accuracy": 0.12985079735517502,
|
|
"num_tokens": 2239867.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.596746301651001,
|
|
"epoch": 1.0653201547056295,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000499438950772416,
|
|
"loss": 6.5299,
|
|
"mean_token_accuracy": 0.1289393313229084,
|
|
"num_tokens": 2248844.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.597238826751709,
|
|
"epoch": 1.0696175333046842,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004994152406362311,
|
|
"loss": 6.4418,
|
|
"mean_token_accuracy": 0.13728953748941422,
|
|
"num_tokens": 2257599.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.583277225494385,
|
|
"epoch": 1.0739149119037388,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004993910404668586,
|
|
"loss": 6.4782,
|
|
"mean_token_accuracy": 0.13000380396842956,
|
|
"num_tokens": 2266510.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.601662445068359,
|
|
"epoch": 1.0782122905027933,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000499366350317142,
|
|
"loss": 6.4784,
|
|
"mean_token_accuracy": 0.13205466270446778,
|
|
"num_tokens": 2275462.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.531731128692627,
|
|
"epoch": 1.0825096691018479,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004993411702409948,
|
|
"loss": 6.4464,
|
|
"mean_token_accuracy": 0.13146719262003898,
|
|
"num_tokens": 2283826.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.632365322113037,
|
|
"epoch": 1.0868070477009024,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004993155002934002,
|
|
"loss": 6.462,
|
|
"mean_token_accuracy": 0.13637664988636972,
|
|
"num_tokens": 2292967.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.631603384017945,
|
|
"epoch": 1.091104426299957,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0004992893405304111,
|
|
"loss": 6.6112,
|
|
"mean_token_accuracy": 0.12938740327954293,
|
|
"num_tokens": 2302336.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.602935266494751,
|
|
"epoch": 1.0954018048990115,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00049926269100915,
|
|
"loss": 6.5089,
|
|
"mean_token_accuracy": 0.13883564844727517,
|
|
"num_tokens": 2311465.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.622984075546265,
|
|
"epoch": 1.0996991834980663,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004992355517878087,
|
|
"loss": 6.6227,
|
|
"mean_token_accuracy": 0.12175218909978866,
|
|
"num_tokens": 2320281.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.5945323467254635,
|
|
"epoch": 1.1039965620971208,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004992079229256484,
|
|
"loss": 6.517,
|
|
"mean_token_accuracy": 0.12854354977607726,
|
|
"num_tokens": 2329755.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.532581329345703,
|
|
"epoch": 1.1082939406961754,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004991798044829996,
|
|
"loss": 6.4556,
|
|
"mean_token_accuracy": 0.13279438391327858,
|
|
"num_tokens": 2338807.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.675839281082153,
|
|
"epoch": 1.11259131929523,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004991511965212618,
|
|
"loss": 6.5491,
|
|
"mean_token_accuracy": 0.13110856264829635,
|
|
"num_tokens": 2348056.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.51497597694397,
|
|
"epoch": 1.1168886978942845,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004991220991029032,
|
|
"loss": 6.5672,
|
|
"mean_token_accuracy": 0.12991769686341287,
|
|
"num_tokens": 2357780.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.638486289978028,
|
|
"epoch": 1.121186076493339,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000499092512291461,
|
|
"loss": 6.5353,
|
|
"mean_token_accuracy": 0.1307793602347374,
|
|
"num_tokens": 2367060.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.606591844558716,
|
|
"epoch": 1.1254834550923936,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000499062436151541,
|
|
"loss": 6.5301,
|
|
"mean_token_accuracy": 0.13085475713014602,
|
|
"num_tokens": 2375751.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.622772312164306,
|
|
"epoch": 1.129780833691448,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004990318707488173,
|
|
"loss": 6.5766,
|
|
"mean_token_accuracy": 0.12039525136351585,
|
|
"num_tokens": 2385013.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.605566358566284,
|
|
"epoch": 1.1340782122905029,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004990008161500327,
|
|
"loss": 6.4701,
|
|
"mean_token_accuracy": 0.13520175516605376,
|
|
"num_tokens": 2392935.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.560553312301636,
|
|
"epoch": 1.1383755908895574,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000498969272422998,
|
|
"loss": 6.4844,
|
|
"mean_token_accuracy": 0.13814345598220826,
|
|
"num_tokens": 2401560.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.537953233718872,
|
|
"epoch": 1.142672969488612,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004989372396365921,
|
|
"loss": 6.4217,
|
|
"mean_token_accuracy": 0.1364870399236679,
|
|
"num_tokens": 2410050.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.616376876831055,
|
|
"epoch": 1.1469703480876665,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004989047178607618,
|
|
"loss": 6.5241,
|
|
"mean_token_accuracy": 0.13191651478409766,
|
|
"num_tokens": 2418980.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.586575651168824,
|
|
"epoch": 1.151267726686721,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004988717071665215,
|
|
"loss": 6.5044,
|
|
"mean_token_accuracy": 0.13132359310984612,
|
|
"num_tokens": 2427992.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.516159582138061,
|
|
"epoch": 1.1555651052857756,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004988382076259537,
|
|
"loss": 6.4354,
|
|
"mean_token_accuracy": 0.1375439666211605,
|
|
"num_tokens": 2436368.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.502273654937744,
|
|
"epoch": 1.1598624838848304,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004988042193122077,
|
|
"loss": 6.4125,
|
|
"mean_token_accuracy": 0.139194605499506,
|
|
"num_tokens": 2445499.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.631434583663941,
|
|
"epoch": 1.164159862483885,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004987697422995005,
|
|
"loss": 6.4497,
|
|
"mean_token_accuracy": 0.1331012539565563,
|
|
"num_tokens": 2454312.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.533642435073853,
|
|
"epoch": 1.1684572410829395,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004987347766631161,
|
|
"loss": 6.5192,
|
|
"mean_token_accuracy": 0.1333911769092083,
|
|
"num_tokens": 2462922.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.590119934082031,
|
|
"epoch": 1.172754619681994,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004986993224794055,
|
|
"loss": 6.558,
|
|
"mean_token_accuracy": 0.12367333769798279,
|
|
"num_tokens": 2472195.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.575766134262085,
|
|
"epoch": 1.1770519982810486,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004986633798257865,
|
|
"loss": 6.4568,
|
|
"mean_token_accuracy": 0.13813585937023162,
|
|
"num_tokens": 2481021.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.527914094924927,
|
|
"epoch": 1.181349376880103,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004986269487807434,
|
|
"loss": 6.4701,
|
|
"mean_token_accuracy": 0.12878329530358315,
|
|
"num_tokens": 2490250.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.606501150131225,
|
|
"epoch": 1.1856467554791577,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000498590029423827,
|
|
"loss": 6.5358,
|
|
"mean_token_accuracy": 0.13699908778071404,
|
|
"num_tokens": 2499122.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.5601091384887695,
|
|
"epoch": 1.1899441340782122,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004985526218356546,
|
|
"loss": 6.4994,
|
|
"mean_token_accuracy": 0.13007601723074913,
|
|
"num_tokens": 2508454.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.563156938552856,
|
|
"epoch": 1.1942415126772667,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004985147260979093,
|
|
"loss": 6.4474,
|
|
"mean_token_accuracy": 0.14253669306635858,
|
|
"num_tokens": 2517353.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.568882608413697,
|
|
"epoch": 1.1985388912763215,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004984763422933402,
|
|
"loss": 6.4666,
|
|
"mean_token_accuracy": 0.13579432666301727,
|
|
"num_tokens": 2526321.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.519547700881958,
|
|
"epoch": 1.202836269875376,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004984374705057623,
|
|
"loss": 6.5024,
|
|
"mean_token_accuracy": 0.13173489049077033,
|
|
"num_tokens": 2535924.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.517468738555908,
|
|
"epoch": 1.2071336484744306,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004983981108200561,
|
|
"loss": 6.4795,
|
|
"mean_token_accuracy": 0.12949901297688485,
|
|
"num_tokens": 2545606.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.484243726730346,
|
|
"epoch": 1.2114310270734852,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004983582633221672,
|
|
"loss": 6.4541,
|
|
"mean_token_accuracy": 0.13506132885813713,
|
|
"num_tokens": 2554947.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.622135829925537,
|
|
"epoch": 1.2157284056725397,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004983179280991068,
|
|
"loss": 6.6325,
|
|
"mean_token_accuracy": 0.12418173924088478,
|
|
"num_tokens": 2564462.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.580014705657959,
|
|
"epoch": 1.2200257842715942,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004982771052389508,
|
|
"loss": 6.4603,
|
|
"mean_token_accuracy": 0.13684669733047486,
|
|
"num_tokens": 2573124.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.583792018890381,
|
|
"epoch": 1.224323162870649,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004982357948308401,
|
|
"loss": 6.5423,
|
|
"mean_token_accuracy": 0.12636070176959038,
|
|
"num_tokens": 2581829.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.592307043075562,
|
|
"epoch": 1.2286205414697036,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004981939969649799,
|
|
"loss": 6.3994,
|
|
"mean_token_accuracy": 0.13814754113554956,
|
|
"num_tokens": 2590631.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.545891141891479,
|
|
"epoch": 1.232917920068758,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004981517117326404,
|
|
"loss": 6.5281,
|
|
"mean_token_accuracy": 0.1337512508034706,
|
|
"num_tokens": 2600684.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.528186511993408,
|
|
"epoch": 1.2372152986678127,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004981089392261553,
|
|
"loss": 6.4338,
|
|
"mean_token_accuracy": 0.13892308846116067,
|
|
"num_tokens": 2609667.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.540981245040894,
|
|
"epoch": 1.2415126772668672,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000498065679538923,
|
|
"loss": 6.5222,
|
|
"mean_token_accuracy": 0.13639934808015824,
|
|
"num_tokens": 2620025.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.587842559814453,
|
|
"epoch": 1.2458100558659218,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004980219327654049,
|
|
"loss": 6.4196,
|
|
"mean_token_accuracy": 0.13706353530287743,
|
|
"num_tokens": 2629032.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.427018499374389,
|
|
"epoch": 1.2501074344649763,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000497977699001127,
|
|
"loss": 6.4228,
|
|
"mean_token_accuracy": 0.13975878730416297,
|
|
"num_tokens": 2638303.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.576618957519531,
|
|
"epoch": 1.2544048130640308,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004979329783426778,
|
|
"loss": 6.4477,
|
|
"mean_token_accuracy": 0.13641118854284287,
|
|
"num_tokens": 2647902.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.471656036376953,
|
|
"epoch": 1.2587021916630854,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004978877708877094,
|
|
"loss": 6.4848,
|
|
"mean_token_accuracy": 0.13354442045092582,
|
|
"num_tokens": 2657902.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.528874826431275,
|
|
"epoch": 1.2629995702621402,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004978420767349368,
|
|
"loss": 6.4302,
|
|
"mean_token_accuracy": 0.13373022601008416,
|
|
"num_tokens": 2667082.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.55286979675293,
|
|
"epoch": 1.2672969488611947,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004977958959841379,
|
|
"loss": 6.5005,
|
|
"mean_token_accuracy": 0.13279795721173288,
|
|
"num_tokens": 2676855.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.536699438095093,
|
|
"epoch": 1.2715943274602493,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000497749228736153,
|
|
"loss": 6.4391,
|
|
"mean_token_accuracy": 0.13942998126149178,
|
|
"num_tokens": 2685750.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.488428926467895,
|
|
"epoch": 1.2758917060593038,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004977020750928845,
|
|
"loss": 6.4966,
|
|
"mean_token_accuracy": 0.13914359211921692,
|
|
"num_tokens": 2695272.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.541899490356445,
|
|
"epoch": 1.2801890846583583,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004976544351572973,
|
|
"loss": 6.4215,
|
|
"mean_token_accuracy": 0.14113220050930977,
|
|
"num_tokens": 2704806.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.387812709808349,
|
|
"epoch": 1.2844864632574131,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004976063090334179,
|
|
"loss": 6.4992,
|
|
"mean_token_accuracy": 0.13385646566748619,
|
|
"num_tokens": 2713521.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.618979215621948,
|
|
"epoch": 1.2887838418564677,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004975576968263346,
|
|
"loss": 6.4657,
|
|
"mean_token_accuracy": 0.12805166989564895,
|
|
"num_tokens": 2721848.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.2887838418564677,
|
|
"eval_entropy": 6.385493223731582,
|
|
"eval_loss": 6.580361366271973,
|
|
"eval_mean_token_accuracy": 0.13450762674816558,
|
|
"eval_num_tokens": 2721848.0,
|
|
"eval_runtime": 2.8577,
|
|
"eval_samples_per_second": 1241.893,
|
|
"eval_steps_per_second": 155.368,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.412834739685058,
|
|
"epoch": 1.2930812204555222,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000497508598642197,
|
|
"loss": 6.4577,
|
|
"mean_token_accuracy": 0.13439935967326164,
|
|
"num_tokens": 2731473.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.588954448699951,
|
|
"epoch": 1.2973785990545768,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000497459014588216,
|
|
"loss": 6.52,
|
|
"mean_token_accuracy": 0.12975201904773712,
|
|
"num_tokens": 2739867.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.519440317153931,
|
|
"epoch": 1.3016759776536313,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000497408944772663,
|
|
"loss": 6.4321,
|
|
"mean_token_accuracy": 0.13900325521826745,
|
|
"num_tokens": 2748903.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.470726060867309,
|
|
"epoch": 1.3059733562526858,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004973583893048707,
|
|
"loss": 6.418,
|
|
"mean_token_accuracy": 0.1378275766968727,
|
|
"num_tokens": 2757711.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.597595596313477,
|
|
"epoch": 1.3102707348517404,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004973073482952321,
|
|
"loss": 6.4339,
|
|
"mean_token_accuracy": 0.13434917181730271,
|
|
"num_tokens": 2765633.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.413757419586181,
|
|
"epoch": 1.314568113450795,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004972558218552004,
|
|
"loss": 6.4602,
|
|
"mean_token_accuracy": 0.1351465404033661,
|
|
"num_tokens": 2774495.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.588878774642945,
|
|
"epoch": 1.3188654920498495,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004972038100972885,
|
|
"loss": 6.4914,
|
|
"mean_token_accuracy": 0.13447514474391936,
|
|
"num_tokens": 2782665.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.44925446510315,
|
|
"epoch": 1.323162870648904,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004971513131350697,
|
|
"loss": 6.4269,
|
|
"mean_token_accuracy": 0.13261223956942558,
|
|
"num_tokens": 2791394.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.476827239990234,
|
|
"epoch": 1.3274602492479588,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004970983310831759,
|
|
"loss": 6.4219,
|
|
"mean_token_accuracy": 0.1362217642366886,
|
|
"num_tokens": 2800488.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.473262977600098,
|
|
"epoch": 1.3317576278470133,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004970448640572989,
|
|
"loss": 6.5419,
|
|
"mean_token_accuracy": 0.13125557377934455,
|
|
"num_tokens": 2810116.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.539748430252075,
|
|
"epoch": 1.336055006446068,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004969909121741895,
|
|
"loss": 6.3387,
|
|
"mean_token_accuracy": 0.13630159422755242,
|
|
"num_tokens": 2819205.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.491139221191406,
|
|
"epoch": 1.3403523850451224,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004969364755516569,
|
|
"loss": 6.4171,
|
|
"mean_token_accuracy": 0.13611557260155677,
|
|
"num_tokens": 2828017.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.500842714309693,
|
|
"epoch": 1.344649763644177,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004968815543085689,
|
|
"loss": 6.4632,
|
|
"mean_token_accuracy": 0.13603930696845054,
|
|
"num_tokens": 2837125.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.474975538253784,
|
|
"epoch": 1.3489471422432318,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004968261485648516,
|
|
"loss": 6.4718,
|
|
"mean_token_accuracy": 0.1337704412639141,
|
|
"num_tokens": 2845438.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.5300768375396725,
|
|
"epoch": 1.3532445208422863,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000496770258441489,
|
|
"loss": 6.4316,
|
|
"mean_token_accuracy": 0.14049385190010072,
|
|
"num_tokens": 2854389.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.487020444869995,
|
|
"epoch": 1.3575418994413408,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004967138840605228,
|
|
"loss": 6.4089,
|
|
"mean_token_accuracy": 0.13504694476723672,
|
|
"num_tokens": 2863654.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.360049486160278,
|
|
"epoch": 1.3618392780403954,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000496657025545052,
|
|
"loss": 6.2999,
|
|
"mean_token_accuracy": 0.1438663646578789,
|
|
"num_tokens": 2872871.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.42272572517395,
|
|
"epoch": 1.36613665663945,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000496599683019233,
|
|
"loss": 6.42,
|
|
"mean_token_accuracy": 0.1387701526284218,
|
|
"num_tokens": 2881140.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.575308084487915,
|
|
"epoch": 1.3704340352385045,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000496541856608279,
|
|
"loss": 6.4135,
|
|
"mean_token_accuracy": 0.13202539682388306,
|
|
"num_tokens": 2889945.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.333611106872558,
|
|
"epoch": 1.374731413837559,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004964835464384595,
|
|
"loss": 6.3367,
|
|
"mean_token_accuracy": 0.1428491897881031,
|
|
"num_tokens": 2898897.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.455891847610474,
|
|
"epoch": 1.3790287924366136,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000496424752637101,
|
|
"loss": 6.3672,
|
|
"mean_token_accuracy": 0.1456029012799263,
|
|
"num_tokens": 2907717.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.4216039180755615,
|
|
"epoch": 1.3833261710356681,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004963654753325853,
|
|
"loss": 6.3559,
|
|
"mean_token_accuracy": 0.13982106298208236,
|
|
"num_tokens": 2916213.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.479219532012939,
|
|
"epoch": 1.387623549634723,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004963057146543505,
|
|
"loss": 6.5087,
|
|
"mean_token_accuracy": 0.1328582361340523,
|
|
"num_tokens": 2925706.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.451479721069336,
|
|
"epoch": 1.3919209282337774,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00049624547073289,
|
|
"loss": 6.4287,
|
|
"mean_token_accuracy": 0.13399232998490335,
|
|
"num_tokens": 2934464.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.463989019393921,
|
|
"epoch": 1.396218306832832,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004961847436997526,
|
|
"loss": 6.3207,
|
|
"mean_token_accuracy": 0.14708311855793,
|
|
"num_tokens": 2944095.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.309212732315063,
|
|
"epoch": 1.4005156854318865,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004961235336875416,
|
|
"loss": 6.3412,
|
|
"mean_token_accuracy": 0.1457139991223812,
|
|
"num_tokens": 2953357.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.458098363876343,
|
|
"epoch": 1.404813064030941,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004960618408299154,
|
|
"loss": 6.4761,
|
|
"mean_token_accuracy": 0.12848138958215713,
|
|
"num_tokens": 2963020.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.458608770370484,
|
|
"epoch": 1.4091104426299956,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004959996652615865,
|
|
"loss": 6.3262,
|
|
"mean_token_accuracy": 0.13877202272415162,
|
|
"num_tokens": 2971955.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.471976852416992,
|
|
"epoch": 1.4134078212290504,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004959370071183216,
|
|
"loss": 6.3798,
|
|
"mean_token_accuracy": 0.13860434293746948,
|
|
"num_tokens": 2980662.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.479498481750488,
|
|
"epoch": 1.417705199828105,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004958738665369407,
|
|
"loss": 6.5272,
|
|
"mean_token_accuracy": 0.12508682459592818,
|
|
"num_tokens": 2990038.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.433883905410767,
|
|
"epoch": 1.4220025784271595,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004958102436553179,
|
|
"loss": 6.4183,
|
|
"mean_token_accuracy": 0.1340093955397606,
|
|
"num_tokens": 2999835.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.503619003295898,
|
|
"epoch": 1.426299957026214,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00049574613861238,
|
|
"loss": 6.407,
|
|
"mean_token_accuracy": 0.13849962279200553,
|
|
"num_tokens": 3009593.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.535696744918823,
|
|
"epoch": 1.4305973356252686,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004956815515481069,
|
|
"loss": 6.4557,
|
|
"mean_token_accuracy": 0.13971493393182755,
|
|
"num_tokens": 3019187.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.430892753601074,
|
|
"epoch": 1.4348947142243231,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004956164826035309,
|
|
"loss": 6.3571,
|
|
"mean_token_accuracy": 0.1340019442141056,
|
|
"num_tokens": 3027875.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.419513416290283,
|
|
"epoch": 1.4391920928233777,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004955509319207363,
|
|
"loss": 6.3898,
|
|
"mean_token_accuracy": 0.1386251926422119,
|
|
"num_tokens": 3036902.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.3557569026947025,
|
|
"epoch": 1.4434894714224322,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004954848996428601,
|
|
"loss": 6.3789,
|
|
"mean_token_accuracy": 0.1433347888290882,
|
|
"num_tokens": 3046653.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.453281450271606,
|
|
"epoch": 1.4477868500214868,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00049541838591409,
|
|
"loss": 6.4503,
|
|
"mean_token_accuracy": 0.13477321714162827,
|
|
"num_tokens": 3056273.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.432342481613159,
|
|
"epoch": 1.4520842286205415,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004953513908796657,
|
|
"loss": 6.3593,
|
|
"mean_token_accuracy": 0.1364903450012207,
|
|
"num_tokens": 3065662.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.50225305557251,
|
|
"epoch": 1.456381607219596,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004952839146858773,
|
|
"loss": 6.4253,
|
|
"mean_token_accuracy": 0.1424226589500904,
|
|
"num_tokens": 3073970.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.386947154998779,
|
|
"epoch": 1.4606789858186506,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004952159574800658,
|
|
"loss": 6.401,
|
|
"mean_token_accuracy": 0.13235734030604362,
|
|
"num_tokens": 3082500.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.470129728317261,
|
|
"epoch": 1.4649763644177052,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004951475194106229,
|
|
"loss": 6.3651,
|
|
"mean_token_accuracy": 0.1414753317832947,
|
|
"num_tokens": 3091574.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.408327865600586,
|
|
"epoch": 1.4692737430167597,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004950786006269898,
|
|
"loss": 6.4654,
|
|
"mean_token_accuracy": 0.1278534062206745,
|
|
"num_tokens": 3102402.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.451749992370606,
|
|
"epoch": 1.4735711216158143,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004950092012796576,
|
|
"loss": 6.2813,
|
|
"mean_token_accuracy": 0.14528158232569693,
|
|
"num_tokens": 3111347.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.4014458656311035,
|
|
"epoch": 1.477868500214869,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004949393215201666,
|
|
"loss": 6.3471,
|
|
"mean_token_accuracy": 0.13892963156104088,
|
|
"num_tokens": 3120018.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.461446380615234,
|
|
"epoch": 1.4821658788139236,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004948689615011065,
|
|
"loss": 6.4143,
|
|
"mean_token_accuracy": 0.1317427448928356,
|
|
"num_tokens": 3129669.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.427045392990112,
|
|
"epoch": 1.4864632574129781,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004947981213761154,
|
|
"loss": 6.3543,
|
|
"mean_token_accuracy": 0.14193105772137643,
|
|
"num_tokens": 3139112.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.3959253311157225,
|
|
"epoch": 1.4907606360120327,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004947268012998797,
|
|
"loss": 6.3047,
|
|
"mean_token_accuracy": 0.14767742902040482,
|
|
"num_tokens": 3148437.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.386318588256836,
|
|
"epoch": 1.4950580146110872,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000494655001428134,
|
|
"loss": 6.3062,
|
|
"mean_token_accuracy": 0.14006873443722725,
|
|
"num_tokens": 3158165.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.411319255828857,
|
|
"epoch": 1.4993553932101418,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004945827219176604,
|
|
"loss": 6.3906,
|
|
"mean_token_accuracy": 0.14335802048444748,
|
|
"num_tokens": 3167262.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.351555633544922,
|
|
"epoch": 1.5036527718091963,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004945099629262888,
|
|
"loss": 6.3776,
|
|
"mean_token_accuracy": 0.14216644018888475,
|
|
"num_tokens": 3176696.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.561797857284546,
|
|
"epoch": 1.5079501504082509,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004944367246128954,
|
|
"loss": 6.4608,
|
|
"mean_token_accuracy": 0.13308502212166787,
|
|
"num_tokens": 3185857.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.3458091735839846,
|
|
"epoch": 1.5122475290073054,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004943630071374036,
|
|
"loss": 6.294,
|
|
"mean_token_accuracy": 0.14438677504658698,
|
|
"num_tokens": 3194687.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.426305198669434,
|
|
"epoch": 1.51654490760636,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004942888106607828,
|
|
"loss": 6.3514,
|
|
"mean_token_accuracy": 0.1347072847187519,
|
|
"num_tokens": 3204913.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.46712646484375,
|
|
"epoch": 1.5208422862054147,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004942141353450486,
|
|
"loss": 6.321,
|
|
"mean_token_accuracy": 0.14456397369503976,
|
|
"num_tokens": 3213312.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.291261672973633,
|
|
"epoch": 1.5251396648044693,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004941389813532619,
|
|
"loss": 6.25,
|
|
"mean_token_accuracy": 0.14945072531700135,
|
|
"num_tokens": 3222992.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.338441944122314,
|
|
"epoch": 1.5294370434035238,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000494063348849529,
|
|
"loss": 6.3064,
|
|
"mean_token_accuracy": 0.14241133108735085,
|
|
"num_tokens": 3232836.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.429257488250732,
|
|
"epoch": 1.5337344220025786,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004939872379990011,
|
|
"loss": 6.4578,
|
|
"mean_token_accuracy": 0.13550666123628616,
|
|
"num_tokens": 3243171.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.530871820449829,
|
|
"epoch": 1.5380318006016331,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004939106489678739,
|
|
"loss": 6.3859,
|
|
"mean_token_accuracy": 0.13990562558174133,
|
|
"num_tokens": 3251995.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.312516927719116,
|
|
"epoch": 1.5423291792006877,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000493833581923387,
|
|
"loss": 6.3353,
|
|
"mean_token_accuracy": 0.1444368615746498,
|
|
"num_tokens": 3260841.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.4656861305236815,
|
|
"epoch": 1.5466265577997422,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004937560370338244,
|
|
"loss": 6.4734,
|
|
"mean_token_accuracy": 0.13109952881932257,
|
|
"num_tokens": 3270979.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.515646362304688,
|
|
"epoch": 1.5509239363987968,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000493678014468513,
|
|
"loss": 6.394,
|
|
"mean_token_accuracy": 0.14090235903859138,
|
|
"num_tokens": 3279848.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.3691669464111325,
|
|
"epoch": 1.5552213149978513,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004935995143978227,
|
|
"loss": 6.3926,
|
|
"mean_token_accuracy": 0.1406005583703518,
|
|
"num_tokens": 3289172.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.388995122909546,
|
|
"epoch": 1.5595186935969059,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004935205369931664,
|
|
"loss": 6.2689,
|
|
"mean_token_accuracy": 0.14269133359193803,
|
|
"num_tokens": 3297432.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.503047513961792,
|
|
"epoch": 1.5638160721959604,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004934410824269992,
|
|
"loss": 6.3165,
|
|
"mean_token_accuracy": 0.1426151879131794,
|
|
"num_tokens": 3307486.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.330971145629883,
|
|
"epoch": 1.568113450795015,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004933611508728182,
|
|
"loss": 6.2791,
|
|
"mean_token_accuracy": 0.14864852502942086,
|
|
"num_tokens": 3316296.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.371285772323608,
|
|
"epoch": 1.5724108293940695,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000493280742505162,
|
|
"loss": 6.3225,
|
|
"mean_token_accuracy": 0.1385447084903717,
|
|
"num_tokens": 3326080.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.3141227722167965,
|
|
"epoch": 1.576708207993124,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004931998574996102,
|
|
"loss": 6.2469,
|
|
"mean_token_accuracy": 0.14493944942951204,
|
|
"num_tokens": 3334826.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.316582584381104,
|
|
"epoch": 1.5810055865921788,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004931184960327832,
|
|
"loss": 6.2489,
|
|
"mean_token_accuracy": 0.15021354407072068,
|
|
"num_tokens": 3343261.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.321848964691162,
|
|
"epoch": 1.5853029651912334,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.0004930366582823421,
|
|
"loss": 6.2997,
|
|
"mean_token_accuracy": 0.14123934581875802,
|
|
"num_tokens": 3352513.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.369439888000488,
|
|
"epoch": 1.589600343790288,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004929543444269879,
|
|
"loss": 6.3258,
|
|
"mean_token_accuracy": 0.14584648087620736,
|
|
"num_tokens": 3361577.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.393662023544311,
|
|
"epoch": 1.5938977223893425,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000492871554646461,
|
|
"loss": 6.4149,
|
|
"mean_token_accuracy": 0.13381075039505957,
|
|
"num_tokens": 3370591.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.351645469665527,
|
|
"epoch": 1.5981951009883972,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004927882891215413,
|
|
"loss": 6.3108,
|
|
"mean_token_accuracy": 0.14434999078512192,
|
|
"num_tokens": 3379761.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.452319765090943,
|
|
"epoch": 1.6024924795874518,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004927045480340475,
|
|
"loss": 6.4137,
|
|
"mean_token_accuracy": 0.13306379094719886,
|
|
"num_tokens": 3388974.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.364160060882568,
|
|
"epoch": 1.6067898581865063,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004926203315668363,
|
|
"loss": 6.3018,
|
|
"mean_token_accuracy": 0.1436440296471119,
|
|
"num_tokens": 3398339.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.3650538444519045,
|
|
"epoch": 1.6110872367855609,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004925356399038032,
|
|
"loss": 6.3056,
|
|
"mean_token_accuracy": 0.1379922404885292,
|
|
"num_tokens": 3408292.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.3104856491088865,
|
|
"epoch": 1.6153846153846154,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004924504732298808,
|
|
"loss": 6.2562,
|
|
"mean_token_accuracy": 0.14880003780126572,
|
|
"num_tokens": 3417057.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.3915791511535645,
|
|
"epoch": 1.61968199398367,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004923648317310391,
|
|
"loss": 6.3722,
|
|
"mean_token_accuracy": 0.1378080278635025,
|
|
"num_tokens": 3425830.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.340031433105469,
|
|
"epoch": 1.6239793725827245,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004922787155942849,
|
|
"loss": 6.401,
|
|
"mean_token_accuracy": 0.13392841145396234,
|
|
"num_tokens": 3435513.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.372339200973511,
|
|
"epoch": 1.628276751181779,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004921921250076611,
|
|
"loss": 6.3095,
|
|
"mean_token_accuracy": 0.14621842429041862,
|
|
"num_tokens": 3444684.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.32946548461914,
|
|
"epoch": 1.6325741297808336,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004921050601602475,
|
|
"loss": 6.3778,
|
|
"mean_token_accuracy": 0.14542506039142608,
|
|
"num_tokens": 3453454.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.35501766204834,
|
|
"epoch": 1.6368715083798882,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004920175212421587,
|
|
"loss": 6.3114,
|
|
"mean_token_accuracy": 0.13968964144587517,
|
|
"num_tokens": 3463228.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.37910737991333,
|
|
"epoch": 1.6411688869789427,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004919295084445445,
|
|
"loss": 6.2069,
|
|
"mean_token_accuracy": 0.14652811288833617,
|
|
"num_tokens": 3472131.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.234958696365356,
|
|
"epoch": 1.6454662655779975,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004918410219595899,
|
|
"loss": 6.2739,
|
|
"mean_token_accuracy": 0.14837560430169106,
|
|
"num_tokens": 3480642.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.476282596588135,
|
|
"epoch": 1.649763644177052,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000491752061980514,
|
|
"loss": 6.2585,
|
|
"mean_token_accuracy": 0.14290489107370377,
|
|
"num_tokens": 3489346.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.235327243804932,
|
|
"epoch": 1.6540610227761066,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004916626287015697,
|
|
"loss": 6.3001,
|
|
"mean_token_accuracy": 0.14698836505413054,
|
|
"num_tokens": 3498473.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.40929479598999,
|
|
"epoch": 1.658358401375161,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004915727223180436,
|
|
"loss": 6.3032,
|
|
"mean_token_accuracy": 0.14206484630703925,
|
|
"num_tokens": 3507415.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.342353248596192,
|
|
"epoch": 1.6626557799742159,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004914823430262554,
|
|
"loss": 6.4182,
|
|
"mean_token_accuracy": 0.13057646304368972,
|
|
"num_tokens": 3516873.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.342631149291992,
|
|
"epoch": 1.6669531585732704,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004913914910235573,
|
|
"loss": 6.2573,
|
|
"mean_token_accuracy": 0.14509516283869744,
|
|
"num_tokens": 3525047.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.319601535797119,
|
|
"epoch": 1.671250537172325,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004913001665083337,
|
|
"loss": 6.2815,
|
|
"mean_token_accuracy": 0.1427992671728134,
|
|
"num_tokens": 3534354.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.29944486618042,
|
|
"epoch": 1.6755479157713795,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004912083696800008,
|
|
"loss": 6.3282,
|
|
"mean_token_accuracy": 0.1369565322995186,
|
|
"num_tokens": 3543830.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.390294408798217,
|
|
"epoch": 1.679845294370434,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004911161007390063,
|
|
"loss": 6.2241,
|
|
"mean_token_accuracy": 0.1440868392586708,
|
|
"num_tokens": 3552314.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.262455129623413,
|
|
"epoch": 1.6841426729694886,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004910233598868287,
|
|
"loss": 6.302,
|
|
"mean_token_accuracy": 0.1431473486125469,
|
|
"num_tokens": 3561656.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.357747316360474,
|
|
"epoch": 1.6884400515685432,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004909301473259769,
|
|
"loss": 6.2899,
|
|
"mean_token_accuracy": 0.14251541048288346,
|
|
"num_tokens": 3571784.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.3709187507629395,
|
|
"epoch": 1.6927374301675977,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004908364632599899,
|
|
"loss": 6.257,
|
|
"mean_token_accuracy": 0.14677265286445618,
|
|
"num_tokens": 3580626.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.291310262680054,
|
|
"epoch": 1.6970348087666522,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004907423078934362,
|
|
"loss": 6.2593,
|
|
"mean_token_accuracy": 0.14560385495424272,
|
|
"num_tokens": 3589916.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.346375846862793,
|
|
"epoch": 1.7013321873657068,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004906476814319134,
|
|
"loss": 6.2762,
|
|
"mean_token_accuracy": 0.14539931267499923,
|
|
"num_tokens": 3599128.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.240916728973389,
|
|
"epoch": 1.7056295659647613,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004905525840820481,
|
|
"loss": 6.2935,
|
|
"mean_token_accuracy": 0.14389279931783677,
|
|
"num_tokens": 3608764.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.434264993667602,
|
|
"epoch": 1.709926944563816,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004904570160514948,
|
|
"loss": 6.3375,
|
|
"mean_token_accuracy": 0.13460767194628714,
|
|
"num_tokens": 3619082.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.353186368942261,
|
|
"epoch": 1.7142243231628707,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004903609775489358,
|
|
"loss": 6.306,
|
|
"mean_token_accuracy": 0.1431412361562252,
|
|
"num_tokens": 3628695.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.308540105819702,
|
|
"epoch": 1.7185217017619252,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004902644687840809,
|
|
"loss": 6.2744,
|
|
"mean_token_accuracy": 0.14257324934005738,
|
|
"num_tokens": 3637599.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.7185217017619252,
|
|
"eval_entropy": 6.102365418597385,
|
|
"eval_loss": 6.377512454986572,
|
|
"eval_mean_token_accuracy": 0.14513525596677182,
|
|
"eval_num_tokens": 3637599.0,
|
|
"eval_runtime": 2.652,
|
|
"eval_samples_per_second": 1338.25,
|
|
"eval_steps_per_second": 167.423,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.3316725730896,
|
|
"epoch": 1.7228190803609797,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004901674899676667,
|
|
"loss": 6.2783,
|
|
"mean_token_accuracy": 0.1429784931242466,
|
|
"num_tokens": 3647406.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.3138439655303955,
|
|
"epoch": 1.7271164589600345,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004900700413114561,
|
|
"loss": 6.1653,
|
|
"mean_token_accuracy": 0.1461011752486229,
|
|
"num_tokens": 3656531.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.25796856880188,
|
|
"epoch": 1.731413837559089,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000489972123028238,
|
|
"loss": 6.2734,
|
|
"mean_token_accuracy": 0.1430951438844204,
|
|
"num_tokens": 3664922.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.352964925765991,
|
|
"epoch": 1.7357112161581436,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004898737353318268,
|
|
"loss": 6.1886,
|
|
"mean_token_accuracy": 0.14680340215563775,
|
|
"num_tokens": 3673283.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.228963136672974,
|
|
"epoch": 1.7400085947571982,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000489774878437062,
|
|
"loss": 6.342,
|
|
"mean_token_accuracy": 0.14178226068615912,
|
|
"num_tokens": 3681760.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.376891946792602,
|
|
"epoch": 1.7443059733562527,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004896755525598074,
|
|
"loss": 6.1431,
|
|
"mean_token_accuracy": 0.14990449845790862,
|
|
"num_tokens": 3689408.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.2323527336120605,
|
|
"epoch": 1.7486033519553073,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004895757579169511,
|
|
"loss": 6.2487,
|
|
"mean_token_accuracy": 0.14669205546379088,
|
|
"num_tokens": 3697904.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.3964447498321535,
|
|
"epoch": 1.7529007305543618,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004894754947264047,
|
|
"loss": 6.2952,
|
|
"mean_token_accuracy": 0.14583562612533568,
|
|
"num_tokens": 3706704.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.348710727691651,
|
|
"epoch": 1.7571981091534163,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000489374763207103,
|
|
"loss": 6.3569,
|
|
"mean_token_accuracy": 0.14076642096042633,
|
|
"num_tokens": 3715690.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.3549293041229244,
|
|
"epoch": 1.761495487752471,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004892735635790033,
|
|
"loss": 6.1596,
|
|
"mean_token_accuracy": 0.15285283625125884,
|
|
"num_tokens": 3724835.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 6.255563449859619,
|
|
"epoch": 1.7657928663515254,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000489171896063085,
|
|
"loss": 6.182,
|
|
"mean_token_accuracy": 0.1519356772303581,
|
|
"num_tokens": 3733977.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.31369194984436,
|
|
"epoch": 1.77009024495058,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004890697608813495,
|
|
"loss": 6.303,
|
|
"mean_token_accuracy": 0.13368857130408288,
|
|
"num_tokens": 3742665.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.462381839752197,
|
|
"epoch": 1.7743876235496348,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004889671582568193,
|
|
"loss": 6.3852,
|
|
"mean_token_accuracy": 0.1396317407488823,
|
|
"num_tokens": 3751647.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.238660097122192,
|
|
"epoch": 1.7786850021486893,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004888640884135374,
|
|
"loss": 6.2769,
|
|
"mean_token_accuracy": 0.14214187860488892,
|
|
"num_tokens": 3760852.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 6.324005889892578,
|
|
"epoch": 1.7829823807477438,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004887605515765671,
|
|
"loss": 6.2103,
|
|
"mean_token_accuracy": 0.14887373968958856,
|
|
"num_tokens": 3768640.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.3109969139099125,
|
|
"epoch": 1.7872797593467986,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004886565479719914,
|
|
"loss": 6.2494,
|
|
"mean_token_accuracy": 0.13794242814183236,
|
|
"num_tokens": 3776504.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.407247066497803,
|
|
"epoch": 1.7915771379458532,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004885520778269128,
|
|
"loss": 6.295,
|
|
"mean_token_accuracy": 0.14428260549902916,
|
|
"num_tokens": 3786353.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.293655490875244,
|
|
"epoch": 1.7958745165449077,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004884471413694523,
|
|
"loss": 6.3331,
|
|
"mean_token_accuracy": 0.13942929208278657,
|
|
"num_tokens": 3795902.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 6.322744464874267,
|
|
"epoch": 1.8001718951439623,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004883417388287491,
|
|
"loss": 6.2366,
|
|
"mean_token_accuracy": 0.13808404356241227,
|
|
"num_tokens": 3805986.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.341541337966919,
|
|
"epoch": 1.8044692737430168,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004882358704349603,
|
|
"loss": 6.3694,
|
|
"mean_token_accuracy": 0.14676245599985122,
|
|
"num_tokens": 3814915.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.315998220443726,
|
|
"epoch": 1.8087666523420713,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004881295364192601,
|
|
"loss": 6.246,
|
|
"mean_token_accuracy": 0.14990996494889258,
|
|
"num_tokens": 3823966.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 6.381868028640747,
|
|
"epoch": 1.813064030941126,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004880227370138394,
|
|
"loss": 6.3049,
|
|
"mean_token_accuracy": 0.13948933705687522,
|
|
"num_tokens": 3832775.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.281226778030396,
|
|
"epoch": 1.8173614095401804,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004879154724519057,
|
|
"loss": 6.2229,
|
|
"mean_token_accuracy": 0.14959346354007722,
|
|
"num_tokens": 3842808.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 6.376323270797729,
|
|
"epoch": 1.821658788139235,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004878077429676816,
|
|
"loss": 6.3574,
|
|
"mean_token_accuracy": 0.13974557369947432,
|
|
"num_tokens": 3853303.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.250073766708374,
|
|
"epoch": 1.8259561667382895,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004876995487964054,
|
|
"loss": 6.2633,
|
|
"mean_token_accuracy": 0.13307938054203988,
|
|
"num_tokens": 3862462.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.35753984451294,
|
|
"epoch": 1.830253545337344,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048759089017432996,
|
|
"loss": 6.3825,
|
|
"mean_token_accuracy": 0.13758554458618164,
|
|
"num_tokens": 3871596.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.3715184211730955,
|
|
"epoch": 1.8345509239363988,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004874817673387222,
|
|
"loss": 6.285,
|
|
"mean_token_accuracy": 0.14220449030399324,
|
|
"num_tokens": 3881276.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.2460695743560795,
|
|
"epoch": 1.8388483025354534,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048737218052786275,
|
|
"loss": 6.3559,
|
|
"mean_token_accuracy": 0.1396686427295208,
|
|
"num_tokens": 3891610.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.420099306106567,
|
|
"epoch": 1.843145681134508,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048726212998104554,
|
|
"loss": 6.2918,
|
|
"mean_token_accuracy": 0.13814298957586288,
|
|
"num_tokens": 3900584.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.248833560943604,
|
|
"epoch": 1.8474430597335625,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004871516159385768,
|
|
"loss": 6.2394,
|
|
"mean_token_accuracy": 0.14645890071988105,
|
|
"num_tokens": 3910208.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.186929512023926,
|
|
"epoch": 1.8517404383326173,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004870406386417752,
|
|
"loss": 6.1619,
|
|
"mean_token_accuracy": 0.15340124666690827,
|
|
"num_tokens": 3918424.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.2605485916137695,
|
|
"epoch": 1.8560378169316718,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004869291983329707,
|
|
"loss": 6.1026,
|
|
"mean_token_accuracy": 0.15957342684268952,
|
|
"num_tokens": 3926206.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 6.292481088638306,
|
|
"epoch": 1.8603351955307263,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004868172952555044,
|
|
"loss": 6.2,
|
|
"mean_token_accuracy": 0.1387283407151699,
|
|
"num_tokens": 3935769.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.234233331680298,
|
|
"epoch": 1.864632574129781,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004867049296537278,
|
|
"loss": 6.1804,
|
|
"mean_token_accuracy": 0.14563867971301078,
|
|
"num_tokens": 3945118.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.340856981277466,
|
|
"epoch": 1.8689299527288354,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004865921017730027,
|
|
"loss": 6.2703,
|
|
"mean_token_accuracy": 0.1473103553056717,
|
|
"num_tokens": 3954012.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 6.260055828094482,
|
|
"epoch": 1.87322733132789,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048647881185969995,
|
|
"loss": 6.2845,
|
|
"mean_token_accuracy": 0.14095541685819626,
|
|
"num_tokens": 3964239.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 6.3267253875732425,
|
|
"epoch": 1.8775247099269445,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004863650601611994,
|
|
"loss": 6.1875,
|
|
"mean_token_accuracy": 0.1507116213440895,
|
|
"num_tokens": 3973694.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.264223241806031,
|
|
"epoch": 1.881822088525999,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00048625084692588937,
|
|
"loss": 6.2352,
|
|
"mean_token_accuracy": 0.15302565097808837,
|
|
"num_tokens": 3982706.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.271808385848999,
|
|
"epoch": 1.8861194671250536,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00048613617240316593,
|
|
"loss": 6.1827,
|
|
"mean_token_accuracy": 0.1553429678082466,
|
|
"num_tokens": 3990934.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.317896461486816,
|
|
"epoch": 1.8904168457241082,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004860210368434323,
|
|
"loss": 6.2366,
|
|
"mean_token_accuracy": 0.14802313670516015,
|
|
"num_tokens": 3999864.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 6.255628681182861,
|
|
"epoch": 1.8947142243231627,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00048590544049809857,
|
|
"loss": 6.2416,
|
|
"mean_token_accuracy": 0.15196478515863418,
|
|
"num_tokens": 4008273.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 6.317771577835083,
|
|
"epoch": 1.8990116029222175,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000485789383619581,
|
|
"loss": 6.2643,
|
|
"mean_token_accuracy": 0.14630038738250734,
|
|
"num_tokens": 4017697.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.305907392501831,
|
|
"epoch": 1.903308981521272,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004856728664613015,
|
|
"loss": 6.2672,
|
|
"mean_token_accuracy": 0.14127509519457818,
|
|
"num_tokens": 4026775.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.2479931831359865,
|
|
"epoch": 1.9076063601203266,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00048555588927768674,
|
|
"loss": 6.2505,
|
|
"mean_token_accuracy": 0.14888640120625496,
|
|
"num_tokens": 4036476.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.362626457214356,
|
|
"epoch": 1.9119037387193811,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004854384523241683,
|
|
"loss": 6.2388,
|
|
"mean_token_accuracy": 0.15212999284267426,
|
|
"num_tokens": 4045221.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 6.2020368576049805,
|
|
"epoch": 1.916201117318436,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00048532055585718143,
|
|
"loss": 6.1573,
|
|
"mean_token_accuracy": 0.15003958493471145,
|
|
"num_tokens": 4053754.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 6.241902828216553,
|
|
"epoch": 1.9204984959174904,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00048520220013416505,
|
|
"loss": 6.1908,
|
|
"mean_token_accuracy": 0.1493961751461029,
|
|
"num_tokens": 4061730.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 6.299831438064575,
|
|
"epoch": 1.924795874516545,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004850833854135607,
|
|
"loss": 6.2578,
|
|
"mean_token_accuracy": 0.14219411388039588,
|
|
"num_tokens": 4070501.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.280361223220825,
|
|
"epoch": 1.9290932531155995,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004849641119548122,
|
|
"loss": 6.3251,
|
|
"mean_token_accuracy": 0.14482801854610444,
|
|
"num_tokens": 4079621.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.433962106704712,
|
|
"epoch": 1.933390631714654,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000484844380018365,
|
|
"loss": 6.3239,
|
|
"mean_token_accuracy": 0.1438091665506363,
|
|
"num_tokens": 4090106.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.3048840999603275,
|
|
"epoch": 1.9376880103137086,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000484724189865666,
|
|
"loss": 6.2464,
|
|
"mean_token_accuracy": 0.14323291927576065,
|
|
"num_tokens": 4099269.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 6.118443775177002,
|
|
"epoch": 1.9419853889127632,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004846035417591624,
|
|
"loss": 6.1918,
|
|
"mean_token_accuracy": 0.14526203274726868,
|
|
"num_tokens": 4108414.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 6.354838895797729,
|
|
"epoch": 1.9462827675118177,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004844824359623014,
|
|
"loss": 6.3136,
|
|
"mean_token_accuracy": 0.14043026119470597,
|
|
"num_tokens": 4117731.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.369909286499023,
|
|
"epoch": 1.9505801461108723,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00048436087273952966,
|
|
"loss": 6.2906,
|
|
"mean_token_accuracy": 0.13964733406901358,
|
|
"num_tokens": 4127194.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 6.302741193771363,
|
|
"epoch": 1.9548775247099268,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00048423885235629265,
|
|
"loss": 6.2467,
|
|
"mean_token_accuracy": 0.1489931106567383,
|
|
"num_tokens": 4135594.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 6.264612770080566,
|
|
"epoch": 1.9591749033089814,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004841163750790342,
|
|
"loss": 6.2771,
|
|
"mean_token_accuracy": 0.14228039383888244,
|
|
"num_tokens": 4145027.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.2522119045257565,
|
|
"epoch": 1.9634722819080361,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048399344117519555,
|
|
"loss": 6.1568,
|
|
"mean_token_accuracy": 0.1510104939341545,
|
|
"num_tokens": 4153754.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 6.2635422706604,
|
|
"epoch": 1.9677696605070907,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048387005091321544,
|
|
"loss": 6.2129,
|
|
"mean_token_accuracy": 0.14985550343990325,
|
|
"num_tokens": 4162765.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.344277286529541,
|
|
"epoch": 1.9720670391061452,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00048374620456252877,
|
|
"loss": 6.2344,
|
|
"mean_token_accuracy": 0.14524247571825982,
|
|
"num_tokens": 4171589.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 6.2353309154510494,
|
|
"epoch": 1.9763644177052,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00048362190239356644,
|
|
"loss": 6.2283,
|
|
"mean_token_accuracy": 0.14546259343624116,
|
|
"num_tokens": 4181817.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 6.261869239807129,
|
|
"epoch": 1.9806617963042545,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048349714467775474,
|
|
"loss": 6.1782,
|
|
"mean_token_accuracy": 0.15101497769355773,
|
|
"num_tokens": 4191350.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 6.270454025268554,
|
|
"epoch": 1.984959174903309,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00048337193168751464,
|
|
"loss": 6.2502,
|
|
"mean_token_accuracy": 0.14235945269465447,
|
|
"num_tokens": 4199888.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.295589876174927,
|
|
"epoch": 1.9892565535023636,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004832462636962613,
|
|
"loss": 6.2102,
|
|
"mean_token_accuracy": 0.14260616898536682,
|
|
"num_tokens": 4209509.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 6.26480188369751,
|
|
"epoch": 1.9935539321014182,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004831201409784034,
|
|
"loss": 6.1744,
|
|
"mean_token_accuracy": 0.15239215418696403,
|
|
"num_tokens": 4218496.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 6.292611980438233,
|
|
"epoch": 1.9978513107004727,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004829935638093424,
|
|
"loss": 6.1855,
|
|
"mean_token_accuracy": 0.14760839939117432,
|
|
"num_tokens": 4227504.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 6.27884938981798,
|
|
"epoch": 2.0017189514396216,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004828665324654724,
|
|
"loss": 6.1031,
|
|
"mean_token_accuracy": 0.14647970845301947,
|
|
"num_tokens": 4235338.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 6.286802339553833,
|
|
"epoch": 2.006016330038676,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004827390472241791,
|
|
"loss": 5.8689,
|
|
"mean_token_accuracy": 0.15517483800649642,
|
|
"num_tokens": 4244905.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 6.213595294952393,
|
|
"epoch": 2.010313708637731,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004826111083638392,
|
|
"loss": 5.9719,
|
|
"mean_token_accuracy": 0.1514093331992626,
|
|
"num_tokens": 4254533.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 6.248905611038208,
|
|
"epoch": 2.0146110872367857,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00048248271616382,
|
|
"loss": 5.9193,
|
|
"mean_token_accuracy": 0.15519882887601852,
|
|
"num_tokens": 4264023.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 6.172601699829102,
|
|
"epoch": 2.0189084658358403,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00048235387090447894,
|
|
"loss": 5.9718,
|
|
"mean_token_accuracy": 0.1513834610581398,
|
|
"num_tokens": 4273298.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 6.2314698696136475,
|
|
"epoch": 2.023205844434895,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00048222457286716235,
|
|
"loss": 5.9363,
|
|
"mean_token_accuracy": 0.1507216587662697,
|
|
"num_tokens": 4283244.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 6.17480731010437,
|
|
"epoch": 2.0275032230339494,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00048209482233420564,
|
|
"loss": 5.8574,
|
|
"mean_token_accuracy": 0.16718780845403672,
|
|
"num_tokens": 4291677.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 6.225293064117432,
|
|
"epoch": 2.031800601633004,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000481964619588932,
|
|
"loss": 5.928,
|
|
"mean_token_accuracy": 0.15908985882997512,
|
|
"num_tokens": 4300822.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 6.164669036865234,
|
|
"epoch": 2.0360979802320585,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004818339649156523,
|
|
"loss": 5.9267,
|
|
"mean_token_accuracy": 0.1617436647415161,
|
|
"num_tokens": 4310149.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 6.098308897018432,
|
|
"epoch": 2.040395358831113,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00048170285859966395,
|
|
"loss": 5.854,
|
|
"mean_token_accuracy": 0.16342242807149887,
|
|
"num_tokens": 4319109.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 6.168161201477051,
|
|
"epoch": 2.0446927374301676,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048157130092725087,
|
|
"loss": 5.8165,
|
|
"mean_token_accuracy": 0.16630419790744783,
|
|
"num_tokens": 4327921.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 6.206757879257202,
|
|
"epoch": 2.048990116029222,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004814392921856824,
|
|
"loss": 5.9859,
|
|
"mean_token_accuracy": 0.1574866846203804,
|
|
"num_tokens": 4338026.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 6.138569498062134,
|
|
"epoch": 2.0532874946282766,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004813068326632128,
|
|
"loss": 5.83,
|
|
"mean_token_accuracy": 0.1653517082333565,
|
|
"num_tokens": 4347794.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 6.188538217544556,
|
|
"epoch": 2.057584873227331,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004811739226490809,
|
|
"loss": 6.0081,
|
|
"mean_token_accuracy": 0.15463653430342675,
|
|
"num_tokens": 4357249.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 6.174555063247681,
|
|
"epoch": 2.0618822518263857,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00048104056243350896,
|
|
"loss": 5.964,
|
|
"mean_token_accuracy": 0.1523067057132721,
|
|
"num_tokens": 4366053.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.169906854629517,
|
|
"epoch": 2.0661796304254403,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004809067523077023,
|
|
"loss": 5.9581,
|
|
"mean_token_accuracy": 0.1572214901447296,
|
|
"num_tokens": 4375543.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.105737686157227,
|
|
"epoch": 2.0704770090244953,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048077249256384884,
|
|
"loss": 5.8465,
|
|
"mean_token_accuracy": 0.16420940905809403,
|
|
"num_tokens": 4384332.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 6.132162141799927,
|
|
"epoch": 2.07477438762355,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004806377834951182,
|
|
"loss": 5.9415,
|
|
"mean_token_accuracy": 0.15044854432344437,
|
|
"num_tokens": 4393670.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 6.286304092407226,
|
|
"epoch": 2.0790717662226044,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00048050262539566104,
|
|
"loss": 5.9517,
|
|
"mean_token_accuracy": 0.1577576845884323,
|
|
"num_tokens": 4402763.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 6.14480562210083,
|
|
"epoch": 2.083369144821659,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004803670185606087,
|
|
"loss": 5.8515,
|
|
"mean_token_accuracy": 0.16330406665802003,
|
|
"num_tokens": 4411863.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 6.126660680770874,
|
|
"epoch": 2.0876665234207135,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004802309632860724,
|
|
"loss": 5.9308,
|
|
"mean_token_accuracy": 0.158506877720356,
|
|
"num_tokens": 4421110.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 6.16443362236023,
|
|
"epoch": 2.091963902019768,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00048009445986914236,
|
|
"loss": 5.9423,
|
|
"mean_token_accuracy": 0.15614678859710693,
|
|
"num_tokens": 4430249.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.127264070510864,
|
|
"epoch": 2.0962612806188226,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047995750860788756,
|
|
"loss": 5.9005,
|
|
"mean_token_accuracy": 0.15375098139047622,
|
|
"num_tokens": 4439686.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 6.098284101486206,
|
|
"epoch": 2.100558659217877,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004798201098013547,
|
|
"loss": 5.8973,
|
|
"mean_token_accuracy": 0.16047993451356887,
|
|
"num_tokens": 4448645.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 6.151400423049926,
|
|
"epoch": 2.1048560378169316,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047968226374956797,
|
|
"loss": 5.8721,
|
|
"mean_token_accuracy": 0.15785705149173737,
|
|
"num_tokens": 4456870.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 6.046982669830323,
|
|
"epoch": 2.109153416415986,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00047954397075352794,
|
|
"loss": 5.9192,
|
|
"mean_token_accuracy": 0.1597234606742859,
|
|
"num_tokens": 4466287.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 6.125206804275512,
|
|
"epoch": 2.1134507950150407,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00047940523111521136,
|
|
"loss": 5.7955,
|
|
"mean_token_accuracy": 0.16594544798135757,
|
|
"num_tokens": 4474461.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 6.123905992507934,
|
|
"epoch": 2.1177481736140953,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004792660451375701,
|
|
"loss": 5.8938,
|
|
"mean_token_accuracy": 0.1566778302192688,
|
|
"num_tokens": 4483002.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 6.071023559570312,
|
|
"epoch": 2.12204555221315,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00047912641312453064,
|
|
"loss": 5.8859,
|
|
"mean_token_accuracy": 0.159532330930233,
|
|
"num_tokens": 4492061.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 6.157165908813477,
|
|
"epoch": 2.1263429308122044,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047898633538099363,
|
|
"loss": 5.9413,
|
|
"mean_token_accuracy": 0.15300000160932542,
|
|
"num_tokens": 4501829.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 6.122453737258911,
|
|
"epoch": 2.130640309411259,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004788458122128327,
|
|
"loss": 5.9604,
|
|
"mean_token_accuracy": 0.15562360361218452,
|
|
"num_tokens": 4511539.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 6.076842832565307,
|
|
"epoch": 2.134937688010314,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00047870484392689434,
|
|
"loss": 5.8336,
|
|
"mean_token_accuracy": 0.15930114388465882,
|
|
"num_tokens": 4520425.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 6.140858602523804,
|
|
"epoch": 2.1392350666093685,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000478563430830997,
|
|
"loss": 5.9241,
|
|
"mean_token_accuracy": 0.15476028025150299,
|
|
"num_tokens": 4529474.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 6.156933403015136,
|
|
"epoch": 2.143532445208423,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00047842157323393035,
|
|
"loss": 5.864,
|
|
"mean_token_accuracy": 0.15560978651046753,
|
|
"num_tokens": 4538082.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 6.0952434062957765,
|
|
"epoch": 2.1478298238074776,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004782792714454547,
|
|
"loss": 6.0423,
|
|
"mean_token_accuracy": 0.15425097942352295,
|
|
"num_tokens": 4547340.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.1478298238074776,
|
|
"eval_entropy": 6.033101958197516,
|
|
"eval_loss": 6.296128749847412,
|
|
"eval_mean_token_accuracy": 0.14961527512819917,
|
|
"eval_num_tokens": 4547340.0,
|
|
"eval_runtime": 2.6476,
|
|
"eval_samples_per_second": 1340.472,
|
|
"eval_steps_per_second": 167.701,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 6.127610158920288,
|
|
"epoch": 2.152127202406532,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004781365257763002,
|
|
"loss": 5.9053,
|
|
"mean_token_accuracy": 0.16055295020341873,
|
|
"num_tokens": 4556415.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 6.023889684677124,
|
|
"epoch": 2.1564245810055866,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00047799333653816633,
|
|
"loss": 5.7855,
|
|
"mean_token_accuracy": 0.16898827254772186,
|
|
"num_tokens": 4565156.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 6.11927752494812,
|
|
"epoch": 2.160721959604641,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047784970404372124,
|
|
"loss": 5.8669,
|
|
"mean_token_accuracy": 0.1618487223982811,
|
|
"num_tokens": 4574678.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 6.0734045028686525,
|
|
"epoch": 2.1650193382036957,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00047770562860660083,
|
|
"loss": 5.9102,
|
|
"mean_token_accuracy": 0.15316808074712754,
|
|
"num_tokens": 4583253.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 6.155279493331909,
|
|
"epoch": 2.1693167168027503,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004775611105414083,
|
|
"loss": 5.9532,
|
|
"mean_token_accuracy": 0.15269838720560075,
|
|
"num_tokens": 4594042.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 6.081975936889648,
|
|
"epoch": 2.173614095401805,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004774161501637133,
|
|
"loss": 5.9304,
|
|
"mean_token_accuracy": 0.1585197150707245,
|
|
"num_tokens": 4603128.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 6.122641563415527,
|
|
"epoch": 2.1779114740008594,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004772707477900514,
|
|
"loss": 5.9152,
|
|
"mean_token_accuracy": 0.15637425184249878,
|
|
"num_tokens": 4611537.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 6.200121688842773,
|
|
"epoch": 2.182208852599914,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004771249037379232,
|
|
"loss": 6.0093,
|
|
"mean_token_accuracy": 0.14852525740861894,
|
|
"num_tokens": 4622481.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 6.118060207366943,
|
|
"epoch": 2.1865062311989685,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004769786183257939,
|
|
"loss": 5.9157,
|
|
"mean_token_accuracy": 0.1607968181371689,
|
|
"num_tokens": 4631259.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 6.09875226020813,
|
|
"epoch": 2.190803609798023,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004768318918730924,
|
|
"loss": 5.8575,
|
|
"mean_token_accuracy": 0.1634613811969757,
|
|
"num_tokens": 4640266.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 6.0824010372161865,
|
|
"epoch": 2.195100988397078,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00047668472470021044,
|
|
"loss": 5.9327,
|
|
"mean_token_accuracy": 0.15035101026296616,
|
|
"num_tokens": 4649520.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 6.145359516143799,
|
|
"epoch": 2.1993983669961326,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004765371171285025,
|
|
"loss": 5.8617,
|
|
"mean_token_accuracy": 0.16391258537769318,
|
|
"num_tokens": 4658501.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 6.012497663497925,
|
|
"epoch": 2.203695745595187,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00047638906948028445,
|
|
"loss": 5.923,
|
|
"mean_token_accuracy": 0.15301997363567352,
|
|
"num_tokens": 4667567.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 6.154702377319336,
|
|
"epoch": 2.2079931241942417,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00047624058207883317,
|
|
"loss": 5.9116,
|
|
"mean_token_accuracy": 0.1582942560315132,
|
|
"num_tokens": 4676618.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 6.210588645935059,
|
|
"epoch": 2.212290502793296,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047609165524838576,
|
|
"loss": 5.9846,
|
|
"mean_token_accuracy": 0.1570892885327339,
|
|
"num_tokens": 4685967.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 6.086702013015747,
|
|
"epoch": 2.2165878813923507,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0004759422893141389,
|
|
"loss": 5.8846,
|
|
"mean_token_accuracy": 0.15947829633951188,
|
|
"num_tokens": 4694568.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 6.054249095916748,
|
|
"epoch": 2.2208852599914053,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004757924846022482,
|
|
"loss": 5.937,
|
|
"mean_token_accuracy": 0.15623788684606552,
|
|
"num_tokens": 4703648.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 6.05418906211853,
|
|
"epoch": 2.22518263859046,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00047564224143982714,
|
|
"loss": 5.8011,
|
|
"mean_token_accuracy": 0.16867249310016633,
|
|
"num_tokens": 4712444.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 6.138991594314575,
|
|
"epoch": 2.2294800171895144,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00047549156015494676,
|
|
"loss": 5.9601,
|
|
"mean_token_accuracy": 0.15550855100154876,
|
|
"num_tokens": 4722034.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 6.025436592102051,
|
|
"epoch": 2.233777395788569,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00047534044107663484,
|
|
"loss": 5.9558,
|
|
"mean_token_accuracy": 0.15787288099527358,
|
|
"num_tokens": 4731344.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 6.163779401779175,
|
|
"epoch": 2.2380747743876235,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00047518888453487496,
|
|
"loss": 5.8752,
|
|
"mean_token_accuracy": 0.16263986229896546,
|
|
"num_tokens": 4739302.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 6.050823783874511,
|
|
"epoch": 2.242372152986678,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004750368908606061,
|
|
"loss": 5.9957,
|
|
"mean_token_accuracy": 0.15456610471010207,
|
|
"num_tokens": 4748848.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 6.149853229522705,
|
|
"epoch": 2.2466695315857326,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047488446038572164,
|
|
"loss": 6.049,
|
|
"mean_token_accuracy": 0.14931472390890121,
|
|
"num_tokens": 4758194.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 6.252527523040771,
|
|
"epoch": 2.250966910184787,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004747315934430688,
|
|
"loss": 5.9514,
|
|
"mean_token_accuracy": 0.1540973365306854,
|
|
"num_tokens": 4768081.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.992271137237549,
|
|
"epoch": 2.2552642887838417,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000474578290366448,
|
|
"loss": 5.8836,
|
|
"mean_token_accuracy": 0.15656516551971436,
|
|
"num_tokens": 4776471.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 6.104082918167114,
|
|
"epoch": 2.259561667382896,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004744245514906117,
|
|
"loss": 5.8769,
|
|
"mean_token_accuracy": 0.1567678838968277,
|
|
"num_tokens": 4784403.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 6.032165384292602,
|
|
"epoch": 2.263859045981951,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00047427037715126426,
|
|
"loss": 5.8498,
|
|
"mean_token_accuracy": 0.15365731716156006,
|
|
"num_tokens": 4792779.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 6.039053583145142,
|
|
"epoch": 2.2681564245810057,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004741157676850608,
|
|
"loss": 5.8591,
|
|
"mean_token_accuracy": 0.16261037290096284,
|
|
"num_tokens": 4801426.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 6.1057905673980715,
|
|
"epoch": 2.2724538031800603,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00047396072342960663,
|
|
"loss": 5.9017,
|
|
"mean_token_accuracy": 0.15529569759964942,
|
|
"num_tokens": 4810329.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 6.082610988616944,
|
|
"epoch": 2.276751181779115,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047380524472345645,
|
|
"loss": 5.9538,
|
|
"mean_token_accuracy": 0.15293473303318023,
|
|
"num_tokens": 4819544.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 6.126261377334595,
|
|
"epoch": 2.2810485603781694,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004736493319061134,
|
|
"loss": 5.9565,
|
|
"mean_token_accuracy": 0.15545238852500914,
|
|
"num_tokens": 4828113.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 6.035648345947266,
|
|
"epoch": 2.285345938977224,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004734929853180291,
|
|
"loss": 5.9397,
|
|
"mean_token_accuracy": 0.15296606793999673,
|
|
"num_tokens": 4836989.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 6.138926267623901,
|
|
"epoch": 2.2896433175762785,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047333620530060175,
|
|
"loss": 5.9777,
|
|
"mean_token_accuracy": 0.154761403799057,
|
|
"num_tokens": 4847103.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 6.138887548446656,
|
|
"epoch": 2.293940696175333,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004731789921961764,
|
|
"loss": 5.9843,
|
|
"mean_token_accuracy": 0.1548290252685547,
|
|
"num_tokens": 4856238.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 6.0893370628356935,
|
|
"epoch": 2.2982380747743876,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004730213463480434,
|
|
"loss": 5.8884,
|
|
"mean_token_accuracy": 0.1642418771982193,
|
|
"num_tokens": 4864608.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 6.106429195404052,
|
|
"epoch": 2.302535453373442,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00047286326810043857,
|
|
"loss": 5.8418,
|
|
"mean_token_accuracy": 0.15996210426092147,
|
|
"num_tokens": 4873889.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 6.01440019607544,
|
|
"epoch": 2.3068328319724967,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00047270475779854137,
|
|
"loss": 5.9107,
|
|
"mean_token_accuracy": 0.16317110285162925,
|
|
"num_tokens": 4882902.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 6.197646999359131,
|
|
"epoch": 2.311130210571551,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00047254581578847507,
|
|
"loss": 5.9147,
|
|
"mean_token_accuracy": 0.15373896956443786,
|
|
"num_tokens": 4892390.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 6.042403793334961,
|
|
"epoch": 2.3154275891706058,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004723864424173055,
|
|
"loss": 6.0281,
|
|
"mean_token_accuracy": 0.15310725271701814,
|
|
"num_tokens": 4901625.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 6.0336872100830075,
|
|
"epoch": 2.3197249677696608,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004722266380330403,
|
|
"loss": 5.8228,
|
|
"mean_token_accuracy": 0.1654062405228615,
|
|
"num_tokens": 4910804.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 6.099563455581665,
|
|
"epoch": 2.3240223463687153,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00047206640298462857,
|
|
"loss": 5.8967,
|
|
"mean_token_accuracy": 0.16032486781477928,
|
|
"num_tokens": 4920441.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 6.088391447067261,
|
|
"epoch": 2.32831972496777,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.00047190573762195945,
|
|
"loss": 5.9458,
|
|
"mean_token_accuracy": 0.15295807868242264,
|
|
"num_tokens": 4930204.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 6.05660605430603,
|
|
"epoch": 2.3326171035668244,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047174464229586186,
|
|
"loss": 6.066,
|
|
"mean_token_accuracy": 0.14822829514741898,
|
|
"num_tokens": 4941191.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 6.219072914123535,
|
|
"epoch": 2.336914482165879,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0004715831173581036,
|
|
"loss": 6.0287,
|
|
"mean_token_accuracy": 0.15552351772785186,
|
|
"num_tokens": 4951825.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 6.073605632781982,
|
|
"epoch": 2.3412118607649335,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047142116316139073,
|
|
"loss": 5.9344,
|
|
"mean_token_accuracy": 0.1615937739610672,
|
|
"num_tokens": 4960632.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 6.153278970718384,
|
|
"epoch": 2.345509239363988,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004712587800593663,
|
|
"loss": 5.9718,
|
|
"mean_token_accuracy": 0.15712961852550505,
|
|
"num_tokens": 4969455.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 6.085105562210083,
|
|
"epoch": 2.3498066179630426,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004710959684066102,
|
|
"loss": 5.9009,
|
|
"mean_token_accuracy": 0.16213798969984056,
|
|
"num_tokens": 4978997.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 6.1117143630981445,
|
|
"epoch": 2.354103996562097,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00047093272855863803,
|
|
"loss": 5.9419,
|
|
"mean_token_accuracy": 0.157034994661808,
|
|
"num_tokens": 4988305.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 6.004328060150146,
|
|
"epoch": 2.3584013751611517,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004707690608719003,
|
|
"loss": 5.8896,
|
|
"mean_token_accuracy": 0.165739107131958,
|
|
"num_tokens": 4997022.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 6.077336597442627,
|
|
"epoch": 2.362698753760206,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004706049657037818,
|
|
"loss": 5.9453,
|
|
"mean_token_accuracy": 0.15341522246599198,
|
|
"num_tokens": 5005664.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 6.062661123275757,
|
|
"epoch": 2.3669961323592608,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004704404434126009,
|
|
"loss": 5.9149,
|
|
"mean_token_accuracy": 0.15545564740896226,
|
|
"num_tokens": 5014769.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 6.118867540359497,
|
|
"epoch": 2.3712935109583153,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00047027549435760843,
|
|
"loss": 5.9795,
|
|
"mean_token_accuracy": 0.15703830868005753,
|
|
"num_tokens": 5024060.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 6.256097888946533,
|
|
"epoch": 2.37559088955737,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004701101188989872,
|
|
"loss": 6.0227,
|
|
"mean_token_accuracy": 0.15264038890600204,
|
|
"num_tokens": 5033046.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 6.012174510955811,
|
|
"epoch": 2.3798882681564244,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00046994431739785114,
|
|
"loss": 5.8591,
|
|
"mean_token_accuracy": 0.165789370238781,
|
|
"num_tokens": 5040894.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 6.080564498901367,
|
|
"epoch": 2.384185646755479,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00046977809021624454,
|
|
"loss": 6.0365,
|
|
"mean_token_accuracy": 0.15661823451519014,
|
|
"num_tokens": 5050961.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 6.206455230712891,
|
|
"epoch": 2.3884830253545335,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004696114377171409,
|
|
"loss": 5.9721,
|
|
"mean_token_accuracy": 0.1517358347773552,
|
|
"num_tokens": 5060226.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 6.005672883987427,
|
|
"epoch": 2.3927804039535885,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004694443602644429,
|
|
"loss": 5.9229,
|
|
"mean_token_accuracy": 0.15407629013061525,
|
|
"num_tokens": 5069225.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 6.100782060623169,
|
|
"epoch": 2.397077782552643,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004692768582229808,
|
|
"loss": 5.8976,
|
|
"mean_token_accuracy": 0.1572364941239357,
|
|
"num_tokens": 5078386.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 6.084141874313355,
|
|
"epoch": 2.4013751611516976,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046910893195851213,
|
|
"loss": 5.8251,
|
|
"mean_token_accuracy": 0.15767684280872346,
|
|
"num_tokens": 5087161.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 6.0314923286437985,
|
|
"epoch": 2.405672539750752,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00046894058183772074,
|
|
"loss": 5.9989,
|
|
"mean_token_accuracy": 0.1523453876376152,
|
|
"num_tokens": 5096613.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 6.068730735778809,
|
|
"epoch": 2.4099699183498067,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000468771808228216,
|
|
"loss": 5.957,
|
|
"mean_token_accuracy": 0.15711593627929688,
|
|
"num_tokens": 5106534.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 6.090402460098266,
|
|
"epoch": 2.414267296948861,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00046860261149853197,
|
|
"loss": 5.9858,
|
|
"mean_token_accuracy": 0.15301669836044313,
|
|
"num_tokens": 5115975.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 6.016880321502685,
|
|
"epoch": 2.4185646755479158,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004684329920181268,
|
|
"loss": 5.8847,
|
|
"mean_token_accuracy": 0.1612982854247093,
|
|
"num_tokens": 5124635.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 6.091065454483032,
|
|
"epoch": 2.4228620541469703,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00046826295015738154,
|
|
"loss": 5.8621,
|
|
"mean_token_accuracy": 0.16809643805027008,
|
|
"num_tokens": 5133226.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 6.060954332351685,
|
|
"epoch": 2.427159432746025,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004680924862875996,
|
|
"loss": 5.9355,
|
|
"mean_token_accuracy": 0.1615917131304741,
|
|
"num_tokens": 5142257.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 6.082175540924072,
|
|
"epoch": 2.4314568113450794,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046792160078100605,
|
|
"loss": 5.9374,
|
|
"mean_token_accuracy": 0.16282177120447158,
|
|
"num_tokens": 5150752.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 6.057878684997559,
|
|
"epoch": 2.435754189944134,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00046775029401074653,
|
|
"loss": 5.8223,
|
|
"mean_token_accuracy": 0.16410571783781053,
|
|
"num_tokens": 5160237.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 6.0860391616821286,
|
|
"epoch": 2.4400515685431885,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00046757856635088645,
|
|
"loss": 5.9287,
|
|
"mean_token_accuracy": 0.16295535117387772,
|
|
"num_tokens": 5169752.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 6.080923175811767,
|
|
"epoch": 2.444348947142243,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004674064181764105,
|
|
"loss": 5.9577,
|
|
"mean_token_accuracy": 0.1593508243560791,
|
|
"num_tokens": 5178892.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 6.101225185394287,
|
|
"epoch": 2.448646325741298,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046723384986322147,
|
|
"loss": 5.9398,
|
|
"mean_token_accuracy": 0.15817923545837403,
|
|
"num_tokens": 5188468.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 6.022679233551026,
|
|
"epoch": 2.4529437043403526,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004670608617881395,
|
|
"loss": 5.8537,
|
|
"mean_token_accuracy": 0.16385465562343599,
|
|
"num_tokens": 5197565.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 6.028839731216431,
|
|
"epoch": 2.457241082939407,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004668874543289014,
|
|
"loss": 5.8512,
|
|
"mean_token_accuracy": 0.16994685530662537,
|
|
"num_tokens": 5205791.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 6.052807712554932,
|
|
"epoch": 2.4615384615384617,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00046671362786415986,
|
|
"loss": 5.8548,
|
|
"mean_token_accuracy": 0.17034601420164108,
|
|
"num_tokens": 5214773.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.9589917182922365,
|
|
"epoch": 2.465835840137516,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00046653938277348237,
|
|
"loss": 5.8802,
|
|
"mean_token_accuracy": 0.1698584884405136,
|
|
"num_tokens": 5223734.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 6.205320119857788,
|
|
"epoch": 2.4701332187365708,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004663647194373505,
|
|
"loss": 5.9559,
|
|
"mean_token_accuracy": 0.15378482192754744,
|
|
"num_tokens": 5231742.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.968310594558716,
|
|
"epoch": 2.4744305973356253,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00046618963823715913,
|
|
"loss": 5.9368,
|
|
"mean_token_accuracy": 0.1594274491071701,
|
|
"num_tokens": 5241673.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 6.081673860549927,
|
|
"epoch": 2.47872797593468,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00046601413955521575,
|
|
"loss": 5.8978,
|
|
"mean_token_accuracy": 0.15689592957496643,
|
|
"num_tokens": 5250082.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 6.075262832641601,
|
|
"epoch": 2.4830253545337344,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004658382237747393,
|
|
"loss": 5.9756,
|
|
"mean_token_accuracy": 0.15185276046395302,
|
|
"num_tokens": 5259680.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 6.099707412719726,
|
|
"epoch": 2.487322733132789,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00046566189127985946,
|
|
"loss": 5.9313,
|
|
"mean_token_accuracy": 0.15997371077537537,
|
|
"num_tokens": 5269561.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 6.068522167205811,
|
|
"epoch": 2.4916201117318435,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000465485142455616,
|
|
"loss": 5.9083,
|
|
"mean_token_accuracy": 0.16538114696741105,
|
|
"num_tokens": 5278659.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.990010023117065,
|
|
"epoch": 2.495917490330898,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00046530797768795765,
|
|
"loss": 5.8721,
|
|
"mean_token_accuracy": 0.16713655591011048,
|
|
"num_tokens": 5287619.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 6.0739173412323,
|
|
"epoch": 2.5002148689299526,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00046513039736374153,
|
|
"loss": 6.0041,
|
|
"mean_token_accuracy": 0.15224301517009736,
|
|
"num_tokens": 5297334.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 6.124326419830322,
|
|
"epoch": 2.504512247529007,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004649524018707319,
|
|
"loss": 5.8877,
|
|
"mean_token_accuracy": 0.16338504552841188,
|
|
"num_tokens": 5306208.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 6.036029243469239,
|
|
"epoch": 2.5088096261280617,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00046477399159759996,
|
|
"loss": 5.8352,
|
|
"mean_token_accuracy": 0.16601199358701707,
|
|
"num_tokens": 5314754.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.907727909088135,
|
|
"epoch": 2.5131070047271162,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00046459516693392246,
|
|
"loss": 5.8583,
|
|
"mean_token_accuracy": 0.16147059053182602,
|
|
"num_tokens": 5324000.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 6.112986707687378,
|
|
"epoch": 2.517404383326171,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004644159282701808,
|
|
"loss": 5.9248,
|
|
"mean_token_accuracy": 0.1625194489955902,
|
|
"num_tokens": 5332478.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 6.05595965385437,
|
|
"epoch": 2.5217017619252258,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046423627599776076,
|
|
"loss": 5.9717,
|
|
"mean_token_accuracy": 0.1506637826561928,
|
|
"num_tokens": 5341635.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 6.100180435180664,
|
|
"epoch": 2.5259991405242803,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000464056210508951,
|
|
"loss": 5.9933,
|
|
"mean_token_accuracy": 0.15302485674619676,
|
|
"num_tokens": 5350144.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 6.0765571117401125,
|
|
"epoch": 2.530296519123335,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004638757321969426,
|
|
"loss": 5.8928,
|
|
"mean_token_accuracy": 0.15683405697345734,
|
|
"num_tokens": 5358788.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 6.018848657608032,
|
|
"epoch": 2.5345938977223894,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00046369484145582815,
|
|
"loss": 6.0033,
|
|
"mean_token_accuracy": 0.15430229604244233,
|
|
"num_tokens": 5368057.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 6.042166662216187,
|
|
"epoch": 2.538891276321444,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00046351353868060054,
|
|
"loss": 5.8321,
|
|
"mean_token_accuracy": 0.16401326805353164,
|
|
"num_tokens": 5376739.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 6.060759830474853,
|
|
"epoch": 2.5431886549204985,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00046333182426715273,
|
|
"loss": 5.9493,
|
|
"mean_token_accuracy": 0.151986326277256,
|
|
"num_tokens": 5385967.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 6.06056399345398,
|
|
"epoch": 2.547486033519553,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00046314969861227626,
|
|
"loss": 5.956,
|
|
"mean_token_accuracy": 0.15500457882881163,
|
|
"num_tokens": 5395192.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 6.068591594696045,
|
|
"epoch": 2.5517834121186076,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004629671621136608,
|
|
"loss": 5.9361,
|
|
"mean_token_accuracy": 0.15863018184900285,
|
|
"num_tokens": 5404694.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 6.044081449508667,
|
|
"epoch": 2.556080790717662,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0004627842151698931,
|
|
"loss": 5.9529,
|
|
"mean_token_accuracy": 0.15535407066345214,
|
|
"num_tokens": 5413102.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 6.012300252914429,
|
|
"epoch": 2.5603781693167167,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00046260085818045625,
|
|
"loss": 5.9567,
|
|
"mean_token_accuracy": 0.15557909458875657,
|
|
"num_tokens": 5423339.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 6.141985654830933,
|
|
"epoch": 2.5646755479157712,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004624170915457284,
|
|
"loss": 5.9245,
|
|
"mean_token_accuracy": 0.16289076060056687,
|
|
"num_tokens": 5432377.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 6.0600813865661625,
|
|
"epoch": 2.5689729265148262,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00046223291566698264,
|
|
"loss": 5.8526,
|
|
"mean_token_accuracy": 0.16326662972569467,
|
|
"num_tokens": 5441038.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.992218255996704,
|
|
"epoch": 2.5732703051138808,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004620483309463855,
|
|
"loss": 5.8912,
|
|
"mean_token_accuracy": 0.16024321168661118,
|
|
"num_tokens": 5449557.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 6.137758588790893,
|
|
"epoch": 2.5775676837129353,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004618633377869961,
|
|
"loss": 6.0125,
|
|
"mean_token_accuracy": 0.15263121500611304,
|
|
"num_tokens": 5458931.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.5775676837129353,
|
|
"eval_entropy": 6.06476079558467,
|
|
"eval_loss": 6.226695537567139,
|
|
"eval_mean_token_accuracy": 0.15326246213134345,
|
|
"eval_num_tokens": 5458931.0,
|
|
"eval_runtime": 2.8763,
|
|
"eval_samples_per_second": 1233.891,
|
|
"eval_steps_per_second": 154.367,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 6.083834648132324,
|
|
"epoch": 2.58186506231199,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004616779365927656,
|
|
"loss": 5.8478,
|
|
"mean_token_accuracy": 0.1706843316555023,
|
|
"num_tokens": 5468539.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.819710111618042,
|
|
"epoch": 2.5861624409110444,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0004614921277685361,
|
|
"loss": 5.7696,
|
|
"mean_token_accuracy": 0.17440510392189026,
|
|
"num_tokens": 5475710.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 6.058649778366089,
|
|
"epoch": 2.590459819510099,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00046130591172003976,
|
|
"loss": 5.9175,
|
|
"mean_token_accuracy": 0.1590907484292984,
|
|
"num_tokens": 5484597.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 6.096326303482056,
|
|
"epoch": 2.5947571981091535,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004611192888538981,
|
|
"loss": 5.9887,
|
|
"mean_token_accuracy": 0.15526969879865646,
|
|
"num_tokens": 5493213.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 6.048277997970581,
|
|
"epoch": 2.599054576708208,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00046093225957762084,
|
|
"loss": 5.9974,
|
|
"mean_token_accuracy": 0.15169204026460648,
|
|
"num_tokens": 5502556.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 6.0936554908752445,
|
|
"epoch": 2.6033519553072626,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004607448242996051,
|
|
"loss": 5.9175,
|
|
"mean_token_accuracy": 0.15795919448137283,
|
|
"num_tokens": 5511779.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 6.135501289367676,
|
|
"epoch": 2.607649333906317,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004605569834291347,
|
|
"loss": 5.8942,
|
|
"mean_token_accuracy": 0.16351283490657806,
|
|
"num_tokens": 5520836.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.940966415405273,
|
|
"epoch": 2.6119467125053717,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00046036873737637904,
|
|
"loss": 5.911,
|
|
"mean_token_accuracy": 0.1636594921350479,
|
|
"num_tokens": 5529285.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 6.08675708770752,
|
|
"epoch": 2.6162440911044262,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004601800865523921,
|
|
"loss": 5.9225,
|
|
"mean_token_accuracy": 0.15881798714399337,
|
|
"num_tokens": 5538160.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 6.036073446273804,
|
|
"epoch": 2.620541469703481,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00045999103136911204,
|
|
"loss": 5.8985,
|
|
"mean_token_accuracy": 0.1553686276078224,
|
|
"num_tokens": 5547355.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 6.0549663543701175,
|
|
"epoch": 2.6248388483025353,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00045980157223935965,
|
|
"loss": 5.9368,
|
|
"mean_token_accuracy": 0.1550423875451088,
|
|
"num_tokens": 5557299.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 6.000140953063965,
|
|
"epoch": 2.62913622690159,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00045961170957683806,
|
|
"loss": 5.8642,
|
|
"mean_token_accuracy": 0.16117126494646072,
|
|
"num_tokens": 5565469.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 6.080613803863526,
|
|
"epoch": 2.6334336055006444,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00045942144379613147,
|
|
"loss": 5.9624,
|
|
"mean_token_accuracy": 0.15438342243432998,
|
|
"num_tokens": 5574740.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 6.067436456680298,
|
|
"epoch": 2.637730984099699,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00045923077531270426,
|
|
"loss": 5.9427,
|
|
"mean_token_accuracy": 0.16129356026649475,
|
|
"num_tokens": 5583438.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 6.049432992935181,
|
|
"epoch": 2.6420283626987535,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0004590397045429001,
|
|
"loss": 5.9207,
|
|
"mean_token_accuracy": 0.16217081993818283,
|
|
"num_tokens": 5592389.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 6.02578330039978,
|
|
"epoch": 2.646325741297808,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045884823190394134,
|
|
"loss": 5.8331,
|
|
"mean_token_accuracy": 0.161735899746418,
|
|
"num_tokens": 5601598.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 6.085376787185669,
|
|
"epoch": 2.650623119896863,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004586563578139275,
|
|
"loss": 5.9278,
|
|
"mean_token_accuracy": 0.15633778274059296,
|
|
"num_tokens": 5610498.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.987744140625,
|
|
"epoch": 2.6549204984959176,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00045846408269183505,
|
|
"loss": 5.8222,
|
|
"mean_token_accuracy": 0.1664077639579773,
|
|
"num_tokens": 5620082.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 6.11291241645813,
|
|
"epoch": 2.659217877094972,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00045827140695751603,
|
|
"loss": 5.923,
|
|
"mean_token_accuracy": 0.16166959404945375,
|
|
"num_tokens": 5630291.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 6.003821516036988,
|
|
"epoch": 2.6635152556940267,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004580783310316971,
|
|
"loss": 5.9046,
|
|
"mean_token_accuracy": 0.1637366384267807,
|
|
"num_tokens": 5638784.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.96187949180603,
|
|
"epoch": 2.6678126342930812,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00045788485533597895,
|
|
"loss": 5.7758,
|
|
"mean_token_accuracy": 0.169245745241642,
|
|
"num_tokens": 5647968.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 6.023313808441162,
|
|
"epoch": 2.672110012892136,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00045769098029283526,
|
|
"loss": 5.9912,
|
|
"mean_token_accuracy": 0.1502842426300049,
|
|
"num_tokens": 5657543.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 6.036089277267456,
|
|
"epoch": 2.6764073914911903,
|
|
"grad_norm": 2.203125,
|
|
"learning_rate": 0.0004574967063256115,
|
|
"loss": 5.8912,
|
|
"mean_token_accuracy": 0.16653026044368743,
|
|
"num_tokens": 5666535.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 6.038121747970581,
|
|
"epoch": 2.680704770090245,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00045730203385852447,
|
|
"loss": 5.9891,
|
|
"mean_token_accuracy": 0.156244920194149,
|
|
"num_tokens": 5676273.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.995073223114014,
|
|
"epoch": 2.6850021486892994,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000457106963316661,
|
|
"loss": 5.8877,
|
|
"mean_token_accuracy": 0.1617273211479187,
|
|
"num_tokens": 5684888.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 6.0655543327331545,
|
|
"epoch": 2.689299527288354,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00045691149512597717,
|
|
"loss": 5.9349,
|
|
"mean_token_accuracy": 0.1592724457383156,
|
|
"num_tokens": 5693626.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 6.095427751541138,
|
|
"epoch": 2.6935969058874085,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 0.00045671562971329736,
|
|
"loss": 5.8591,
|
|
"mean_token_accuracy": 0.1641289085149765,
|
|
"num_tokens": 5702542.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 6.010550117492675,
|
|
"epoch": 2.6978942844864635,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00045651936750631337,
|
|
"loss": 5.8762,
|
|
"mean_token_accuracy": 0.16455970853567123,
|
|
"num_tokens": 5711440.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 6.109865999221801,
|
|
"epoch": 2.702191663085518,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00045632270893358333,
|
|
"loss": 5.9553,
|
|
"mean_token_accuracy": 0.1571758583188057,
|
|
"num_tokens": 5721495.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 6.105884838104248,
|
|
"epoch": 2.7064890416845726,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004561256544245312,
|
|
"loss": 5.9774,
|
|
"mean_token_accuracy": 0.15268160849809648,
|
|
"num_tokens": 5730664.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.983286333084107,
|
|
"epoch": 2.710786420283627,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000455928204409445,
|
|
"loss": 5.8767,
|
|
"mean_token_accuracy": 0.16583027988672255,
|
|
"num_tokens": 5740229.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.995842981338501,
|
|
"epoch": 2.7150837988826817,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00045573035931947684,
|
|
"loss": 5.8521,
|
|
"mean_token_accuracy": 0.165335813164711,
|
|
"num_tokens": 5748549.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 6.039884805679321,
|
|
"epoch": 2.7193811774817362,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004555321195866411,
|
|
"loss": 5.8073,
|
|
"mean_token_accuracy": 0.16034042686223984,
|
|
"num_tokens": 5757603.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 6.014684391021729,
|
|
"epoch": 2.723678556080791,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004553334856438143,
|
|
"loss": 5.9799,
|
|
"mean_token_accuracy": 0.15787654370069504,
|
|
"num_tokens": 5767520.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 6.103806591033935,
|
|
"epoch": 2.7279759346798453,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00045513445792473356,
|
|
"loss": 5.9752,
|
|
"mean_token_accuracy": 0.15519406497478486,
|
|
"num_tokens": 5776778.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 6.121107959747315,
|
|
"epoch": 2.7322733132789,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004549350368639958,
|
|
"loss": 6.017,
|
|
"mean_token_accuracy": 0.14941470623016356,
|
|
"num_tokens": 5785652.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 6.127715063095093,
|
|
"epoch": 2.7365706918779544,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00045473522289705693,
|
|
"loss": 5.9586,
|
|
"mean_token_accuracy": 0.15768330097198485,
|
|
"num_tokens": 5795766.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.985169172286987,
|
|
"epoch": 2.740868070477009,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00045453501646023085,
|
|
"loss": 6.0074,
|
|
"mean_token_accuracy": 0.15443310737609864,
|
|
"num_tokens": 5804504.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 6.056064414978027,
|
|
"epoch": 2.7451654490760635,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00045433441799068837,
|
|
"loss": 5.9318,
|
|
"mean_token_accuracy": 0.16102925091981887,
|
|
"num_tokens": 5814161.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 6.089243841171265,
|
|
"epoch": 2.749462827675118,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004541334279264562,
|
|
"loss": 5.8213,
|
|
"mean_token_accuracy": 0.1706912562251091,
|
|
"num_tokens": 5822235.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.9845123291015625,
|
|
"epoch": 2.7537602062741726,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00045393204670641656,
|
|
"loss": 5.8208,
|
|
"mean_token_accuracy": 0.16322228834033012,
|
|
"num_tokens": 5831572.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.950841474533081,
|
|
"epoch": 2.758057584873227,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004537302747703055,
|
|
"loss": 5.8472,
|
|
"mean_token_accuracy": 0.16879533231258392,
|
|
"num_tokens": 5839694.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 6.131230783462525,
|
|
"epoch": 2.7623549634722817,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00045352811255871216,
|
|
"loss": 5.977,
|
|
"mean_token_accuracy": 0.1576570153236389,
|
|
"num_tokens": 5849131.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 6.06053147315979,
|
|
"epoch": 2.7666523420713363,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00045332556051307804,
|
|
"loss": 5.8961,
|
|
"mean_token_accuracy": 0.15734031647443772,
|
|
"num_tokens": 5858861.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 6.070972490310669,
|
|
"epoch": 2.770949720670391,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00045312261907569585,
|
|
"loss": 5.8975,
|
|
"mean_token_accuracy": 0.16149673759937286,
|
|
"num_tokens": 5867585.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 6.065789651870728,
|
|
"epoch": 2.775247099269446,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00045291928868970867,
|
|
"loss": 5.9368,
|
|
"mean_token_accuracy": 0.15555136650800705,
|
|
"num_tokens": 5876256.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.986635446548462,
|
|
"epoch": 2.7795444778685003,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004527155697991087,
|
|
"loss": 5.9878,
|
|
"mean_token_accuracy": 0.15262294709682464,
|
|
"num_tokens": 5885302.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 6.065853786468506,
|
|
"epoch": 2.783841856467555,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004525114628487365,
|
|
"loss": 5.9882,
|
|
"mean_token_accuracy": 0.15331604927778245,
|
|
"num_tokens": 5895066.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 6.082369232177735,
|
|
"epoch": 2.7881392350666094,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00045230696828428026,
|
|
"loss": 5.9681,
|
|
"mean_token_accuracy": 0.15811365395784377,
|
|
"num_tokens": 5903258.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 6.095545625686645,
|
|
"epoch": 2.792436613665664,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0004521020865522742,
|
|
"loss": 5.8736,
|
|
"mean_token_accuracy": 0.16074307411909103,
|
|
"num_tokens": 5911714.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.9788854122161865,
|
|
"epoch": 2.7967339922647185,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00045189681810009827,
|
|
"loss": 5.9362,
|
|
"mean_token_accuracy": 0.1553533896803856,
|
|
"num_tokens": 5920432.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 6.08413987159729,
|
|
"epoch": 2.801031370863773,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00045169116337597653,
|
|
"loss": 5.9512,
|
|
"mean_token_accuracy": 0.15794725120067596,
|
|
"num_tokens": 5929202.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 6.150643682479858,
|
|
"epoch": 2.8053287494628276,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000451485122828977,
|
|
"loss": 5.9992,
|
|
"mean_token_accuracy": 0.1512179285287857,
|
|
"num_tokens": 5938034.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 6.035185527801514,
|
|
"epoch": 2.809626128061882,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00045127869690900956,
|
|
"loss": 5.8427,
|
|
"mean_token_accuracy": 0.16318582147359847,
|
|
"num_tokens": 5946944.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 6.003673124313354,
|
|
"epoch": 2.8139235066609367,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00045107188606682613,
|
|
"loss": 5.9278,
|
|
"mean_token_accuracy": 0.16485971510410308,
|
|
"num_tokens": 5956475.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 6.074523639678955,
|
|
"epoch": 2.8182208852599913,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0004508646907540188,
|
|
"loss": 5.9141,
|
|
"mean_token_accuracy": 0.15292692929506302,
|
|
"num_tokens": 5965814.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 6.039639902114868,
|
|
"epoch": 2.8225182638590463,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004506571114230195,
|
|
"loss": 5.9433,
|
|
"mean_token_accuracy": 0.15330400615930556,
|
|
"num_tokens": 5973850.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.9470912456512455,
|
|
"epoch": 2.826815642458101,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00045044914852709824,
|
|
"loss": 5.9089,
|
|
"mean_token_accuracy": 0.16015242487192155,
|
|
"num_tokens": 5982987.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 6.110384130477906,
|
|
"epoch": 2.8311130210571553,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0004502408025203631,
|
|
"loss": 5.8842,
|
|
"mean_token_accuracy": 0.16538529396057128,
|
|
"num_tokens": 5992227.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 6.019767904281617,
|
|
"epoch": 2.83541039965621,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004500320738577584,
|
|
"loss": 5.8761,
|
|
"mean_token_accuracy": 0.16317329704761505,
|
|
"num_tokens": 6000243.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 6.030108404159546,
|
|
"epoch": 2.8397077782552644,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00044982296299506407,
|
|
"loss": 5.8914,
|
|
"mean_token_accuracy": 0.1607070490717888,
|
|
"num_tokens": 6009771.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 6.118225860595703,
|
|
"epoch": 2.844005156854319,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004496134703888948,
|
|
"loss": 5.9607,
|
|
"mean_token_accuracy": 0.1556072324514389,
|
|
"num_tokens": 6018683.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 6.020996952056885,
|
|
"epoch": 2.8483025354533735,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00044940359649669846,
|
|
"loss": 5.8087,
|
|
"mean_token_accuracy": 0.1732095018029213,
|
|
"num_tokens": 6027422.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.9873603820800785,
|
|
"epoch": 2.852599914052428,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00044919334177675595,
|
|
"loss": 5.9097,
|
|
"mean_token_accuracy": 0.15791536569595338,
|
|
"num_tokens": 6035670.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 6.055611944198608,
|
|
"epoch": 2.8568972926514826,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00044898270668817955,
|
|
"loss": 5.8263,
|
|
"mean_token_accuracy": 0.16295338571071624,
|
|
"num_tokens": 6044092.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.976633167266845,
|
|
"epoch": 2.861194671250537,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000448771691690912,
|
|
"loss": 5.8944,
|
|
"mean_token_accuracy": 0.15649735927581787,
|
|
"num_tokens": 6053970.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 6.052254343032837,
|
|
"epoch": 2.8654920498495917,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004485602972457257,
|
|
"loss": 5.8554,
|
|
"mean_token_accuracy": 0.16512034088373184,
|
|
"num_tokens": 6062965.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 6.056798696517944,
|
|
"epoch": 2.8697894284486463,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00044834852381422165,
|
|
"loss": 5.9353,
|
|
"mean_token_accuracy": 0.1571454830467701,
|
|
"num_tokens": 6072420.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 6.001439142227173,
|
|
"epoch": 2.874086807047701,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00044813637185882836,
|
|
"loss": 5.8361,
|
|
"mean_token_accuracy": 0.15966329127550125,
|
|
"num_tokens": 6080915.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 6.056124019622803,
|
|
"epoch": 2.8783841856467554,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00044792384184280106,
|
|
"loss": 5.9615,
|
|
"mean_token_accuracy": 0.15560488998889924,
|
|
"num_tokens": 6090453.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 6.047571516036987,
|
|
"epoch": 2.88268156424581,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00044771093423022013,
|
|
"loss": 6.0116,
|
|
"mean_token_accuracy": 0.15179160833358765,
|
|
"num_tokens": 6099390.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 6.016407108306884,
|
|
"epoch": 2.8869789428448644,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004474976494859909,
|
|
"loss": 5.9221,
|
|
"mean_token_accuracy": 0.1655833527445793,
|
|
"num_tokens": 6108677.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 6.0324239253997805,
|
|
"epoch": 2.891276321443919,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004472839880758419,
|
|
"loss": 5.8471,
|
|
"mean_token_accuracy": 0.16346813291311263,
|
|
"num_tokens": 6117151.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 6.125880289077759,
|
|
"epoch": 2.8955737000429735,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004470699504663242,
|
|
"loss": 5.958,
|
|
"mean_token_accuracy": 0.15495969355106354,
|
|
"num_tokens": 6127167.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.970075178146362,
|
|
"epoch": 2.899871078642028,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004468555371248104,
|
|
"loss": 5.8519,
|
|
"mean_token_accuracy": 0.16259023696184158,
|
|
"num_tokens": 6136487.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 6.040617656707764,
|
|
"epoch": 2.904168457241083,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004466407485194937,
|
|
"loss": 5.9488,
|
|
"mean_token_accuracy": 0.15836277604103088,
|
|
"num_tokens": 6145334.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 6.061579990386963,
|
|
"epoch": 2.9084658358401376,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0004464255851193864,
|
|
"loss": 5.8812,
|
|
"mean_token_accuracy": 0.1616512954235077,
|
|
"num_tokens": 6155062.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 6.0243054866790775,
|
|
"epoch": 2.912763214439192,
|
|
"grad_norm": 2.171875,
|
|
"learning_rate": 0.0004462100473943194,
|
|
"loss": 5.8465,
|
|
"mean_token_accuracy": 0.1651116468012333,
|
|
"num_tokens": 6164313.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.990101289749146,
|
|
"epoch": 2.9170605930382467,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.000445994135814941,
|
|
"loss": 5.893,
|
|
"mean_token_accuracy": 0.1572832629084587,
|
|
"num_tokens": 6173513.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 6.036487340927124,
|
|
"epoch": 2.9213579716373013,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00044577785085271566,
|
|
"loss": 5.9094,
|
|
"mean_token_accuracy": 0.1575427159667015,
|
|
"num_tokens": 6182000.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 6.066547584533692,
|
|
"epoch": 2.925655350236356,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004455611929799235,
|
|
"loss": 5.9565,
|
|
"mean_token_accuracy": 0.1518727272748947,
|
|
"num_tokens": 6191887.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.954565000534058,
|
|
"epoch": 2.9299527288354104,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004453441626696585,
|
|
"loss": 5.992,
|
|
"mean_token_accuracy": 0.1506217211484909,
|
|
"num_tokens": 6202897.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 6.113213729858399,
|
|
"epoch": 2.934250107434465,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00044512676039582823,
|
|
"loss": 5.8712,
|
|
"mean_token_accuracy": 0.16455415189266204,
|
|
"num_tokens": 6211811.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 6.095485687255859,
|
|
"epoch": 2.9385474860335195,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004449089866331524,
|
|
"loss": 5.8746,
|
|
"mean_token_accuracy": 0.16725364476442336,
|
|
"num_tokens": 6219896.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.957816171646118,
|
|
"epoch": 2.942844864632574,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004446908418571617,
|
|
"loss": 5.8484,
|
|
"mean_token_accuracy": 0.16169391572475433,
|
|
"num_tokens": 6228212.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 6.073782682418823,
|
|
"epoch": 2.9471422432316285,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004444723265441973,
|
|
"loss": 6.0233,
|
|
"mean_token_accuracy": 0.15530604422092437,
|
|
"num_tokens": 6238133.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.997539377212524,
|
|
"epoch": 2.9514396218306835,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004442534411714092,
|
|
"loss": 5.9421,
|
|
"mean_token_accuracy": 0.1516396000981331,
|
|
"num_tokens": 6247331.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 6.130997800827027,
|
|
"epoch": 2.955737000429738,
|
|
"grad_norm": 1.9140625,
|
|
"learning_rate": 0.00044403418621675555,
|
|
"loss": 5.9203,
|
|
"mean_token_accuracy": 0.15922869890928268,
|
|
"num_tokens": 6255280.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 6.053282690048218,
|
|
"epoch": 2.9600343790287926,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004438145621590017,
|
|
"loss": 5.874,
|
|
"mean_token_accuracy": 0.16302104890346528,
|
|
"num_tokens": 6264752.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.927190065383911,
|
|
"epoch": 2.964331757627847,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00044359456947771857,
|
|
"loss": 5.8244,
|
|
"mean_token_accuracy": 0.1627937912940979,
|
|
"num_tokens": 6273258.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.865182542800904,
|
|
"epoch": 2.9686291362269017,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004433742086532824,
|
|
"loss": 5.749,
|
|
"mean_token_accuracy": 0.17724997848272322,
|
|
"num_tokens": 6281584.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 6.036783647537232,
|
|
"epoch": 2.9729265148259563,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00044315348016687317,
|
|
"loss": 5.8699,
|
|
"mean_token_accuracy": 0.16180484518408775,
|
|
"num_tokens": 6290016.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 6.031026220321655,
|
|
"epoch": 2.977223893425011,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004429323845004736,
|
|
"loss": 5.7715,
|
|
"mean_token_accuracy": 0.16854195296764374,
|
|
"num_tokens": 6298569.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.935258579254151,
|
|
"epoch": 2.9815212720240654,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00044271092213686824,
|
|
"loss": 5.8229,
|
|
"mean_token_accuracy": 0.16938660740852357,
|
|
"num_tokens": 6307684.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 6.056712579727173,
|
|
"epoch": 2.98581865062312,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00044248909355964247,
|
|
"loss": 5.9448,
|
|
"mean_token_accuracy": 0.16194785684347152,
|
|
"num_tokens": 6317767.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 6.139704084396362,
|
|
"epoch": 2.9901160292221745,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00044226689925318117,
|
|
"loss": 5.983,
|
|
"mean_token_accuracy": 0.14899415671825408,
|
|
"num_tokens": 6327457.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 6.000520706176758,
|
|
"epoch": 2.994413407821229,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00044204433970266785,
|
|
"loss": 5.7789,
|
|
"mean_token_accuracy": 0.17212653458118438,
|
|
"num_tokens": 6335747.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.991744089126587,
|
|
"epoch": 2.9987107864202835,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004418214153940837,
|
|
"loss": 5.8526,
|
|
"mean_token_accuracy": 0.15984081029891967,
|
|
"num_tokens": 6344750.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.980947123633491,
|
|
"epoch": 3.002578427159433,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00044159812681420624,
|
|
"loss": 5.8216,
|
|
"mean_token_accuracy": 0.16503895819187164,
|
|
"num_tokens": 6354779.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 6.031644868850708,
|
|
"epoch": 3.0068758057584875,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004413744744506086,
|
|
"loss": 5.6015,
|
|
"mean_token_accuracy": 0.1682058244943619,
|
|
"num_tokens": 6363809.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 3.0068758057584875,
|
|
"eval_entropy": 5.87646267972551,
|
|
"eval_loss": 6.164710998535156,
|
|
"eval_mean_token_accuracy": 0.15751764235330057,
|
|
"eval_num_tokens": 6363809.0,
|
|
"eval_runtime": 2.6524,
|
|
"eval_samples_per_second": 1338.05,
|
|
"eval_steps_per_second": 167.398,
|
|
"step": 3500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 11630,
|
|
"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": 9678727905408000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|