3068 lines
82 KiB
JSON
3068 lines
82 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.2887838418564677,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.691965675354004,
|
|
"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
|
|
}
|
|
],
|
|
"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": 4136966316288000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|