15200 lines
415 KiB
JSON
15200 lines
415 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 6.443489471422432,
|
|
"eval_steps": 500,
|
|
"global_step": 7500,
|
|
"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
|
|
},
|
|
{
|
|
"entropy": 6.015894985198974,
|
|
"epoch": 3.011173184357542,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00044115045879165806,
|
|
"loss": 5.684,
|
|
"mean_token_accuracy": 0.16529116332530974,
|
|
"num_tokens": 6373082.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.9731237411499025,
|
|
"epoch": 3.0154705629565965,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00044092608032651515,
|
|
"loss": 5.6293,
|
|
"mean_token_accuracy": 0.16774954795837402,
|
|
"num_tokens": 6381286.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.953437948226929,
|
|
"epoch": 3.019767941555651,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00044070133954513305,
|
|
"loss": 5.5925,
|
|
"mean_token_accuracy": 0.17507788687944412,
|
|
"num_tokens": 6390217.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 6.063429689407348,
|
|
"epoch": 3.0240653201547056,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0004404762369382555,
|
|
"loss": 5.6172,
|
|
"mean_token_accuracy": 0.16711781471967696,
|
|
"num_tokens": 6399276.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.934738111495972,
|
|
"epoch": 3.02836269875376,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00044025077299741683,
|
|
"loss": 5.598,
|
|
"mean_token_accuracy": 0.17819776833057405,
|
|
"num_tokens": 6407981.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.873000812530518,
|
|
"epoch": 3.0326600773528147,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.00044002494821494007,
|
|
"loss": 5.6096,
|
|
"mean_token_accuracy": 0.16942654699087142,
|
|
"num_tokens": 6416159.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.9321716785430905,
|
|
"epoch": 3.0369574559518693,
|
|
"grad_norm": 1.84375,
|
|
"learning_rate": 0.00043979876308393635,
|
|
"loss": 5.6456,
|
|
"mean_token_accuracy": 0.16875908076763152,
|
|
"num_tokens": 6424564.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 6.044365787506104,
|
|
"epoch": 3.041254834550924,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0004395722180983036,
|
|
"loss": 5.6842,
|
|
"mean_token_accuracy": 0.16706381291151046,
|
|
"num_tokens": 6434163.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.885945844650268,
|
|
"epoch": 3.0455522131499784,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00043934531375272535,
|
|
"loss": 5.5156,
|
|
"mean_token_accuracy": 0.18382880687713624,
|
|
"num_tokens": 6443372.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.890963315963745,
|
|
"epoch": 3.049849591749033,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00043911805054267015,
|
|
"loss": 5.603,
|
|
"mean_token_accuracy": 0.17347687631845474,
|
|
"num_tokens": 6452638.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 6.013320207595825,
|
|
"epoch": 3.0541469703480875,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00043889042896439004,
|
|
"loss": 5.6054,
|
|
"mean_token_accuracy": 0.17310492396354676,
|
|
"num_tokens": 6461319.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.980799150466919,
|
|
"epoch": 3.0584443489471425,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00043866244951491946,
|
|
"loss": 5.5212,
|
|
"mean_token_accuracy": 0.18536942601203918,
|
|
"num_tokens": 6469506.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.918030881881714,
|
|
"epoch": 3.062741727546197,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00043843411269207445,
|
|
"loss": 5.6022,
|
|
"mean_token_accuracy": 0.17848867923021317,
|
|
"num_tokens": 6478404.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.859469270706176,
|
|
"epoch": 3.0670391061452515,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004382054189944514,
|
|
"loss": 5.5275,
|
|
"mean_token_accuracy": 0.16954101473093033,
|
|
"num_tokens": 6487447.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.928318214416504,
|
|
"epoch": 3.071336484744306,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004379763689214259,
|
|
"loss": 5.5771,
|
|
"mean_token_accuracy": 0.17103980779647826,
|
|
"num_tokens": 6496738.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.9367999076843265,
|
|
"epoch": 3.0756338633433606,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004377469629731518,
|
|
"loss": 5.5846,
|
|
"mean_token_accuracy": 0.17155404388904572,
|
|
"num_tokens": 6505848.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.927693223953247,
|
|
"epoch": 3.079931241942415,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004375172016505599,
|
|
"loss": 5.5666,
|
|
"mean_token_accuracy": 0.1732747122645378,
|
|
"num_tokens": 6515731.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.933186864852905,
|
|
"epoch": 3.0842286205414697,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004372870854553572,
|
|
"loss": 5.6321,
|
|
"mean_token_accuracy": 0.17319453209638597,
|
|
"num_tokens": 6524914.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.973398923873901,
|
|
"epoch": 3.0885259991405243,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004370566148900255,
|
|
"loss": 5.6117,
|
|
"mean_token_accuracy": 0.1755769744515419,
|
|
"num_tokens": 6533712.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.945407104492188,
|
|
"epoch": 3.092823377739579,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00043682579045782024,
|
|
"loss": 5.6883,
|
|
"mean_token_accuracy": 0.16778713166713716,
|
|
"num_tokens": 6543313.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.8294957160949705,
|
|
"epoch": 3.0971207563386334,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0004365946126627699,
|
|
"loss": 5.5898,
|
|
"mean_token_accuracy": 0.18004253804683684,
|
|
"num_tokens": 6551634.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.959538412094116,
|
|
"epoch": 3.101418134937688,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00043636308200967433,
|
|
"loss": 5.5841,
|
|
"mean_token_accuracy": 0.17596693336963654,
|
|
"num_tokens": 6560695.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.855001211166382,
|
|
"epoch": 3.1057155135367425,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004361311990041039,
|
|
"loss": 5.4911,
|
|
"mean_token_accuracy": 0.1822124481201172,
|
|
"num_tokens": 6569086.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.857224369049073,
|
|
"epoch": 3.110012892135797,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00043589896415239843,
|
|
"loss": 5.5732,
|
|
"mean_token_accuracy": 0.18176483660936354,
|
|
"num_tokens": 6578287.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.954432964324951,
|
|
"epoch": 3.1143102707348516,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00043566637796166595,
|
|
"loss": 5.6391,
|
|
"mean_token_accuracy": 0.17255895137786864,
|
|
"num_tokens": 6587015.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.947692966461181,
|
|
"epoch": 3.118607649333906,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00043543344093978186,
|
|
"loss": 5.6755,
|
|
"mean_token_accuracy": 0.16395042687654496,
|
|
"num_tokens": 6596187.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.905115127563477,
|
|
"epoch": 3.122905027932961,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00043520015359538745,
|
|
"loss": 5.5528,
|
|
"mean_token_accuracy": 0.17874250262975694,
|
|
"num_tokens": 6605226.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.842601776123047,
|
|
"epoch": 3.1272024065320156,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004349665164378891,
|
|
"loss": 5.5742,
|
|
"mean_token_accuracy": 0.17076538652181625,
|
|
"num_tokens": 6613232.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.845444011688232,
|
|
"epoch": 3.13149978513107,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00043473252997745684,
|
|
"loss": 5.5979,
|
|
"mean_token_accuracy": 0.1720021113753319,
|
|
"num_tokens": 6622247.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.979873275756836,
|
|
"epoch": 3.1357971637301247,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.00043449819472502366,
|
|
"loss": 5.5604,
|
|
"mean_token_accuracy": 0.1729029208421707,
|
|
"num_tokens": 6630883.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.91176176071167,
|
|
"epoch": 3.1400945423291793,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004342635111922841,
|
|
"loss": 5.6813,
|
|
"mean_token_accuracy": 0.1710082083940506,
|
|
"num_tokens": 6639399.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.905380296707153,
|
|
"epoch": 3.144391920928234,
|
|
"grad_norm": 1.796875,
|
|
"learning_rate": 0.0004340284798916931,
|
|
"loss": 5.61,
|
|
"mean_token_accuracy": 0.17253366112709045,
|
|
"num_tokens": 6649288.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.920901727676392,
|
|
"epoch": 3.1486892995272884,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004337931013364653,
|
|
"loss": 5.5288,
|
|
"mean_token_accuracy": 0.17396813333034516,
|
|
"num_tokens": 6658670.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.912379932403565,
|
|
"epoch": 3.152986678126343,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000433557376040573,
|
|
"loss": 5.5899,
|
|
"mean_token_accuracy": 0.17683041244745254,
|
|
"num_tokens": 6667302.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.9615960121154785,
|
|
"epoch": 3.1572840567253975,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00043332130451874645,
|
|
"loss": 5.6428,
|
|
"mean_token_accuracy": 0.1759822517633438,
|
|
"num_tokens": 6677393.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.95091872215271,
|
|
"epoch": 3.161581435324452,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00043308488728647127,
|
|
"loss": 5.6342,
|
|
"mean_token_accuracy": 0.1701880618929863,
|
|
"num_tokens": 6686727.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.840176820755005,
|
|
"epoch": 3.1658788139235066,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004328481248599882,
|
|
"loss": 5.5449,
|
|
"mean_token_accuracy": 0.17628217339515687,
|
|
"num_tokens": 6696116.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.963242101669311,
|
|
"epoch": 3.170176192522561,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004326110177562918,
|
|
"loss": 5.6446,
|
|
"mean_token_accuracy": 0.1671574518084526,
|
|
"num_tokens": 6704640.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.824881553649902,
|
|
"epoch": 3.1744735711216157,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00043237356649312926,
|
|
"loss": 5.5193,
|
|
"mean_token_accuracy": 0.18162815719842912,
|
|
"num_tokens": 6713663.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.866562271118164,
|
|
"epoch": 3.17877094972067,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004321357715889991,
|
|
"loss": 5.6668,
|
|
"mean_token_accuracy": 0.16941877454519272,
|
|
"num_tokens": 6722965.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.899015760421753,
|
|
"epoch": 3.1830683283197247,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0004318976335631505,
|
|
"loss": 5.6111,
|
|
"mean_token_accuracy": 0.17989031970500946,
|
|
"num_tokens": 6732776.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.929231309890747,
|
|
"epoch": 3.1873657069187797,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00043165915293558155,
|
|
"loss": 5.5999,
|
|
"mean_token_accuracy": 0.17258843183517455,
|
|
"num_tokens": 6741309.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.877465438842774,
|
|
"epoch": 3.1916630855178343,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004314203302270388,
|
|
"loss": 5.6631,
|
|
"mean_token_accuracy": 0.16744825690984727,
|
|
"num_tokens": 6750584.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.876456165313721,
|
|
"epoch": 3.195960464116889,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004311811659590154,
|
|
"loss": 5.6316,
|
|
"mean_token_accuracy": 0.17229558080434798,
|
|
"num_tokens": 6759344.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.984838008880615,
|
|
"epoch": 3.2002578427159434,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004309416606537507,
|
|
"loss": 5.7678,
|
|
"mean_token_accuracy": 0.16722186505794526,
|
|
"num_tokens": 6770345.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.941770744323731,
|
|
"epoch": 3.204555221314998,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00043070181483422843,
|
|
"loss": 5.6582,
|
|
"mean_token_accuracy": 0.16996753960847855,
|
|
"num_tokens": 6779991.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.878832578659058,
|
|
"epoch": 3.2088525999140525,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000430461629024176,
|
|
"loss": 5.6153,
|
|
"mean_token_accuracy": 0.1736441284418106,
|
|
"num_tokens": 6788972.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.871813106536865,
|
|
"epoch": 3.213149978513107,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004302211037480634,
|
|
"loss": 5.5313,
|
|
"mean_token_accuracy": 0.17377722412347793,
|
|
"num_tokens": 6796967.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.8901526927948,
|
|
"epoch": 3.2174473571121616,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004299802395311015,
|
|
"loss": 5.6466,
|
|
"mean_token_accuracy": 0.17091624289751053,
|
|
"num_tokens": 6805961.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.819453954696655,
|
|
"epoch": 3.221744735711216,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0004297390368992414,
|
|
"loss": 5.5531,
|
|
"mean_token_accuracy": 0.17513273805379867,
|
|
"num_tokens": 6814657.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.874273490905762,
|
|
"epoch": 3.2260421143102707,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00042949749637917353,
|
|
"loss": 5.5683,
|
|
"mean_token_accuracy": 0.17742332816123962,
|
|
"num_tokens": 6823095.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.943544006347656,
|
|
"epoch": 3.230339492909325,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004292556184983256,
|
|
"loss": 5.5998,
|
|
"mean_token_accuracy": 0.1751185715198517,
|
|
"num_tokens": 6832195.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.970035457611084,
|
|
"epoch": 3.2346368715083798,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0004290134037848623,
|
|
"loss": 5.7291,
|
|
"mean_token_accuracy": 0.16783777326345445,
|
|
"num_tokens": 6840922.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.885451602935791,
|
|
"epoch": 3.2389342501074343,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00042877085276768386,
|
|
"loss": 5.5785,
|
|
"mean_token_accuracy": 0.17677924335002898,
|
|
"num_tokens": 6849182.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.899662208557129,
|
|
"epoch": 3.243231628706489,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00042852796597642455,
|
|
"loss": 5.5591,
|
|
"mean_token_accuracy": 0.17737246453762054,
|
|
"num_tokens": 6857932.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 6.003991031646729,
|
|
"epoch": 3.247529007305544,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004282847439414522,
|
|
"loss": 5.7175,
|
|
"mean_token_accuracy": 0.1637961134314537,
|
|
"num_tokens": 6867283.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.907202100753784,
|
|
"epoch": 3.2518263859045984,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004280411871938664,
|
|
"loss": 5.6856,
|
|
"mean_token_accuracy": 0.17289682179689408,
|
|
"num_tokens": 6876123.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.9634599685668945,
|
|
"epoch": 3.256123764503653,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004277972962654979,
|
|
"loss": 5.6322,
|
|
"mean_token_accuracy": 0.1731411874294281,
|
|
"num_tokens": 6885239.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.924293422698975,
|
|
"epoch": 3.2604211431027075,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0004275530716889069,
|
|
"loss": 5.6646,
|
|
"mean_token_accuracy": 0.17210501581430435,
|
|
"num_tokens": 6895061.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.930949592590332,
|
|
"epoch": 3.264718521701762,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0004273085139973822,
|
|
"loss": 5.7252,
|
|
"mean_token_accuracy": 0.16514313966035843,
|
|
"num_tokens": 6903828.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.97692174911499,
|
|
"epoch": 3.2690159003008166,
|
|
"grad_norm": 1.984375,
|
|
"learning_rate": 0.0004270636237249401,
|
|
"loss": 5.6161,
|
|
"mean_token_accuracy": 0.1754364401102066,
|
|
"num_tokens": 6912805.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.92157392501831,
|
|
"epoch": 3.273313278899871,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00042681840140632314,
|
|
"loss": 5.6871,
|
|
"mean_token_accuracy": 0.16453767269849778,
|
|
"num_tokens": 6922165.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.899108171463013,
|
|
"epoch": 3.2776106574989257,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004265728475769989,
|
|
"loss": 5.6671,
|
|
"mean_token_accuracy": 0.16593455374240876,
|
|
"num_tokens": 6931677.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.943723392486572,
|
|
"epoch": 3.28190803609798,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004263269627731586,
|
|
"loss": 5.6095,
|
|
"mean_token_accuracy": 0.17542295902967453,
|
|
"num_tokens": 6940486.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.843339300155639,
|
|
"epoch": 3.2862054146970348,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0004260807475317164,
|
|
"loss": 5.6407,
|
|
"mean_token_accuracy": 0.16801389753818513,
|
|
"num_tokens": 6948990.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.9931563377380375,
|
|
"epoch": 3.2905027932960893,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004258342023903081,
|
|
"loss": 5.7578,
|
|
"mean_token_accuracy": 0.1611199975013733,
|
|
"num_tokens": 6959311.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.908085584640503,
|
|
"epoch": 3.294800171895144,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00042558732788728975,
|
|
"loss": 5.5339,
|
|
"mean_token_accuracy": 0.17625602185726166,
|
|
"num_tokens": 6968619.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.82570743560791,
|
|
"epoch": 3.2990975504941984,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00042534012456173643,
|
|
"loss": 5.5896,
|
|
"mean_token_accuracy": 0.1780730351805687,
|
|
"num_tokens": 6977469.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.8205639839172365,
|
|
"epoch": 3.303394929093253,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00042509259295344157,
|
|
"loss": 5.5753,
|
|
"mean_token_accuracy": 0.1723286271095276,
|
|
"num_tokens": 6986772.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.927529001235962,
|
|
"epoch": 3.3076923076923075,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00042484473360291514,
|
|
"loss": 5.5949,
|
|
"mean_token_accuracy": 0.16374675631523133,
|
|
"num_tokens": 6993937.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.877727699279785,
|
|
"epoch": 3.311989686291362,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00042459654705138294,
|
|
"loss": 5.6445,
|
|
"mean_token_accuracy": 0.17454762309789656,
|
|
"num_tokens": 7003222.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.823605346679687,
|
|
"epoch": 3.316287064890417,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.0004243480338407853,
|
|
"loss": 5.6121,
|
|
"mean_token_accuracy": 0.17623748630285263,
|
|
"num_tokens": 7012055.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.951910734176636,
|
|
"epoch": 3.3205844434894716,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004240991945137755,
|
|
"loss": 5.618,
|
|
"mean_token_accuracy": 0.1724048599600792,
|
|
"num_tokens": 7021036.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.845747280120849,
|
|
"epoch": 3.324881822088526,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00042385002961371944,
|
|
"loss": 5.6124,
|
|
"mean_token_accuracy": 0.1768123984336853,
|
|
"num_tokens": 7030450.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.891090822219849,
|
|
"epoch": 3.3291792006875807,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004236005396846935,
|
|
"loss": 5.7088,
|
|
"mean_token_accuracy": 0.16993040442466736,
|
|
"num_tokens": 7039740.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.988770532608032,
|
|
"epoch": 3.333476579286635,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00042335072527148406,
|
|
"loss": 5.6964,
|
|
"mean_token_accuracy": 0.17072457522153855,
|
|
"num_tokens": 7050430.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.8443334102630615,
|
|
"epoch": 3.3377739578856898,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0004231005869195859,
|
|
"loss": 5.6788,
|
|
"mean_token_accuracy": 0.16346832513809204,
|
|
"num_tokens": 7059477.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.877062797546387,
|
|
"epoch": 3.3420713364847443,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0004228501251752011,
|
|
"loss": 5.6384,
|
|
"mean_token_accuracy": 0.17213663309812546,
|
|
"num_tokens": 7067805.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.900841093063354,
|
|
"epoch": 3.346368715083799,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00042259934058523814,
|
|
"loss": 5.6272,
|
|
"mean_token_accuracy": 0.17160079777240753,
|
|
"num_tokens": 7077606.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.889443397521973,
|
|
"epoch": 3.3506660936828534,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00042234823369731027,
|
|
"loss": 5.6009,
|
|
"mean_token_accuracy": 0.17515387386083603,
|
|
"num_tokens": 7085647.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.867230081558228,
|
|
"epoch": 3.354963472281908,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00042209680505973465,
|
|
"loss": 5.6146,
|
|
"mean_token_accuracy": 0.17526530474424362,
|
|
"num_tokens": 7095298.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.842862987518311,
|
|
"epoch": 3.3592608508809625,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004218450552215308,
|
|
"loss": 5.6755,
|
|
"mean_token_accuracy": 0.1685987025499344,
|
|
"num_tokens": 7105207.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.902641105651855,
|
|
"epoch": 3.363558229480017,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004215929847324199,
|
|
"loss": 5.7311,
|
|
"mean_token_accuracy": 0.1662062719464302,
|
|
"num_tokens": 7114833.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.988971614837647,
|
|
"epoch": 3.3678556080790716,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000421340594142823,
|
|
"loss": 5.5249,
|
|
"mean_token_accuracy": 0.18613329231739045,
|
|
"num_tokens": 7123608.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.90705714225769,
|
|
"epoch": 3.3721529866781266,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00042108788400386035,
|
|
"loss": 5.5904,
|
|
"mean_token_accuracy": 0.17238757908344268,
|
|
"num_tokens": 7132250.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.792404079437256,
|
|
"epoch": 3.376450365277181,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004208348548673498,
|
|
"loss": 5.7009,
|
|
"mean_token_accuracy": 0.16918646693229675,
|
|
"num_tokens": 7142086.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.949172496795654,
|
|
"epoch": 3.3807477438762357,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000420581507285806,
|
|
"loss": 5.6596,
|
|
"mean_token_accuracy": 0.16378311216831207,
|
|
"num_tokens": 7152434.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.887945604324341,
|
|
"epoch": 3.38504512247529,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004203278418124386,
|
|
"loss": 5.575,
|
|
"mean_token_accuracy": 0.18023645877838135,
|
|
"num_tokens": 7163041.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.804234600067138,
|
|
"epoch": 3.3893425010743448,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004200738590011518,
|
|
"loss": 5.5398,
|
|
"mean_token_accuracy": 0.18725930899381638,
|
|
"num_tokens": 7171875.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.8566234588623045,
|
|
"epoch": 3.3936398796733993,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00041981955940654245,
|
|
"loss": 5.6844,
|
|
"mean_token_accuracy": 0.16930943876504898,
|
|
"num_tokens": 7180803.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.904590606689453,
|
|
"epoch": 3.397937258272454,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004195649435838992,
|
|
"loss": 5.6848,
|
|
"mean_token_accuracy": 0.16652224510908126,
|
|
"num_tokens": 7190661.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.834088945388794,
|
|
"epoch": 3.4022346368715084,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004193100120892013,
|
|
"loss": 5.5524,
|
|
"mean_token_accuracy": 0.18242962062358856,
|
|
"num_tokens": 7199357.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.93598427772522,
|
|
"epoch": 3.406532015470563,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004190547654791172,
|
|
"loss": 5.709,
|
|
"mean_token_accuracy": 0.1664261445403099,
|
|
"num_tokens": 7209856.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.938437223434448,
|
|
"epoch": 3.4108293940696175,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00041879920431100347,
|
|
"loss": 5.6738,
|
|
"mean_token_accuracy": 0.164241161942482,
|
|
"num_tokens": 7218778.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.844049024581909,
|
|
"epoch": 3.415126772668672,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004185433291429036,
|
|
"loss": 5.6757,
|
|
"mean_token_accuracy": 0.17473727464675903,
|
|
"num_tokens": 7228442.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.896280431747437,
|
|
"epoch": 3.4194241512677266,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00041828714053354665,
|
|
"loss": 5.6889,
|
|
"mean_token_accuracy": 0.16805070340633393,
|
|
"num_tokens": 7238724.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.891405868530273,
|
|
"epoch": 3.423721529866781,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004180306390423462,
|
|
"loss": 5.6373,
|
|
"mean_token_accuracy": 0.18036131858825682,
|
|
"num_tokens": 7247844.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.900328779220581,
|
|
"epoch": 3.4280189084658357,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00041777382522939884,
|
|
"loss": 5.6854,
|
|
"mean_token_accuracy": 0.17353073060512542,
|
|
"num_tokens": 7257260.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.885912942886352,
|
|
"epoch": 3.4323162870648902,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00041751669965548344,
|
|
"loss": 5.7127,
|
|
"mean_token_accuracy": 0.1682794451713562,
|
|
"num_tokens": 7266890.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.970642328262329,
|
|
"epoch": 3.4366136656639448,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00041725926288205945,
|
|
"loss": 5.7342,
|
|
"mean_token_accuracy": 0.16623314768075942,
|
|
"num_tokens": 7276114.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 3.4366136656639448,
|
|
"eval_entropy": 5.80956880036775,
|
|
"eval_loss": 6.141725540161133,
|
|
"eval_mean_token_accuracy": 0.15913750622253697,
|
|
"eval_num_tokens": 7276114.0,
|
|
"eval_runtime": 2.6531,
|
|
"eval_samples_per_second": 1337.66,
|
|
"eval_steps_per_second": 167.349,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"entropy": 5.945313453674316,
|
|
"epoch": 3.4409110442629998,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004170015154712658,
|
|
"loss": 5.6549,
|
|
"mean_token_accuracy": 0.1715627580881119,
|
|
"num_tokens": 7284426.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 5.851800489425659,
|
|
"epoch": 3.4452084228620543,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00041674345798591993,
|
|
"loss": 5.6862,
|
|
"mean_token_accuracy": 0.16800928711891175,
|
|
"num_tokens": 7294813.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 5.941851615905762,
|
|
"epoch": 3.449505801461109,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004164850909895161,
|
|
"loss": 5.6742,
|
|
"mean_token_accuracy": 0.17423058599233626,
|
|
"num_tokens": 7304655.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 5.844107723236084,
|
|
"epoch": 3.4538031800601634,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004162264150462247,
|
|
"loss": 5.6599,
|
|
"mean_token_accuracy": 0.16926209628582,
|
|
"num_tokens": 7313610.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 5.881008768081665,
|
|
"epoch": 3.458100558659218,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00041596743072089065,
|
|
"loss": 5.6638,
|
|
"mean_token_accuracy": 0.1754756286740303,
|
|
"num_tokens": 7322243.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 6.026263904571533,
|
|
"epoch": 3.4623979372582725,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000415708138579032,
|
|
"loss": 5.6347,
|
|
"mean_token_accuracy": 0.16741572916507722,
|
|
"num_tokens": 7331040.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 5.880664825439453,
|
|
"epoch": 3.466695315857327,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00041544853918683923,
|
|
"loss": 5.7161,
|
|
"mean_token_accuracy": 0.16357140839099885,
|
|
"num_tokens": 7340771.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 5.874464750289917,
|
|
"epoch": 3.4709926944563816,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004151886331111737,
|
|
"loss": 5.748,
|
|
"mean_token_accuracy": 0.16120901852846145,
|
|
"num_tokens": 7349960.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 5.928933048248291,
|
|
"epoch": 3.475290073055436,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00041492842091956646,
|
|
"loss": 5.601,
|
|
"mean_token_accuracy": 0.17824786752462388,
|
|
"num_tokens": 7357983.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 5.940189504623413,
|
|
"epoch": 3.4795874516544907,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004146679031802167,
|
|
"loss": 5.7001,
|
|
"mean_token_accuracy": 0.1747121438384056,
|
|
"num_tokens": 7366814.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 5.88372917175293,
|
|
"epoch": 3.4838848302535452,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00041440708046199123,
|
|
"loss": 5.5787,
|
|
"mean_token_accuracy": 0.17489822953939438,
|
|
"num_tokens": 7374773.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 5.857002401351929,
|
|
"epoch": 3.4881822088525998,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004141459533344226,
|
|
"loss": 5.6802,
|
|
"mean_token_accuracy": 0.16947068572044371,
|
|
"num_tokens": 7383937.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 5.892751169204712,
|
|
"epoch": 3.4924795874516543,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00041388452236770795,
|
|
"loss": 5.6426,
|
|
"mean_token_accuracy": 0.16677659898996353,
|
|
"num_tokens": 7392577.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 5.778850460052491,
|
|
"epoch": 3.4967769660507093,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 0.00041362278813270823,
|
|
"loss": 5.5193,
|
|
"mean_token_accuracy": 0.1853351354598999,
|
|
"num_tokens": 7401473.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 5.932539796829223,
|
|
"epoch": 3.501074344649764,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00041336075120094616,
|
|
"loss": 5.7326,
|
|
"mean_token_accuracy": 0.1593330055475235,
|
|
"num_tokens": 7410831.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 5.937400960922242,
|
|
"epoch": 3.5053717232488184,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00041309841214460586,
|
|
"loss": 5.7444,
|
|
"mean_token_accuracy": 0.16660909727215767,
|
|
"num_tokens": 7421563.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 5.8440385341644285,
|
|
"epoch": 3.509669101847873,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004128357715365309,
|
|
"loss": 5.6536,
|
|
"mean_token_accuracy": 0.17547903209924698,
|
|
"num_tokens": 7430174.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 5.864094829559326,
|
|
"epoch": 3.5139664804469275,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00041257282995022345,
|
|
"loss": 5.6395,
|
|
"mean_token_accuracy": 0.17667886912822722,
|
|
"num_tokens": 7439034.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 5.834180307388306,
|
|
"epoch": 3.518263859045982,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0004123095879598426,
|
|
"loss": 5.6481,
|
|
"mean_token_accuracy": 0.16979644894599916,
|
|
"num_tokens": 7447663.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 5.928986501693726,
|
|
"epoch": 3.5225612376450366,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00041204604614020397,
|
|
"loss": 5.7427,
|
|
"mean_token_accuracy": 0.16193264126777648,
|
|
"num_tokens": 7456615.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 5.9707435131072994,
|
|
"epoch": 3.526858616244091,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004117822050667773,
|
|
"loss": 5.7479,
|
|
"mean_token_accuracy": 0.16694117784500123,
|
|
"num_tokens": 7466203.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 5.906427669525146,
|
|
"epoch": 3.5311559948431457,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00041151806531568617,
|
|
"loss": 5.6938,
|
|
"mean_token_accuracy": 0.16936880797147752,
|
|
"num_tokens": 7475411.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 5.87329797744751,
|
|
"epoch": 3.5354533734422002,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00041125362746370625,
|
|
"loss": 5.7088,
|
|
"mean_token_accuracy": 0.16725610196590424,
|
|
"num_tokens": 7484965.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 5.954242277145386,
|
|
"epoch": 3.5397507520412548,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004109888920882639,
|
|
"loss": 5.6398,
|
|
"mean_token_accuracy": 0.17583244293928146,
|
|
"num_tokens": 7494240.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 5.8576123237609865,
|
|
"epoch": 3.5440481306403093,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004107238597674356,
|
|
"loss": 5.6831,
|
|
"mean_token_accuracy": 0.16786444336175918,
|
|
"num_tokens": 7503560.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 5.869033670425415,
|
|
"epoch": 3.548345509239364,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000410458531079946,
|
|
"loss": 5.6005,
|
|
"mean_token_accuracy": 0.17719124406576156,
|
|
"num_tokens": 7512650.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 5.9289051532745365,
|
|
"epoch": 3.5526428878384184,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004101929066051668,
|
|
"loss": 5.7036,
|
|
"mean_token_accuracy": 0.17097264677286148,
|
|
"num_tokens": 7521864.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 5.84982328414917,
|
|
"epoch": 3.556940266437473,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 0.0004099269869231157,
|
|
"loss": 5.6011,
|
|
"mean_token_accuracy": 0.17773334980010985,
|
|
"num_tokens": 7531013.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 5.842282152175903,
|
|
"epoch": 3.5612376450365275,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00040966077261445495,
|
|
"loss": 5.6317,
|
|
"mean_token_accuracy": 0.16675826907157898,
|
|
"num_tokens": 7539959.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 5.937725257873535,
|
|
"epoch": 3.565535023635582,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.0004093942642604904,
|
|
"loss": 5.5699,
|
|
"mean_token_accuracy": 0.17509743124246596,
|
|
"num_tokens": 7548354.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 5.933695459365845,
|
|
"epoch": 3.5698324022346366,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00040912746244316944,
|
|
"loss": 5.7411,
|
|
"mean_token_accuracy": 0.16647439002990722,
|
|
"num_tokens": 7558321.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 5.855962228775025,
|
|
"epoch": 3.5741297808336916,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00040886036774508095,
|
|
"loss": 5.5984,
|
|
"mean_token_accuracy": 0.17487595230340958,
|
|
"num_tokens": 7567889.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 5.892229986190796,
|
|
"epoch": 3.578427159432746,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0004085929807494527,
|
|
"loss": 5.6735,
|
|
"mean_token_accuracy": 0.1672454982995987,
|
|
"num_tokens": 7576752.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 5.8532696723937985,
|
|
"epoch": 3.5827245380318007,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0004083253020401512,
|
|
"loss": 5.5769,
|
|
"mean_token_accuracy": 0.18205645382404329,
|
|
"num_tokens": 7585413.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 5.851567888259888,
|
|
"epoch": 3.5870219166308552,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004080573322016797,
|
|
"loss": 5.5133,
|
|
"mean_token_accuracy": 0.18339248448610307,
|
|
"num_tokens": 7593966.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 5.8663169860839846,
|
|
"epoch": 3.59131929522991,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004077890718191773,
|
|
"loss": 5.5531,
|
|
"mean_token_accuracy": 0.17229120880365373,
|
|
"num_tokens": 7602746.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 5.859878444671631,
|
|
"epoch": 3.5956166738289643,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00040752052147841733,
|
|
"loss": 5.6103,
|
|
"mean_token_accuracy": 0.17114459723234177,
|
|
"num_tokens": 7611245.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 5.816100597381592,
|
|
"epoch": 3.599914052428019,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004072516817658065,
|
|
"loss": 5.6304,
|
|
"mean_token_accuracy": 0.17312704175710678,
|
|
"num_tokens": 7620234.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 5.882503318786621,
|
|
"epoch": 3.6042114310270734,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004069825532683831,
|
|
"loss": 5.6514,
|
|
"mean_token_accuracy": 0.1738997742533684,
|
|
"num_tokens": 7629794.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 5.836305379867554,
|
|
"epoch": 3.608508809626128,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00040671313657381645,
|
|
"loss": 5.5813,
|
|
"mean_token_accuracy": 0.18098288029432297,
|
|
"num_tokens": 7639497.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 5.750227785110473,
|
|
"epoch": 3.6128061882251825,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00040644343227040473,
|
|
"loss": 5.5416,
|
|
"mean_token_accuracy": 0.1755503609776497,
|
|
"num_tokens": 7647647.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 5.906509447097778,
|
|
"epoch": 3.617103566824237,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004061734409470745,
|
|
"loss": 5.7239,
|
|
"mean_token_accuracy": 0.17304449528455734,
|
|
"num_tokens": 7657988.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 5.948587131500244,
|
|
"epoch": 3.621400945423292,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004059031631933788,
|
|
"loss": 5.6442,
|
|
"mean_token_accuracy": 0.16922481805086137,
|
|
"num_tokens": 7667498.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 5.8439648151397705,
|
|
"epoch": 3.6256983240223466,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00040563259959949615,
|
|
"loss": 5.7762,
|
|
"mean_token_accuracy": 0.161246457695961,
|
|
"num_tokens": 7677386.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 5.95087080001831,
|
|
"epoch": 3.629995702621401,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004053617507562295,
|
|
"loss": 5.612,
|
|
"mean_token_accuracy": 0.173043355345726,
|
|
"num_tokens": 7686643.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 5.895866870880127,
|
|
"epoch": 3.6342930812204557,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.00040509061725500426,
|
|
"loss": 5.6337,
|
|
"mean_token_accuracy": 0.1729728639125824,
|
|
"num_tokens": 7695089.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 5.811243104934692,
|
|
"epoch": 3.6385904598195102,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004048191996878677,
|
|
"loss": 5.6375,
|
|
"mean_token_accuracy": 0.17418366819620132,
|
|
"num_tokens": 7703854.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 5.8532263278961185,
|
|
"epoch": 3.642887838418565,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00040454749864748734,
|
|
"loss": 5.5804,
|
|
"mean_token_accuracy": 0.17380526959896087,
|
|
"num_tokens": 7712903.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 5.82907280921936,
|
|
"epoch": 3.6471852170176193,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004042755147271496,
|
|
"loss": 5.5323,
|
|
"mean_token_accuracy": 0.17679498344659805,
|
|
"num_tokens": 7721701.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 5.705156469345093,
|
|
"epoch": 3.651482595616674,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004040032485207587,
|
|
"loss": 5.6347,
|
|
"mean_token_accuracy": 0.1726035177707672,
|
|
"num_tokens": 7731318.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 5.885972023010254,
|
|
"epoch": 3.6557799742157284,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004037307006228352,
|
|
"loss": 5.5643,
|
|
"mean_token_accuracy": 0.17485362589359282,
|
|
"num_tokens": 7740413.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 5.8497552394866945,
|
|
"epoch": 3.660077352814783,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004034578716285147,
|
|
"loss": 5.5409,
|
|
"mean_token_accuracy": 0.18239518254995346,
|
|
"num_tokens": 7749054.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 5.804295063018799,
|
|
"epoch": 3.6643747314138375,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004031847621335467,
|
|
"loss": 5.5851,
|
|
"mean_token_accuracy": 0.1809212550520897,
|
|
"num_tokens": 7757366.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 5.8172869205474855,
|
|
"epoch": 3.668672110012892,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004029113727342933,
|
|
"loss": 5.6087,
|
|
"mean_token_accuracy": 0.17429128736257554,
|
|
"num_tokens": 7766471.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 5.789394521713257,
|
|
"epoch": 3.6729694886119466,
|
|
"grad_norm": 2.03125,
|
|
"learning_rate": 0.00040263770402772746,
|
|
"loss": 5.5994,
|
|
"mean_token_accuracy": 0.17218794971704482,
|
|
"num_tokens": 7775920.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 5.851673603057861,
|
|
"epoch": 3.677266867211001,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0004023637566114325,
|
|
"loss": 5.6566,
|
|
"mean_token_accuracy": 0.1731671839952469,
|
|
"num_tokens": 7784530.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 5.892326545715332,
|
|
"epoch": 3.6815642458100557,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004020895310835999,
|
|
"loss": 5.5783,
|
|
"mean_token_accuracy": 0.1740590050816536,
|
|
"num_tokens": 7793656.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 5.841686058044433,
|
|
"epoch": 3.6858616244091102,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00040181502804302865,
|
|
"loss": 5.6079,
|
|
"mean_token_accuracy": 0.1703685387969017,
|
|
"num_tokens": 7802185.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 5.803410863876342,
|
|
"epoch": 3.690159003008165,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00040154024808912377,
|
|
"loss": 5.6087,
|
|
"mean_token_accuracy": 0.1748747169971466,
|
|
"num_tokens": 7810345.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 5.848033142089844,
|
|
"epoch": 3.6944563816072193,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0004012651918218947,
|
|
"loss": 5.6695,
|
|
"mean_token_accuracy": 0.1688275918364525,
|
|
"num_tokens": 7818998.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 5.8963113784790036,
|
|
"epoch": 3.6987537602062743,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004009898598419544,
|
|
"loss": 5.7696,
|
|
"mean_token_accuracy": 0.15907692313194274,
|
|
"num_tokens": 7828638.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 5.872035169601441,
|
|
"epoch": 3.703051138805329,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000400714252750518,
|
|
"loss": 5.7285,
|
|
"mean_token_accuracy": 0.16805977523326873,
|
|
"num_tokens": 7838812.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 5.981037616729736,
|
|
"epoch": 3.7073485174043834,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004004383711494011,
|
|
"loss": 5.6447,
|
|
"mean_token_accuracy": 0.17727971971035003,
|
|
"num_tokens": 7847458.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 5.9139772891998295,
|
|
"epoch": 3.711645896003438,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004001622156410189,
|
|
"loss": 5.6829,
|
|
"mean_token_accuracy": 0.16504186540842056,
|
|
"num_tokens": 7856553.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 5.786783981323242,
|
|
"epoch": 3.7159432746024925,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00039988578682838467,
|
|
"loss": 5.6261,
|
|
"mean_token_accuracy": 0.17588848173618316,
|
|
"num_tokens": 7864788.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 5.83101077079773,
|
|
"epoch": 3.720240653201547,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00039960908531510843,
|
|
"loss": 5.5906,
|
|
"mean_token_accuracy": 0.1767484351992607,
|
|
"num_tokens": 7873850.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 5.906121778488159,
|
|
"epoch": 3.7245380318006016,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003993321117053956,
|
|
"loss": 5.7387,
|
|
"mean_token_accuracy": 0.16848002076148988,
|
|
"num_tokens": 7882775.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 5.968108892440796,
|
|
"epoch": 3.728835410399656,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00039905486660404604,
|
|
"loss": 5.6602,
|
|
"mean_token_accuracy": 0.16529427766799926,
|
|
"num_tokens": 7890570.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 5.820059299468994,
|
|
"epoch": 3.7331327889987107,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00039877735061645206,
|
|
"loss": 5.6341,
|
|
"mean_token_accuracy": 0.18253806978464127,
|
|
"num_tokens": 7900090.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 5.87975811958313,
|
|
"epoch": 3.7374301675977653,
|
|
"grad_norm": 4.6875,
|
|
"learning_rate": 0.0003984995643485977,
|
|
"loss": 5.6514,
|
|
"mean_token_accuracy": 0.1710534170269966,
|
|
"num_tokens": 7908077.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 5.919925594329834,
|
|
"epoch": 3.74172754619682,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.00039822150840705716,
|
|
"loss": 5.6638,
|
|
"mean_token_accuracy": 0.1758440315723419,
|
|
"num_tokens": 7916290.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 5.9543579578399655,
|
|
"epoch": 3.746024924795875,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00039794318339899347,
|
|
"loss": 5.7413,
|
|
"mean_token_accuracy": 0.1703621730208397,
|
|
"num_tokens": 7925835.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 5.906834125518799,
|
|
"epoch": 3.7503223033949293,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00039766458993215726,
|
|
"loss": 5.6931,
|
|
"mean_token_accuracy": 0.17016691565513611,
|
|
"num_tokens": 7935076.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 5.862006330490113,
|
|
"epoch": 3.754619681993984,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00039738572861488527,
|
|
"loss": 5.615,
|
|
"mean_token_accuracy": 0.17408015877008437,
|
|
"num_tokens": 7943958.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 5.828081035614014,
|
|
"epoch": 3.7589170605930384,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000397106600056099,
|
|
"loss": 5.6516,
|
|
"mean_token_accuracy": 0.17019706666469575,
|
|
"num_tokens": 7953189.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 5.807856559753418,
|
|
"epoch": 3.763214439192093,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003968272048653039,
|
|
"loss": 5.5611,
|
|
"mean_token_accuracy": 0.18258637934923172,
|
|
"num_tokens": 7962927.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 5.772855567932129,
|
|
"epoch": 3.7675118177911475,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003965475436525873,
|
|
"loss": 5.5854,
|
|
"mean_token_accuracy": 0.17843964099884033,
|
|
"num_tokens": 7973087.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 5.789877653121948,
|
|
"epoch": 3.771809196390202,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003962676170286174,
|
|
"loss": 5.5481,
|
|
"mean_token_accuracy": 0.17716174721717834,
|
|
"num_tokens": 7982535.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 5.850610685348511,
|
|
"epoch": 3.7761065749892566,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00039598742560464223,
|
|
"loss": 5.6337,
|
|
"mean_token_accuracy": 0.18029867857694626,
|
|
"num_tokens": 7990740.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 5.864075326919556,
|
|
"epoch": 3.780403953588311,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0003957069699924877,
|
|
"loss": 5.61,
|
|
"mean_token_accuracy": 0.16865527033805847,
|
|
"num_tokens": 7999349.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 5.886800861358642,
|
|
"epoch": 3.7847013321873657,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000395426250804557,
|
|
"loss": 5.6337,
|
|
"mean_token_accuracy": 0.18143983781337739,
|
|
"num_tokens": 8007615.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 5.838603496551514,
|
|
"epoch": 3.7889987107864203,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00039514526865382847,
|
|
"loss": 5.6228,
|
|
"mean_token_accuracy": 0.17375304400920868,
|
|
"num_tokens": 8017545.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 5.86598572731018,
|
|
"epoch": 3.793296089385475,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003948640241538548,
|
|
"loss": 5.567,
|
|
"mean_token_accuracy": 0.17485908418893814,
|
|
"num_tokens": 8026381.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 5.860766315460205,
|
|
"epoch": 3.7975934679845293,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0003945825179187617,
|
|
"loss": 5.6728,
|
|
"mean_token_accuracy": 0.1720847576856613,
|
|
"num_tokens": 8034745.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 5.890695714950562,
|
|
"epoch": 3.801890846583584,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.00039430075056324604,
|
|
"loss": 5.6121,
|
|
"mean_token_accuracy": 0.18080423325300216,
|
|
"num_tokens": 8043995.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 5.954921388626099,
|
|
"epoch": 3.8061882251826384,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00039401872270257546,
|
|
"loss": 5.7048,
|
|
"mean_token_accuracy": 0.168908504396677,
|
|
"num_tokens": 8053059.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 5.893271398544312,
|
|
"epoch": 3.810485603781693,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00039373643495258567,
|
|
"loss": 5.7191,
|
|
"mean_token_accuracy": 0.17044830471277236,
|
|
"num_tokens": 8062160.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 5.778645277023315,
|
|
"epoch": 3.8147829823807475,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.00039345388792968056,
|
|
"loss": 5.6098,
|
|
"mean_token_accuracy": 0.1839832752943039,
|
|
"num_tokens": 8071260.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 5.94516019821167,
|
|
"epoch": 3.819080360979802,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00039317108225082984,
|
|
"loss": 5.7379,
|
|
"mean_token_accuracy": 0.1626624658703804,
|
|
"num_tokens": 8081540.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 5.9059412479400635,
|
|
"epoch": 3.8233777395788566,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00039288801853356806,
|
|
"loss": 5.6955,
|
|
"mean_token_accuracy": 0.17502384185791015,
|
|
"num_tokens": 8089785.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 5.8967235565185545,
|
|
"epoch": 3.8276751181779116,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0003926046973959932,
|
|
"loss": 5.5368,
|
|
"mean_token_accuracy": 0.17684618532657623,
|
|
"num_tokens": 8098097.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 5.817251539230346,
|
|
"epoch": 3.831972496776966,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0003923211194567654,
|
|
"loss": 5.7853,
|
|
"mean_token_accuracy": 0.16855504512786865,
|
|
"num_tokens": 8108693.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 5.936377763748169,
|
|
"epoch": 3.8362698753760207,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00039203728533510556,
|
|
"loss": 5.6202,
|
|
"mean_token_accuracy": 0.17411938160657883,
|
|
"num_tokens": 8117181.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 5.883836030960083,
|
|
"epoch": 3.8405672539750753,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.000391753195650794,
|
|
"loss": 5.6274,
|
|
"mean_token_accuracy": 0.17058600783348082,
|
|
"num_tokens": 8125398.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 5.868739318847656,
|
|
"epoch": 3.84486463257413,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 0.00039146885102416895,
|
|
"loss": 5.645,
|
|
"mean_token_accuracy": 0.17607929259538652,
|
|
"num_tokens": 8135320.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 5.881189346313477,
|
|
"epoch": 3.8491620111731844,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 0.00039118425207612553,
|
|
"loss": 5.703,
|
|
"mean_token_accuracy": 0.17280273139476776,
|
|
"num_tokens": 8144320.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 5.780130767822266,
|
|
"epoch": 3.853459389772239,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00039089939942811396,
|
|
"loss": 5.5951,
|
|
"mean_token_accuracy": 0.1795727789402008,
|
|
"num_tokens": 8153653.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 5.929383611679077,
|
|
"epoch": 3.8577567683712934,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00039061429370213863,
|
|
"loss": 5.6336,
|
|
"mean_token_accuracy": 0.17494203001260758,
|
|
"num_tokens": 8162741.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 5.822046375274658,
|
|
"epoch": 3.862054146970348,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00039032893552075646,
|
|
"loss": 5.5311,
|
|
"mean_token_accuracy": 0.1902599960565567,
|
|
"num_tokens": 8171078.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 5.801896953582764,
|
|
"epoch": 3.8663515255694025,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003900433255070758,
|
|
"loss": 5.62,
|
|
"mean_token_accuracy": 0.18065827935934067,
|
|
"num_tokens": 8179968.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"epoch": 3.8663515255694025,
|
|
"eval_entropy": 5.773520280648996,
|
|
"eval_loss": 6.096230983734131,
|
|
"eval_mean_token_accuracy": 0.1635508523700205,
|
|
"eval_num_tokens": 8179968.0,
|
|
"eval_runtime": 2.8246,
|
|
"eval_samples_per_second": 1256.476,
|
|
"eval_steps_per_second": 157.192,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"entropy": 5.866156673431396,
|
|
"epoch": 3.870648904168457,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00038975746428475454,
|
|
"loss": 5.5754,
|
|
"mean_token_accuracy": 0.17498472034931184,
|
|
"num_tokens": 8189261.0,
|
|
"step": 4505
|
|
},
|
|
{
|
|
"entropy": 5.900315570831299,
|
|
"epoch": 3.874946282767512,
|
|
"grad_norm": 2.25,
|
|
"learning_rate": 0.00038947135247799955,
|
|
"loss": 5.5931,
|
|
"mean_token_accuracy": 0.178064002096653,
|
|
"num_tokens": 8198302.0,
|
|
"step": 4510
|
|
},
|
|
{
|
|
"entropy": 5.850727176666259,
|
|
"epoch": 3.8792436613665666,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00038918499071156443,
|
|
"loss": 5.5906,
|
|
"mean_token_accuracy": 0.1818610206246376,
|
|
"num_tokens": 8207098.0,
|
|
"step": 4515
|
|
},
|
|
{
|
|
"entropy": 5.925307655334473,
|
|
"epoch": 3.883541039965621,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000388898379610749,
|
|
"loss": 5.6397,
|
|
"mean_token_accuracy": 0.17176988124847412,
|
|
"num_tokens": 8216831.0,
|
|
"step": 4520
|
|
},
|
|
{
|
|
"entropy": 5.824055910110474,
|
|
"epoch": 3.8878384185646757,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003886115198013973,
|
|
"loss": 5.639,
|
|
"mean_token_accuracy": 0.1793515294790268,
|
|
"num_tokens": 8225369.0,
|
|
"step": 4525
|
|
},
|
|
{
|
|
"entropy": 5.814682149887085,
|
|
"epoch": 3.8921357971637303,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.0003883244119098965,
|
|
"loss": 5.766,
|
|
"mean_token_accuracy": 0.15825398564338683,
|
|
"num_tokens": 8234440.0,
|
|
"step": 4530
|
|
},
|
|
{
|
|
"entropy": 5.864934873580933,
|
|
"epoch": 3.896433175762785,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0003880370565631754,
|
|
"loss": 5.5592,
|
|
"mean_token_accuracy": 0.1808269739151001,
|
|
"num_tokens": 8243707.0,
|
|
"step": 4535
|
|
},
|
|
{
|
|
"entropy": 5.894186115264892,
|
|
"epoch": 3.9007305543618394,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00038774945438870337,
|
|
"loss": 5.7209,
|
|
"mean_token_accuracy": 0.1657774493098259,
|
|
"num_tokens": 8254223.0,
|
|
"step": 4540
|
|
},
|
|
{
|
|
"entropy": 5.8490087509155275,
|
|
"epoch": 3.905027932960894,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00038746160601448845,
|
|
"loss": 5.5653,
|
|
"mean_token_accuracy": 0.1781230702996254,
|
|
"num_tokens": 8263105.0,
|
|
"step": 4545
|
|
},
|
|
{
|
|
"entropy": 5.8097310066223145,
|
|
"epoch": 3.9093253115599484,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003871735120690766,
|
|
"loss": 5.641,
|
|
"mean_token_accuracy": 0.178280271589756,
|
|
"num_tokens": 8271478.0,
|
|
"step": 4550
|
|
},
|
|
{
|
|
"entropy": 5.9077965259552006,
|
|
"epoch": 3.913622690159003,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003868851731815497,
|
|
"loss": 5.6698,
|
|
"mean_token_accuracy": 0.1717348054051399,
|
|
"num_tokens": 8280396.0,
|
|
"step": 4555
|
|
},
|
|
{
|
|
"entropy": 5.912223625183105,
|
|
"epoch": 3.9179200687580575,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0003865965899815247,
|
|
"loss": 5.6713,
|
|
"mean_token_accuracy": 0.17081275582313538,
|
|
"num_tokens": 8290371.0,
|
|
"step": 4560
|
|
},
|
|
{
|
|
"entropy": 5.809206581115722,
|
|
"epoch": 3.922217447357112,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0003863077630991518,
|
|
"loss": 5.569,
|
|
"mean_token_accuracy": 0.18193934857845306,
|
|
"num_tokens": 8298976.0,
|
|
"step": 4565
|
|
},
|
|
{
|
|
"entropy": 5.8157744884490965,
|
|
"epoch": 3.9265148259561666,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003860186931651139,
|
|
"loss": 5.6118,
|
|
"mean_token_accuracy": 0.1693265035748482,
|
|
"num_tokens": 8308752.0,
|
|
"step": 4570
|
|
},
|
|
{
|
|
"entropy": 5.849201774597168,
|
|
"epoch": 3.930812204555221,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003857293808106238,
|
|
"loss": 5.6911,
|
|
"mean_token_accuracy": 0.17183010876178742,
|
|
"num_tokens": 8317343.0,
|
|
"step": 4575
|
|
},
|
|
{
|
|
"entropy": 5.866435098648071,
|
|
"epoch": 3.9351095831542757,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003854398266674241,
|
|
"loss": 5.5285,
|
|
"mean_token_accuracy": 0.18550786823034288,
|
|
"num_tokens": 8326956.0,
|
|
"step": 4580
|
|
},
|
|
{
|
|
"entropy": 5.833143281936645,
|
|
"epoch": 3.9394069617533303,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00038515003136778544,
|
|
"loss": 5.6607,
|
|
"mean_token_accuracy": 0.1690329134464264,
|
|
"num_tokens": 8335589.0,
|
|
"step": 4585
|
|
},
|
|
{
|
|
"entropy": 5.821763515472412,
|
|
"epoch": 3.943704340352385,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00038485999554450483,
|
|
"loss": 5.6247,
|
|
"mean_token_accuracy": 0.1734362944960594,
|
|
"num_tokens": 8345517.0,
|
|
"step": 4590
|
|
},
|
|
{
|
|
"entropy": 5.73569917678833,
|
|
"epoch": 3.9480017189514394,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00038456971983090454,
|
|
"loss": 5.5589,
|
|
"mean_token_accuracy": 0.18558432757854462,
|
|
"num_tokens": 8354702.0,
|
|
"step": 4595
|
|
},
|
|
{
|
|
"entropy": 5.8099524021148685,
|
|
"epoch": 3.9522990975504944,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0003842792048608309,
|
|
"loss": 5.5985,
|
|
"mean_token_accuracy": 0.18032405376434327,
|
|
"num_tokens": 8362940.0,
|
|
"step": 4600
|
|
},
|
|
{
|
|
"entropy": 5.925025367736817,
|
|
"epoch": 3.956596476149549,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.0003839884512686523,
|
|
"loss": 5.6452,
|
|
"mean_token_accuracy": 0.17352996915578842,
|
|
"num_tokens": 8372034.0,
|
|
"step": 4605
|
|
},
|
|
{
|
|
"entropy": 5.856717681884765,
|
|
"epoch": 3.9608938547486034,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00038369745968925846,
|
|
"loss": 5.6574,
|
|
"mean_token_accuracy": 0.17035734802484512,
|
|
"num_tokens": 8381673.0,
|
|
"step": 4610
|
|
},
|
|
{
|
|
"entropy": 5.88683295249939,
|
|
"epoch": 3.965191233347658,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00038340623075805875,
|
|
"loss": 5.6109,
|
|
"mean_token_accuracy": 0.17607690244913102,
|
|
"num_tokens": 8390804.0,
|
|
"step": 4615
|
|
},
|
|
{
|
|
"entropy": 5.857405328750611,
|
|
"epoch": 3.9694886119467125,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00038311476511098053,
|
|
"loss": 5.6583,
|
|
"mean_token_accuracy": 0.17764031887054443,
|
|
"num_tokens": 8399644.0,
|
|
"step": 4620
|
|
},
|
|
{
|
|
"entropy": 5.846802282333374,
|
|
"epoch": 3.973785990545767,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0003828230633844685,
|
|
"loss": 5.6701,
|
|
"mean_token_accuracy": 0.17315337657928467,
|
|
"num_tokens": 8409264.0,
|
|
"step": 4625
|
|
},
|
|
{
|
|
"entropy": 5.8850417137146,
|
|
"epoch": 3.9780833691448216,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00038253112621548243,
|
|
"loss": 5.6275,
|
|
"mean_token_accuracy": 0.16971271783113479,
|
|
"num_tokens": 8418383.0,
|
|
"step": 4630
|
|
},
|
|
{
|
|
"entropy": 5.8646105289459225,
|
|
"epoch": 3.982380747743876,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003822389542414966,
|
|
"loss": 5.6257,
|
|
"mean_token_accuracy": 0.17218141853809357,
|
|
"num_tokens": 8427411.0,
|
|
"step": 4635
|
|
},
|
|
{
|
|
"entropy": 5.8065088272094725,
|
|
"epoch": 3.9866781263429307,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00038194654810049775,
|
|
"loss": 5.5887,
|
|
"mean_token_accuracy": 0.1758611410856247,
|
|
"num_tokens": 8435537.0,
|
|
"step": 4640
|
|
},
|
|
{
|
|
"entropy": 5.842695474624634,
|
|
"epoch": 3.9909755049419853,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000381653908430984,
|
|
"loss": 5.6647,
|
|
"mean_token_accuracy": 0.16946116387844085,
|
|
"num_tokens": 8444400.0,
|
|
"step": 4645
|
|
},
|
|
{
|
|
"entropy": 5.9148087978363035,
|
|
"epoch": 3.99527288354104,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003813610358719634,
|
|
"loss": 5.6587,
|
|
"mean_token_accuracy": 0.170665642619133,
|
|
"num_tokens": 8453830.0,
|
|
"step": 4650
|
|
},
|
|
{
|
|
"entropy": 5.816837453842163,
|
|
"epoch": 3.999570262140095,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00038106793106295266,
|
|
"loss": 5.5942,
|
|
"mean_token_accuracy": 0.18440621644258498,
|
|
"num_tokens": 8463033.0,
|
|
"step": 4655
|
|
},
|
|
{
|
|
"entropy": 5.7983916070726185,
|
|
"epoch": 4.003437902879243,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003807745946439754,
|
|
"loss": 5.3975,
|
|
"mean_token_accuracy": 0.18456794652673933,
|
|
"num_tokens": 8470740.0,
|
|
"step": 4660
|
|
},
|
|
{
|
|
"entropy": 5.842042922973633,
|
|
"epoch": 4.007735281478298,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003804810272555612,
|
|
"loss": 5.392,
|
|
"mean_token_accuracy": 0.18096994161605834,
|
|
"num_tokens": 8480480.0,
|
|
"step": 4665
|
|
},
|
|
{
|
|
"entropy": 5.791851234436035,
|
|
"epoch": 4.012032660077352,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0003801872295387439,
|
|
"loss": 5.3486,
|
|
"mean_token_accuracy": 0.18860224783420562,
|
|
"num_tokens": 8489047.0,
|
|
"step": 4670
|
|
},
|
|
{
|
|
"entropy": 5.842134189605713,
|
|
"epoch": 4.016330038676408,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0003798932021350603,
|
|
"loss": 5.4415,
|
|
"mean_token_accuracy": 0.18516335934400557,
|
|
"num_tokens": 8497763.0,
|
|
"step": 4675
|
|
},
|
|
{
|
|
"entropy": 5.814905071258545,
|
|
"epoch": 4.020627417275462,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00037959894568654864,
|
|
"loss": 5.4218,
|
|
"mean_token_accuracy": 0.18434918075799941,
|
|
"num_tokens": 8506814.0,
|
|
"step": 4680
|
|
},
|
|
{
|
|
"entropy": 5.917452049255371,
|
|
"epoch": 4.024924795874517,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 0.0003793044608357474,
|
|
"loss": 5.5227,
|
|
"mean_token_accuracy": 0.1723128989338875,
|
|
"num_tokens": 8516384.0,
|
|
"step": 4685
|
|
},
|
|
{
|
|
"entropy": 5.8502603530883786,
|
|
"epoch": 4.0292221744735714,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 0.0003790097482256939,
|
|
"loss": 5.3729,
|
|
"mean_token_accuracy": 0.1855533540248871,
|
|
"num_tokens": 8524822.0,
|
|
"step": 4690
|
|
},
|
|
{
|
|
"entropy": 5.837485504150391,
|
|
"epoch": 4.033519553072626,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003787148084999225,
|
|
"loss": 5.4009,
|
|
"mean_token_accuracy": 0.1880905494093895,
|
|
"num_tokens": 8534129.0,
|
|
"step": 4695
|
|
},
|
|
{
|
|
"entropy": 5.810588788986206,
|
|
"epoch": 4.0378169316716805,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00037841964230246394,
|
|
"loss": 5.4689,
|
|
"mean_token_accuracy": 0.17597413510084153,
|
|
"num_tokens": 8543235.0,
|
|
"step": 4700
|
|
},
|
|
{
|
|
"entropy": 5.764852046966553,
|
|
"epoch": 4.042114310270735,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0003781242502778429,
|
|
"loss": 5.336,
|
|
"mean_token_accuracy": 0.19322870671749115,
|
|
"num_tokens": 8551903.0,
|
|
"step": 4705
|
|
},
|
|
{
|
|
"entropy": 5.792543745040893,
|
|
"epoch": 4.04641168886979,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00037782863307107785,
|
|
"loss": 5.4435,
|
|
"mean_token_accuracy": 0.1793735593557358,
|
|
"num_tokens": 8561173.0,
|
|
"step": 4710
|
|
},
|
|
{
|
|
"entropy": 5.854628419876098,
|
|
"epoch": 4.050709067468844,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00037753279132767833,
|
|
"loss": 5.3401,
|
|
"mean_token_accuracy": 0.1945023715496063,
|
|
"num_tokens": 8569789.0,
|
|
"step": 4715
|
|
},
|
|
{
|
|
"entropy": 5.788295412063599,
|
|
"epoch": 4.055006446067899,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.00037723672569364453,
|
|
"loss": 5.3443,
|
|
"mean_token_accuracy": 0.19259989261627197,
|
|
"num_tokens": 8577971.0,
|
|
"step": 4720
|
|
},
|
|
{
|
|
"entropy": 5.8331249237060545,
|
|
"epoch": 4.059303824666953,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00037694043681546545,
|
|
"loss": 5.4189,
|
|
"mean_token_accuracy": 0.1816018119454384,
|
|
"num_tokens": 8587299.0,
|
|
"step": 4725
|
|
},
|
|
{
|
|
"entropy": 5.771479558944702,
|
|
"epoch": 4.063601203266008,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.0003766439253401177,
|
|
"loss": 5.3947,
|
|
"mean_token_accuracy": 0.18417158722877502,
|
|
"num_tokens": 8595813.0,
|
|
"step": 4730
|
|
},
|
|
{
|
|
"entropy": 5.7723979473114015,
|
|
"epoch": 4.067898581865062,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00037634719191506367,
|
|
"loss": 5.3985,
|
|
"mean_token_accuracy": 0.19036534875631334,
|
|
"num_tokens": 8604552.0,
|
|
"step": 4735
|
|
},
|
|
{
|
|
"entropy": 5.755882120132446,
|
|
"epoch": 4.072195960464117,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.00037605023718825065,
|
|
"loss": 5.3484,
|
|
"mean_token_accuracy": 0.18971681445837021,
|
|
"num_tokens": 8612701.0,
|
|
"step": 4740
|
|
},
|
|
{
|
|
"entropy": 5.840817975997925,
|
|
"epoch": 4.0764933390631715,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.000375753061808109,
|
|
"loss": 5.4096,
|
|
"mean_token_accuracy": 0.18549485504627228,
|
|
"num_tokens": 8622699.0,
|
|
"step": 4745
|
|
},
|
|
{
|
|
"entropy": 5.775212574005127,
|
|
"epoch": 4.080790717662226,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 0.00037545566642355107,
|
|
"loss": 5.3935,
|
|
"mean_token_accuracy": 0.179714597761631,
|
|
"num_tokens": 8631821.0,
|
|
"step": 4750
|
|
},
|
|
{
|
|
"entropy": 5.790747880935669,
|
|
"epoch": 4.0850880962612806,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0003751580516839695,
|
|
"loss": 5.3569,
|
|
"mean_token_accuracy": 0.18772245794534684,
|
|
"num_tokens": 8641814.0,
|
|
"step": 4755
|
|
},
|
|
{
|
|
"entropy": 5.849743461608886,
|
|
"epoch": 4.089385474860335,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00037486021823923574,
|
|
"loss": 5.4293,
|
|
"mean_token_accuracy": 0.18185275048017502,
|
|
"num_tokens": 8649649.0,
|
|
"step": 4760
|
|
},
|
|
{
|
|
"entropy": 5.728197622299194,
|
|
"epoch": 4.09368285345939,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00037456216673969925,
|
|
"loss": 5.3746,
|
|
"mean_token_accuracy": 0.1894873335957527,
|
|
"num_tokens": 8658216.0,
|
|
"step": 4765
|
|
},
|
|
{
|
|
"entropy": 5.818626546859742,
|
|
"epoch": 4.097980232058444,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003742638978361851,
|
|
"loss": 5.4489,
|
|
"mean_token_accuracy": 0.1774630144238472,
|
|
"num_tokens": 8667725.0,
|
|
"step": 4770
|
|
},
|
|
{
|
|
"entropy": 5.75709171295166,
|
|
"epoch": 4.102277610657499,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00037396541217999367,
|
|
"loss": 5.3083,
|
|
"mean_token_accuracy": 0.18892404288053513,
|
|
"num_tokens": 8675739.0,
|
|
"step": 4775
|
|
},
|
|
{
|
|
"entropy": 5.773554944992066,
|
|
"epoch": 4.106574989256553,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0003736667104228981,
|
|
"loss": 5.3761,
|
|
"mean_token_accuracy": 0.18949810564517974,
|
|
"num_tokens": 8685764.0,
|
|
"step": 4780
|
|
},
|
|
{
|
|
"entropy": 5.820678567886352,
|
|
"epoch": 4.110872367855608,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00037336779321714376,
|
|
"loss": 5.3569,
|
|
"mean_token_accuracy": 0.19179614782333373,
|
|
"num_tokens": 8695476.0,
|
|
"step": 4785
|
|
},
|
|
{
|
|
"entropy": 5.7612114429473875,
|
|
"epoch": 4.115169746454662,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00037306866121544633,
|
|
"loss": 5.4082,
|
|
"mean_token_accuracy": 0.18679875284433364,
|
|
"num_tokens": 8705544.0,
|
|
"step": 4790
|
|
},
|
|
{
|
|
"entropy": 5.827223634719848,
|
|
"epoch": 4.119467125053717,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0003727693150709904,
|
|
"loss": 5.4121,
|
|
"mean_token_accuracy": 0.1911407306790352,
|
|
"num_tokens": 8714883.0,
|
|
"step": 4795
|
|
},
|
|
{
|
|
"entropy": 5.845416450500489,
|
|
"epoch": 4.1237645036527715,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00037246975543742843,
|
|
"loss": 5.4624,
|
|
"mean_token_accuracy": 0.1837080627679825,
|
|
"num_tokens": 8724589.0,
|
|
"step": 4800
|
|
},
|
|
{
|
|
"entropy": 5.715303039550781,
|
|
"epoch": 4.128061882251826,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000372169982968879,
|
|
"loss": 5.3371,
|
|
"mean_token_accuracy": 0.18826118409633635,
|
|
"num_tokens": 8733771.0,
|
|
"step": 4805
|
|
},
|
|
{
|
|
"entropy": 5.795674514770508,
|
|
"epoch": 4.132359260850881,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0003718699983199252,
|
|
"loss": 5.4181,
|
|
"mean_token_accuracy": 0.18492107689380646,
|
|
"num_tokens": 8742348.0,
|
|
"step": 4810
|
|
},
|
|
{
|
|
"entropy": 5.724571228027344,
|
|
"epoch": 4.136656639449935,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.0003715698021456137,
|
|
"loss": 5.3568,
|
|
"mean_token_accuracy": 0.19371217340230942,
|
|
"num_tokens": 8751357.0,
|
|
"step": 4815
|
|
},
|
|
{
|
|
"entropy": 5.782418298721313,
|
|
"epoch": 4.1409540180489905,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00037126939510145294,
|
|
"loss": 5.4045,
|
|
"mean_token_accuracy": 0.19003892689943314,
|
|
"num_tokens": 8760813.0,
|
|
"step": 4820
|
|
},
|
|
{
|
|
"entropy": 5.834395456314087,
|
|
"epoch": 4.145251396648045,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0003709687778434118,
|
|
"loss": 5.4657,
|
|
"mean_token_accuracy": 0.17690672278404235,
|
|
"num_tokens": 8770228.0,
|
|
"step": 4825
|
|
},
|
|
{
|
|
"entropy": 5.690515184402466,
|
|
"epoch": 4.1495487752471,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0003706679510279183,
|
|
"loss": 5.2961,
|
|
"mean_token_accuracy": 0.18667328953742982,
|
|
"num_tokens": 8779351.0,
|
|
"step": 4830
|
|
},
|
|
{
|
|
"entropy": 5.77583532333374,
|
|
"epoch": 4.153846153846154,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 0.0003703669153118578,
|
|
"loss": 5.4439,
|
|
"mean_token_accuracy": 0.1770554691553116,
|
|
"num_tokens": 8789116.0,
|
|
"step": 4835
|
|
},
|
|
{
|
|
"entropy": 5.755214405059815,
|
|
"epoch": 4.158143532445209,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00037006567135257216,
|
|
"loss": 5.426,
|
|
"mean_token_accuracy": 0.18286389261484146,
|
|
"num_tokens": 8797790.0,
|
|
"step": 4840
|
|
},
|
|
{
|
|
"entropy": 5.8320070743560795,
|
|
"epoch": 4.162440911044263,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00036976421980785764,
|
|
"loss": 5.4731,
|
|
"mean_token_accuracy": 0.1745737001299858,
|
|
"num_tokens": 8808067.0,
|
|
"step": 4845
|
|
},
|
|
{
|
|
"entropy": 5.744848108291626,
|
|
"epoch": 4.166738289643318,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0003694625613359641,
|
|
"loss": 5.361,
|
|
"mean_token_accuracy": 0.19239065945148467,
|
|
"num_tokens": 8816587.0,
|
|
"step": 4850
|
|
},
|
|
{
|
|
"entropy": 5.793048477172851,
|
|
"epoch": 4.171035668242372,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.0003691606965955929,
|
|
"loss": 5.4467,
|
|
"mean_token_accuracy": 0.18077430874109268,
|
|
"num_tokens": 8826045.0,
|
|
"step": 4855
|
|
},
|
|
{
|
|
"entropy": 5.735131502151489,
|
|
"epoch": 4.175333046841427,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000368858626245896,
|
|
"loss": 5.4116,
|
|
"mean_token_accuracy": 0.18889937847852706,
|
|
"num_tokens": 8835427.0,
|
|
"step": 4860
|
|
},
|
|
{
|
|
"entropy": 5.738683366775513,
|
|
"epoch": 4.1796304254404815,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003685563509464744,
|
|
"loss": 5.3581,
|
|
"mean_token_accuracy": 0.1888204649090767,
|
|
"num_tokens": 8845167.0,
|
|
"step": 4865
|
|
},
|
|
{
|
|
"entropy": 5.839661455154419,
|
|
"epoch": 4.183927804039536,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00036825387135737647,
|
|
"loss": 5.3764,
|
|
"mean_token_accuracy": 0.18816280514001846,
|
|
"num_tokens": 8853591.0,
|
|
"step": 4870
|
|
},
|
|
{
|
|
"entropy": 5.801888132095337,
|
|
"epoch": 4.188225182638591,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00036795118813909674,
|
|
"loss": 5.4729,
|
|
"mean_token_accuracy": 0.17597800940275193,
|
|
"num_tokens": 8863647.0,
|
|
"step": 4875
|
|
},
|
|
{
|
|
"entropy": 5.847436714172363,
|
|
"epoch": 4.192522561237645,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00036764830195257437,
|
|
"loss": 5.4173,
|
|
"mean_token_accuracy": 0.17815880179405214,
|
|
"num_tokens": 8872911.0,
|
|
"step": 4880
|
|
},
|
|
{
|
|
"entropy": 5.829396724700928,
|
|
"epoch": 4.1968199398367,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.0003673452134591918,
|
|
"loss": 5.4408,
|
|
"mean_token_accuracy": 0.18240633606910706,
|
|
"num_tokens": 8881001.0,
|
|
"step": 4885
|
|
},
|
|
{
|
|
"entropy": 5.782381868362426,
|
|
"epoch": 4.201117318435754,
|
|
"grad_norm": 1.9609375,
|
|
"learning_rate": 0.000367041923320773,
|
|
"loss": 5.3376,
|
|
"mean_token_accuracy": 0.19203079342842103,
|
|
"num_tokens": 8890323.0,
|
|
"step": 4890
|
|
},
|
|
{
|
|
"entropy": 5.8106194019317625,
|
|
"epoch": 4.205414697034809,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00036673843219958257,
|
|
"loss": 5.4051,
|
|
"mean_token_accuracy": 0.18914072811603547,
|
|
"num_tokens": 8900471.0,
|
|
"step": 4895
|
|
},
|
|
{
|
|
"entropy": 5.794140338897705,
|
|
"epoch": 4.209712075633863,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.0003664347407583238,
|
|
"loss": 5.454,
|
|
"mean_token_accuracy": 0.18029122948646545,
|
|
"num_tokens": 8909320.0,
|
|
"step": 4900
|
|
},
|
|
{
|
|
"entropy": 5.756102800369263,
|
|
"epoch": 4.214009454232918,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0003661308496601373,
|
|
"loss": 5.3642,
|
|
"mean_token_accuracy": 0.18891047388315202,
|
|
"num_tokens": 8917453.0,
|
|
"step": 4905
|
|
},
|
|
{
|
|
"entropy": 5.785586977005005,
|
|
"epoch": 4.218306832831972,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00036582675956859983,
|
|
"loss": 5.4345,
|
|
"mean_token_accuracy": 0.18389338999986649,
|
|
"num_tokens": 8925737.0,
|
|
"step": 4910
|
|
},
|
|
{
|
|
"entropy": 5.721272134780884,
|
|
"epoch": 4.222604211431027,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00036552247114772263,
|
|
"loss": 5.3709,
|
|
"mean_token_accuracy": 0.18165080696344377,
|
|
"num_tokens": 8935475.0,
|
|
"step": 4915
|
|
},
|
|
{
|
|
"entropy": 5.817439126968384,
|
|
"epoch": 4.2269015900300815,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00036521798506194996,
|
|
"loss": 5.3959,
|
|
"mean_token_accuracy": 0.19157968014478682,
|
|
"num_tokens": 8944683.0,
|
|
"step": 4920
|
|
},
|
|
{
|
|
"entropy": 5.790067720413208,
|
|
"epoch": 4.231198968629136,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00036491330197615775,
|
|
"loss": 5.4484,
|
|
"mean_token_accuracy": 0.17825245559215547,
|
|
"num_tokens": 8953837.0,
|
|
"step": 4925
|
|
},
|
|
{
|
|
"entropy": 5.738772678375244,
|
|
"epoch": 4.235496347228191,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00036460842255565197,
|
|
"loss": 5.4774,
|
|
"mean_token_accuracy": 0.1834802284836769,
|
|
"num_tokens": 8964822.0,
|
|
"step": 4930
|
|
},
|
|
{
|
|
"entropy": 5.796894359588623,
|
|
"epoch": 4.239793725827245,
|
|
"grad_norm": 1.84375,
|
|
"learning_rate": 0.0003643033474661676,
|
|
"loss": 5.4537,
|
|
"mean_token_accuracy": 0.17824995517730713,
|
|
"num_tokens": 8974363.0,
|
|
"step": 4935
|
|
},
|
|
{
|
|
"entropy": 5.859163665771485,
|
|
"epoch": 4.2440911044263,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00036399807737386657,
|
|
"loss": 5.3405,
|
|
"mean_token_accuracy": 0.19169752895832062,
|
|
"num_tokens": 8983122.0,
|
|
"step": 4940
|
|
},
|
|
{
|
|
"entropy": 5.863666582107544,
|
|
"epoch": 4.248388483025354,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0003636926129453368,
|
|
"loss": 5.468,
|
|
"mean_token_accuracy": 0.17703534364700318,
|
|
"num_tokens": 8991618.0,
|
|
"step": 4945
|
|
},
|
|
{
|
|
"entropy": 5.780852937698365,
|
|
"epoch": 4.252685861624409,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0003633869548475904,
|
|
"loss": 5.3796,
|
|
"mean_token_accuracy": 0.18670552223920822,
|
|
"num_tokens": 9000128.0,
|
|
"step": 4950
|
|
},
|
|
{
|
|
"entropy": 5.751747989654541,
|
|
"epoch": 4.256983240223463,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003630811037480627,
|
|
"loss": 5.3649,
|
|
"mean_token_accuracy": 0.19108584225177766,
|
|
"num_tokens": 9008951.0,
|
|
"step": 4955
|
|
},
|
|
{
|
|
"entropy": 5.811222076416016,
|
|
"epoch": 4.261280618822518,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003627750603146101,
|
|
"loss": 5.414,
|
|
"mean_token_accuracy": 0.1837746262550354,
|
|
"num_tokens": 9018949.0,
|
|
"step": 4960
|
|
},
|
|
{
|
|
"entropy": 5.808954477310181,
|
|
"epoch": 4.265577997421573,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003624688252155091,
|
|
"loss": 5.4126,
|
|
"mean_token_accuracy": 0.18410263508558272,
|
|
"num_tokens": 9028910.0,
|
|
"step": 4965
|
|
},
|
|
{
|
|
"entropy": 5.759587383270263,
|
|
"epoch": 4.269875376020628,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003621623991194549,
|
|
"loss": 5.4695,
|
|
"mean_token_accuracy": 0.17817870974540712,
|
|
"num_tokens": 9039012.0,
|
|
"step": 4970
|
|
},
|
|
{
|
|
"entropy": 5.840623998641968,
|
|
"epoch": 4.274172754619682,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003618557826955594,
|
|
"loss": 5.448,
|
|
"mean_token_accuracy": 0.18455753922462464,
|
|
"num_tokens": 9048639.0,
|
|
"step": 4975
|
|
},
|
|
{
|
|
"entropy": 5.79478087425232,
|
|
"epoch": 4.278470133218737,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00036154897661335063,
|
|
"loss": 5.3827,
|
|
"mean_token_accuracy": 0.18554842621088027,
|
|
"num_tokens": 9057453.0,
|
|
"step": 4980
|
|
},
|
|
{
|
|
"entropy": 5.762424564361572,
|
|
"epoch": 4.2827675118177915,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 0.0003612419815427702,
|
|
"loss": 5.4292,
|
|
"mean_token_accuracy": 0.17840946614742278,
|
|
"num_tokens": 9066761.0,
|
|
"step": 4985
|
|
},
|
|
{
|
|
"entropy": 5.812322425842285,
|
|
"epoch": 4.287064890416846,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0003609347981541726,
|
|
"loss": 5.4933,
|
|
"mean_token_accuracy": 0.17946255803108216,
|
|
"num_tokens": 9075535.0,
|
|
"step": 4990
|
|
},
|
|
{
|
|
"entropy": 5.829926776885986,
|
|
"epoch": 4.291362269015901,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00036062742711832376,
|
|
"loss": 5.4073,
|
|
"mean_token_accuracy": 0.18371862769126893,
|
|
"num_tokens": 9084559.0,
|
|
"step": 4995
|
|
},
|
|
{
|
|
"entropy": 5.8012488842010494,
|
|
"epoch": 4.295659647614955,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0003603198691063991,
|
|
"loss": 5.3923,
|
|
"mean_token_accuracy": 0.18513423651456834,
|
|
"num_tokens": 9093069.0,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"epoch": 4.295659647614955,
|
|
"eval_entropy": 5.645841325725521,
|
|
"eval_loss": 6.082663059234619,
|
|
"eval_mean_token_accuracy": 0.1659068378141603,
|
|
"eval_num_tokens": 9093069.0,
|
|
"eval_runtime": 2.6427,
|
|
"eval_samples_per_second": 1342.939,
|
|
"eval_steps_per_second": 168.009,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"entropy": 5.751983594894409,
|
|
"epoch": 4.29995702621401,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0003600121247899824,
|
|
"loss": 5.3573,
|
|
"mean_token_accuracy": 0.19086044281721115,
|
|
"num_tokens": 9101914.0,
|
|
"step": 5005
|
|
},
|
|
{
|
|
"entropy": 5.778422880172729,
|
|
"epoch": 4.304254404813064,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00035970419484106404,
|
|
"loss": 5.46,
|
|
"mean_token_accuracy": 0.18139050155878067,
|
|
"num_tokens": 9110967.0,
|
|
"step": 5010
|
|
},
|
|
{
|
|
"entropy": 5.830139064788819,
|
|
"epoch": 4.308551783412119,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0003593960799320402,
|
|
"loss": 5.5309,
|
|
"mean_token_accuracy": 0.1760479047894478,
|
|
"num_tokens": 9119774.0,
|
|
"step": 5015
|
|
},
|
|
{
|
|
"entropy": 5.832511234283447,
|
|
"epoch": 4.312849162011173,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0003590877807357107,
|
|
"loss": 5.4255,
|
|
"mean_token_accuracy": 0.18054159283638,
|
|
"num_tokens": 9127738.0,
|
|
"step": 5020
|
|
},
|
|
{
|
|
"entropy": 5.733070516586304,
|
|
"epoch": 4.317146540610228,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003587792979252776,
|
|
"loss": 5.4212,
|
|
"mean_token_accuracy": 0.18748513609170914,
|
|
"num_tokens": 9137060.0,
|
|
"step": 5025
|
|
},
|
|
{
|
|
"entropy": 5.7518134117126465,
|
|
"epoch": 4.321443919209282,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0003584706321743442,
|
|
"loss": 5.3294,
|
|
"mean_token_accuracy": 0.18704721778631211,
|
|
"num_tokens": 9145169.0,
|
|
"step": 5030
|
|
},
|
|
{
|
|
"entropy": 5.773259878158569,
|
|
"epoch": 4.325741297808337,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000358161784156913,
|
|
"loss": 5.4,
|
|
"mean_token_accuracy": 0.18436064571142197,
|
|
"num_tokens": 9154092.0,
|
|
"step": 5035
|
|
},
|
|
{
|
|
"entropy": 5.780962133407593,
|
|
"epoch": 4.3300386764073915,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00035785275454738456,
|
|
"loss": 5.423,
|
|
"mean_token_accuracy": 0.17839469909667968,
|
|
"num_tokens": 9162824.0,
|
|
"step": 5040
|
|
},
|
|
{
|
|
"entropy": 5.748840427398681,
|
|
"epoch": 4.334336055006446,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00035754354402055635,
|
|
"loss": 5.3253,
|
|
"mean_token_accuracy": 0.19027216732501984,
|
|
"num_tokens": 9170977.0,
|
|
"step": 5045
|
|
},
|
|
{
|
|
"entropy": 5.757723617553711,
|
|
"epoch": 4.338633433605501,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003572341532516202,
|
|
"loss": 5.3925,
|
|
"mean_token_accuracy": 0.18991930037736893,
|
|
"num_tokens": 9179539.0,
|
|
"step": 5050
|
|
},
|
|
{
|
|
"entropy": 5.725891828536987,
|
|
"epoch": 4.342930812204555,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003569245829161622,
|
|
"loss": 5.4591,
|
|
"mean_token_accuracy": 0.18510087281465532,
|
|
"num_tokens": 9188861.0,
|
|
"step": 5055
|
|
},
|
|
{
|
|
"entropy": 5.779156923294067,
|
|
"epoch": 4.34722819080361,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00035661483369016004,
|
|
"loss": 5.395,
|
|
"mean_token_accuracy": 0.18243012726306915,
|
|
"num_tokens": 9197724.0,
|
|
"step": 5060
|
|
},
|
|
{
|
|
"entropy": 5.766918468475342,
|
|
"epoch": 4.351525569402664,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0003563049062499822,
|
|
"loss": 5.4307,
|
|
"mean_token_accuracy": 0.18592111766338348,
|
|
"num_tokens": 9206375.0,
|
|
"step": 5065
|
|
},
|
|
{
|
|
"entropy": 5.767550706863403,
|
|
"epoch": 4.355822948001719,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.0003559948012723865,
|
|
"loss": 5.3805,
|
|
"mean_token_accuracy": 0.18851764053106307,
|
|
"num_tokens": 9214675.0,
|
|
"step": 5070
|
|
},
|
|
{
|
|
"entropy": 5.708806324005127,
|
|
"epoch": 4.360120326600773,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003556845194345181,
|
|
"loss": 5.399,
|
|
"mean_token_accuracy": 0.1869270160794258,
|
|
"num_tokens": 9224128.0,
|
|
"step": 5075
|
|
},
|
|
{
|
|
"entropy": 5.68547477722168,
|
|
"epoch": 4.364417705199828,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0003553740614139086,
|
|
"loss": 5.3175,
|
|
"mean_token_accuracy": 0.19330300241708756,
|
|
"num_tokens": 9232568.0,
|
|
"step": 5080
|
|
},
|
|
{
|
|
"entropy": 5.772458839416504,
|
|
"epoch": 4.368715083798882,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0003550634278884742,
|
|
"loss": 5.4421,
|
|
"mean_token_accuracy": 0.1865006282925606,
|
|
"num_tokens": 9241809.0,
|
|
"step": 5085
|
|
},
|
|
{
|
|
"entropy": 5.783112716674805,
|
|
"epoch": 4.373012462397937,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00035475261953651433,
|
|
"loss": 5.4258,
|
|
"mean_token_accuracy": 0.1853316143155098,
|
|
"num_tokens": 9250845.0,
|
|
"step": 5090
|
|
},
|
|
{
|
|
"entropy": 5.7176933765411375,
|
|
"epoch": 4.3773098409969915,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00035444163703671026,
|
|
"loss": 5.3914,
|
|
"mean_token_accuracy": 0.19202740788459777,
|
|
"num_tokens": 9259465.0,
|
|
"step": 5095
|
|
},
|
|
{
|
|
"entropy": 5.7700214862823485,
|
|
"epoch": 4.381607219596046,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00035413048106812357,
|
|
"loss": 5.3692,
|
|
"mean_token_accuracy": 0.2002510607242584,
|
|
"num_tokens": 9267853.0,
|
|
"step": 5100
|
|
},
|
|
{
|
|
"entropy": 5.831751823425293,
|
|
"epoch": 4.385904598195101,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00035381915231019425,
|
|
"loss": 5.5593,
|
|
"mean_token_accuracy": 0.16938029676675798,
|
|
"num_tokens": 9276664.0,
|
|
"step": 5105
|
|
},
|
|
{
|
|
"entropy": 5.768694305419922,
|
|
"epoch": 4.390201976794156,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003535076514427401,
|
|
"loss": 5.3643,
|
|
"mean_token_accuracy": 0.1853679820895195,
|
|
"num_tokens": 9285482.0,
|
|
"step": 5110
|
|
},
|
|
{
|
|
"entropy": 5.821120071411133,
|
|
"epoch": 4.39449935539321,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00035319597914595436,
|
|
"loss": 5.477,
|
|
"mean_token_accuracy": 0.17474496960639954,
|
|
"num_tokens": 9293936.0,
|
|
"step": 5115
|
|
},
|
|
{
|
|
"entropy": 5.808133840560913,
|
|
"epoch": 4.398796733992265,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0003528841361004049,
|
|
"loss": 5.508,
|
|
"mean_token_accuracy": 0.17144946306943892,
|
|
"num_tokens": 9303998.0,
|
|
"step": 5120
|
|
},
|
|
{
|
|
"entropy": 5.743658590316772,
|
|
"epoch": 4.40309411259132,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003525721229870323,
|
|
"loss": 5.4208,
|
|
"mean_token_accuracy": 0.18494210690259932,
|
|
"num_tokens": 9313117.0,
|
|
"step": 5125
|
|
},
|
|
{
|
|
"entropy": 5.814204978942871,
|
|
"epoch": 4.407391491190374,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00035225994048714823,
|
|
"loss": 5.4268,
|
|
"mean_token_accuracy": 0.1809573084115982,
|
|
"num_tokens": 9321446.0,
|
|
"step": 5130
|
|
},
|
|
{
|
|
"entropy": 5.721041345596314,
|
|
"epoch": 4.411688869789429,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0003519475892824348,
|
|
"loss": 5.4285,
|
|
"mean_token_accuracy": 0.18174952268600464,
|
|
"num_tokens": 9330752.0,
|
|
"step": 5135
|
|
},
|
|
{
|
|
"entropy": 5.768051433563232,
|
|
"epoch": 4.415986248388483,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003516350700549419,
|
|
"loss": 5.446,
|
|
"mean_token_accuracy": 0.18225976079702377,
|
|
"num_tokens": 9339322.0,
|
|
"step": 5140
|
|
},
|
|
{
|
|
"entropy": 5.852107238769531,
|
|
"epoch": 4.420283626987538,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00035132238348708697,
|
|
"loss": 5.4705,
|
|
"mean_token_accuracy": 0.18594772070646287,
|
|
"num_tokens": 9349024.0,
|
|
"step": 5145
|
|
},
|
|
{
|
|
"entropy": 5.85985164642334,
|
|
"epoch": 4.424581005586592,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00035100953026165224,
|
|
"loss": 5.57,
|
|
"mean_token_accuracy": 0.1765264689922333,
|
|
"num_tokens": 9358833.0,
|
|
"step": 5150
|
|
},
|
|
{
|
|
"entropy": 5.828482389450073,
|
|
"epoch": 4.428878384185647,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003506965110617841,
|
|
"loss": 5.4011,
|
|
"mean_token_accuracy": 0.1904045358300209,
|
|
"num_tokens": 9368276.0,
|
|
"step": 5155
|
|
},
|
|
{
|
|
"entropy": 5.83995475769043,
|
|
"epoch": 4.4331757627847015,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003503833265709915,
|
|
"loss": 5.4934,
|
|
"mean_token_accuracy": 0.1783384919166565,
|
|
"num_tokens": 9378501.0,
|
|
"step": 5160
|
|
},
|
|
{
|
|
"entropy": 5.844538736343384,
|
|
"epoch": 4.437473141383756,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00035006997747314404,
|
|
"loss": 5.4741,
|
|
"mean_token_accuracy": 0.17655717581510544,
|
|
"num_tokens": 9387789.0,
|
|
"step": 5165
|
|
},
|
|
{
|
|
"entropy": 5.817201757431031,
|
|
"epoch": 4.441770519982811,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00034975646445247106,
|
|
"loss": 5.5393,
|
|
"mean_token_accuracy": 0.17986075729131698,
|
|
"num_tokens": 9397041.0,
|
|
"step": 5170
|
|
},
|
|
{
|
|
"entropy": 5.77465705871582,
|
|
"epoch": 4.446067898581865,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0003494427881935596,
|
|
"loss": 5.4257,
|
|
"mean_token_accuracy": 0.18991996198892594,
|
|
"num_tokens": 9405393.0,
|
|
"step": 5175
|
|
},
|
|
{
|
|
"entropy": 5.763102245330811,
|
|
"epoch": 4.45036527718092,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00034912894938135325,
|
|
"loss": 5.425,
|
|
"mean_token_accuracy": 0.17877031713724137,
|
|
"num_tokens": 9415127.0,
|
|
"step": 5180
|
|
},
|
|
{
|
|
"entropy": 5.812909126281738,
|
|
"epoch": 4.454662655779974,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0003488149487011506,
|
|
"loss": 5.5255,
|
|
"mean_token_accuracy": 0.17933376878499985,
|
|
"num_tokens": 9424416.0,
|
|
"step": 5185
|
|
},
|
|
{
|
|
"entropy": 5.853853988647461,
|
|
"epoch": 4.458960034379029,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00034850078683860346,
|
|
"loss": 5.4739,
|
|
"mean_token_accuracy": 0.17511797249317168,
|
|
"num_tokens": 9434523.0,
|
|
"step": 5190
|
|
},
|
|
{
|
|
"entropy": 5.836376714706421,
|
|
"epoch": 4.463257412978083,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003481864644797159,
|
|
"loss": 5.4815,
|
|
"mean_token_accuracy": 0.1875310242176056,
|
|
"num_tokens": 9443605.0,
|
|
"step": 5195
|
|
},
|
|
{
|
|
"entropy": 5.799241399765014,
|
|
"epoch": 4.467554791577138,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003478719823108424,
|
|
"loss": 5.4664,
|
|
"mean_token_accuracy": 0.17694201022386552,
|
|
"num_tokens": 9453268.0,
|
|
"step": 5200
|
|
},
|
|
{
|
|
"entropy": 5.734842443466187,
|
|
"epoch": 4.471852170176192,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00034755734101868613,
|
|
"loss": 5.3474,
|
|
"mean_token_accuracy": 0.18722489178180696,
|
|
"num_tokens": 9461578.0,
|
|
"step": 5205
|
|
},
|
|
{
|
|
"entropy": 5.719850301742554,
|
|
"epoch": 4.476149548775247,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00034724254129029795,
|
|
"loss": 5.3751,
|
|
"mean_token_accuracy": 0.18702739328145981,
|
|
"num_tokens": 9470722.0,
|
|
"step": 5210
|
|
},
|
|
{
|
|
"entropy": 5.8327727794647215,
|
|
"epoch": 4.4804469273743015,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0003469275838130748,
|
|
"loss": 5.4911,
|
|
"mean_token_accuracy": 0.1831965833902359,
|
|
"num_tokens": 9479695.0,
|
|
"step": 5215
|
|
},
|
|
{
|
|
"entropy": 5.773561906814575,
|
|
"epoch": 4.484744305973356,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003466124692747577,
|
|
"loss": 5.4214,
|
|
"mean_token_accuracy": 0.18437404334545135,
|
|
"num_tokens": 9488444.0,
|
|
"step": 5220
|
|
},
|
|
{
|
|
"entropy": 5.700850105285644,
|
|
"epoch": 4.489041684572411,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00034629719836343106,
|
|
"loss": 5.3691,
|
|
"mean_token_accuracy": 0.1867497444152832,
|
|
"num_tokens": 9497413.0,
|
|
"step": 5225
|
|
},
|
|
{
|
|
"entropy": 5.758021974563599,
|
|
"epoch": 4.493339063171465,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0003459817717675203,
|
|
"loss": 5.4049,
|
|
"mean_token_accuracy": 0.19463123977184296,
|
|
"num_tokens": 9506135.0,
|
|
"step": 5230
|
|
},
|
|
{
|
|
"entropy": 5.786157035827637,
|
|
"epoch": 4.49763644177052,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003456661901757913,
|
|
"loss": 5.4866,
|
|
"mean_token_accuracy": 0.17635240852832795,
|
|
"num_tokens": 9516918.0,
|
|
"step": 5235
|
|
},
|
|
{
|
|
"entropy": 5.873275089263916,
|
|
"epoch": 4.501933820369574,
|
|
"grad_norm": 1.84375,
|
|
"learning_rate": 0.00034535045427734796,
|
|
"loss": 5.4344,
|
|
"mean_token_accuracy": 0.18653480857610702,
|
|
"num_tokens": 9526052.0,
|
|
"step": 5240
|
|
},
|
|
{
|
|
"entropy": 5.737301445007324,
|
|
"epoch": 4.506231198968629,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003450345647616313,
|
|
"loss": 5.4785,
|
|
"mean_token_accuracy": 0.1859130308032036,
|
|
"num_tokens": 9535200.0,
|
|
"step": 5245
|
|
},
|
|
{
|
|
"entropy": 5.724278545379638,
|
|
"epoch": 4.510528577567683,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0003447185223184177,
|
|
"loss": 5.4519,
|
|
"mean_token_accuracy": 0.18500425815582275,
|
|
"num_tokens": 9544786.0,
|
|
"step": 5250
|
|
},
|
|
{
|
|
"entropy": 5.802084589004517,
|
|
"epoch": 4.514825956166739,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00034440232763781765,
|
|
"loss": 5.4039,
|
|
"mean_token_accuracy": 0.18684340715408326,
|
|
"num_tokens": 9553694.0,
|
|
"step": 5255
|
|
},
|
|
{
|
|
"entropy": 5.775804281234741,
|
|
"epoch": 4.519123334765792,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000344085981410274,
|
|
"loss": 5.4643,
|
|
"mean_token_accuracy": 0.18814241141080856,
|
|
"num_tokens": 9563332.0,
|
|
"step": 5260
|
|
},
|
|
{
|
|
"entropy": 5.685090065002441,
|
|
"epoch": 4.523420713364848,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00034376948432656036,
|
|
"loss": 5.3812,
|
|
"mean_token_accuracy": 0.1896350249648094,
|
|
"num_tokens": 9572367.0,
|
|
"step": 5265
|
|
},
|
|
{
|
|
"entropy": 5.82707896232605,
|
|
"epoch": 4.527718091963902,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003434528370777798,
|
|
"loss": 5.4705,
|
|
"mean_token_accuracy": 0.17945346981287003,
|
|
"num_tokens": 9582535.0,
|
|
"step": 5270
|
|
},
|
|
{
|
|
"entropy": 5.821118688583374,
|
|
"epoch": 4.532015470562957,
|
|
"grad_norm": 2.40625,
|
|
"learning_rate": 0.00034313604035536344,
|
|
"loss": 5.4158,
|
|
"mean_token_accuracy": 0.18645084351301194,
|
|
"num_tokens": 9590688.0,
|
|
"step": 5275
|
|
},
|
|
{
|
|
"entropy": 5.694446611404419,
|
|
"epoch": 4.5363128491620115,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003428190948510687,
|
|
"loss": 5.469,
|
|
"mean_token_accuracy": 0.1815877929329872,
|
|
"num_tokens": 9599209.0,
|
|
"step": 5280
|
|
},
|
|
{
|
|
"entropy": 5.786466073989868,
|
|
"epoch": 4.540610227761066,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003425020012569778,
|
|
"loss": 5.5005,
|
|
"mean_token_accuracy": 0.17607855647802353,
|
|
"num_tokens": 9608575.0,
|
|
"step": 5285
|
|
},
|
|
{
|
|
"entropy": 5.89086275100708,
|
|
"epoch": 4.544907606360121,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00034218476026549665,
|
|
"loss": 5.4477,
|
|
"mean_token_accuracy": 0.18247157335281372,
|
|
"num_tokens": 9617312.0,
|
|
"step": 5290
|
|
},
|
|
{
|
|
"entropy": 5.751475381851196,
|
|
"epoch": 4.549204984959175,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003418673725693524,
|
|
"loss": 5.426,
|
|
"mean_token_accuracy": 0.1912728428840637,
|
|
"num_tokens": 9626398.0,
|
|
"step": 5295
|
|
},
|
|
{
|
|
"entropy": 5.738027286529541,
|
|
"epoch": 4.55350236355823,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003415498388615932,
|
|
"loss": 5.4188,
|
|
"mean_token_accuracy": 0.18312942683696748,
|
|
"num_tokens": 9635470.0,
|
|
"step": 5300
|
|
},
|
|
{
|
|
"entropy": 5.8186338424682615,
|
|
"epoch": 4.557799742157284,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.0003412321598355857,
|
|
"loss": 5.3787,
|
|
"mean_token_accuracy": 0.18646468222141266,
|
|
"num_tokens": 9644728.0,
|
|
"step": 5305
|
|
},
|
|
{
|
|
"entropy": 5.777834177017212,
|
|
"epoch": 4.562097120756339,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003409143361850139,
|
|
"loss": 5.42,
|
|
"mean_token_accuracy": 0.18390751183032988,
|
|
"num_tokens": 9654129.0,
|
|
"step": 5310
|
|
},
|
|
{
|
|
"entropy": 5.753074026107788,
|
|
"epoch": 4.566394499355393,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003405963686038775,
|
|
"loss": 5.4869,
|
|
"mean_token_accuracy": 0.18370855152606963,
|
|
"num_tokens": 9662648.0,
|
|
"step": 5315
|
|
},
|
|
{
|
|
"entropy": 5.809203624725342,
|
|
"epoch": 4.570691877954448,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.0003402782577864908,
|
|
"loss": 5.4739,
|
|
"mean_token_accuracy": 0.18376611918210983,
|
|
"num_tokens": 9672082.0,
|
|
"step": 5320
|
|
},
|
|
{
|
|
"entropy": 5.833519744873047,
|
|
"epoch": 4.574989256553502,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00033996000442748056,
|
|
"loss": 5.3935,
|
|
"mean_token_accuracy": 0.1896258920431137,
|
|
"num_tokens": 9681422.0,
|
|
"step": 5325
|
|
},
|
|
{
|
|
"entropy": 5.761556482315063,
|
|
"epoch": 4.579286635152557,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00033964160922178495,
|
|
"loss": 5.4305,
|
|
"mean_token_accuracy": 0.186528979241848,
|
|
"num_tokens": 9690675.0,
|
|
"step": 5330
|
|
},
|
|
{
|
|
"entropy": 5.717699146270752,
|
|
"epoch": 4.5835840137516115,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003393230728646518,
|
|
"loss": 5.413,
|
|
"mean_token_accuracy": 0.18689863085746766,
|
|
"num_tokens": 9700200.0,
|
|
"step": 5335
|
|
},
|
|
{
|
|
"entropy": 5.699422597885132,
|
|
"epoch": 4.587881392350666,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00033900439605163724,
|
|
"loss": 5.423,
|
|
"mean_token_accuracy": 0.18302758783102036,
|
|
"num_tokens": 9709533.0,
|
|
"step": 5340
|
|
},
|
|
{
|
|
"entropy": 5.745850563049316,
|
|
"epoch": 4.592178770949721,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00033868557947860407,
|
|
"loss": 5.4402,
|
|
"mean_token_accuracy": 0.185868239402771,
|
|
"num_tokens": 9719250.0,
|
|
"step": 5345
|
|
},
|
|
{
|
|
"entropy": 5.780076551437378,
|
|
"epoch": 4.596476149548775,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00033836662384172014,
|
|
"loss": 5.377,
|
|
"mean_token_accuracy": 0.19353508353233337,
|
|
"num_tokens": 9727837.0,
|
|
"step": 5350
|
|
},
|
|
{
|
|
"entropy": 5.734384632110595,
|
|
"epoch": 4.60077352814783,
|
|
"grad_norm": 2.140625,
|
|
"learning_rate": 0.0003380475298374573,
|
|
"loss": 5.4721,
|
|
"mean_token_accuracy": 0.1821171373128891,
|
|
"num_tokens": 9737125.0,
|
|
"step": 5355
|
|
},
|
|
{
|
|
"entropy": 5.774573087692261,
|
|
"epoch": 4.605070906746884,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.000337728298162589,
|
|
"loss": 5.4879,
|
|
"mean_token_accuracy": 0.17716386020183564,
|
|
"num_tokens": 9746309.0,
|
|
"step": 5360
|
|
},
|
|
{
|
|
"entropy": 5.8264130592346195,
|
|
"epoch": 4.609368285345939,
|
|
"grad_norm": 1.890625,
|
|
"learning_rate": 0.00033740892951418993,
|
|
"loss": 5.3815,
|
|
"mean_token_accuracy": 0.18831967562437057,
|
|
"num_tokens": 9755633.0,
|
|
"step": 5365
|
|
},
|
|
{
|
|
"entropy": 5.807550096511841,
|
|
"epoch": 4.613665663944993,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0003370894245896333,
|
|
"loss": 5.422,
|
|
"mean_token_accuracy": 0.17816249877214432,
|
|
"num_tokens": 9765179.0,
|
|
"step": 5370
|
|
},
|
|
{
|
|
"entropy": 5.848112964630127,
|
|
"epoch": 4.617963042544048,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00033676978408659047,
|
|
"loss": 5.4337,
|
|
"mean_token_accuracy": 0.1831375315785408,
|
|
"num_tokens": 9774085.0,
|
|
"step": 5375
|
|
},
|
|
{
|
|
"entropy": 5.8263157367706295,
|
|
"epoch": 4.622260421143102,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.0003364500087030283,
|
|
"loss": 5.5484,
|
|
"mean_token_accuracy": 0.16986390799283982,
|
|
"num_tokens": 9784650.0,
|
|
"step": 5380
|
|
},
|
|
{
|
|
"entropy": 5.811712312698364,
|
|
"epoch": 4.626557799742157,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00033613009913720845,
|
|
"loss": 5.3989,
|
|
"mean_token_accuracy": 0.18403580784797668,
|
|
"num_tokens": 9793947.0,
|
|
"step": 5385
|
|
},
|
|
{
|
|
"entropy": 5.627312660217285,
|
|
"epoch": 4.6308551783412115,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.00033581005608768563,
|
|
"loss": 5.3963,
|
|
"mean_token_accuracy": 0.1883517861366272,
|
|
"num_tokens": 9803593.0,
|
|
"step": 5390
|
|
},
|
|
{
|
|
"entropy": 5.699808931350708,
|
|
"epoch": 4.635152556940266,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003354898802533058,
|
|
"loss": 5.4264,
|
|
"mean_token_accuracy": 0.18414541929960251,
|
|
"num_tokens": 9812295.0,
|
|
"step": 5395
|
|
},
|
|
{
|
|
"entropy": 5.82594518661499,
|
|
"epoch": 4.6394499355393215,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0003351695723332051,
|
|
"loss": 5.4384,
|
|
"mean_token_accuracy": 0.18271765410900115,
|
|
"num_tokens": 9820586.0,
|
|
"step": 5400
|
|
},
|
|
{
|
|
"entropy": 5.754842710494995,
|
|
"epoch": 4.643747314138375,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00033484913302680807,
|
|
"loss": 5.3511,
|
|
"mean_token_accuracy": 0.1845670998096466,
|
|
"num_tokens": 9829080.0,
|
|
"step": 5405
|
|
},
|
|
{
|
|
"entropy": 5.761151599884033,
|
|
"epoch": 4.648044692737431,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00033452856303382595,
|
|
"loss": 5.381,
|
|
"mean_token_accuracy": 0.18969232738018035,
|
|
"num_tokens": 9838421.0,
|
|
"step": 5410
|
|
},
|
|
{
|
|
"entropy": 5.74131817817688,
|
|
"epoch": 4.652342071336484,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0003342078630542555,
|
|
"loss": 5.371,
|
|
"mean_token_accuracy": 0.19116330593824388,
|
|
"num_tokens": 9847151.0,
|
|
"step": 5415
|
|
},
|
|
{
|
|
"entropy": 5.7510984420776365,
|
|
"epoch": 4.65663944993554,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00033388703378837737,
|
|
"loss": 5.4373,
|
|
"mean_token_accuracy": 0.18419362604618073,
|
|
"num_tokens": 9856803.0,
|
|
"step": 5420
|
|
},
|
|
{
|
|
"entropy": 5.717685747146606,
|
|
"epoch": 4.660936828534594,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0003335660759367544,
|
|
"loss": 5.3278,
|
|
"mean_token_accuracy": 0.20107688307762145,
|
|
"num_tokens": 9865617.0,
|
|
"step": 5425
|
|
},
|
|
{
|
|
"entropy": 5.7392292499542235,
|
|
"epoch": 4.665234207133649,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00033324499020023025,
|
|
"loss": 5.3938,
|
|
"mean_token_accuracy": 0.19347498565912247,
|
|
"num_tokens": 9875454.0,
|
|
"step": 5430
|
|
},
|
|
{
|
|
"entropy": 5.780171680450439,
|
|
"epoch": 4.669531585732703,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0003329237772799277,
|
|
"loss": 5.3954,
|
|
"mean_token_accuracy": 0.1843357041478157,
|
|
"num_tokens": 9884770.0,
|
|
"step": 5435
|
|
},
|
|
{
|
|
"entropy": 5.72673544883728,
|
|
"epoch": 4.673828964331758,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0003326024378772477,
|
|
"loss": 5.4005,
|
|
"mean_token_accuracy": 0.1911465644836426,
|
|
"num_tokens": 9893594.0,
|
|
"step": 5440
|
|
},
|
|
{
|
|
"entropy": 5.811465883255005,
|
|
"epoch": 4.678126342930812,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.0003322809726938667,
|
|
"loss": 5.525,
|
|
"mean_token_accuracy": 0.17573501765727997,
|
|
"num_tokens": 9902260.0,
|
|
"step": 5445
|
|
},
|
|
{
|
|
"entropy": 5.8122059345245365,
|
|
"epoch": 4.682423721529867,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00033195938243173645,
|
|
"loss": 5.4135,
|
|
"mean_token_accuracy": 0.18877927958965302,
|
|
"num_tokens": 9911020.0,
|
|
"step": 5450
|
|
},
|
|
{
|
|
"entropy": 5.814115381240844,
|
|
"epoch": 4.6867211001289215,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0003316376677930814,
|
|
"loss": 5.4137,
|
|
"mean_token_accuracy": 0.18664776682853698,
|
|
"num_tokens": 9918696.0,
|
|
"step": 5455
|
|
},
|
|
{
|
|
"entropy": 5.678168821334839,
|
|
"epoch": 4.691018478727976,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0003313158294803977,
|
|
"loss": 5.4551,
|
|
"mean_token_accuracy": 0.17881893068552018,
|
|
"num_tokens": 9927638.0,
|
|
"step": 5460
|
|
},
|
|
{
|
|
"entropy": 5.778184986114502,
|
|
"epoch": 4.695315857327031,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00033099386819645176,
|
|
"loss": 5.4218,
|
|
"mean_token_accuracy": 0.18223019540309907,
|
|
"num_tokens": 9936969.0,
|
|
"step": 5465
|
|
},
|
|
{
|
|
"entropy": 5.821640682220459,
|
|
"epoch": 4.699613235926085,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0003306717846442782,
|
|
"loss": 5.3423,
|
|
"mean_token_accuracy": 0.18939315378665925,
|
|
"num_tokens": 9945229.0,
|
|
"step": 5470
|
|
},
|
|
{
|
|
"entropy": 5.761671543121338,
|
|
"epoch": 4.70391061452514,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003303495795271788,
|
|
"loss": 5.339,
|
|
"mean_token_accuracy": 0.18879547119140624,
|
|
"num_tokens": 9953759.0,
|
|
"step": 5475
|
|
},
|
|
{
|
|
"entropy": 5.733751392364502,
|
|
"epoch": 4.708207993124194,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00033002725354872075,
|
|
"loss": 5.458,
|
|
"mean_token_accuracy": 0.18100002110004426,
|
|
"num_tokens": 9962771.0,
|
|
"step": 5480
|
|
},
|
|
{
|
|
"entropy": 5.772163772583008,
|
|
"epoch": 4.712505371723249,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00032970480741273514,
|
|
"loss": 5.47,
|
|
"mean_token_accuracy": 0.17284650206565857,
|
|
"num_tokens": 9972481.0,
|
|
"step": 5485
|
|
},
|
|
{
|
|
"entropy": 5.869122457504273,
|
|
"epoch": 4.716802750322303,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 0.0003293822418233155,
|
|
"loss": 5.408,
|
|
"mean_token_accuracy": 0.1872358277440071,
|
|
"num_tokens": 9980773.0,
|
|
"step": 5490
|
|
},
|
|
{
|
|
"entropy": 5.792991781234742,
|
|
"epoch": 4.721100128921358,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.0003290595574848161,
|
|
"loss": 5.4888,
|
|
"mean_token_accuracy": 0.17654326558113098,
|
|
"num_tokens": 9989830.0,
|
|
"step": 5495
|
|
},
|
|
{
|
|
"entropy": 5.68425669670105,
|
|
"epoch": 4.725397507520412,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003287367551018505,
|
|
"loss": 5.4222,
|
|
"mean_token_accuracy": 0.19247835725545884,
|
|
"num_tokens": 9999234.0,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"epoch": 4.725397507520412,
|
|
"eval_entropy": 5.5955866059741455,
|
|
"eval_loss": 6.048821926116943,
|
|
"eval_mean_token_accuracy": 0.16785702025433918,
|
|
"eval_num_tokens": 9999234.0,
|
|
"eval_runtime": 2.6542,
|
|
"eval_samples_per_second": 1337.109,
|
|
"eval_steps_per_second": 167.28,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"entropy": 5.7583198070526125,
|
|
"epoch": 4.729694886119467,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003284138353792903,
|
|
"loss": 5.468,
|
|
"mean_token_accuracy": 0.1807752311229706,
|
|
"num_tokens": 10008671.0,
|
|
"step": 5505
|
|
},
|
|
{
|
|
"entropy": 5.76370325088501,
|
|
"epoch": 4.7339922647185215,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0003280907990222628,
|
|
"loss": 5.4336,
|
|
"mean_token_accuracy": 0.18588260561227798,
|
|
"num_tokens": 10017170.0,
|
|
"step": 5510
|
|
},
|
|
{
|
|
"entropy": 5.741564083099365,
|
|
"epoch": 4.738289643317576,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.00032776764673615055,
|
|
"loss": 5.46,
|
|
"mean_token_accuracy": 0.17722293436527253,
|
|
"num_tokens": 10025712.0,
|
|
"step": 5515
|
|
},
|
|
{
|
|
"entropy": 5.734073162078857,
|
|
"epoch": 4.742587021916631,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003274443792265888,
|
|
"loss": 5.4502,
|
|
"mean_token_accuracy": 0.1886144295334816,
|
|
"num_tokens": 10035297.0,
|
|
"step": 5520
|
|
},
|
|
{
|
|
"entropy": 5.779096269607544,
|
|
"epoch": 4.746884400515685,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00032712099719946474,
|
|
"loss": 5.4312,
|
|
"mean_token_accuracy": 0.18919673711061477,
|
|
"num_tokens": 10043903.0,
|
|
"step": 5525
|
|
},
|
|
{
|
|
"entropy": 5.779982995986939,
|
|
"epoch": 4.75118177911474,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00032679750136091533,
|
|
"loss": 5.4485,
|
|
"mean_token_accuracy": 0.18333171755075456,
|
|
"num_tokens": 10053035.0,
|
|
"step": 5530
|
|
},
|
|
{
|
|
"entropy": 5.698885917663574,
|
|
"epoch": 4.755479157713794,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003264738924173262,
|
|
"loss": 5.3868,
|
|
"mean_token_accuracy": 0.18878008276224137,
|
|
"num_tokens": 10061911.0,
|
|
"step": 5535
|
|
},
|
|
{
|
|
"entropy": 5.727502393722534,
|
|
"epoch": 4.759776536312849,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00032615017107533,
|
|
"loss": 5.4134,
|
|
"mean_token_accuracy": 0.1958436921238899,
|
|
"num_tokens": 10070738.0,
|
|
"step": 5540
|
|
},
|
|
{
|
|
"entropy": 5.78805832862854,
|
|
"epoch": 4.764073914911903,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003258263380418047,
|
|
"loss": 5.4407,
|
|
"mean_token_accuracy": 0.18714932799339296,
|
|
"num_tokens": 10080638.0,
|
|
"step": 5545
|
|
},
|
|
{
|
|
"entropy": 5.860460805892944,
|
|
"epoch": 4.768371293510958,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00032550239402387226,
|
|
"loss": 5.4767,
|
|
"mean_token_accuracy": 0.18044149428606032,
|
|
"num_tokens": 10089429.0,
|
|
"step": 5550
|
|
},
|
|
{
|
|
"entropy": 5.679690313339234,
|
|
"epoch": 4.772668672110013,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00032517833972889695,
|
|
"loss": 5.3468,
|
|
"mean_token_accuracy": 0.18653945326805116,
|
|
"num_tokens": 10098109.0,
|
|
"step": 5555
|
|
},
|
|
{
|
|
"entropy": 5.755960845947266,
|
|
"epoch": 4.776966050709067,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00032485417586448375,
|
|
"loss": 5.4486,
|
|
"mean_token_accuracy": 0.18553524017333983,
|
|
"num_tokens": 10106808.0,
|
|
"step": 5560
|
|
},
|
|
{
|
|
"entropy": 5.769974708557129,
|
|
"epoch": 4.781263429308122,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000324529903138477,
|
|
"loss": 5.4518,
|
|
"mean_token_accuracy": 0.18162155598402024,
|
|
"num_tokens": 10116372.0,
|
|
"step": 5565
|
|
},
|
|
{
|
|
"entropy": 5.729024028778076,
|
|
"epoch": 4.785560807907177,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0003242055222589587,
|
|
"loss": 5.368,
|
|
"mean_token_accuracy": 0.19198417514562607,
|
|
"num_tokens": 10125256.0,
|
|
"step": 5570
|
|
},
|
|
{
|
|
"entropy": 5.770532274246216,
|
|
"epoch": 4.7898581865062315,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.000323881033934247,
|
|
"loss": 5.4873,
|
|
"mean_token_accuracy": 0.17426208257675171,
|
|
"num_tokens": 10134784.0,
|
|
"step": 5575
|
|
},
|
|
{
|
|
"entropy": 5.823772192001343,
|
|
"epoch": 4.794155565105286,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00032355643887289486,
|
|
"loss": 5.4385,
|
|
"mean_token_accuracy": 0.18774469643831254,
|
|
"num_tokens": 10144324.0,
|
|
"step": 5580
|
|
},
|
|
{
|
|
"entropy": 5.771240282058716,
|
|
"epoch": 4.798452943704341,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0003232317377836881,
|
|
"loss": 5.4788,
|
|
"mean_token_accuracy": 0.17641482949256898,
|
|
"num_tokens": 10152866.0,
|
|
"step": 5585
|
|
},
|
|
{
|
|
"entropy": 5.747385025024414,
|
|
"epoch": 4.802750322303395,
|
|
"grad_norm": 1.796875,
|
|
"learning_rate": 0.000322906931375644,
|
|
"loss": 5.3892,
|
|
"mean_token_accuracy": 0.1895698770880699,
|
|
"num_tokens": 10162457.0,
|
|
"step": 5590
|
|
},
|
|
{
|
|
"entropy": 5.744189691543579,
|
|
"epoch": 4.80704770090245,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00032258202035801,
|
|
"loss": 5.4903,
|
|
"mean_token_accuracy": 0.18102981448173522,
|
|
"num_tokens": 10171604.0,
|
|
"step": 5595
|
|
},
|
|
{
|
|
"entropy": 5.815001106262207,
|
|
"epoch": 4.811345079501504,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000322257005440262,
|
|
"loss": 5.4134,
|
|
"mean_token_accuracy": 0.19356087297201158,
|
|
"num_tokens": 10180762.0,
|
|
"step": 5600
|
|
},
|
|
{
|
|
"entropy": 5.7168957710266115,
|
|
"epoch": 4.815642458100559,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003219318873321025,
|
|
"loss": 5.3481,
|
|
"mean_token_accuracy": 0.20167198330163955,
|
|
"num_tokens": 10189122.0,
|
|
"step": 5605
|
|
},
|
|
{
|
|
"entropy": 5.758258438110351,
|
|
"epoch": 4.819939836699613,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00032160666674345954,
|
|
"loss": 5.4253,
|
|
"mean_token_accuracy": 0.18134229332208635,
|
|
"num_tokens": 10197280.0,
|
|
"step": 5610
|
|
},
|
|
{
|
|
"entropy": 5.783818531036377,
|
|
"epoch": 4.824237215298668,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00032128134438448504,
|
|
"loss": 5.4771,
|
|
"mean_token_accuracy": 0.18173396736383438,
|
|
"num_tokens": 10207507.0,
|
|
"step": 5615
|
|
},
|
|
{
|
|
"entropy": 5.806025743484497,
|
|
"epoch": 4.828534593897722,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00032095592096555284,
|
|
"loss": 5.4602,
|
|
"mean_token_accuracy": 0.18259961307048797,
|
|
"num_tokens": 10217584.0,
|
|
"step": 5620
|
|
},
|
|
{
|
|
"entropy": 5.7541428565979,
|
|
"epoch": 4.832831972496777,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003206303971972577,
|
|
"loss": 5.4073,
|
|
"mean_token_accuracy": 0.18918070197105408,
|
|
"num_tokens": 10226388.0,
|
|
"step": 5625
|
|
},
|
|
{
|
|
"entropy": 5.755873918533325,
|
|
"epoch": 4.8371293510958315,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0003203047737904134,
|
|
"loss": 5.4327,
|
|
"mean_token_accuracy": 0.18559988886117934,
|
|
"num_tokens": 10235333.0,
|
|
"step": 5630
|
|
},
|
|
{
|
|
"entropy": 5.787179088592529,
|
|
"epoch": 4.841426729694886,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00031997905145605135,
|
|
"loss": 5.4603,
|
|
"mean_token_accuracy": 0.18634885996580125,
|
|
"num_tokens": 10243985.0,
|
|
"step": 5635
|
|
},
|
|
{
|
|
"entropy": 5.769427824020386,
|
|
"epoch": 4.845724108293941,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00031965323090541874,
|
|
"loss": 5.4722,
|
|
"mean_token_accuracy": 0.17648543417453766,
|
|
"num_tokens": 10252968.0,
|
|
"step": 5640
|
|
},
|
|
{
|
|
"entropy": 5.755967330932617,
|
|
"epoch": 4.850021486892995,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0003193273128499777,
|
|
"loss": 5.3186,
|
|
"mean_token_accuracy": 0.1852602019906044,
|
|
"num_tokens": 10261890.0,
|
|
"step": 5645
|
|
},
|
|
{
|
|
"entropy": 5.759440135955811,
|
|
"epoch": 4.85431886549205,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00031900129800140287,
|
|
"loss": 5.4285,
|
|
"mean_token_accuracy": 0.18027369678020477,
|
|
"num_tokens": 10271363.0,
|
|
"step": 5650
|
|
},
|
|
{
|
|
"entropy": 5.77560977935791,
|
|
"epoch": 4.858616244091104,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00031867518707158027,
|
|
"loss": 5.476,
|
|
"mean_token_accuracy": 0.1777102008461952,
|
|
"num_tokens": 10280608.0,
|
|
"step": 5655
|
|
},
|
|
{
|
|
"entropy": 5.751598310470581,
|
|
"epoch": 4.862913622690159,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 0.000318348980772606,
|
|
"loss": 5.3987,
|
|
"mean_token_accuracy": 0.19135443717241288,
|
|
"num_tokens": 10289972.0,
|
|
"step": 5660
|
|
},
|
|
{
|
|
"entropy": 5.797596025466919,
|
|
"epoch": 4.867211001289213,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00031802267981678414,
|
|
"loss": 5.461,
|
|
"mean_token_accuracy": 0.183905990421772,
|
|
"num_tokens": 10298740.0,
|
|
"step": 5665
|
|
},
|
|
{
|
|
"entropy": 5.792353725433349,
|
|
"epoch": 4.871508379888268,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00031769628491662563,
|
|
"loss": 5.4185,
|
|
"mean_token_accuracy": 0.18972143083810805,
|
|
"num_tokens": 10307706.0,
|
|
"step": 5670
|
|
},
|
|
{
|
|
"entropy": 5.79081506729126,
|
|
"epoch": 4.8758057584873224,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00031736979678484634,
|
|
"loss": 5.4647,
|
|
"mean_token_accuracy": 0.18548725843429564,
|
|
"num_tokens": 10317549.0,
|
|
"step": 5675
|
|
},
|
|
{
|
|
"entropy": 5.7501555442810055,
|
|
"epoch": 4.880103137086377,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 0.00031704321613436597,
|
|
"loss": 5.4856,
|
|
"mean_token_accuracy": 0.1795574128627777,
|
|
"num_tokens": 10327681.0,
|
|
"step": 5680
|
|
},
|
|
{
|
|
"entropy": 5.669476127624511,
|
|
"epoch": 4.8844005156854315,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003167165436783061,
|
|
"loss": 5.4296,
|
|
"mean_token_accuracy": 0.18604628145694732,
|
|
"num_tokens": 10336261.0,
|
|
"step": 5685
|
|
},
|
|
{
|
|
"entropy": 5.669170475006103,
|
|
"epoch": 4.888697894284486,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00031638978012998875,
|
|
"loss": 5.3478,
|
|
"mean_token_accuracy": 0.20169953405857086,
|
|
"num_tokens": 10344770.0,
|
|
"step": 5690
|
|
},
|
|
{
|
|
"entropy": 5.780188274383545,
|
|
"epoch": 4.892995272883541,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000316062926202935,
|
|
"loss": 5.509,
|
|
"mean_token_accuracy": 0.175765922665596,
|
|
"num_tokens": 10354246.0,
|
|
"step": 5695
|
|
},
|
|
{
|
|
"entropy": 5.831387567520141,
|
|
"epoch": 4.897292651482596,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0003157359826108632,
|
|
"loss": 5.4178,
|
|
"mean_token_accuracy": 0.1851250186562538,
|
|
"num_tokens": 10362693.0,
|
|
"step": 5700
|
|
},
|
|
{
|
|
"entropy": 5.785976934432983,
|
|
"epoch": 4.90159003008165,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00031540895006768727,
|
|
"loss": 5.4205,
|
|
"mean_token_accuracy": 0.18571031391620635,
|
|
"num_tokens": 10371639.0,
|
|
"step": 5705
|
|
},
|
|
{
|
|
"entropy": 5.73499116897583,
|
|
"epoch": 4.905887408680705,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003150818292875158,
|
|
"loss": 5.444,
|
|
"mean_token_accuracy": 0.18135066330432892,
|
|
"num_tokens": 10381237.0,
|
|
"step": 5710
|
|
},
|
|
{
|
|
"entropy": 5.777145195007324,
|
|
"epoch": 4.91018478727976,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0003147546209846497,
|
|
"loss": 5.406,
|
|
"mean_token_accuracy": 0.18948156237602234,
|
|
"num_tokens": 10389932.0,
|
|
"step": 5715
|
|
},
|
|
{
|
|
"entropy": 5.695976591110229,
|
|
"epoch": 4.914482165878814,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003144273258735812,
|
|
"loss": 5.3064,
|
|
"mean_token_accuracy": 0.18776963353157045,
|
|
"num_tokens": 10398938.0,
|
|
"step": 5720
|
|
},
|
|
{
|
|
"entropy": 5.699533605575562,
|
|
"epoch": 4.918779544477869,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.0003140999446689919,
|
|
"loss": 5.4052,
|
|
"mean_token_accuracy": 0.18290529996156693,
|
|
"num_tokens": 10407980.0,
|
|
"step": 5725
|
|
},
|
|
{
|
|
"entropy": 5.685202360153198,
|
|
"epoch": 4.923076923076923,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0003137724780857516,
|
|
"loss": 5.4792,
|
|
"mean_token_accuracy": 0.1808568373322487,
|
|
"num_tokens": 10416990.0,
|
|
"step": 5730
|
|
},
|
|
{
|
|
"entropy": 5.84086389541626,
|
|
"epoch": 4.927374301675978,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00031344492683891634,
|
|
"loss": 5.4907,
|
|
"mean_token_accuracy": 0.18662243634462355,
|
|
"num_tokens": 10426573.0,
|
|
"step": 5735
|
|
},
|
|
{
|
|
"entropy": 5.8604381561279295,
|
|
"epoch": 4.931671680275032,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003131172916437272,
|
|
"loss": 5.4528,
|
|
"mean_token_accuracy": 0.185541370511055,
|
|
"num_tokens": 10435162.0,
|
|
"step": 5740
|
|
},
|
|
{
|
|
"entropy": 5.748629808425903,
|
|
"epoch": 4.935969058874087,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00031278957321560845,
|
|
"loss": 5.4482,
|
|
"mean_token_accuracy": 0.18387553095817566,
|
|
"num_tokens": 10444374.0,
|
|
"step": 5745
|
|
},
|
|
{
|
|
"entropy": 5.803906059265136,
|
|
"epoch": 4.9402664374731415,
|
|
"grad_norm": 2.84375,
|
|
"learning_rate": 0.00031246177227016615,
|
|
"loss": 5.4859,
|
|
"mean_token_accuracy": 0.1740569144487381,
|
|
"num_tokens": 10452679.0,
|
|
"step": 5750
|
|
},
|
|
{
|
|
"entropy": 5.744943141937256,
|
|
"epoch": 4.944563816072196,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 0.00031213388952318653,
|
|
"loss": 5.429,
|
|
"mean_token_accuracy": 0.19203304648399352,
|
|
"num_tokens": 10461801.0,
|
|
"step": 5755
|
|
},
|
|
{
|
|
"entropy": 5.779998779296875,
|
|
"epoch": 4.948861194671251,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0003118059256906345,
|
|
"loss": 5.4162,
|
|
"mean_token_accuracy": 0.1845451921224594,
|
|
"num_tokens": 10471176.0,
|
|
"step": 5760
|
|
},
|
|
{
|
|
"entropy": 5.792189311981201,
|
|
"epoch": 4.953158573270305,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00031147788148865204,
|
|
"loss": 5.4773,
|
|
"mean_token_accuracy": 0.17578926533460618,
|
|
"num_tokens": 10480403.0,
|
|
"step": 5765
|
|
},
|
|
{
|
|
"entropy": 5.775957775115967,
|
|
"epoch": 4.95745595186936,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0003111497576335564,
|
|
"loss": 5.4188,
|
|
"mean_token_accuracy": 0.1873456656932831,
|
|
"num_tokens": 10489574.0,
|
|
"step": 5770
|
|
},
|
|
{
|
|
"entropy": 5.725253582000732,
|
|
"epoch": 4.961753330468414,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003108215548418391,
|
|
"loss": 5.4295,
|
|
"mean_token_accuracy": 0.18676181882619858,
|
|
"num_tokens": 10499631.0,
|
|
"step": 5775
|
|
},
|
|
{
|
|
"entropy": 5.7812906265258786,
|
|
"epoch": 4.966050709067469,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0003104932738301637,
|
|
"loss": 5.3936,
|
|
"mean_token_accuracy": 0.1881558582186699,
|
|
"num_tokens": 10508128.0,
|
|
"step": 5780
|
|
},
|
|
{
|
|
"entropy": 5.755512762069702,
|
|
"epoch": 4.970348087666523,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00031016491531536477,
|
|
"loss": 5.3822,
|
|
"mean_token_accuracy": 0.18803726881742477,
|
|
"num_tokens": 10517544.0,
|
|
"step": 5785
|
|
},
|
|
{
|
|
"entropy": 5.69546070098877,
|
|
"epoch": 4.974645466265578,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003098364800144462,
|
|
"loss": 5.4411,
|
|
"mean_token_accuracy": 0.18596616983413697,
|
|
"num_tokens": 10526244.0,
|
|
"step": 5790
|
|
},
|
|
{
|
|
"entropy": 5.7579303741455075,
|
|
"epoch": 4.9789428448646325,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0003095079686445792,
|
|
"loss": 5.5269,
|
|
"mean_token_accuracy": 0.18171786069869994,
|
|
"num_tokens": 10535887.0,
|
|
"step": 5795
|
|
},
|
|
{
|
|
"entropy": 5.772639369964599,
|
|
"epoch": 4.983240223463687,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00030917938192310146,
|
|
"loss": 5.3508,
|
|
"mean_token_accuracy": 0.19144437909126283,
|
|
"num_tokens": 10544420.0,
|
|
"step": 5800
|
|
},
|
|
{
|
|
"entropy": 5.8085168361663815,
|
|
"epoch": 4.9875376020627415,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00030885072056751494,
|
|
"loss": 5.4523,
|
|
"mean_token_accuracy": 0.18296891897916795,
|
|
"num_tokens": 10553114.0,
|
|
"step": 5805
|
|
},
|
|
{
|
|
"entropy": 5.71706223487854,
|
|
"epoch": 4.991834980661796,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00030852198529548476,
|
|
"loss": 5.4556,
|
|
"mean_token_accuracy": 0.1879746973514557,
|
|
"num_tokens": 10562272.0,
|
|
"step": 5810
|
|
},
|
|
{
|
|
"entropy": 5.752150583267212,
|
|
"epoch": 4.996132359260851,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0003081931768248373,
|
|
"loss": 5.411,
|
|
"mean_token_accuracy": 0.18645633310079573,
|
|
"num_tokens": 10571757.0,
|
|
"step": 5815
|
|
},
|
|
{
|
|
"entropy": 5.764840761820476,
|
|
"epoch": 5.0,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 0.0003078642958735588,
|
|
"loss": 5.3869,
|
|
"mean_token_accuracy": 0.19194019668632084,
|
|
"num_tokens": 10579660.0,
|
|
"step": 5820
|
|
},
|
|
{
|
|
"entropy": 5.785556221008301,
|
|
"epoch": 5.0042973785990545,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00030753534315979393,
|
|
"loss": 5.2723,
|
|
"mean_token_accuracy": 0.1888723000884056,
|
|
"num_tokens": 10589139.0,
|
|
"step": 5825
|
|
},
|
|
{
|
|
"entropy": 5.71416163444519,
|
|
"epoch": 5.008594757198109,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0003072063194018438,
|
|
"loss": 5.0853,
|
|
"mean_token_accuracy": 0.20962987542152406,
|
|
"num_tokens": 10597915.0,
|
|
"step": 5830
|
|
},
|
|
{
|
|
"entropy": 5.735292434692383,
|
|
"epoch": 5.012892135797164,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0003068772253181648,
|
|
"loss": 5.2925,
|
|
"mean_token_accuracy": 0.18797790855169297,
|
|
"num_tokens": 10606491.0,
|
|
"step": 5835
|
|
},
|
|
{
|
|
"entropy": 5.7417901992797855,
|
|
"epoch": 5.017189514396218,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0003065480616273671,
|
|
"loss": 5.2905,
|
|
"mean_token_accuracy": 0.19306957423686982,
|
|
"num_tokens": 10615852.0,
|
|
"step": 5840
|
|
},
|
|
{
|
|
"entropy": 5.864093160629272,
|
|
"epoch": 5.021486892995273,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0003062188290482123,
|
|
"loss": 5.2635,
|
|
"mean_token_accuracy": 0.19216784834861755,
|
|
"num_tokens": 10625442.0,
|
|
"step": 5845
|
|
},
|
|
{
|
|
"entropy": 5.748987483978271,
|
|
"epoch": 5.025784271594327,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00030588952829961304,
|
|
"loss": 5.1911,
|
|
"mean_token_accuracy": 0.2049729496240616,
|
|
"num_tokens": 10634972.0,
|
|
"step": 5850
|
|
},
|
|
{
|
|
"entropy": 5.731730365753174,
|
|
"epoch": 5.030081650193382,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0003055601601006303,
|
|
"loss": 5.1795,
|
|
"mean_token_accuracy": 0.19543102532625198,
|
|
"num_tokens": 10644487.0,
|
|
"step": 5855
|
|
},
|
|
{
|
|
"entropy": 5.776778936386108,
|
|
"epoch": 5.034379028792436,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0003052307251704728,
|
|
"loss": 5.19,
|
|
"mean_token_accuracy": 0.21007055342197417,
|
|
"num_tokens": 10654144.0,
|
|
"step": 5860
|
|
},
|
|
{
|
|
"entropy": 5.730387496948242,
|
|
"epoch": 5.038676407391491,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.0003049012242284946,
|
|
"loss": 5.2343,
|
|
"mean_token_accuracy": 0.19403138458728791,
|
|
"num_tokens": 10663023.0,
|
|
"step": 5865
|
|
},
|
|
{
|
|
"entropy": 5.789040231704712,
|
|
"epoch": 5.0429737859905455,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0003045716579941941,
|
|
"loss": 5.3058,
|
|
"mean_token_accuracy": 0.19056892544031143,
|
|
"num_tokens": 10672001.0,
|
|
"step": 5870
|
|
},
|
|
{
|
|
"entropy": 5.744835710525512,
|
|
"epoch": 5.0472711645896,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00030424202718721215,
|
|
"loss": 5.2541,
|
|
"mean_token_accuracy": 0.18971439450979233,
|
|
"num_tokens": 10682654.0,
|
|
"step": 5875
|
|
},
|
|
{
|
|
"entropy": 5.76101336479187,
|
|
"epoch": 5.051568543188655,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00030391233252733085,
|
|
"loss": 5.2563,
|
|
"mean_token_accuracy": 0.19590554833412172,
|
|
"num_tokens": 10691429.0,
|
|
"step": 5880
|
|
},
|
|
{
|
|
"entropy": 5.715771722793579,
|
|
"epoch": 5.055865921787709,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00030358257473447144,
|
|
"loss": 5.2079,
|
|
"mean_token_accuracy": 0.19552415907382964,
|
|
"num_tokens": 10701130.0,
|
|
"step": 5885
|
|
},
|
|
{
|
|
"entropy": 5.7014076709747314,
|
|
"epoch": 5.060163300386764,
|
|
"grad_norm": 1.9375,
|
|
"learning_rate": 0.00030325275452869316,
|
|
"loss": 5.1993,
|
|
"mean_token_accuracy": 0.20109210014343262,
|
|
"num_tokens": 10709779.0,
|
|
"step": 5890
|
|
},
|
|
{
|
|
"entropy": 5.685995626449585,
|
|
"epoch": 5.064460678985819,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00030292287263019153,
|
|
"loss": 5.2412,
|
|
"mean_token_accuracy": 0.19837577044963836,
|
|
"num_tokens": 10718795.0,
|
|
"step": 5895
|
|
},
|
|
{
|
|
"entropy": 5.637307453155517,
|
|
"epoch": 5.068758057584874,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00030259292975929675,
|
|
"loss": 5.1354,
|
|
"mean_token_accuracy": 0.20832081735134125,
|
|
"num_tokens": 10728202.0,
|
|
"step": 5900
|
|
},
|
|
{
|
|
"entropy": 5.727624702453613,
|
|
"epoch": 5.073055436183928,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003022629266364723,
|
|
"loss": 5.1837,
|
|
"mean_token_accuracy": 0.19618202596902848,
|
|
"num_tokens": 10737050.0,
|
|
"step": 5905
|
|
},
|
|
{
|
|
"entropy": 5.761156225204468,
|
|
"epoch": 5.077352814782983,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00030193286398231276,
|
|
"loss": 5.206,
|
|
"mean_token_accuracy": 0.200938780605793,
|
|
"num_tokens": 10745261.0,
|
|
"step": 5910
|
|
},
|
|
{
|
|
"entropy": 5.73181471824646,
|
|
"epoch": 5.081650193382037,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00030160274251754337,
|
|
"loss": 5.2533,
|
|
"mean_token_accuracy": 0.19682066142559052,
|
|
"num_tokens": 10755008.0,
|
|
"step": 5915
|
|
},
|
|
{
|
|
"entropy": 5.698417139053345,
|
|
"epoch": 5.085947571981092,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00030127256296301724,
|
|
"loss": 5.294,
|
|
"mean_token_accuracy": 0.18680242151021959,
|
|
"num_tokens": 10763951.0,
|
|
"step": 5920
|
|
},
|
|
{
|
|
"entropy": 5.704818439483643,
|
|
"epoch": 5.090244950580146,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003009423260397148,
|
|
"loss": 5.2034,
|
|
"mean_token_accuracy": 0.19875070750713347,
|
|
"num_tokens": 10772770.0,
|
|
"step": 5925
|
|
},
|
|
{
|
|
"entropy": 5.784666728973389,
|
|
"epoch": 5.094542329179201,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00030061203246874125,
|
|
"loss": 5.2801,
|
|
"mean_token_accuracy": 0.18822978883981706,
|
|
"num_tokens": 10781827.0,
|
|
"step": 5930
|
|
},
|
|
{
|
|
"entropy": 5.80539083480835,
|
|
"epoch": 5.0988397077782555,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00030028168297132593,
|
|
"loss": 5.3681,
|
|
"mean_token_accuracy": 0.1860760048031807,
|
|
"num_tokens": 10792321.0,
|
|
"step": 5935
|
|
},
|
|
{
|
|
"entropy": 5.684858417510986,
|
|
"epoch": 5.10313708637731,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0002999512782688199,
|
|
"loss": 5.2929,
|
|
"mean_token_accuracy": 0.19473320841789246,
|
|
"num_tokens": 10801689.0,
|
|
"step": 5940
|
|
},
|
|
{
|
|
"entropy": 5.765773010253906,
|
|
"epoch": 5.1074344649763646,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002996208190826951,
|
|
"loss": 5.225,
|
|
"mean_token_accuracy": 0.19382953345775605,
|
|
"num_tokens": 10810513.0,
|
|
"step": 5945
|
|
},
|
|
{
|
|
"entropy": 5.7859851837158205,
|
|
"epoch": 5.111731843575419,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00029929030613454227,
|
|
"loss": 5.2035,
|
|
"mean_token_accuracy": 0.1984575793147087,
|
|
"num_tokens": 10819581.0,
|
|
"step": 5950
|
|
},
|
|
{
|
|
"entropy": 5.6832208156585695,
|
|
"epoch": 5.116029222174474,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002989597401460697,
|
|
"loss": 5.1767,
|
|
"mean_token_accuracy": 0.20104998648166655,
|
|
"num_tokens": 10828139.0,
|
|
"step": 5955
|
|
},
|
|
{
|
|
"entropy": 5.696868228912353,
|
|
"epoch": 5.120326600773528,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00029862912183910105,
|
|
"loss": 5.1862,
|
|
"mean_token_accuracy": 0.20635706037282944,
|
|
"num_tokens": 10836256.0,
|
|
"step": 5960
|
|
},
|
|
{
|
|
"entropy": 5.724944448471069,
|
|
"epoch": 5.124623979372583,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 0.00029829845193557496,
|
|
"loss": 5.2371,
|
|
"mean_token_accuracy": 0.19585344195365906,
|
|
"num_tokens": 10846255.0,
|
|
"step": 5965
|
|
},
|
|
{
|
|
"entropy": 5.724349784851074,
|
|
"epoch": 5.128921357971637,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0002979677311575421,
|
|
"loss": 5.2281,
|
|
"mean_token_accuracy": 0.19752541929483414,
|
|
"num_tokens": 10855546.0,
|
|
"step": 5970
|
|
},
|
|
{
|
|
"entropy": 5.740917587280274,
|
|
"epoch": 5.133218736570692,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002976369602271646,
|
|
"loss": 5.3118,
|
|
"mean_token_accuracy": 0.19228585809469223,
|
|
"num_tokens": 10864417.0,
|
|
"step": 5975
|
|
},
|
|
{
|
|
"entropy": 5.75031099319458,
|
|
"epoch": 5.137516115169746,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002973061398667138,
|
|
"loss": 5.2755,
|
|
"mean_token_accuracy": 0.19699454605579375,
|
|
"num_tokens": 10874527.0,
|
|
"step": 5980
|
|
},
|
|
{
|
|
"entropy": 5.768909645080567,
|
|
"epoch": 5.141813493768801,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00029697527079856916,
|
|
"loss": 5.2908,
|
|
"mean_token_accuracy": 0.19175970405340195,
|
|
"num_tokens": 10883486.0,
|
|
"step": 5985
|
|
},
|
|
{
|
|
"entropy": 5.711429071426392,
|
|
"epoch": 5.1461108723678555,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00029664435374521665,
|
|
"loss": 5.1684,
|
|
"mean_token_accuracy": 0.19666438549757004,
|
|
"num_tokens": 10891972.0,
|
|
"step": 5990
|
|
},
|
|
{
|
|
"entropy": 5.717523288726807,
|
|
"epoch": 5.15040825096691,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00029631338942924664,
|
|
"loss": 5.1882,
|
|
"mean_token_accuracy": 0.199525348842144,
|
|
"num_tokens": 10901350.0,
|
|
"step": 5995
|
|
},
|
|
{
|
|
"entropy": 5.716346025466919,
|
|
"epoch": 5.154705629565965,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0002959823785733531,
|
|
"loss": 5.1953,
|
|
"mean_token_accuracy": 0.1983650267124176,
|
|
"num_tokens": 10910114.0,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"epoch": 5.154705629565965,
|
|
"eval_entropy": 5.571174391755113,
|
|
"eval_loss": 6.033395290374756,
|
|
"eval_mean_token_accuracy": 0.16984019363047304,
|
|
"eval_num_tokens": 10910114.0,
|
|
"eval_runtime": 2.6526,
|
|
"eval_samples_per_second": 1337.912,
|
|
"eval_steps_per_second": 167.38,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"entropy": 5.702062702178955,
|
|
"epoch": 5.159003008165019,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.0002956513219003312,
|
|
"loss": 5.2696,
|
|
"mean_token_accuracy": 0.1978200137615204,
|
|
"num_tokens": 10919781.0,
|
|
"step": 6005
|
|
},
|
|
{
|
|
"entropy": 5.739422225952149,
|
|
"epoch": 5.163300386764074,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00029532022013307666,
|
|
"loss": 5.2667,
|
|
"mean_token_accuracy": 0.19581635147333146,
|
|
"num_tokens": 10929561.0,
|
|
"step": 6010
|
|
},
|
|
{
|
|
"entropy": 5.7579302310943605,
|
|
"epoch": 5.167597765363128,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00029498907399458325,
|
|
"loss": 5.2263,
|
|
"mean_token_accuracy": 0.1889701694250107,
|
|
"num_tokens": 10939123.0,
|
|
"step": 6015
|
|
},
|
|
{
|
|
"entropy": 5.73661470413208,
|
|
"epoch": 5.171895143962183,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0002946578842079418,
|
|
"loss": 5.2745,
|
|
"mean_token_accuracy": 0.19807808697223664,
|
|
"num_tokens": 10947990.0,
|
|
"step": 6020
|
|
},
|
|
{
|
|
"entropy": 5.7330670833587645,
|
|
"epoch": 5.176192522561237,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0002943266514963384,
|
|
"loss": 5.1971,
|
|
"mean_token_accuracy": 0.2052408739924431,
|
|
"num_tokens": 10956569.0,
|
|
"step": 6025
|
|
},
|
|
{
|
|
"entropy": 5.71516637802124,
|
|
"epoch": 5.180489901160292,
|
|
"grad_norm": 4.78125,
|
|
"learning_rate": 0.0002939953765830529,
|
|
"loss": 5.2139,
|
|
"mean_token_accuracy": 0.19366423934698104,
|
|
"num_tokens": 10965466.0,
|
|
"step": 6030
|
|
},
|
|
{
|
|
"entropy": 5.658834075927734,
|
|
"epoch": 5.184787279759346,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00029366406019145735,
|
|
"loss": 5.254,
|
|
"mean_token_accuracy": 0.19226128607988358,
|
|
"num_tokens": 10975051.0,
|
|
"step": 6035
|
|
},
|
|
{
|
|
"entropy": 5.669994163513183,
|
|
"epoch": 5.189084658358402,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0002933327030450143,
|
|
"loss": 5.159,
|
|
"mean_token_accuracy": 0.20337782204151153,
|
|
"num_tokens": 10983940.0,
|
|
"step": 6040
|
|
},
|
|
{
|
|
"entropy": 5.728955554962158,
|
|
"epoch": 5.193382036957456,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00029300130586727545,
|
|
"loss": 5.262,
|
|
"mean_token_accuracy": 0.18812055587768556,
|
|
"num_tokens": 10994029.0,
|
|
"step": 6045
|
|
},
|
|
{
|
|
"entropy": 5.793410778045654,
|
|
"epoch": 5.197679415556511,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00029266986938187943,
|
|
"loss": 5.3182,
|
|
"mean_token_accuracy": 0.18724069595336915,
|
|
"num_tokens": 11003616.0,
|
|
"step": 6050
|
|
},
|
|
{
|
|
"entropy": 5.708478021621704,
|
|
"epoch": 5.2019767941555655,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0002923383943125514,
|
|
"loss": 5.2041,
|
|
"mean_token_accuracy": 0.20091085880994797,
|
|
"num_tokens": 11012068.0,
|
|
"step": 6055
|
|
},
|
|
{
|
|
"entropy": 5.690085458755493,
|
|
"epoch": 5.20627417275462,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002920068813831002,
|
|
"loss": 5.2345,
|
|
"mean_token_accuracy": 0.19673667699098588,
|
|
"num_tokens": 11020510.0,
|
|
"step": 6060
|
|
},
|
|
{
|
|
"entropy": 5.6816661834716795,
|
|
"epoch": 5.210571551353675,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002916753313174178,
|
|
"loss": 5.2566,
|
|
"mean_token_accuracy": 0.19812606424093246,
|
|
"num_tokens": 11029804.0,
|
|
"step": 6065
|
|
},
|
|
{
|
|
"entropy": 5.73789005279541,
|
|
"epoch": 5.214868929952729,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0002913437448394768,
|
|
"loss": 5.2441,
|
|
"mean_token_accuracy": 0.2016511619091034,
|
|
"num_tokens": 11038586.0,
|
|
"step": 6070
|
|
},
|
|
{
|
|
"entropy": 5.804594278335571,
|
|
"epoch": 5.219166308551784,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00029101212267332955,
|
|
"loss": 5.339,
|
|
"mean_token_accuracy": 0.1902748167514801,
|
|
"num_tokens": 11048240.0,
|
|
"step": 6075
|
|
},
|
|
{
|
|
"entropy": 5.770946788787842,
|
|
"epoch": 5.223463687150838,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00029068046554310637,
|
|
"loss": 5.1789,
|
|
"mean_token_accuracy": 0.19486818015575408,
|
|
"num_tokens": 11056703.0,
|
|
"step": 6080
|
|
},
|
|
{
|
|
"entropy": 5.68377914428711,
|
|
"epoch": 5.227761065749893,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002903487741730139,
|
|
"loss": 5.2519,
|
|
"mean_token_accuracy": 0.1914640724658966,
|
|
"num_tokens": 11066246.0,
|
|
"step": 6085
|
|
},
|
|
{
|
|
"entropy": 5.7236161708831785,
|
|
"epoch": 5.232058444348947,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00029001704928733354,
|
|
"loss": 5.2868,
|
|
"mean_token_accuracy": 0.19090876579284669,
|
|
"num_tokens": 11075277.0,
|
|
"step": 6090
|
|
},
|
|
{
|
|
"entropy": 5.672372627258301,
|
|
"epoch": 5.236355822948002,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002896852916104198,
|
|
"loss": 5.1872,
|
|
"mean_token_accuracy": 0.20404913723468782,
|
|
"num_tokens": 11083759.0,
|
|
"step": 6095
|
|
},
|
|
{
|
|
"entropy": 5.736728668212891,
|
|
"epoch": 5.240653201547056,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002893535018666988,
|
|
"loss": 5.158,
|
|
"mean_token_accuracy": 0.20445130616426468,
|
|
"num_tokens": 11091960.0,
|
|
"step": 6100
|
|
},
|
|
{
|
|
"entropy": 5.68701286315918,
|
|
"epoch": 5.244950580146111,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00028902168078066674,
|
|
"loss": 5.2631,
|
|
"mean_token_accuracy": 0.1987990543246269,
|
|
"num_tokens": 11101134.0,
|
|
"step": 6105
|
|
},
|
|
{
|
|
"entropy": 5.725564765930176,
|
|
"epoch": 5.2492479587451655,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0002886898290768883,
|
|
"loss": 5.2126,
|
|
"mean_token_accuracy": 0.1973779931664467,
|
|
"num_tokens": 11110282.0,
|
|
"step": 6110
|
|
},
|
|
{
|
|
"entropy": 5.766754579544068,
|
|
"epoch": 5.25354533734422,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.000288357947479995,
|
|
"loss": 5.2893,
|
|
"mean_token_accuracy": 0.19814405888319014,
|
|
"num_tokens": 11119591.0,
|
|
"step": 6115
|
|
},
|
|
{
|
|
"entropy": 5.725649309158325,
|
|
"epoch": 5.257842715943275,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00028802603671468347,
|
|
"loss": 5.292,
|
|
"mean_token_accuracy": 0.1955984503030777,
|
|
"num_tokens": 11129164.0,
|
|
"step": 6120
|
|
},
|
|
{
|
|
"entropy": 5.711507415771484,
|
|
"epoch": 5.262140094542329,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00028769409750571413,
|
|
"loss": 5.2501,
|
|
"mean_token_accuracy": 0.19035946130752562,
|
|
"num_tokens": 11137973.0,
|
|
"step": 6125
|
|
},
|
|
{
|
|
"entropy": 5.727103424072266,
|
|
"epoch": 5.266437473141384,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00028736213057790975,
|
|
"loss": 5.3404,
|
|
"mean_token_accuracy": 0.19213739782571793,
|
|
"num_tokens": 11147285.0,
|
|
"step": 6130
|
|
},
|
|
{
|
|
"entropy": 5.809697961807251,
|
|
"epoch": 5.270734851740438,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002870301366561533,
|
|
"loss": 5.2992,
|
|
"mean_token_accuracy": 0.19491042643785478,
|
|
"num_tokens": 11155303.0,
|
|
"step": 6135
|
|
},
|
|
{
|
|
"entropy": 5.717234706878662,
|
|
"epoch": 5.275032230339493,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002866981164653867,
|
|
"loss": 5.1685,
|
|
"mean_token_accuracy": 0.20516792982816695,
|
|
"num_tokens": 11163553.0,
|
|
"step": 6140
|
|
},
|
|
{
|
|
"entropy": 5.643297719955444,
|
|
"epoch": 5.279329608938547,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0002863660707306095,
|
|
"loss": 5.1151,
|
|
"mean_token_accuracy": 0.21079060733318328,
|
|
"num_tokens": 11171865.0,
|
|
"step": 6145
|
|
},
|
|
{
|
|
"entropy": 5.732720136642456,
|
|
"epoch": 5.283626987537602,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.00028603400017687675,
|
|
"loss": 5.3085,
|
|
"mean_token_accuracy": 0.190241639316082,
|
|
"num_tokens": 11181137.0,
|
|
"step": 6150
|
|
},
|
|
{
|
|
"entropy": 5.617628002166748,
|
|
"epoch": 5.287924366136656,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00028570190552929794,
|
|
"loss": 5.1681,
|
|
"mean_token_accuracy": 0.21120553612709045,
|
|
"num_tokens": 11190174.0,
|
|
"step": 6155
|
|
},
|
|
{
|
|
"entropy": 5.669597339630127,
|
|
"epoch": 5.292221744735711,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.000285369787513035,
|
|
"loss": 5.1863,
|
|
"mean_token_accuracy": 0.20306089520454407,
|
|
"num_tokens": 11197964.0,
|
|
"step": 6160
|
|
},
|
|
{
|
|
"entropy": 5.778067970275879,
|
|
"epoch": 5.2965191233347655,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00028503764685330077,
|
|
"loss": 5.3219,
|
|
"mean_token_accuracy": 0.1941687807440758,
|
|
"num_tokens": 11207974.0,
|
|
"step": 6165
|
|
},
|
|
{
|
|
"entropy": 5.71546049118042,
|
|
"epoch": 5.30081650193382,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.00028470548427535794,
|
|
"loss": 5.2274,
|
|
"mean_token_accuracy": 0.19352466911077498,
|
|
"num_tokens": 11216430.0,
|
|
"step": 6170
|
|
},
|
|
{
|
|
"entropy": 5.773167419433594,
|
|
"epoch": 5.305113880532875,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00028437330050451654,
|
|
"loss": 5.3277,
|
|
"mean_token_accuracy": 0.18593378067016603,
|
|
"num_tokens": 11226189.0,
|
|
"step": 6175
|
|
},
|
|
{
|
|
"entropy": 5.747695207595825,
|
|
"epoch": 5.309411259131929,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0002840410962661334,
|
|
"loss": 5.2792,
|
|
"mean_token_accuracy": 0.19144221097230912,
|
|
"num_tokens": 11234691.0,
|
|
"step": 6180
|
|
},
|
|
{
|
|
"entropy": 5.636720132827759,
|
|
"epoch": 5.313708637730985,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 0.0002837088722856098,
|
|
"loss": 5.1319,
|
|
"mean_token_accuracy": 0.2054472416639328,
|
|
"num_tokens": 11243852.0,
|
|
"step": 6185
|
|
},
|
|
{
|
|
"entropy": 5.7381391525268555,
|
|
"epoch": 5.318006016330038,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00028337662928838996,
|
|
"loss": 5.2967,
|
|
"mean_token_accuracy": 0.19290618300437928,
|
|
"num_tokens": 11253416.0,
|
|
"step": 6190
|
|
},
|
|
{
|
|
"entropy": 5.684747648239136,
|
|
"epoch": 5.322303394929094,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00028304436799995986,
|
|
"loss": 5.1978,
|
|
"mean_token_accuracy": 0.20399291664361954,
|
|
"num_tokens": 11262869.0,
|
|
"step": 6195
|
|
},
|
|
{
|
|
"entropy": 5.697920131683349,
|
|
"epoch": 5.326600773528148,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00028271208914584534,
|
|
"loss": 5.3173,
|
|
"mean_token_accuracy": 0.18693045526742935,
|
|
"num_tokens": 11272386.0,
|
|
"step": 6200
|
|
},
|
|
{
|
|
"entropy": 5.736681890487671,
|
|
"epoch": 5.330898152127203,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00028237979345161065,
|
|
"loss": 5.1712,
|
|
"mean_token_accuracy": 0.20304832458496094,
|
|
"num_tokens": 11281590.0,
|
|
"step": 6205
|
|
},
|
|
{
|
|
"entropy": 5.670126008987427,
|
|
"epoch": 5.335195530726257,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0002820474816428568,
|
|
"loss": 5.2276,
|
|
"mean_token_accuracy": 0.19725264012813568,
|
|
"num_tokens": 11290873.0,
|
|
"step": 6210
|
|
},
|
|
{
|
|
"entropy": 5.694404888153076,
|
|
"epoch": 5.339492909325312,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002817151544452198,
|
|
"loss": 5.1903,
|
|
"mean_token_accuracy": 0.19650624096393585,
|
|
"num_tokens": 11299064.0,
|
|
"step": 6215
|
|
},
|
|
{
|
|
"entropy": 5.680583715438843,
|
|
"epoch": 5.343790287924366,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00028138281258436947,
|
|
"loss": 5.2412,
|
|
"mean_token_accuracy": 0.19210307002067567,
|
|
"num_tokens": 11307390.0,
|
|
"step": 6220
|
|
},
|
|
{
|
|
"entropy": 5.622157335281372,
|
|
"epoch": 5.348087666523421,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0002810504567860078,
|
|
"loss": 5.18,
|
|
"mean_token_accuracy": 0.20140269696712493,
|
|
"num_tokens": 11315606.0,
|
|
"step": 6225
|
|
},
|
|
{
|
|
"entropy": 5.79349045753479,
|
|
"epoch": 5.3523850451224755,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0002807180877758667,
|
|
"loss": 5.3228,
|
|
"mean_token_accuracy": 0.18875322341918946,
|
|
"num_tokens": 11323821.0,
|
|
"step": 6230
|
|
},
|
|
{
|
|
"entropy": 5.741941308975219,
|
|
"epoch": 5.35668242372153,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00028038570627970754,
|
|
"loss": 5.253,
|
|
"mean_token_accuracy": 0.1991521954536438,
|
|
"num_tokens": 11331850.0,
|
|
"step": 6235
|
|
},
|
|
{
|
|
"entropy": 5.649916315078736,
|
|
"epoch": 5.360979802320585,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0002800533130233184,
|
|
"loss": 5.1923,
|
|
"mean_token_accuracy": 0.19856758862733842,
|
|
"num_tokens": 11340125.0,
|
|
"step": 6240
|
|
},
|
|
{
|
|
"entropy": 5.73130464553833,
|
|
"epoch": 5.365277180919639,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0002797209087325135,
|
|
"loss": 5.2697,
|
|
"mean_token_accuracy": 0.19176995605230332,
|
|
"num_tokens": 11349184.0,
|
|
"step": 6245
|
|
},
|
|
{
|
|
"entropy": 5.804795598983764,
|
|
"epoch": 5.369574559518694,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00027938849413313083,
|
|
"loss": 5.2922,
|
|
"mean_token_accuracy": 0.19152362793684005,
|
|
"num_tokens": 11357536.0,
|
|
"step": 6250
|
|
},
|
|
{
|
|
"entropy": 5.729605102539063,
|
|
"epoch": 5.373871938117748,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000279056069951031,
|
|
"loss": 5.2837,
|
|
"mean_token_accuracy": 0.1947401210665703,
|
|
"num_tokens": 11367242.0,
|
|
"step": 6255
|
|
},
|
|
{
|
|
"entropy": 5.723223495483398,
|
|
"epoch": 5.378169316716803,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00027872363691209564,
|
|
"loss": 5.2244,
|
|
"mean_token_accuracy": 0.19894828796386718,
|
|
"num_tokens": 11374932.0,
|
|
"step": 6260
|
|
},
|
|
{
|
|
"entropy": 5.682012367248535,
|
|
"epoch": 5.382466695315857,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002783911957422256,
|
|
"loss": 5.2444,
|
|
"mean_token_accuracy": 0.1929147273302078,
|
|
"num_tokens": 11383575.0,
|
|
"step": 6265
|
|
},
|
|
{
|
|
"entropy": 5.724541664123535,
|
|
"epoch": 5.386764073914912,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0002780587471673394,
|
|
"loss": 5.2721,
|
|
"mean_token_accuracy": 0.1960109919309616,
|
|
"num_tokens": 11392285.0,
|
|
"step": 6270
|
|
},
|
|
{
|
|
"entropy": 5.722800588607788,
|
|
"epoch": 5.391061452513966,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00027772629191337206,
|
|
"loss": 5.2558,
|
|
"mean_token_accuracy": 0.19811440408229827,
|
|
"num_tokens": 11401054.0,
|
|
"step": 6275
|
|
},
|
|
{
|
|
"entropy": 5.670466518402099,
|
|
"epoch": 5.395358831113021,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00027739383070627283,
|
|
"loss": 5.2586,
|
|
"mean_token_accuracy": 0.19449169486761092,
|
|
"num_tokens": 11410529.0,
|
|
"step": 6280
|
|
},
|
|
{
|
|
"entropy": 5.684411239624024,
|
|
"epoch": 5.3996562097120755,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002770613642720041,
|
|
"loss": 5.2052,
|
|
"mean_token_accuracy": 0.19831081479787827,
|
|
"num_tokens": 11419961.0,
|
|
"step": 6285
|
|
},
|
|
{
|
|
"entropy": 5.760186433792114,
|
|
"epoch": 5.40395358831113,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00027672889333653984,
|
|
"loss": 5.3472,
|
|
"mean_token_accuracy": 0.18055398017168045,
|
|
"num_tokens": 11429529.0,
|
|
"step": 6290
|
|
},
|
|
{
|
|
"entropy": 5.723178720474243,
|
|
"epoch": 5.408250966910185,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002763964186258635,
|
|
"loss": 5.216,
|
|
"mean_token_accuracy": 0.19768301248550416,
|
|
"num_tokens": 11438254.0,
|
|
"step": 6295
|
|
},
|
|
{
|
|
"entropy": 5.726367521286011,
|
|
"epoch": 5.412548345509239,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002760639408659671,
|
|
"loss": 5.2741,
|
|
"mean_token_accuracy": 0.18960459232330323,
|
|
"num_tokens": 11447587.0,
|
|
"step": 6300
|
|
},
|
|
{
|
|
"entropy": 5.681859922409058,
|
|
"epoch": 5.416845724108294,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.0002757314607828489,
|
|
"loss": 5.2297,
|
|
"mean_token_accuracy": 0.19529251903295516,
|
|
"num_tokens": 11455493.0,
|
|
"step": 6305
|
|
},
|
|
{
|
|
"entropy": 5.685959196090698,
|
|
"epoch": 5.421143102707348,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00027539897910251293,
|
|
"loss": 5.2063,
|
|
"mean_token_accuracy": 0.20866367518901824,
|
|
"num_tokens": 11464143.0,
|
|
"step": 6310
|
|
},
|
|
{
|
|
"entropy": 5.700673913955688,
|
|
"epoch": 5.425440481306403,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00027506649655096595,
|
|
"loss": 5.1591,
|
|
"mean_token_accuracy": 0.1953020215034485,
|
|
"num_tokens": 11471813.0,
|
|
"step": 6315
|
|
},
|
|
{
|
|
"entropy": 5.678422117233277,
|
|
"epoch": 5.429737859905457,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0002747340138542171,
|
|
"loss": 5.2241,
|
|
"mean_token_accuracy": 0.1963564544916153,
|
|
"num_tokens": 11481374.0,
|
|
"step": 6320
|
|
},
|
|
{
|
|
"entropy": 5.763215446472168,
|
|
"epoch": 5.434035238504512,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002744015317382757,
|
|
"loss": 5.3604,
|
|
"mean_token_accuracy": 0.19148044437170028,
|
|
"num_tokens": 11490575.0,
|
|
"step": 6325
|
|
},
|
|
{
|
|
"entropy": 5.7217549800872805,
|
|
"epoch": 5.438332617103566,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0002740690509291498,
|
|
"loss": 5.2576,
|
|
"mean_token_accuracy": 0.19364129453897477,
|
|
"num_tokens": 11499898.0,
|
|
"step": 6330
|
|
},
|
|
{
|
|
"entropy": 5.695327615737915,
|
|
"epoch": 5.442629995702621,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0002737365721528445,
|
|
"loss": 5.2442,
|
|
"mean_token_accuracy": 0.18990284204483032,
|
|
"num_tokens": 11508544.0,
|
|
"step": 6335
|
|
},
|
|
{
|
|
"entropy": 5.698026847839356,
|
|
"epoch": 5.446927374301676,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0002734040961353607,
|
|
"loss": 5.2541,
|
|
"mean_token_accuracy": 0.19756245464086533,
|
|
"num_tokens": 11519239.0,
|
|
"step": 6340
|
|
},
|
|
{
|
|
"entropy": 5.772835683822632,
|
|
"epoch": 5.451224752900731,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000273071623602693,
|
|
"loss": 5.2064,
|
|
"mean_token_accuracy": 0.2003451868891716,
|
|
"num_tokens": 11529014.0,
|
|
"step": 6345
|
|
},
|
|
{
|
|
"entropy": 5.7379475116729735,
|
|
"epoch": 5.4555221314997855,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00027273915528082865,
|
|
"loss": 5.2209,
|
|
"mean_token_accuracy": 0.1987021416425705,
|
|
"num_tokens": 11538367.0,
|
|
"step": 6350
|
|
},
|
|
{
|
|
"entropy": 5.7248560905456545,
|
|
"epoch": 5.45981951009884,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002724066918957455,
|
|
"loss": 5.3718,
|
|
"mean_token_accuracy": 0.17996017038822174,
|
|
"num_tokens": 11548166.0,
|
|
"step": 6355
|
|
},
|
|
{
|
|
"entropy": 5.611713695526123,
|
|
"epoch": 5.464116888697895,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002720742341734107,
|
|
"loss": 5.2309,
|
|
"mean_token_accuracy": 0.20304029881954194,
|
|
"num_tokens": 11557187.0,
|
|
"step": 6360
|
|
},
|
|
{
|
|
"entropy": 5.764591264724731,
|
|
"epoch": 5.468414267296949,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00027174178283977904,
|
|
"loss": 5.2767,
|
|
"mean_token_accuracy": 0.19075549989938737,
|
|
"num_tokens": 11566181.0,
|
|
"step": 6365
|
|
},
|
|
{
|
|
"entropy": 5.683410120010376,
|
|
"epoch": 5.472711645896004,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00027140933862079136,
|
|
"loss": 5.2482,
|
|
"mean_token_accuracy": 0.19976761490106582,
|
|
"num_tokens": 11576157.0,
|
|
"step": 6370
|
|
},
|
|
{
|
|
"entropy": 5.6435911655426025,
|
|
"epoch": 5.477009024495058,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.000271076902242373,
|
|
"loss": 5.2048,
|
|
"mean_token_accuracy": 0.19636543542146684,
|
|
"num_tokens": 11585325.0,
|
|
"step": 6375
|
|
},
|
|
{
|
|
"entropy": 5.726132154464722,
|
|
"epoch": 5.481306403094113,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000270744474430432,
|
|
"loss": 5.1822,
|
|
"mean_token_accuracy": 0.20143266022205353,
|
|
"num_tokens": 11594623.0,
|
|
"step": 6380
|
|
},
|
|
{
|
|
"entropy": 5.789005994796753,
|
|
"epoch": 5.485603781693167,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000270412055910858,
|
|
"loss": 5.3679,
|
|
"mean_token_accuracy": 0.18628330677747726,
|
|
"num_tokens": 11604370.0,
|
|
"step": 6385
|
|
},
|
|
{
|
|
"entropy": 5.601118278503418,
|
|
"epoch": 5.489901160292222,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 0.0002700796474095201,
|
|
"loss": 5.174,
|
|
"mean_token_accuracy": 0.2114277422428131,
|
|
"num_tokens": 11613779.0,
|
|
"step": 6390
|
|
},
|
|
{
|
|
"entropy": 5.701408767700196,
|
|
"epoch": 5.494198538891276,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0002697472496522656,
|
|
"loss": 5.2398,
|
|
"mean_token_accuracy": 0.1997276723384857,
|
|
"num_tokens": 11623037.0,
|
|
"step": 6395
|
|
},
|
|
{
|
|
"entropy": 5.770825958251953,
|
|
"epoch": 5.498495917490331,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002694148633649184,
|
|
"loss": 5.2812,
|
|
"mean_token_accuracy": 0.19196936190128328,
|
|
"num_tokens": 11631640.0,
|
|
"step": 6400
|
|
},
|
|
{
|
|
"entropy": 5.708783435821533,
|
|
"epoch": 5.5027932960893855,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002690824892732772,
|
|
"loss": 5.2365,
|
|
"mean_token_accuracy": 0.19937683492898942,
|
|
"num_tokens": 11640500.0,
|
|
"step": 6405
|
|
},
|
|
{
|
|
"entropy": 5.749935626983643,
|
|
"epoch": 5.50709067468844,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002687501281031142,
|
|
"loss": 5.3059,
|
|
"mean_token_accuracy": 0.1923367291688919,
|
|
"num_tokens": 11649173.0,
|
|
"step": 6410
|
|
},
|
|
{
|
|
"entropy": 5.584640550613403,
|
|
"epoch": 5.511388053287495,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0002684177805801734,
|
|
"loss": 5.1255,
|
|
"mean_token_accuracy": 0.2054922863841057,
|
|
"num_tokens": 11658808.0,
|
|
"step": 6415
|
|
},
|
|
{
|
|
"entropy": 5.663260555267334,
|
|
"epoch": 5.515685431886549,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.00026808544743016886,
|
|
"loss": 5.2359,
|
|
"mean_token_accuracy": 0.19006944745779036,
|
|
"num_tokens": 11667600.0,
|
|
"step": 6420
|
|
},
|
|
{
|
|
"entropy": 5.676917171478271,
|
|
"epoch": 5.519982810485604,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002677531293787835,
|
|
"loss": 5.2539,
|
|
"mean_token_accuracy": 0.19116365760564805,
|
|
"num_tokens": 11675597.0,
|
|
"step": 6425
|
|
},
|
|
{
|
|
"entropy": 5.591699838638306,
|
|
"epoch": 5.524280189084658,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.000267420827151667,
|
|
"loss": 5.1704,
|
|
"mean_token_accuracy": 0.20831066220998765,
|
|
"num_tokens": 11684549.0,
|
|
"step": 6430
|
|
},
|
|
{
|
|
"entropy": 5.692930507659912,
|
|
"epoch": 5.528577567683713,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0002670885414744347,
|
|
"loss": 5.2529,
|
|
"mean_token_accuracy": 0.1958039253950119,
|
|
"num_tokens": 11693043.0,
|
|
"step": 6435
|
|
},
|
|
{
|
|
"entropy": 5.777565622329712,
|
|
"epoch": 5.532874946282767,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.0002667562730726655,
|
|
"loss": 5.3573,
|
|
"mean_token_accuracy": 0.1849025934934616,
|
|
"num_tokens": 11702982.0,
|
|
"step": 6440
|
|
},
|
|
{
|
|
"entropy": 5.767904281616211,
|
|
"epoch": 5.537172324881822,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00026642402267190095,
|
|
"loss": 5.3544,
|
|
"mean_token_accuracy": 0.18599240332841874,
|
|
"num_tokens": 11711994.0,
|
|
"step": 6445
|
|
},
|
|
{
|
|
"entropy": 5.749307918548584,
|
|
"epoch": 5.541469703480876,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00026609179099764313,
|
|
"loss": 5.2802,
|
|
"mean_token_accuracy": 0.19736941754817963,
|
|
"num_tokens": 11722165.0,
|
|
"step": 6450
|
|
},
|
|
{
|
|
"entropy": 5.723725461959839,
|
|
"epoch": 5.545767082079931,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00026575957877535323,
|
|
"loss": 5.2594,
|
|
"mean_token_accuracy": 0.19194465279579162,
|
|
"num_tokens": 11731265.0,
|
|
"step": 6455
|
|
},
|
|
{
|
|
"entropy": 5.768704509735107,
|
|
"epoch": 5.5500644606789855,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00026542738673044985,
|
|
"loss": 5.2858,
|
|
"mean_token_accuracy": 0.19620778411626816,
|
|
"num_tokens": 11741779.0,
|
|
"step": 6460
|
|
},
|
|
{
|
|
"entropy": 5.661592578887939,
|
|
"epoch": 5.55436183927804,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.0002650952155883077,
|
|
"loss": 5.2482,
|
|
"mean_token_accuracy": 0.1982031360268593,
|
|
"num_tokens": 11749976.0,
|
|
"step": 6465
|
|
},
|
|
{
|
|
"entropy": 5.658786153793335,
|
|
"epoch": 5.558659217877095,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002647630660742559,
|
|
"loss": 5.2445,
|
|
"mean_token_accuracy": 0.19488973617553712,
|
|
"num_tokens": 11759781.0,
|
|
"step": 6470
|
|
},
|
|
{
|
|
"entropy": 5.690324926376343,
|
|
"epoch": 5.56295659647615,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000264430938913576,
|
|
"loss": 5.2388,
|
|
"mean_token_accuracy": 0.1998610481619835,
|
|
"num_tokens": 11769544.0,
|
|
"step": 6475
|
|
},
|
|
{
|
|
"entropy": 5.733021879196167,
|
|
"epoch": 5.567253975075204,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00026409883483150123,
|
|
"loss": 5.2195,
|
|
"mean_token_accuracy": 0.19633202105760575,
|
|
"num_tokens": 11778831.0,
|
|
"step": 6480
|
|
},
|
|
{
|
|
"entropy": 5.6924807071685795,
|
|
"epoch": 5.571551353674259,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.000263766754553214,
|
|
"loss": 5.2959,
|
|
"mean_token_accuracy": 0.19168850630521775,
|
|
"num_tokens": 11788813.0,
|
|
"step": 6485
|
|
},
|
|
{
|
|
"entropy": 5.756376934051514,
|
|
"epoch": 5.575848732273313,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002634346988038448,
|
|
"loss": 5.2391,
|
|
"mean_token_accuracy": 0.19819915741682054,
|
|
"num_tokens": 11797335.0,
|
|
"step": 6490
|
|
},
|
|
{
|
|
"entropy": 5.7295080661773685,
|
|
"epoch": 5.580146110872368,
|
|
"grad_norm": 2.71875,
|
|
"learning_rate": 0.00026310266830847093,
|
|
"loss": 5.2491,
|
|
"mean_token_accuracy": 0.19368611127138138,
|
|
"num_tokens": 11806741.0,
|
|
"step": 6495
|
|
},
|
|
{
|
|
"entropy": 5.73190712928772,
|
|
"epoch": 5.584443489471423,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00026277066379211406,
|
|
"loss": 5.2571,
|
|
"mean_token_accuracy": 0.1975494831800461,
|
|
"num_tokens": 11815551.0,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"epoch": 5.584443489471423,
|
|
"eval_entropy": 5.543760545081921,
|
|
"eval_loss": 6.006871223449707,
|
|
"eval_mean_token_accuracy": 0.17120395820621434,
|
|
"eval_num_tokens": 11815551.0,
|
|
"eval_runtime": 2.6431,
|
|
"eval_samples_per_second": 1342.757,
|
|
"eval_steps_per_second": 167.987,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"entropy": 5.727163648605346,
|
|
"epoch": 5.588740868070477,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0002624386859797396,
|
|
"loss": 5.3018,
|
|
"mean_token_accuracy": 0.1953060582280159,
|
|
"num_tokens": 11824483.0,
|
|
"step": 6505
|
|
},
|
|
{
|
|
"entropy": 5.640713262557983,
|
|
"epoch": 5.593038246669532,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00026210673559625406,
|
|
"loss": 5.1064,
|
|
"mean_token_accuracy": 0.2080324560403824,
|
|
"num_tokens": 11832383.0,
|
|
"step": 6510
|
|
},
|
|
{
|
|
"entropy": 5.7270731925964355,
|
|
"epoch": 5.597335625268586,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.0002617748133665047,
|
|
"loss": 5.3426,
|
|
"mean_token_accuracy": 0.19230563640594484,
|
|
"num_tokens": 11841430.0,
|
|
"step": 6515
|
|
},
|
|
{
|
|
"entropy": 5.656913375854492,
|
|
"epoch": 5.601633003867641,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0002614429200152768,
|
|
"loss": 5.3195,
|
|
"mean_token_accuracy": 0.18978479653596877,
|
|
"num_tokens": 11850863.0,
|
|
"step": 6520
|
|
},
|
|
{
|
|
"entropy": 5.669814395904541,
|
|
"epoch": 5.6059303824666955,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000261111056267293,
|
|
"loss": 5.1364,
|
|
"mean_token_accuracy": 0.2016347110271454,
|
|
"num_tokens": 11859392.0,
|
|
"step": 6525
|
|
},
|
|
{
|
|
"entropy": 5.727662467956543,
|
|
"epoch": 5.61022776106575,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00026077922284721084,
|
|
"loss": 5.21,
|
|
"mean_token_accuracy": 0.20116600394248962,
|
|
"num_tokens": 11868762.0,
|
|
"step": 6530
|
|
},
|
|
{
|
|
"entropy": 5.709690427780151,
|
|
"epoch": 5.614525139664805,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00026044742047962206,
|
|
"loss": 5.1777,
|
|
"mean_token_accuracy": 0.20647604763507843,
|
|
"num_tokens": 11876722.0,
|
|
"step": 6535
|
|
},
|
|
{
|
|
"entropy": 5.696219062805175,
|
|
"epoch": 5.618822518263859,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00026011564988905023,
|
|
"loss": 5.31,
|
|
"mean_token_accuracy": 0.19389723390340804,
|
|
"num_tokens": 11885614.0,
|
|
"step": 6540
|
|
},
|
|
{
|
|
"entropy": 5.784447908401489,
|
|
"epoch": 5.623119896862914,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0002597839117999499,
|
|
"loss": 5.3272,
|
|
"mean_token_accuracy": 0.1921556770801544,
|
|
"num_tokens": 11894702.0,
|
|
"step": 6545
|
|
},
|
|
{
|
|
"entropy": 5.646423387527466,
|
|
"epoch": 5.627417275461968,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0002594522069367044,
|
|
"loss": 5.2023,
|
|
"mean_token_accuracy": 0.20706268697977065,
|
|
"num_tokens": 11902829.0,
|
|
"step": 6550
|
|
},
|
|
{
|
|
"entropy": 5.686774587631225,
|
|
"epoch": 5.631714654061023,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002591205360236245,
|
|
"loss": 5.2397,
|
|
"mean_token_accuracy": 0.19559515863656998,
|
|
"num_tokens": 11912377.0,
|
|
"step": 6555
|
|
},
|
|
{
|
|
"entropy": 5.7376320362091064,
|
|
"epoch": 5.636012032660077,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.000258788899784947,
|
|
"loss": 5.2461,
|
|
"mean_token_accuracy": 0.19445491582155228,
|
|
"num_tokens": 11920563.0,
|
|
"step": 6560
|
|
},
|
|
{
|
|
"entropy": 5.720705032348633,
|
|
"epoch": 5.640309411259132,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00025845729894483283,
|
|
"loss": 5.2787,
|
|
"mean_token_accuracy": 0.19352309852838517,
|
|
"num_tokens": 11930190.0,
|
|
"step": 6565
|
|
},
|
|
{
|
|
"entropy": 5.6869124412536625,
|
|
"epoch": 5.644606789858186,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002581257342273657,
|
|
"loss": 5.3306,
|
|
"mean_token_accuracy": 0.19242538064718245,
|
|
"num_tokens": 11939840.0,
|
|
"step": 6570
|
|
},
|
|
{
|
|
"entropy": 5.738491201400757,
|
|
"epoch": 5.648904168457241,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002577942063565504,
|
|
"loss": 5.2953,
|
|
"mean_token_accuracy": 0.1939183861017227,
|
|
"num_tokens": 11948260.0,
|
|
"step": 6575
|
|
},
|
|
{
|
|
"entropy": 5.74149341583252,
|
|
"epoch": 5.6532015470562955,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0002574627160563114,
|
|
"loss": 5.3042,
|
|
"mean_token_accuracy": 0.1983557492494583,
|
|
"num_tokens": 11956776.0,
|
|
"step": 6580
|
|
},
|
|
{
|
|
"entropy": 5.780646514892578,
|
|
"epoch": 5.65749892565535,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002571312640504909,
|
|
"loss": 5.3362,
|
|
"mean_token_accuracy": 0.19133733510971068,
|
|
"num_tokens": 11966375.0,
|
|
"step": 6585
|
|
},
|
|
{
|
|
"entropy": 5.814681768417358,
|
|
"epoch": 5.661796304254405,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002567998510628476,
|
|
"loss": 5.3593,
|
|
"mean_token_accuracy": 0.19273963570594788,
|
|
"num_tokens": 11975835.0,
|
|
"step": 6590
|
|
},
|
|
{
|
|
"entropy": 5.68937611579895,
|
|
"epoch": 5.666093682853459,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00025646847781705506,
|
|
"loss": 5.2361,
|
|
"mean_token_accuracy": 0.20095739960670472,
|
|
"num_tokens": 11984672.0,
|
|
"step": 6595
|
|
},
|
|
{
|
|
"entropy": 5.7210855960845945,
|
|
"epoch": 5.670391061452514,
|
|
"grad_norm": 1.8203125,
|
|
"learning_rate": 0.0002561371450367,
|
|
"loss": 5.2528,
|
|
"mean_token_accuracy": 0.19773562699556352,
|
|
"num_tokens": 11993954.0,
|
|
"step": 6600
|
|
},
|
|
{
|
|
"entropy": 5.746278858184814,
|
|
"epoch": 5.674688440051568,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00025580585344528076,
|
|
"loss": 5.3181,
|
|
"mean_token_accuracy": 0.18820740133523942,
|
|
"num_tokens": 12002523.0,
|
|
"step": 6605
|
|
},
|
|
{
|
|
"entropy": 5.721010971069336,
|
|
"epoch": 5.678985818650623,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002554746037662058,
|
|
"loss": 5.3521,
|
|
"mean_token_accuracy": 0.18585011065006257,
|
|
"num_tokens": 12011638.0,
|
|
"step": 6610
|
|
},
|
|
{
|
|
"entropy": 5.755534267425537,
|
|
"epoch": 5.683283197249677,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002551433967227919,
|
|
"loss": 5.2875,
|
|
"mean_token_accuracy": 0.18850458413362503,
|
|
"num_tokens": 12021319.0,
|
|
"step": 6615
|
|
},
|
|
{
|
|
"entropy": 5.713330936431885,
|
|
"epoch": 5.687580575848732,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000254812233038263,
|
|
"loss": 5.2301,
|
|
"mean_token_accuracy": 0.1979709580540657,
|
|
"num_tokens": 12030255.0,
|
|
"step": 6620
|
|
},
|
|
{
|
|
"entropy": 5.718487787246704,
|
|
"epoch": 5.691877954447786,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00025448111343574813,
|
|
"loss": 5.2395,
|
|
"mean_token_accuracy": 0.19640317410230637,
|
|
"num_tokens": 12038884.0,
|
|
"step": 6625
|
|
},
|
|
{
|
|
"entropy": 5.69110860824585,
|
|
"epoch": 5.696175333046842,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002541500386382802,
|
|
"loss": 5.2064,
|
|
"mean_token_accuracy": 0.1965741366147995,
|
|
"num_tokens": 12047477.0,
|
|
"step": 6630
|
|
},
|
|
{
|
|
"entropy": 5.692670249938965,
|
|
"epoch": 5.7004727116458955,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00025381900936879433,
|
|
"loss": 5.3065,
|
|
"mean_token_accuracy": 0.19132504910230635,
|
|
"num_tokens": 12056902.0,
|
|
"step": 6635
|
|
},
|
|
{
|
|
"entropy": 5.7387948513031,
|
|
"epoch": 5.704770090244951,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002534880263501259,
|
|
"loss": 5.2484,
|
|
"mean_token_accuracy": 0.1949061155319214,
|
|
"num_tokens": 12065721.0,
|
|
"step": 6640
|
|
},
|
|
{
|
|
"entropy": 5.678815031051636,
|
|
"epoch": 5.7090674688440055,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0002531570903050097,
|
|
"loss": 5.2389,
|
|
"mean_token_accuracy": 0.1989353448152542,
|
|
"num_tokens": 12074870.0,
|
|
"step": 6645
|
|
},
|
|
{
|
|
"entropy": 5.7372067928314205,
|
|
"epoch": 5.71336484744306,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002528262019560776,
|
|
"loss": 5.2793,
|
|
"mean_token_accuracy": 0.19413130581378937,
|
|
"num_tokens": 12084557.0,
|
|
"step": 6650
|
|
},
|
|
{
|
|
"entropy": 5.670536470413208,
|
|
"epoch": 5.717662226042115,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002524953620258579,
|
|
"loss": 5.1274,
|
|
"mean_token_accuracy": 0.20778495222330093,
|
|
"num_tokens": 12093074.0,
|
|
"step": 6655
|
|
},
|
|
{
|
|
"entropy": 5.674096488952637,
|
|
"epoch": 5.721959604641169,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002521645712367724,
|
|
"loss": 5.1765,
|
|
"mean_token_accuracy": 0.20832548439502716,
|
|
"num_tokens": 12102785.0,
|
|
"step": 6660
|
|
},
|
|
{
|
|
"entropy": 5.672455167770385,
|
|
"epoch": 5.726256983240224,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00025183383031113606,
|
|
"loss": 5.2033,
|
|
"mean_token_accuracy": 0.20182351469993592,
|
|
"num_tokens": 12112535.0,
|
|
"step": 6665
|
|
},
|
|
{
|
|
"entropy": 5.589156818389893,
|
|
"epoch": 5.730554361839278,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00025150313997115476,
|
|
"loss": 5.1555,
|
|
"mean_token_accuracy": 0.20321130603551865,
|
|
"num_tokens": 12121604.0,
|
|
"step": 6670
|
|
},
|
|
{
|
|
"entropy": 5.690019321441651,
|
|
"epoch": 5.734851740438333,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0002511725009389244,
|
|
"loss": 5.3347,
|
|
"mean_token_accuracy": 0.18688519150018693,
|
|
"num_tokens": 12131276.0,
|
|
"step": 6675
|
|
},
|
|
{
|
|
"entropy": 5.697007465362549,
|
|
"epoch": 5.739149119037387,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000250841913936428,
|
|
"loss": 5.2764,
|
|
"mean_token_accuracy": 0.2016531065106392,
|
|
"num_tokens": 12140180.0,
|
|
"step": 6680
|
|
},
|
|
{
|
|
"entropy": 5.716262006759644,
|
|
"epoch": 5.743446497636442,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002505113796855357,
|
|
"loss": 5.2564,
|
|
"mean_token_accuracy": 0.1987043410539627,
|
|
"num_tokens": 12149635.0,
|
|
"step": 6685
|
|
},
|
|
{
|
|
"entropy": 5.712231016159057,
|
|
"epoch": 5.747743876235496,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00025018089890800225,
|
|
"loss": 5.2141,
|
|
"mean_token_accuracy": 0.2032635509967804,
|
|
"num_tokens": 12157565.0,
|
|
"step": 6690
|
|
},
|
|
{
|
|
"entropy": 5.687877988815307,
|
|
"epoch": 5.752041254834551,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00024985047232546544,
|
|
"loss": 5.3114,
|
|
"mean_token_accuracy": 0.19476086497306824,
|
|
"num_tokens": 12166647.0,
|
|
"step": 6695
|
|
},
|
|
{
|
|
"entropy": 5.706553173065186,
|
|
"epoch": 5.7563386334336055,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00024952010065944485,
|
|
"loss": 5.2751,
|
|
"mean_token_accuracy": 0.19282412827014922,
|
|
"num_tokens": 12175554.0,
|
|
"step": 6700
|
|
},
|
|
{
|
|
"entropy": 5.764104270935059,
|
|
"epoch": 5.76063601203266,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002491897846313402,
|
|
"loss": 5.2685,
|
|
"mean_token_accuracy": 0.199551260471344,
|
|
"num_tokens": 12184756.0,
|
|
"step": 6705
|
|
},
|
|
{
|
|
"entropy": 5.755422925949096,
|
|
"epoch": 5.764933390631715,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002488595249624297,
|
|
"loss": 5.3528,
|
|
"mean_token_accuracy": 0.18964100033044815,
|
|
"num_tokens": 12194724.0,
|
|
"step": 6710
|
|
},
|
|
{
|
|
"entropy": 5.716435050964355,
|
|
"epoch": 5.769230769230769,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00024852932237386837,
|
|
"loss": 5.3008,
|
|
"mean_token_accuracy": 0.19272180050611495,
|
|
"num_tokens": 12203804.0,
|
|
"step": 6715
|
|
},
|
|
{
|
|
"entropy": 5.739119148254394,
|
|
"epoch": 5.773528147829824,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00024819917758668673,
|
|
"loss": 5.2631,
|
|
"mean_token_accuracy": 0.1970479354262352,
|
|
"num_tokens": 12212868.0,
|
|
"step": 6720
|
|
},
|
|
{
|
|
"entropy": 5.748186016082764,
|
|
"epoch": 5.777825526428878,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00024786909132178906,
|
|
"loss": 5.3202,
|
|
"mean_token_accuracy": 0.19409209191799165,
|
|
"num_tokens": 12221650.0,
|
|
"step": 6725
|
|
},
|
|
{
|
|
"entropy": 5.731725645065308,
|
|
"epoch": 5.782122905027933,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00024753906429995194,
|
|
"loss": 5.2072,
|
|
"mean_token_accuracy": 0.20286511182785033,
|
|
"num_tokens": 12231541.0,
|
|
"step": 6730
|
|
},
|
|
{
|
|
"entropy": 5.693597078323364,
|
|
"epoch": 5.786420283626987,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002472090972418222,
|
|
"loss": 5.3085,
|
|
"mean_token_accuracy": 0.19542299211025238,
|
|
"num_tokens": 12240899.0,
|
|
"step": 6735
|
|
},
|
|
{
|
|
"entropy": 5.717250013351441,
|
|
"epoch": 5.790717662226042,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002468791908679163,
|
|
"loss": 5.2792,
|
|
"mean_token_accuracy": 0.189197401702404,
|
|
"num_tokens": 12250352.0,
|
|
"step": 6740
|
|
},
|
|
{
|
|
"entropy": 5.777727508544922,
|
|
"epoch": 5.795015040825096,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002465493458986175,
|
|
"loss": 5.3387,
|
|
"mean_token_accuracy": 0.1892807438969612,
|
|
"num_tokens": 12259975.0,
|
|
"step": 6745
|
|
},
|
|
{
|
|
"entropy": 5.6279644012451175,
|
|
"epoch": 5.799312419424151,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00024621956305417587,
|
|
"loss": 5.1723,
|
|
"mean_token_accuracy": 0.20524910390377044,
|
|
"num_tokens": 12269203.0,
|
|
"step": 6750
|
|
},
|
|
{
|
|
"entropy": 5.733295536041259,
|
|
"epoch": 5.8036097980232055,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000245889843054705,
|
|
"loss": 5.2981,
|
|
"mean_token_accuracy": 0.19974669069051743,
|
|
"num_tokens": 12279481.0,
|
|
"step": 6755
|
|
},
|
|
{
|
|
"entropy": 5.701198387145996,
|
|
"epoch": 5.80790717662226,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00024556018662018163,
|
|
"loss": 5.2755,
|
|
"mean_token_accuracy": 0.1926541268825531,
|
|
"num_tokens": 12288848.0,
|
|
"step": 6760
|
|
},
|
|
{
|
|
"entropy": 5.777185106277466,
|
|
"epoch": 5.812204555221315,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00024523059447044377,
|
|
"loss": 5.2382,
|
|
"mean_token_accuracy": 0.19783916026353837,
|
|
"num_tokens": 12297346.0,
|
|
"step": 6765
|
|
},
|
|
{
|
|
"entropy": 5.748342561721802,
|
|
"epoch": 5.816501933820369,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002449010673251887,
|
|
"loss": 5.262,
|
|
"mean_token_accuracy": 0.1865608349442482,
|
|
"num_tokens": 12306233.0,
|
|
"step": 6770
|
|
},
|
|
{
|
|
"entropy": 5.72390570640564,
|
|
"epoch": 5.820799312419425,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002445716059039723,
|
|
"loss": 5.3628,
|
|
"mean_token_accuracy": 0.18657551109790801,
|
|
"num_tokens": 12315609.0,
|
|
"step": 6775
|
|
},
|
|
{
|
|
"entropy": 5.736563587188721,
|
|
"epoch": 5.825096691018478,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00024424221092620644,
|
|
"loss": 5.312,
|
|
"mean_token_accuracy": 0.19203596115112304,
|
|
"num_tokens": 12323915.0,
|
|
"step": 6780
|
|
},
|
|
{
|
|
"entropy": 5.661197090148926,
|
|
"epoch": 5.829394069617534,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00024391288311115822,
|
|
"loss": 5.2188,
|
|
"mean_token_accuracy": 0.19612112641334534,
|
|
"num_tokens": 12334077.0,
|
|
"step": 6785
|
|
},
|
|
{
|
|
"entropy": 5.715623760223389,
|
|
"epoch": 5.833691448216588,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002435836231779478,
|
|
"loss": 5.2908,
|
|
"mean_token_accuracy": 0.19464734047651291,
|
|
"num_tokens": 12342411.0,
|
|
"step": 6790
|
|
},
|
|
{
|
|
"entropy": 5.654740953445435,
|
|
"epoch": 5.837988826815643,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00024325443184554724,
|
|
"loss": 5.1519,
|
|
"mean_token_accuracy": 0.2052411288022995,
|
|
"num_tokens": 12351308.0,
|
|
"step": 6795
|
|
},
|
|
{
|
|
"entropy": 5.657415914535522,
|
|
"epoch": 5.842286205414697,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00024292530983277904,
|
|
"loss": 5.3293,
|
|
"mean_token_accuracy": 0.1911078616976738,
|
|
"num_tokens": 12359673.0,
|
|
"step": 6800
|
|
},
|
|
{
|
|
"entropy": 5.68788104057312,
|
|
"epoch": 5.846583584013752,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00024259625785831408,
|
|
"loss": 5.1709,
|
|
"mean_token_accuracy": 0.20246697962284088,
|
|
"num_tokens": 12367876.0,
|
|
"step": 6805
|
|
},
|
|
{
|
|
"entropy": 5.740637874603271,
|
|
"epoch": 5.850880962612806,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00024226727664067023,
|
|
"loss": 5.3231,
|
|
"mean_token_accuracy": 0.1934667184948921,
|
|
"num_tokens": 12377040.0,
|
|
"step": 6810
|
|
},
|
|
{
|
|
"entropy": 5.788400888442993,
|
|
"epoch": 5.855178341211861,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00024193836689821109,
|
|
"loss": 5.3988,
|
|
"mean_token_accuracy": 0.18783645629882811,
|
|
"num_tokens": 12387622.0,
|
|
"step": 6815
|
|
},
|
|
{
|
|
"entropy": 5.642332696914673,
|
|
"epoch": 5.8594757198109155,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002416095293491439,
|
|
"loss": 5.1672,
|
|
"mean_token_accuracy": 0.20579997152090074,
|
|
"num_tokens": 12396447.0,
|
|
"step": 6820
|
|
},
|
|
{
|
|
"entropy": 5.716213703155518,
|
|
"epoch": 5.86377309840997,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 0.0002412807647115186,
|
|
"loss": 5.1823,
|
|
"mean_token_accuracy": 0.20663501918315888,
|
|
"num_tokens": 12405887.0,
|
|
"step": 6825
|
|
},
|
|
{
|
|
"entropy": 5.671609258651733,
|
|
"epoch": 5.868070477009025,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00024095207370322574,
|
|
"loss": 5.218,
|
|
"mean_token_accuracy": 0.19455846697092055,
|
|
"num_tokens": 12414543.0,
|
|
"step": 6830
|
|
},
|
|
{
|
|
"entropy": 5.698519659042359,
|
|
"epoch": 5.872367855608079,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00024062345704199507,
|
|
"loss": 5.2235,
|
|
"mean_token_accuracy": 0.2004447713494301,
|
|
"num_tokens": 12423370.0,
|
|
"step": 6835
|
|
},
|
|
{
|
|
"entropy": 5.7754487037658695,
|
|
"epoch": 5.876665234207134,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00024029491544539405,
|
|
"loss": 5.3273,
|
|
"mean_token_accuracy": 0.187202025949955,
|
|
"num_tokens": 12433980.0,
|
|
"step": 6840
|
|
},
|
|
{
|
|
"entropy": 5.716495847702026,
|
|
"epoch": 5.880962612806188,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00023996644963082616,
|
|
"loss": 5.2601,
|
|
"mean_token_accuracy": 0.20953239500522614,
|
|
"num_tokens": 12443300.0,
|
|
"step": 6845
|
|
},
|
|
{
|
|
"entropy": 5.718546295166016,
|
|
"epoch": 5.885259991405243,
|
|
"grad_norm": 2.71875,
|
|
"learning_rate": 0.00023963806031552948,
|
|
"loss": 5.3183,
|
|
"mean_token_accuracy": 0.1878659173846245,
|
|
"num_tokens": 12452462.0,
|
|
"step": 6850
|
|
},
|
|
{
|
|
"entropy": 5.641903352737427,
|
|
"epoch": 5.889557370004297,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00023930974821657504,
|
|
"loss": 5.2249,
|
|
"mean_token_accuracy": 0.1982801601290703,
|
|
"num_tokens": 12461605.0,
|
|
"step": 6855
|
|
},
|
|
{
|
|
"entropy": 5.799238491058349,
|
|
"epoch": 5.893854748603352,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00023898151405086533,
|
|
"loss": 5.2937,
|
|
"mean_token_accuracy": 0.19286635965108873,
|
|
"num_tokens": 12470905.0,
|
|
"step": 6860
|
|
},
|
|
{
|
|
"entropy": 5.766983079910278,
|
|
"epoch": 5.8981521272024064,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00023865335853513232,
|
|
"loss": 5.2831,
|
|
"mean_token_accuracy": 0.19688766151666642,
|
|
"num_tokens": 12478913.0,
|
|
"step": 6865
|
|
},
|
|
{
|
|
"entropy": 5.742091226577759,
|
|
"epoch": 5.902449505801461,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00023832528238593677,
|
|
"loss": 5.3891,
|
|
"mean_token_accuracy": 0.18582995384931564,
|
|
"num_tokens": 12487561.0,
|
|
"step": 6870
|
|
},
|
|
{
|
|
"entropy": 5.679473829269409,
|
|
"epoch": 5.9067468844005155,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00023799728631966556,
|
|
"loss": 5.244,
|
|
"mean_token_accuracy": 0.19819188714027405,
|
|
"num_tokens": 12496781.0,
|
|
"step": 6875
|
|
},
|
|
{
|
|
"entropy": 5.711652946472168,
|
|
"epoch": 5.91104426299957,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.0002376693710525313,
|
|
"loss": 5.3311,
|
|
"mean_token_accuracy": 0.1853245958685875,
|
|
"num_tokens": 12505716.0,
|
|
"step": 6880
|
|
},
|
|
{
|
|
"entropy": 5.786692047119141,
|
|
"epoch": 5.915341641598625,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00023734153730056967,
|
|
"loss": 5.3006,
|
|
"mean_token_accuracy": 0.189934578537941,
|
|
"num_tokens": 12515594.0,
|
|
"step": 6885
|
|
},
|
|
{
|
|
"entropy": 5.729607725143433,
|
|
"epoch": 5.919639020197679,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00023701378577963873,
|
|
"loss": 5.121,
|
|
"mean_token_accuracy": 0.20797530561685562,
|
|
"num_tokens": 12523439.0,
|
|
"step": 6890
|
|
},
|
|
{
|
|
"entropy": 5.672238159179687,
|
|
"epoch": 5.923936398796734,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002366861172054166,
|
|
"loss": 5.2196,
|
|
"mean_token_accuracy": 0.20252593606710434,
|
|
"num_tokens": 12532100.0,
|
|
"step": 6895
|
|
},
|
|
{
|
|
"entropy": 5.70272479057312,
|
|
"epoch": 5.928233777395788,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00023635853229340054,
|
|
"loss": 5.252,
|
|
"mean_token_accuracy": 0.19352464228868485,
|
|
"num_tokens": 12539689.0,
|
|
"step": 6900
|
|
},
|
|
{
|
|
"entropy": 5.707230854034424,
|
|
"epoch": 5.932531155994843,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00023603103175890512,
|
|
"loss": 5.2915,
|
|
"mean_token_accuracy": 0.18372513055801393,
|
|
"num_tokens": 12548486.0,
|
|
"step": 6905
|
|
},
|
|
{
|
|
"entropy": 5.713245677947998,
|
|
"epoch": 5.936828534593897,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00023570361631706062,
|
|
"loss": 5.159,
|
|
"mean_token_accuracy": 0.2086911678314209,
|
|
"num_tokens": 12557423.0,
|
|
"step": 6910
|
|
},
|
|
{
|
|
"entropy": 5.7005983829498295,
|
|
"epoch": 5.941125913192952,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00023537628668281142,
|
|
"loss": 5.324,
|
|
"mean_token_accuracy": 0.1907919690012932,
|
|
"num_tokens": 12566086.0,
|
|
"step": 6915
|
|
},
|
|
{
|
|
"entropy": 5.6613325595855715,
|
|
"epoch": 5.945423291792007,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00023504904357091468,
|
|
"loss": 5.2085,
|
|
"mean_token_accuracy": 0.1961720034480095,
|
|
"num_tokens": 12575827.0,
|
|
"step": 6920
|
|
},
|
|
{
|
|
"entropy": 5.7310045719146725,
|
|
"epoch": 5.949720670391061,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0002347218876959384,
|
|
"loss": 5.1938,
|
|
"mean_token_accuracy": 0.20821031779050828,
|
|
"num_tokens": 12585044.0,
|
|
"step": 6925
|
|
},
|
|
{
|
|
"entropy": 5.704947137832642,
|
|
"epoch": 5.954018048990116,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002343948197722604,
|
|
"loss": 5.2216,
|
|
"mean_token_accuracy": 0.19832299202680587,
|
|
"num_tokens": 12594677.0,
|
|
"step": 6930
|
|
},
|
|
{
|
|
"entropy": 5.68612756729126,
|
|
"epoch": 5.958315427589171,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00023406784051406638,
|
|
"loss": 5.2691,
|
|
"mean_token_accuracy": 0.19611845761537552,
|
|
"num_tokens": 12604829.0,
|
|
"step": 6935
|
|
},
|
|
{
|
|
"entropy": 5.812674808502197,
|
|
"epoch": 5.9626128061882255,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00023374095063534816,
|
|
"loss": 5.2611,
|
|
"mean_token_accuracy": 0.19945968836545944,
|
|
"num_tokens": 12613869.0,
|
|
"step": 6940
|
|
},
|
|
{
|
|
"entropy": 5.71683087348938,
|
|
"epoch": 5.96691018478728,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.00023341415084990276,
|
|
"loss": 5.2381,
|
|
"mean_token_accuracy": 0.20038213431835175,
|
|
"num_tokens": 12623248.0,
|
|
"step": 6945
|
|
},
|
|
{
|
|
"entropy": 5.609876537322998,
|
|
"epoch": 5.971207563386335,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00023308744187132996,
|
|
"loss": 5.1787,
|
|
"mean_token_accuracy": 0.2036821648478508,
|
|
"num_tokens": 12631973.0,
|
|
"step": 6950
|
|
},
|
|
{
|
|
"entropy": 5.7167661666870115,
|
|
"epoch": 5.975504941985389,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.00023276082441303197,
|
|
"loss": 5.294,
|
|
"mean_token_accuracy": 0.19542943090200424,
|
|
"num_tokens": 12641435.0,
|
|
"step": 6955
|
|
},
|
|
{
|
|
"entropy": 5.738678693771362,
|
|
"epoch": 5.979802320584444,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00023243429918821056,
|
|
"loss": 5.2681,
|
|
"mean_token_accuracy": 0.19121294766664504,
|
|
"num_tokens": 12651077.0,
|
|
"step": 6960
|
|
},
|
|
{
|
|
"entropy": 5.8076066970825195,
|
|
"epoch": 5.984099699183498,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00023210786690986646,
|
|
"loss": 5.3425,
|
|
"mean_token_accuracy": 0.19524374306201936,
|
|
"num_tokens": 12659929.0,
|
|
"step": 6965
|
|
},
|
|
{
|
|
"entropy": 5.7342523574829105,
|
|
"epoch": 5.988397077782553,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00023178152829079712,
|
|
"loss": 5.2048,
|
|
"mean_token_accuracy": 0.20237503051757813,
|
|
"num_tokens": 12670725.0,
|
|
"step": 6970
|
|
},
|
|
{
|
|
"entropy": 5.699238443374634,
|
|
"epoch": 5.992694456381607,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00023145528404359562,
|
|
"loss": 5.2448,
|
|
"mean_token_accuracy": 0.2052067905664444,
|
|
"num_tokens": 12680820.0,
|
|
"step": 6975
|
|
},
|
|
{
|
|
"entropy": 5.748762941360473,
|
|
"epoch": 5.996991834980662,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002311291348806492,
|
|
"loss": 5.3309,
|
|
"mean_token_accuracy": 0.18711166083812714,
|
|
"num_tokens": 12689785.0,
|
|
"step": 6980
|
|
},
|
|
{
|
|
"entropy": 5.67414304945204,
|
|
"epoch": 6.000859475719811,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0002308030815141372,
|
|
"loss": 5.2339,
|
|
"mean_token_accuracy": 0.1927154047621621,
|
|
"num_tokens": 12697221.0,
|
|
"step": 6985
|
|
},
|
|
{
|
|
"entropy": 5.644857788085938,
|
|
"epoch": 6.005156854318866,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00023047712465602976,
|
|
"loss": 5.0695,
|
|
"mean_token_accuracy": 0.20997209548950196,
|
|
"num_tokens": 12707127.0,
|
|
"step": 6990
|
|
},
|
|
{
|
|
"entropy": 5.651292705535889,
|
|
"epoch": 6.00945423291792,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00023015126501808641,
|
|
"loss": 5.0595,
|
|
"mean_token_accuracy": 0.20088990777730942,
|
|
"num_tokens": 12715364.0,
|
|
"step": 6995
|
|
},
|
|
{
|
|
"entropy": 5.716405582427979,
|
|
"epoch": 6.013751611516975,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00022982550331185437,
|
|
"loss": 5.0943,
|
|
"mean_token_accuracy": 0.20499059855937957,
|
|
"num_tokens": 12724914.0,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"epoch": 6.013751611516975,
|
|
"eval_entropy": 5.554691632051726,
|
|
"eval_loss": 5.983340740203857,
|
|
"eval_mean_token_accuracy": 0.17420890928992816,
|
|
"eval_num_tokens": 12724914.0,
|
|
"eval_runtime": 2.6502,
|
|
"eval_samples_per_second": 1339.119,
|
|
"eval_steps_per_second": 167.531,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"entropy": 5.660956859588623,
|
|
"epoch": 6.0180489901160295,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00022949984024866704,
|
|
"loss": 5.0791,
|
|
"mean_token_accuracy": 0.20920842438936232,
|
|
"num_tokens": 12735193.0,
|
|
"step": 7005
|
|
},
|
|
{
|
|
"entropy": 5.729982233047485,
|
|
"epoch": 6.022346368715084,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0002291742765396424,
|
|
"loss": 5.1408,
|
|
"mean_token_accuracy": 0.20760069340467452,
|
|
"num_tokens": 12743945.0,
|
|
"step": 7010
|
|
},
|
|
{
|
|
"entropy": 5.727019357681274,
|
|
"epoch": 6.0266437473141385,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00022884881289568133,
|
|
"loss": 5.1573,
|
|
"mean_token_accuracy": 0.19925844222307204,
|
|
"num_tokens": 12753130.0,
|
|
"step": 7015
|
|
},
|
|
{
|
|
"entropy": 5.669996213912964,
|
|
"epoch": 6.030941125913193,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002285234500274665,
|
|
"loss": 5.1341,
|
|
"mean_token_accuracy": 0.20169465690851213,
|
|
"num_tokens": 12762108.0,
|
|
"step": 7020
|
|
},
|
|
{
|
|
"entropy": 5.792676067352295,
|
|
"epoch": 6.035238504512248,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00022819818864546016,
|
|
"loss": 5.199,
|
|
"mean_token_accuracy": 0.19825805425643922,
|
|
"num_tokens": 12772102.0,
|
|
"step": 7025
|
|
},
|
|
{
|
|
"entropy": 5.731275320053101,
|
|
"epoch": 6.039535883111302,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00022787302945990345,
|
|
"loss": 5.1325,
|
|
"mean_token_accuracy": 0.2114907592535019,
|
|
"num_tokens": 12781225.0,
|
|
"step": 7030
|
|
},
|
|
{
|
|
"entropy": 5.682911729812622,
|
|
"epoch": 6.043833261710357,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00022754797318081383,
|
|
"loss": 5.0995,
|
|
"mean_token_accuracy": 0.20623902678489686,
|
|
"num_tokens": 12789896.0,
|
|
"step": 7035
|
|
},
|
|
{
|
|
"entropy": 5.645928907394409,
|
|
"epoch": 6.048130640309411,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00022722302051798442,
|
|
"loss": 5.0325,
|
|
"mean_token_accuracy": 0.20997481644153596,
|
|
"num_tokens": 12798596.0,
|
|
"step": 7040
|
|
},
|
|
{
|
|
"entropy": 5.626775169372559,
|
|
"epoch": 6.052428018908466,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002268981721809819,
|
|
"loss": 5.038,
|
|
"mean_token_accuracy": 0.21463392823934554,
|
|
"num_tokens": 12807285.0,
|
|
"step": 7045
|
|
},
|
|
{
|
|
"entropy": 5.704924535751343,
|
|
"epoch": 6.05672539750752,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0002265734288791451,
|
|
"loss": 5.1129,
|
|
"mean_token_accuracy": 0.2053753986954689,
|
|
"num_tokens": 12816668.0,
|
|
"step": 7050
|
|
},
|
|
{
|
|
"entropy": 5.747388410568237,
|
|
"epoch": 6.061022776106575,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00022624879132158377,
|
|
"loss": 5.2106,
|
|
"mean_token_accuracy": 0.1975909262895584,
|
|
"num_tokens": 12825943.0,
|
|
"step": 7055
|
|
},
|
|
{
|
|
"entropy": 5.6859784603118895,
|
|
"epoch": 6.0653201547056295,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00022592426021717654,
|
|
"loss": 5.019,
|
|
"mean_token_accuracy": 0.2167255163192749,
|
|
"num_tokens": 12835693.0,
|
|
"step": 7060
|
|
},
|
|
{
|
|
"entropy": 5.6669517993927006,
|
|
"epoch": 6.069617533304684,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002255998362745696,
|
|
"loss": 4.924,
|
|
"mean_token_accuracy": 0.22433971166610717,
|
|
"num_tokens": 12844201.0,
|
|
"step": 7065
|
|
},
|
|
{
|
|
"entropy": 5.657796382904053,
|
|
"epoch": 6.073914911903739,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00022527552020217513,
|
|
"loss": 5.0722,
|
|
"mean_token_accuracy": 0.21229760795831681,
|
|
"num_tokens": 12853220.0,
|
|
"step": 7070
|
|
},
|
|
{
|
|
"entropy": 5.6575384616851805,
|
|
"epoch": 6.078212290502793,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002249513127081697,
|
|
"loss": 5.1487,
|
|
"mean_token_accuracy": 0.20583109855651854,
|
|
"num_tokens": 12862486.0,
|
|
"step": 7075
|
|
},
|
|
{
|
|
"entropy": 5.753157329559326,
|
|
"epoch": 6.082509669101848,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00022462721450049316,
|
|
"loss": 5.1318,
|
|
"mean_token_accuracy": 0.20287663489580154,
|
|
"num_tokens": 12871717.0,
|
|
"step": 7080
|
|
},
|
|
{
|
|
"entropy": 5.744639301300049,
|
|
"epoch": 6.086807047700902,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0002243032262868464,
|
|
"loss": 5.1134,
|
|
"mean_token_accuracy": 0.19600057303905488,
|
|
"num_tokens": 12881278.0,
|
|
"step": 7085
|
|
},
|
|
{
|
|
"entropy": 5.705004405975342,
|
|
"epoch": 6.091104426299957,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00022397934877469,
|
|
"loss": 5.1445,
|
|
"mean_token_accuracy": 0.19870713502168655,
|
|
"num_tokens": 12890720.0,
|
|
"step": 7090
|
|
},
|
|
{
|
|
"entropy": 5.692745542526245,
|
|
"epoch": 6.095401804899011,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002236555826712432,
|
|
"loss": 5.1612,
|
|
"mean_token_accuracy": 0.19816917032003403,
|
|
"num_tokens": 12900428.0,
|
|
"step": 7095
|
|
},
|
|
{
|
|
"entropy": 5.7419703006744385,
|
|
"epoch": 6.099699183498066,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00022333192868348152,
|
|
"loss": 5.1461,
|
|
"mean_token_accuracy": 0.2057621270418167,
|
|
"num_tokens": 12910177.0,
|
|
"step": 7100
|
|
},
|
|
{
|
|
"entropy": 5.73578987121582,
|
|
"epoch": 6.10399656209712,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00022300838751813606,
|
|
"loss": 5.177,
|
|
"mean_token_accuracy": 0.207381971180439,
|
|
"num_tokens": 12920734.0,
|
|
"step": 7105
|
|
},
|
|
{
|
|
"entropy": 5.70026741027832,
|
|
"epoch": 6.108293940696175,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00022268495988169145,
|
|
"loss": 5.061,
|
|
"mean_token_accuracy": 0.20728667825460434,
|
|
"num_tokens": 12929585.0,
|
|
"step": 7110
|
|
},
|
|
{
|
|
"entropy": 5.610219430923462,
|
|
"epoch": 6.1125913192952295,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00022236164648038433,
|
|
"loss": 5.064,
|
|
"mean_token_accuracy": 0.20866932421922685,
|
|
"num_tokens": 12938933.0,
|
|
"step": 7115
|
|
},
|
|
{
|
|
"entropy": 5.619111204147339,
|
|
"epoch": 6.116888697894285,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0002220384480202019,
|
|
"loss": 5.0288,
|
|
"mean_token_accuracy": 0.21320623606443406,
|
|
"num_tokens": 12947461.0,
|
|
"step": 7120
|
|
},
|
|
{
|
|
"entropy": 5.6642773151397705,
|
|
"epoch": 6.1211860764933395,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00022171536520688046,
|
|
"loss": 5.0756,
|
|
"mean_token_accuracy": 0.2107524961233139,
|
|
"num_tokens": 12956507.0,
|
|
"step": 7125
|
|
},
|
|
{
|
|
"entropy": 5.748213100433349,
|
|
"epoch": 6.125483455092394,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.00022139239874590362,
|
|
"loss": 5.2089,
|
|
"mean_token_accuracy": 0.19881164282560349,
|
|
"num_tokens": 12965740.0,
|
|
"step": 7130
|
|
},
|
|
{
|
|
"entropy": 5.7308104038238525,
|
|
"epoch": 6.1297808336914485,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002210695493425013,
|
|
"loss": 5.1126,
|
|
"mean_token_accuracy": 0.20537542253732682,
|
|
"num_tokens": 12975057.0,
|
|
"step": 7135
|
|
},
|
|
{
|
|
"entropy": 5.609074831008911,
|
|
"epoch": 6.134078212290503,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00022074681770164735,
|
|
"loss": 5.0254,
|
|
"mean_token_accuracy": 0.2130096212029457,
|
|
"num_tokens": 12984087.0,
|
|
"step": 7140
|
|
},
|
|
{
|
|
"entropy": 5.614879131317139,
|
|
"epoch": 6.138375590889558,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00022042420452805868,
|
|
"loss": 5.05,
|
|
"mean_token_accuracy": 0.21960145980119705,
|
|
"num_tokens": 12992793.0,
|
|
"step": 7145
|
|
},
|
|
{
|
|
"entropy": 5.720534706115723,
|
|
"epoch": 6.142672969488612,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00022010171052619365,
|
|
"loss": 5.1343,
|
|
"mean_token_accuracy": 0.21084621995687486,
|
|
"num_tokens": 13000769.0,
|
|
"step": 7150
|
|
},
|
|
{
|
|
"entropy": 5.72202205657959,
|
|
"epoch": 6.146970348087667,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00021977933640025,
|
|
"loss": 5.1247,
|
|
"mean_token_accuracy": 0.20645512342453004,
|
|
"num_tokens": 13010677.0,
|
|
"step": 7155
|
|
},
|
|
{
|
|
"entropy": 5.63046441078186,
|
|
"epoch": 6.151267726686721,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00021945708285416434,
|
|
"loss": 4.9872,
|
|
"mean_token_accuracy": 0.21690686494112016,
|
|
"num_tokens": 13019791.0,
|
|
"step": 7160
|
|
},
|
|
{
|
|
"entropy": 5.687076139450073,
|
|
"epoch": 6.155565105285776,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0002191349505916093,
|
|
"loss": 5.1401,
|
|
"mean_token_accuracy": 0.2043655186891556,
|
|
"num_tokens": 13029223.0,
|
|
"step": 7165
|
|
},
|
|
{
|
|
"entropy": 5.699808073043823,
|
|
"epoch": 6.15986248388483,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00021881294031599318,
|
|
"loss": 5.1021,
|
|
"mean_token_accuracy": 0.20606594085693358,
|
|
"num_tokens": 13038716.0,
|
|
"step": 7170
|
|
},
|
|
{
|
|
"entropy": 5.687010097503662,
|
|
"epoch": 6.164159862483885,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002184910527304576,
|
|
"loss": 5.1429,
|
|
"mean_token_accuracy": 0.20721092075109482,
|
|
"num_tokens": 13047915.0,
|
|
"step": 7175
|
|
},
|
|
{
|
|
"entropy": 5.623597431182861,
|
|
"epoch": 6.1684572410829395,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00021816928853787636,
|
|
"loss": 5.0868,
|
|
"mean_token_accuracy": 0.21370672732591628,
|
|
"num_tokens": 13056613.0,
|
|
"step": 7180
|
|
},
|
|
{
|
|
"entropy": 5.701534414291382,
|
|
"epoch": 6.172754619681994,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00021784764844085398,
|
|
"loss": 5.1506,
|
|
"mean_token_accuracy": 0.20698235481977462,
|
|
"num_tokens": 13066658.0,
|
|
"step": 7185
|
|
},
|
|
{
|
|
"entropy": 5.7871640682220455,
|
|
"epoch": 6.177051998281049,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0002175261331417238,
|
|
"loss": 5.1578,
|
|
"mean_token_accuracy": 0.20413654744625093,
|
|
"num_tokens": 13074798.0,
|
|
"step": 7190
|
|
},
|
|
{
|
|
"entropy": 5.6923215866088865,
|
|
"epoch": 6.181349376880103,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00021720474334254675,
|
|
"loss": 5.0627,
|
|
"mean_token_accuracy": 0.20658961534500123,
|
|
"num_tokens": 13084173.0,
|
|
"step": 7195
|
|
},
|
|
{
|
|
"entropy": 5.712640857696533,
|
|
"epoch": 6.185646755479158,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00021688347974510962,
|
|
"loss": 5.0679,
|
|
"mean_token_accuracy": 0.21156614869832993,
|
|
"num_tokens": 13093096.0,
|
|
"step": 7200
|
|
},
|
|
{
|
|
"entropy": 5.651481819152832,
|
|
"epoch": 6.189944134078212,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.00021656234305092377,
|
|
"loss": 5.0815,
|
|
"mean_token_accuracy": 0.20865412056446075,
|
|
"num_tokens": 13101191.0,
|
|
"step": 7205
|
|
},
|
|
{
|
|
"entropy": 5.652866172790527,
|
|
"epoch": 6.194241512677267,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0002162413339612234,
|
|
"loss": 5.1174,
|
|
"mean_token_accuracy": 0.20467220097780228,
|
|
"num_tokens": 13109829.0,
|
|
"step": 7210
|
|
},
|
|
{
|
|
"entropy": 5.666905641555786,
|
|
"epoch": 6.198538891276321,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00021592045317696406,
|
|
"loss": 5.0794,
|
|
"mean_token_accuracy": 0.20641382187604904,
|
|
"num_tokens": 13119314.0,
|
|
"step": 7215
|
|
},
|
|
{
|
|
"entropy": 5.672490215301513,
|
|
"epoch": 6.202836269875376,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00021559970139882102,
|
|
"loss": 5.0349,
|
|
"mean_token_accuracy": 0.21516600847244263,
|
|
"num_tokens": 13128113.0,
|
|
"step": 7220
|
|
},
|
|
{
|
|
"entropy": 5.7386664867401125,
|
|
"epoch": 6.20713364847443,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002152790793271881,
|
|
"loss": 5.1912,
|
|
"mean_token_accuracy": 0.1985057532787323,
|
|
"num_tokens": 13136892.0,
|
|
"step": 7225
|
|
},
|
|
{
|
|
"entropy": 5.726948595046997,
|
|
"epoch": 6.211431027073485,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00021495858766217558,
|
|
"loss": 5.1582,
|
|
"mean_token_accuracy": 0.20369147062301635,
|
|
"num_tokens": 13146960.0,
|
|
"step": 7230
|
|
},
|
|
{
|
|
"entropy": 5.65604600906372,
|
|
"epoch": 6.2157284056725395,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00021463822710360932,
|
|
"loss": 5.0339,
|
|
"mean_token_accuracy": 0.21908514499664306,
|
|
"num_tokens": 13156147.0,
|
|
"step": 7235
|
|
},
|
|
{
|
|
"entropy": 5.657547855377198,
|
|
"epoch": 6.220025784271594,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00021431799835102867,
|
|
"loss": 5.1033,
|
|
"mean_token_accuracy": 0.20035499185323716,
|
|
"num_tokens": 13164588.0,
|
|
"step": 7240
|
|
},
|
|
{
|
|
"entropy": 5.621375322341919,
|
|
"epoch": 6.224323162870649,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00021399790210368524,
|
|
"loss": 5.0697,
|
|
"mean_token_accuracy": 0.20748384147882462,
|
|
"num_tokens": 13174361.0,
|
|
"step": 7245
|
|
},
|
|
{
|
|
"entropy": 5.741681146621704,
|
|
"epoch": 6.228620541469703,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00021367793906054133,
|
|
"loss": 5.2595,
|
|
"mean_token_accuracy": 0.19659782499074935,
|
|
"num_tokens": 13185266.0,
|
|
"step": 7250
|
|
},
|
|
{
|
|
"entropy": 5.698254156112671,
|
|
"epoch": 6.232917920068758,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00021335810992026823,
|
|
"loss": 5.1042,
|
|
"mean_token_accuracy": 0.21781918555498123,
|
|
"num_tokens": 13194227.0,
|
|
"step": 7255
|
|
},
|
|
{
|
|
"entropy": 5.718651008605957,
|
|
"epoch": 6.237215298667812,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00021303841538124497,
|
|
"loss": 5.1428,
|
|
"mean_token_accuracy": 0.19917880445718766,
|
|
"num_tokens": 13202569.0,
|
|
"step": 7260
|
|
},
|
|
{
|
|
"entropy": 5.667302227020263,
|
|
"epoch": 6.241512677266867,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00021271885614155685,
|
|
"loss": 5.0953,
|
|
"mean_token_accuracy": 0.21134843081235885,
|
|
"num_tokens": 13212201.0,
|
|
"step": 7265
|
|
},
|
|
{
|
|
"entropy": 5.59873948097229,
|
|
"epoch": 6.245810055865922,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 0.0002123994328989932,
|
|
"loss": 5.0133,
|
|
"mean_token_accuracy": 0.2108217567205429,
|
|
"num_tokens": 13220802.0,
|
|
"step": 7270
|
|
},
|
|
{
|
|
"entropy": 5.743954277038574,
|
|
"epoch": 6.250107434464977,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00021208014635104688,
|
|
"loss": 5.1797,
|
|
"mean_token_accuracy": 0.2005596563220024,
|
|
"num_tokens": 13229519.0,
|
|
"step": 7275
|
|
},
|
|
{
|
|
"entropy": 5.715333318710327,
|
|
"epoch": 6.254404813064031,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00021176099719491209,
|
|
"loss": 5.1169,
|
|
"mean_token_accuracy": 0.20878705978393555,
|
|
"num_tokens": 13238865.0,
|
|
"step": 7280
|
|
},
|
|
{
|
|
"entropy": 5.585339498519898,
|
|
"epoch": 6.258702191663086,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.00021144198612748312,
|
|
"loss": 5.0304,
|
|
"mean_token_accuracy": 0.21656264662742614,
|
|
"num_tokens": 13247259.0,
|
|
"step": 7285
|
|
},
|
|
{
|
|
"entropy": 5.666435956954956,
|
|
"epoch": 6.26299957026214,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00021112311384535243,
|
|
"loss": 5.1515,
|
|
"mean_token_accuracy": 0.20209793448448182,
|
|
"num_tokens": 13256692.0,
|
|
"step": 7290
|
|
},
|
|
{
|
|
"entropy": 5.71600375175476,
|
|
"epoch": 6.267296948861195,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00021080438104480976,
|
|
"loss": 5.0869,
|
|
"mean_token_accuracy": 0.20736092031002046,
|
|
"num_tokens": 13266109.0,
|
|
"step": 7295
|
|
},
|
|
{
|
|
"entropy": 5.739312219619751,
|
|
"epoch": 6.2715943274602495,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00021048578842184019,
|
|
"loss": 5.1893,
|
|
"mean_token_accuracy": 0.20091432183980942,
|
|
"num_tokens": 13275484.0,
|
|
"step": 7300
|
|
},
|
|
{
|
|
"entropy": 5.636427688598633,
|
|
"epoch": 6.275891706059304,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00021016733667212245,
|
|
"loss": 5.0716,
|
|
"mean_token_accuracy": 0.21109066009521485,
|
|
"num_tokens": 13284755.0,
|
|
"step": 7305
|
|
},
|
|
{
|
|
"entropy": 5.632757663726807,
|
|
"epoch": 6.280189084658359,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00020984902649102806,
|
|
"loss": 5.0633,
|
|
"mean_token_accuracy": 0.21345401853322982,
|
|
"num_tokens": 13294386.0,
|
|
"step": 7310
|
|
},
|
|
{
|
|
"entropy": 5.668309688568115,
|
|
"epoch": 6.284486463257413,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00020953085857361924,
|
|
"loss": 5.0759,
|
|
"mean_token_accuracy": 0.2100720912218094,
|
|
"num_tokens": 13303926.0,
|
|
"step": 7315
|
|
},
|
|
{
|
|
"entropy": 5.709054565429687,
|
|
"epoch": 6.288783841856468,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00020921283361464754,
|
|
"loss": 5.1609,
|
|
"mean_token_accuracy": 0.20190484672784806,
|
|
"num_tokens": 13312727.0,
|
|
"step": 7320
|
|
},
|
|
{
|
|
"entropy": 5.70226035118103,
|
|
"epoch": 6.293081220455522,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00020889495230855232,
|
|
"loss": 5.1578,
|
|
"mean_token_accuracy": 0.20113313049077988,
|
|
"num_tokens": 13321706.0,
|
|
"step": 7325
|
|
},
|
|
{
|
|
"entropy": 5.613316059112549,
|
|
"epoch": 6.297378599054577,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00020857721534945923,
|
|
"loss": 5.0583,
|
|
"mean_token_accuracy": 0.21448355317115783,
|
|
"num_tokens": 13330436.0,
|
|
"step": 7330
|
|
},
|
|
{
|
|
"entropy": 5.692767190933227,
|
|
"epoch": 6.301675977653631,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0002082596234311789,
|
|
"loss": 5.1882,
|
|
"mean_token_accuracy": 0.19399737417697907,
|
|
"num_tokens": 13339334.0,
|
|
"step": 7335
|
|
},
|
|
{
|
|
"entropy": 5.763005208969116,
|
|
"epoch": 6.305973356252686,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0002079421772472051,
|
|
"loss": 5.2097,
|
|
"mean_token_accuracy": 0.1995681032538414,
|
|
"num_tokens": 13348969.0,
|
|
"step": 7340
|
|
},
|
|
{
|
|
"entropy": 5.733112382888794,
|
|
"epoch": 6.31027073485174,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002076248774907134,
|
|
"loss": 5.1664,
|
|
"mean_token_accuracy": 0.2021029695868492,
|
|
"num_tokens": 13358467.0,
|
|
"step": 7345
|
|
},
|
|
{
|
|
"entropy": 5.643312120437622,
|
|
"epoch": 6.314568113450795,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00020730772485455962,
|
|
"loss": 4.9741,
|
|
"mean_token_accuracy": 0.21521805226802826,
|
|
"num_tokens": 13366413.0,
|
|
"step": 7350
|
|
},
|
|
{
|
|
"entropy": 5.7070718765258786,
|
|
"epoch": 6.3188654920498495,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002069907200312785,
|
|
"loss": 5.1765,
|
|
"mean_token_accuracy": 0.20171394795179368,
|
|
"num_tokens": 13376620.0,
|
|
"step": 7355
|
|
},
|
|
{
|
|
"entropy": 5.654757356643676,
|
|
"epoch": 6.323162870648904,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00020667386371308162,
|
|
"loss": 5.1436,
|
|
"mean_token_accuracy": 0.2084876760840416,
|
|
"num_tokens": 13385492.0,
|
|
"step": 7360
|
|
},
|
|
{
|
|
"entropy": 5.665206480026245,
|
|
"epoch": 6.327460249247959,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00020635715659185673,
|
|
"loss": 5.1005,
|
|
"mean_token_accuracy": 0.2068571150302887,
|
|
"num_tokens": 13395562.0,
|
|
"step": 7365
|
|
},
|
|
{
|
|
"entropy": 5.629107475280762,
|
|
"epoch": 6.331757627847013,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00020604059935916551,
|
|
"loss": 5.0321,
|
|
"mean_token_accuracy": 0.21338356137275696,
|
|
"num_tokens": 13403357.0,
|
|
"step": 7370
|
|
},
|
|
{
|
|
"entropy": 5.671495056152343,
|
|
"epoch": 6.336055006446068,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00020572419270624255,
|
|
"loss": 5.1399,
|
|
"mean_token_accuracy": 0.2084973394870758,
|
|
"num_tokens": 13412527.0,
|
|
"step": 7375
|
|
},
|
|
{
|
|
"entropy": 5.682686233520508,
|
|
"epoch": 6.340352385045122,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00020540793732399339,
|
|
"loss": 5.2025,
|
|
"mean_token_accuracy": 0.19823017120361328,
|
|
"num_tokens": 13422455.0,
|
|
"step": 7380
|
|
},
|
|
{
|
|
"entropy": 5.785962390899658,
|
|
"epoch": 6.344649763644177,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00020509183390299325,
|
|
"loss": 5.2997,
|
|
"mean_token_accuracy": 0.18935328871011733,
|
|
"num_tokens": 13431677.0,
|
|
"step": 7385
|
|
},
|
|
{
|
|
"entropy": 5.619129610061646,
|
|
"epoch": 6.348947142243231,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00020477588313348594,
|
|
"loss": 5.0029,
|
|
"mean_token_accuracy": 0.22419680804014205,
|
|
"num_tokens": 13440522.0,
|
|
"step": 7390
|
|
},
|
|
{
|
|
"entropy": 5.6984035015106205,
|
|
"epoch": 6.353244520842286,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00020446008570538154,
|
|
"loss": 5.0954,
|
|
"mean_token_accuracy": 0.19963915646076202,
|
|
"num_tokens": 13450021.0,
|
|
"step": 7395
|
|
},
|
|
{
|
|
"entropy": 5.658949899673462,
|
|
"epoch": 6.35754189944134,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002041444423082554,
|
|
"loss": 5.0592,
|
|
"mean_token_accuracy": 0.210471011698246,
|
|
"num_tokens": 13458115.0,
|
|
"step": 7400
|
|
},
|
|
{
|
|
"entropy": 5.7386767864227295,
|
|
"epoch": 6.361839278040395,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00020382895363134652,
|
|
"loss": 5.1346,
|
|
"mean_token_accuracy": 0.19451270401477813,
|
|
"num_tokens": 13466798.0,
|
|
"step": 7405
|
|
},
|
|
{
|
|
"entropy": 5.6361218929290775,
|
|
"epoch": 6.3661366566394495,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00020351362036355602,
|
|
"loss": 5.1537,
|
|
"mean_token_accuracy": 0.20265983492136003,
|
|
"num_tokens": 13476096.0,
|
|
"step": 7410
|
|
},
|
|
{
|
|
"entropy": 5.635464429855347,
|
|
"epoch": 6.370434035238505,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0002031984431934459,
|
|
"loss": 5.0971,
|
|
"mean_token_accuracy": 0.2096142739057541,
|
|
"num_tokens": 13484601.0,
|
|
"step": 7415
|
|
},
|
|
{
|
|
"entropy": 5.693839740753174,
|
|
"epoch": 6.3747314138375595,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00020288342280923695,
|
|
"loss": 5.0661,
|
|
"mean_token_accuracy": 0.21409792304039002,
|
|
"num_tokens": 13493994.0,
|
|
"step": 7420
|
|
},
|
|
{
|
|
"entropy": 5.719204139709473,
|
|
"epoch": 6.379028792436614,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00020256855989880785,
|
|
"loss": 5.1319,
|
|
"mean_token_accuracy": 0.20417070537805557,
|
|
"num_tokens": 13502890.0,
|
|
"step": 7425
|
|
},
|
|
{
|
|
"entropy": 5.651063394546509,
|
|
"epoch": 6.383326171035669,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00020225385514969336,
|
|
"loss": 5.091,
|
|
"mean_token_accuracy": 0.21501939743757248,
|
|
"num_tokens": 13512980.0,
|
|
"step": 7430
|
|
},
|
|
{
|
|
"entropy": 5.681841278076172,
|
|
"epoch": 6.387623549634723,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00020193930924908277,
|
|
"loss": 5.1353,
|
|
"mean_token_accuracy": 0.20528124421834945,
|
|
"num_tokens": 13521558.0,
|
|
"step": 7435
|
|
},
|
|
{
|
|
"entropy": 5.662312698364258,
|
|
"epoch": 6.391920928233778,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00020162492288381867,
|
|
"loss": 5.0735,
|
|
"mean_token_accuracy": 0.20651479214429855,
|
|
"num_tokens": 13531506.0,
|
|
"step": 7440
|
|
},
|
|
{
|
|
"entropy": 5.6301778793334964,
|
|
"epoch": 6.396218306832832,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002013106967403953,
|
|
"loss": 5.0472,
|
|
"mean_token_accuracy": 0.21609198153018952,
|
|
"num_tokens": 13540559.0,
|
|
"step": 7445
|
|
},
|
|
{
|
|
"entropy": 5.706904363632202,
|
|
"epoch": 6.400515685431887,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002009966315049569,
|
|
"loss": 5.115,
|
|
"mean_token_accuracy": 0.20681810826063157,
|
|
"num_tokens": 13550654.0,
|
|
"step": 7450
|
|
},
|
|
{
|
|
"entropy": 5.700806522369385,
|
|
"epoch": 6.404813064030941,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0002006827278632964,
|
|
"loss": 5.1681,
|
|
"mean_token_accuracy": 0.20253819674253465,
|
|
"num_tokens": 13560708.0,
|
|
"step": 7455
|
|
},
|
|
{
|
|
"entropy": 5.671815633773804,
|
|
"epoch": 6.409110442629996,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00020036898650085377,
|
|
"loss": 5.0523,
|
|
"mean_token_accuracy": 0.2059344157576561,
|
|
"num_tokens": 13569330.0,
|
|
"step": 7460
|
|
},
|
|
{
|
|
"entropy": 5.733552742004394,
|
|
"epoch": 6.41340782122905,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00020005540810271493,
|
|
"loss": 5.2014,
|
|
"mean_token_accuracy": 0.1965842977166176,
|
|
"num_tokens": 13577500.0,
|
|
"step": 7465
|
|
},
|
|
{
|
|
"entropy": 5.640770721435547,
|
|
"epoch": 6.417705199828105,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00019974199335360976,
|
|
"loss": 5.1138,
|
|
"mean_token_accuracy": 0.20814297795295716,
|
|
"num_tokens": 13586087.0,
|
|
"step": 7470
|
|
},
|
|
{
|
|
"entropy": 5.642022657394409,
|
|
"epoch": 6.4220025784271595,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00019942874293791068,
|
|
"loss": 5.0878,
|
|
"mean_token_accuracy": 0.20771096795797347,
|
|
"num_tokens": 13595346.0,
|
|
"step": 7475
|
|
},
|
|
{
|
|
"entropy": 5.756491088867188,
|
|
"epoch": 6.426299957026214,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00019911565753963145,
|
|
"loss": 5.2221,
|
|
"mean_token_accuracy": 0.20019467771053315,
|
|
"num_tokens": 13604755.0,
|
|
"step": 7480
|
|
},
|
|
{
|
|
"entropy": 5.694640970230102,
|
|
"epoch": 6.430597335625269,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001988027378424254,
|
|
"loss": 5.1552,
|
|
"mean_token_accuracy": 0.20626308917999267,
|
|
"num_tokens": 13613860.0,
|
|
"step": 7485
|
|
},
|
|
{
|
|
"entropy": 5.625195121765136,
|
|
"epoch": 6.434894714224323,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00019848998452958429,
|
|
"loss": 5.0041,
|
|
"mean_token_accuracy": 0.21522979885339738,
|
|
"num_tokens": 13622574.0,
|
|
"step": 7490
|
|
},
|
|
{
|
|
"entropy": 5.691166973114013,
|
|
"epoch": 6.439192092823378,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00019817739828403602,
|
|
"loss": 5.1486,
|
|
"mean_token_accuracy": 0.20029460191726683,
|
|
"num_tokens": 13632366.0,
|
|
"step": 7495
|
|
},
|
|
{
|
|
"entropy": 5.770128440856934,
|
|
"epoch": 6.443489471422432,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00019786497978834422,
|
|
"loss": 5.0524,
|
|
"mean_token_accuracy": 0.22132157385349274,
|
|
"num_tokens": 13640682.0,
|
|
"step": 7500
|
|
},
|
|
{
|
|
"epoch": 6.443489471422432,
|
|
"eval_entropy": 5.547238680156502,
|
|
"eval_loss": 5.977188587188721,
|
|
"eval_mean_token_accuracy": 0.1755683734039734,
|
|
"eval_num_tokens": 13640682.0,
|
|
"eval_runtime": 2.8621,
|
|
"eval_samples_per_second": 1240.001,
|
|
"eval_steps_per_second": 155.131,
|
|
"step": 7500
|
|
}
|
|
],
|
|
"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": 2.0718053944704e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|