442bd82e67
Model: fpadovani/eus-latn-100mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407 Source: Original Platform
8123 lines
221 KiB
JSON
8123 lines
221 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 3.4366136656639448,
|
|
"eval_steps": 500,
|
|
"global_step": 4000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 4.746951723098755,
|
|
"epoch": 0.004297378599054577,
|
|
"grad_norm": 13.0,
|
|
"learning_rate": 2e-06,
|
|
"loss": 14.3802,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 10107.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 4.743826389312744,
|
|
"epoch": 0.008594757198109154,
|
|
"grad_norm": 14.125,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 14.1741,
|
|
"mean_token_accuracy": 0.00023007302079349756,
|
|
"num_tokens": 18391.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 4.808784341812133,
|
|
"epoch": 0.01289213579716373,
|
|
"grad_norm": 17.5,
|
|
"learning_rate": 7e-06,
|
|
"loss": 13.9391,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 27061.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 5.067050266265869,
|
|
"epoch": 0.017189514396218308,
|
|
"grad_norm": 28.5,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 13.3774,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 36339.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 7.5024512767791744,
|
|
"epoch": 0.021486892995272882,
|
|
"grad_norm": 19.5,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 11.6591,
|
|
"mean_token_accuracy": 0.00038444948149845003,
|
|
"num_tokens": 45770.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.366794681549072,
|
|
"epoch": 0.02578427159432746,
|
|
"grad_norm": 4.0625,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.7655,
|
|
"mean_token_accuracy": 0.00122114981058985,
|
|
"num_tokens": 54575.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.686303997039795,
|
|
"epoch": 0.030081650193382038,
|
|
"grad_norm": 3.09375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.5273,
|
|
"mean_token_accuracy": 0.03648172654211521,
|
|
"num_tokens": 66403.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.672397136688232,
|
|
"epoch": 0.034379028792436615,
|
|
"grad_norm": 2.640625,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.1061,
|
|
"mean_token_accuracy": 0.058901159465312956,
|
|
"num_tokens": 76510.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.515008449554443,
|
|
"epoch": 0.03867640739149119,
|
|
"grad_norm": 2.296875,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.8287,
|
|
"mean_token_accuracy": 0.05614017099142075,
|
|
"num_tokens": 84836.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.411137008666993,
|
|
"epoch": 0.042973785990545764,
|
|
"grad_norm": 1.84375,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.6563,
|
|
"mean_token_accuracy": 0.05396632887423038,
|
|
"num_tokens": 93197.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.430077075958252,
|
|
"epoch": 0.047271164589600345,
|
|
"grad_norm": 2.40625,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.5344,
|
|
"mean_token_accuracy": 0.055140935629606244,
|
|
"num_tokens": 101546.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.39182996749878,
|
|
"epoch": 0.05156854318865492,
|
|
"grad_norm": 1.8203125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.5444,
|
|
"mean_token_accuracy": 0.05580790489912033,
|
|
"num_tokens": 111703.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.338463401794433,
|
|
"epoch": 0.055865921787709494,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.4004,
|
|
"mean_token_accuracy": 0.0589165486395359,
|
|
"num_tokens": 119894.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.395281219482422,
|
|
"epoch": 0.060163300386764075,
|
|
"grad_norm": 1.921875,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.321,
|
|
"mean_token_accuracy": 0.0635790053755045,
|
|
"num_tokens": 128885.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.310928249359131,
|
|
"epoch": 0.06446067898581866,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.2708,
|
|
"mean_token_accuracy": 0.05928411558270454,
|
|
"num_tokens": 138106.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.260056686401366,
|
|
"epoch": 0.06875805758487323,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.2086,
|
|
"mean_token_accuracy": 0.0639796681702137,
|
|
"num_tokens": 146691.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.25273675918579,
|
|
"epoch": 0.0730554361839278,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.1145,
|
|
"mean_token_accuracy": 0.07113314382731914,
|
|
"num_tokens": 155792.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.214042377471923,
|
|
"epoch": 0.07735281478298238,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.1291,
|
|
"mean_token_accuracy": 0.0655292171984911,
|
|
"num_tokens": 166944.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.106289291381836,
|
|
"epoch": 0.08165019338203695,
|
|
"grad_norm": 1.9296875,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.9072,
|
|
"mean_token_accuracy": 0.0739534068852663,
|
|
"num_tokens": 175303.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.038150310516357,
|
|
"epoch": 0.08594757198109153,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.8898,
|
|
"mean_token_accuracy": 0.0677933320403099,
|
|
"num_tokens": 184708.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.01686487197876,
|
|
"epoch": 0.09024495058014612,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.7892,
|
|
"mean_token_accuracy": 0.06626052372157573,
|
|
"num_tokens": 193835.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.002698516845703,
|
|
"epoch": 0.09454232917920069,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.6916,
|
|
"mean_token_accuracy": 0.07468680031597615,
|
|
"num_tokens": 203344.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 9.896427822113036,
|
|
"epoch": 0.09883970777825526,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.6808,
|
|
"mean_token_accuracy": 0.06314092017710209,
|
|
"num_tokens": 213048.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.836742782592774,
|
|
"epoch": 0.10313708637730984,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.5044,
|
|
"mean_token_accuracy": 0.0604202140122652,
|
|
"num_tokens": 221784.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.552239894866943,
|
|
"epoch": 0.10743446497636441,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.3827,
|
|
"mean_token_accuracy": 0.07485125958919525,
|
|
"num_tokens": 230971.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.533222961425782,
|
|
"epoch": 0.11173184357541899,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.2984,
|
|
"mean_token_accuracy": 0.06787660419940948,
|
|
"num_tokens": 240524.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.320106983184814,
|
|
"epoch": 0.11602922217447358,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.2114,
|
|
"mean_token_accuracy": 0.07236036621034145,
|
|
"num_tokens": 249220.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.246935367584229,
|
|
"epoch": 0.12032660077352815,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.205,
|
|
"mean_token_accuracy": 0.06581656485795975,
|
|
"num_tokens": 258934.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.093988609313964,
|
|
"epoch": 0.12462397937258272,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.1081,
|
|
"mean_token_accuracy": 0.07143516950309277,
|
|
"num_tokens": 267680.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.769672107696532,
|
|
"epoch": 0.1289213579716373,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.0668,
|
|
"mean_token_accuracy": 0.07136639766395092,
|
|
"num_tokens": 276227.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.74085454940796,
|
|
"epoch": 0.1332187365706919,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.9531,
|
|
"mean_token_accuracy": 0.07765112966299056,
|
|
"num_tokens": 286342.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.63883180618286,
|
|
"epoch": 0.13751611516974646,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.9755,
|
|
"mean_token_accuracy": 0.06722700893878937,
|
|
"num_tokens": 294994.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.51505479812622,
|
|
"epoch": 0.14181349376880104,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.9472,
|
|
"mean_token_accuracy": 0.07361817210912705,
|
|
"num_tokens": 303882.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.39471378326416,
|
|
"epoch": 0.1461108723678556,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.9246,
|
|
"mean_token_accuracy": 0.07520690858364106,
|
|
"num_tokens": 312515.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.389386463165284,
|
|
"epoch": 0.15040825096691018,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.9233,
|
|
"mean_token_accuracy": 0.07540233507752418,
|
|
"num_tokens": 320801.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.254477882385254,
|
|
"epoch": 0.15470562956596476,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.9058,
|
|
"mean_token_accuracy": 0.08120642676949501,
|
|
"num_tokens": 329382.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.332880115509033,
|
|
"epoch": 0.15900300816501933,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.8804,
|
|
"mean_token_accuracy": 0.07222829759120941,
|
|
"num_tokens": 337894.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.286204624176026,
|
|
"epoch": 0.1633003867640739,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.8675,
|
|
"mean_token_accuracy": 0.0751982007175684,
|
|
"num_tokens": 346380.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.191093826293946,
|
|
"epoch": 0.16759776536312848,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.9246,
|
|
"mean_token_accuracy": 0.0726242907345295,
|
|
"num_tokens": 356305.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.269799137115479,
|
|
"epoch": 0.17189514396218306,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.8754,
|
|
"mean_token_accuracy": 0.07030960507690906,
|
|
"num_tokens": 364899.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.152498531341553,
|
|
"epoch": 0.17619252256123766,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.8478,
|
|
"mean_token_accuracy": 0.07710132971405984,
|
|
"num_tokens": 373663.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.213716077804566,
|
|
"epoch": 0.18048990116029223,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.8339,
|
|
"mean_token_accuracy": 0.07559528909623622,
|
|
"num_tokens": 382730.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.200652503967286,
|
|
"epoch": 0.1847872797593468,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.8751,
|
|
"mean_token_accuracy": 0.06925368532538415,
|
|
"num_tokens": 392676.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.960573291778564,
|
|
"epoch": 0.18908465835840138,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.8586,
|
|
"mean_token_accuracy": 0.08359932452440262,
|
|
"num_tokens": 401050.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.224654483795167,
|
|
"epoch": 0.19338203695745596,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.7944,
|
|
"mean_token_accuracy": 0.0739112664014101,
|
|
"num_tokens": 410009.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.076114654541016,
|
|
"epoch": 0.19767941555651053,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.784,
|
|
"mean_token_accuracy": 0.07774697132408619,
|
|
"num_tokens": 419302.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.134877395629882,
|
|
"epoch": 0.2019767941555651,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.7035,
|
|
"mean_token_accuracy": 0.0846225418150425,
|
|
"num_tokens": 427296.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.007131862640382,
|
|
"epoch": 0.20627417275461968,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.8318,
|
|
"mean_token_accuracy": 0.07436586394906045,
|
|
"num_tokens": 436368.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.062287092208862,
|
|
"epoch": 0.21057155135367425,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.7528,
|
|
"mean_token_accuracy": 0.0802234098315239,
|
|
"num_tokens": 445535.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.11290283203125,
|
|
"epoch": 0.21486892995272883,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.7292,
|
|
"mean_token_accuracy": 0.07871693558990955,
|
|
"num_tokens": 454769.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.053960275650024,
|
|
"epoch": 0.2191663085517834,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.7547,
|
|
"mean_token_accuracy": 0.08017503693699837,
|
|
"num_tokens": 463975.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.906766176223755,
|
|
"epoch": 0.22346368715083798,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.7438,
|
|
"mean_token_accuracy": 0.08368083834648132,
|
|
"num_tokens": 472899.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.039681100845337,
|
|
"epoch": 0.22776106574989258,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.7418,
|
|
"mean_token_accuracy": 0.07471806854009629,
|
|
"num_tokens": 481556.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.029216766357422,
|
|
"epoch": 0.23205844434894715,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.7123,
|
|
"mean_token_accuracy": 0.08060996606945992,
|
|
"num_tokens": 490253.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 8.084418487548827,
|
|
"epoch": 0.23635582294800173,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.7157,
|
|
"mean_token_accuracy": 0.07992063909769058,
|
|
"num_tokens": 498444.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.9231773853302006,
|
|
"epoch": 0.2406532015470563,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.664,
|
|
"mean_token_accuracy": 0.08332347348332406,
|
|
"num_tokens": 508330.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.949900054931641,
|
|
"epoch": 0.24495058014611087,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.7586,
|
|
"mean_token_accuracy": 0.0789482720196247,
|
|
"num_tokens": 517900.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.986289691925049,
|
|
"epoch": 0.24924795874516545,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.7176,
|
|
"mean_token_accuracy": 0.07970602363348007,
|
|
"num_tokens": 527808.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.944300651550293,
|
|
"epoch": 0.25354533734422,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.7025,
|
|
"mean_token_accuracy": 0.074923125654459,
|
|
"num_tokens": 536931.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 8.035796642303467,
|
|
"epoch": 0.2578427159432746,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.7166,
|
|
"mean_token_accuracy": 0.08002732172608376,
|
|
"num_tokens": 545758.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.967339134216308,
|
|
"epoch": 0.26214009454232917,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.735,
|
|
"mean_token_accuracy": 0.08182198479771614,
|
|
"num_tokens": 555165.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.954096126556396,
|
|
"epoch": 0.2664374731413838,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.6505,
|
|
"mean_token_accuracy": 0.08079984113574028,
|
|
"num_tokens": 564719.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 8.011483335494995,
|
|
"epoch": 0.2707348517404383,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.6203,
|
|
"mean_token_accuracy": 0.0855506956577301,
|
|
"num_tokens": 573572.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.915408372879028,
|
|
"epoch": 0.2750322303394929,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.7815,
|
|
"mean_token_accuracy": 0.07522618100047111,
|
|
"num_tokens": 581497.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.911842298507691,
|
|
"epoch": 0.27932960893854747,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.6282,
|
|
"mean_token_accuracy": 0.08246761485934258,
|
|
"num_tokens": 591107.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.908317279815674,
|
|
"epoch": 0.28362698753760207,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.6727,
|
|
"mean_token_accuracy": 0.07925951480865479,
|
|
"num_tokens": 600241.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.992547035217285,
|
|
"epoch": 0.2879243661366566,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.5985,
|
|
"mean_token_accuracy": 0.08225269913673401,
|
|
"num_tokens": 608697.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.899400424957276,
|
|
"epoch": 0.2922217447357112,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.5755,
|
|
"mean_token_accuracy": 0.08702421039342881,
|
|
"num_tokens": 617275.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.919580078125,
|
|
"epoch": 0.29651912333476577,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.6679,
|
|
"mean_token_accuracy": 0.08152761459350585,
|
|
"num_tokens": 626644.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.855055665969848,
|
|
"epoch": 0.30081650193382037,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.5941,
|
|
"mean_token_accuracy": 0.08607768341898918,
|
|
"num_tokens": 635110.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.904864168167114,
|
|
"epoch": 0.30511388053287497,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.731,
|
|
"mean_token_accuracy": 0.08116863258183002,
|
|
"num_tokens": 644746.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 8.003858804702759,
|
|
"epoch": 0.3094112591319295,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.615,
|
|
"mean_token_accuracy": 0.08461246490478516,
|
|
"num_tokens": 654281.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.807305002212525,
|
|
"epoch": 0.3137086377309841,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.6127,
|
|
"mean_token_accuracy": 0.08584560602903366,
|
|
"num_tokens": 663174.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.868645715713501,
|
|
"epoch": 0.31800601633003867,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.5401,
|
|
"mean_token_accuracy": 0.08367656841874123,
|
|
"num_tokens": 672178.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.838464975357056,
|
|
"epoch": 0.32230339492909327,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.5441,
|
|
"mean_token_accuracy": 0.08873862400650978,
|
|
"num_tokens": 681323.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.861468744277954,
|
|
"epoch": 0.3266007735281478,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.4599,
|
|
"mean_token_accuracy": 0.08782187923789024,
|
|
"num_tokens": 690461.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.801782035827637,
|
|
"epoch": 0.3308981521272024,
|
|
"grad_norm": 1.9375,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.5001,
|
|
"mean_token_accuracy": 0.08662289157509803,
|
|
"num_tokens": 699199.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.801443576812744,
|
|
"epoch": 0.33519553072625696,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.4921,
|
|
"mean_token_accuracy": 0.09190349280834198,
|
|
"num_tokens": 707949.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.831652879714966,
|
|
"epoch": 0.33949290932531156,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.5308,
|
|
"mean_token_accuracy": 0.08901702687144279,
|
|
"num_tokens": 715752.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.8149333000183105,
|
|
"epoch": 0.3437902879243661,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.4733,
|
|
"mean_token_accuracy": 0.08961250334978103,
|
|
"num_tokens": 724416.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.810607194900513,
|
|
"epoch": 0.3480876665234207,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.5157,
|
|
"mean_token_accuracy": 0.08556607961654664,
|
|
"num_tokens": 733116.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.813943815231323,
|
|
"epoch": 0.3523850451224753,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.4905,
|
|
"mean_token_accuracy": 0.08932617083191871,
|
|
"num_tokens": 742093.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.833597850799561,
|
|
"epoch": 0.35668242372152986,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.4856,
|
|
"mean_token_accuracy": 0.08866851031780243,
|
|
"num_tokens": 750402.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.651254415512085,
|
|
"epoch": 0.36097980232058446,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.4363,
|
|
"mean_token_accuracy": 0.09088600948452949,
|
|
"num_tokens": 760961.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.848704862594604,
|
|
"epoch": 0.365277180919639,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.5138,
|
|
"mean_token_accuracy": 0.08746314346790314,
|
|
"num_tokens": 770554.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.62291374206543,
|
|
"epoch": 0.3695745595186936,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.4938,
|
|
"mean_token_accuracy": 0.09328901097178459,
|
|
"num_tokens": 779172.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.798684120178223,
|
|
"epoch": 0.37387193811774816,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.5411,
|
|
"mean_token_accuracy": 0.0797388345003128,
|
|
"num_tokens": 788040.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.799862670898437,
|
|
"epoch": 0.37816931671680276,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.4367,
|
|
"mean_token_accuracy": 0.09049512967467307,
|
|
"num_tokens": 796786.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.653696584701538,
|
|
"epoch": 0.3824666953158573,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.4549,
|
|
"mean_token_accuracy": 0.08962092027068139,
|
|
"num_tokens": 805520.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.74529128074646,
|
|
"epoch": 0.3867640739149119,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.3698,
|
|
"mean_token_accuracy": 0.09231138080358506,
|
|
"num_tokens": 814939.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.6202319145202635,
|
|
"epoch": 0.39106145251396646,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.3903,
|
|
"mean_token_accuracy": 0.08881158530712127,
|
|
"num_tokens": 823862.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.745624685287476,
|
|
"epoch": 0.39535883111302106,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.3957,
|
|
"mean_token_accuracy": 0.09272652640938758,
|
|
"num_tokens": 832820.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.745703077316284,
|
|
"epoch": 0.39965620971207566,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.3724,
|
|
"mean_token_accuracy": 0.09544146358966828,
|
|
"num_tokens": 841538.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.645022964477539,
|
|
"epoch": 0.4039535883111302,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.4799,
|
|
"mean_token_accuracy": 0.08852436989545823,
|
|
"num_tokens": 851123.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.723429012298584,
|
|
"epoch": 0.4082509669101848,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.4569,
|
|
"mean_token_accuracy": 0.09157502874732018,
|
|
"num_tokens": 860357.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.702331590652466,
|
|
"epoch": 0.41254834550923936,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.3553,
|
|
"mean_token_accuracy": 0.09141395911574364,
|
|
"num_tokens": 869980.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.594449472427368,
|
|
"epoch": 0.41684572410829396,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.3152,
|
|
"mean_token_accuracy": 0.09529538601636886,
|
|
"num_tokens": 878250.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.647209882736206,
|
|
"epoch": 0.4211431027073485,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.432,
|
|
"mean_token_accuracy": 0.08998423144221306,
|
|
"num_tokens": 887624.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.620798921585083,
|
|
"epoch": 0.4254404813064031,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.3811,
|
|
"mean_token_accuracy": 0.09268759191036224,
|
|
"num_tokens": 897120.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.560871362686157,
|
|
"epoch": 0.42973785990545765,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.3499,
|
|
"mean_token_accuracy": 0.09775018766522407,
|
|
"num_tokens": 906215.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.42973785990545765,
|
|
"eval_entropy": 7.452849494444357,
|
|
"eval_loss": 7.401772499084473,
|
|
"eval_mean_token_accuracy": 0.09653489479610512,
|
|
"eval_num_tokens": 906215.0,
|
|
"eval_runtime": 2.6878,
|
|
"eval_samples_per_second": 1320.408,
|
|
"eval_steps_per_second": 165.191,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.597157382965088,
|
|
"epoch": 0.43403523850451226,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.3332,
|
|
"mean_token_accuracy": 0.09525968879461288,
|
|
"num_tokens": 915181.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.594379711151123,
|
|
"epoch": 0.4383326171035668,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.3459,
|
|
"mean_token_accuracy": 0.09548510313034057,
|
|
"num_tokens": 924377.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.679618883132934,
|
|
"epoch": 0.4426299957026214,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.4198,
|
|
"mean_token_accuracy": 0.09123436212539673,
|
|
"num_tokens": 933114.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.633022212982178,
|
|
"epoch": 0.44692737430167595,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.3843,
|
|
"mean_token_accuracy": 0.09295491799712181,
|
|
"num_tokens": 943306.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.551965427398682,
|
|
"epoch": 0.45122475290073055,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.2733,
|
|
"mean_token_accuracy": 0.09815134108066559,
|
|
"num_tokens": 951515.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.5277495861053465,
|
|
"epoch": 0.45552213149978515,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.4038,
|
|
"mean_token_accuracy": 0.09049785882234573,
|
|
"num_tokens": 962686.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.628147220611572,
|
|
"epoch": 0.4598195100988397,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.398,
|
|
"mean_token_accuracy": 0.0952046200633049,
|
|
"num_tokens": 972136.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.478239583969116,
|
|
"epoch": 0.4641168886978943,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.2984,
|
|
"mean_token_accuracy": 0.09473967850208283,
|
|
"num_tokens": 981301.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.588911247253418,
|
|
"epoch": 0.46841426729694885,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.3291,
|
|
"mean_token_accuracy": 0.09194630309939385,
|
|
"num_tokens": 990360.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.613106918334961,
|
|
"epoch": 0.47271164589600345,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.3639,
|
|
"mean_token_accuracy": 0.09320687279105186,
|
|
"num_tokens": 999415.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.489257049560547,
|
|
"epoch": 0.477009024495058,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.3081,
|
|
"mean_token_accuracy": 0.09810379892587662,
|
|
"num_tokens": 1008762.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.5401557922363285,
|
|
"epoch": 0.4813064030941126,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.2954,
|
|
"mean_token_accuracy": 0.0994754634797573,
|
|
"num_tokens": 1017704.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.514191961288452,
|
|
"epoch": 0.48560378169316715,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.2372,
|
|
"mean_token_accuracy": 0.09821939915418625,
|
|
"num_tokens": 1026251.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.479329919815063,
|
|
"epoch": 0.48990116029222175,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.2328,
|
|
"mean_token_accuracy": 0.09923249781131745,
|
|
"num_tokens": 1036191.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.463119411468506,
|
|
"epoch": 0.4941985388912763,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.265,
|
|
"mean_token_accuracy": 0.10023235529661179,
|
|
"num_tokens": 1044936.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.498543548583984,
|
|
"epoch": 0.4984959174903309,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.2826,
|
|
"mean_token_accuracy": 0.09519179984927177,
|
|
"num_tokens": 1053683.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.506382083892822,
|
|
"epoch": 0.5027932960893855,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.258,
|
|
"mean_token_accuracy": 0.1035758949816227,
|
|
"num_tokens": 1062932.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.397946691513061,
|
|
"epoch": 0.50709067468844,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.2631,
|
|
"mean_token_accuracy": 0.09526635929942132,
|
|
"num_tokens": 1072313.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.519252490997315,
|
|
"epoch": 0.5113880532874946,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.2487,
|
|
"mean_token_accuracy": 0.09590291604399681,
|
|
"num_tokens": 1081675.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.414077806472778,
|
|
"epoch": 0.5156854318865493,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.2657,
|
|
"mean_token_accuracy": 0.09974563643336296,
|
|
"num_tokens": 1091541.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.551970100402832,
|
|
"epoch": 0.5199828104856038,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.2713,
|
|
"mean_token_accuracy": 0.09822080582380295,
|
|
"num_tokens": 1100724.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.332079648971558,
|
|
"epoch": 0.5242801890846583,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.2355,
|
|
"mean_token_accuracy": 0.09697570949792862,
|
|
"num_tokens": 1108869.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.480484294891357,
|
|
"epoch": 0.5285775676837129,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.25,
|
|
"mean_token_accuracy": 0.09792520478367805,
|
|
"num_tokens": 1117314.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.362159061431885,
|
|
"epoch": 0.5328749462827675,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.1859,
|
|
"mean_token_accuracy": 0.10019350796937943,
|
|
"num_tokens": 1126786.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.40272159576416,
|
|
"epoch": 0.5371723248818221,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.1913,
|
|
"mean_token_accuracy": 0.09889030531048774,
|
|
"num_tokens": 1136013.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.388201570510864,
|
|
"epoch": 0.5414697034808766,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.2166,
|
|
"mean_token_accuracy": 0.10013808235526085,
|
|
"num_tokens": 1144970.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.420843648910522,
|
|
"epoch": 0.5457670820799312,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.1756,
|
|
"mean_token_accuracy": 0.10355055630207062,
|
|
"num_tokens": 1153810.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.348835706710815,
|
|
"epoch": 0.5500644606789858,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.1665,
|
|
"mean_token_accuracy": 0.10198655724525452,
|
|
"num_tokens": 1162498.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.412497043609619,
|
|
"epoch": 0.5543618392780404,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.2589,
|
|
"mean_token_accuracy": 0.09946419224143029,
|
|
"num_tokens": 1172091.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.459282636642456,
|
|
"epoch": 0.5586592178770949,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.2226,
|
|
"mean_token_accuracy": 0.10092289820313453,
|
|
"num_tokens": 1181400.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.370707035064697,
|
|
"epoch": 0.5629565964761496,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.1949,
|
|
"mean_token_accuracy": 0.10673824399709701,
|
|
"num_tokens": 1189780.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.305479955673218,
|
|
"epoch": 0.5672539750752041,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.1683,
|
|
"mean_token_accuracy": 0.10197578221559525,
|
|
"num_tokens": 1198671.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.374381971359253,
|
|
"epoch": 0.5715513536742587,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.1977,
|
|
"mean_token_accuracy": 0.10073041394352913,
|
|
"num_tokens": 1207173.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.295852279663086,
|
|
"epoch": 0.5758487322733132,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.1303,
|
|
"mean_token_accuracy": 0.10481933280825614,
|
|
"num_tokens": 1216387.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.426829862594604,
|
|
"epoch": 0.5801461108723679,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.0964,
|
|
"mean_token_accuracy": 0.1035951815545559,
|
|
"num_tokens": 1224461.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.240889406204223,
|
|
"epoch": 0.5844434894714224,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.1706,
|
|
"mean_token_accuracy": 0.10345663502812386,
|
|
"num_tokens": 1233774.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.37571530342102,
|
|
"epoch": 0.588740868070477,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.1333,
|
|
"mean_token_accuracy": 0.10829071477055549,
|
|
"num_tokens": 1242812.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.373822784423828,
|
|
"epoch": 0.5930382466695315,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.2372,
|
|
"mean_token_accuracy": 0.10122949331998825,
|
|
"num_tokens": 1252872.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.192572116851807,
|
|
"epoch": 0.5973356252685862,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.028,
|
|
"mean_token_accuracy": 0.10453386306762695,
|
|
"num_tokens": 1260852.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.270504331588745,
|
|
"epoch": 0.6016330038676407,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.1215,
|
|
"mean_token_accuracy": 0.10755191147327423,
|
|
"num_tokens": 1268925.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.358825540542602,
|
|
"epoch": 0.6059303824666953,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.1678,
|
|
"mean_token_accuracy": 0.09976383447647094,
|
|
"num_tokens": 1278994.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.304379320144653,
|
|
"epoch": 0.6102277610657499,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.1582,
|
|
"mean_token_accuracy": 0.09768114313483238,
|
|
"num_tokens": 1287698.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.309779357910156,
|
|
"epoch": 0.6145251396648045,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.1038,
|
|
"mean_token_accuracy": 0.10614442825317383,
|
|
"num_tokens": 1297475.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.348983478546143,
|
|
"epoch": 0.618822518263859,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.1361,
|
|
"mean_token_accuracy": 0.10617230013012886,
|
|
"num_tokens": 1306836.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.270800399780273,
|
|
"epoch": 0.6231198968629136,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.1698,
|
|
"mean_token_accuracy": 0.09751353785395622,
|
|
"num_tokens": 1315872.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.261401176452637,
|
|
"epoch": 0.6274172754619682,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.1272,
|
|
"mean_token_accuracy": 0.10790985673666001,
|
|
"num_tokens": 1324624.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.305949640274048,
|
|
"epoch": 0.6317146540610228,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.0967,
|
|
"mean_token_accuracy": 0.1090677909553051,
|
|
"num_tokens": 1333058.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.134536504745483,
|
|
"epoch": 0.6360120326600773,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.0539,
|
|
"mean_token_accuracy": 0.10417703315615653,
|
|
"num_tokens": 1342376.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.308330726623535,
|
|
"epoch": 0.6403094112591319,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.1004,
|
|
"mean_token_accuracy": 0.099180655926466,
|
|
"num_tokens": 1351386.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.188440799713135,
|
|
"epoch": 0.6446067898581865,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.0039,
|
|
"mean_token_accuracy": 0.11299571469426155,
|
|
"num_tokens": 1358958.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.3689507961273195,
|
|
"epoch": 0.6489041684572411,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.1512,
|
|
"mean_token_accuracy": 0.1033362664282322,
|
|
"num_tokens": 1368599.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.267584180831909,
|
|
"epoch": 0.6532015470562956,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.1894,
|
|
"mean_token_accuracy": 0.10379417687654495,
|
|
"num_tokens": 1378529.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.118428087234497,
|
|
"epoch": 0.6574989256553503,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.9598,
|
|
"mean_token_accuracy": 0.11283587664365768,
|
|
"num_tokens": 1386993.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.2359050750732425,
|
|
"epoch": 0.6617963042544048,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 7.0655,
|
|
"mean_token_accuracy": 0.11571843326091766,
|
|
"num_tokens": 1395790.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.0956977844238285,
|
|
"epoch": 0.6660936828534594,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.0727,
|
|
"mean_token_accuracy": 0.10878411009907722,
|
|
"num_tokens": 1405587.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.142232990264892,
|
|
"epoch": 0.6703910614525139,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.0347,
|
|
"mean_token_accuracy": 0.1105166994035244,
|
|
"num_tokens": 1414478.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.294425201416016,
|
|
"epoch": 0.6746884400515686,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.0946,
|
|
"mean_token_accuracy": 0.112546656280756,
|
|
"num_tokens": 1423791.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.22660493850708,
|
|
"epoch": 0.6789858186506231,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.1307,
|
|
"mean_token_accuracy": 0.09776996597647666,
|
|
"num_tokens": 1432955.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.262840032577515,
|
|
"epoch": 0.6832831972496777,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.1067,
|
|
"mean_token_accuracy": 0.1036052331328392,
|
|
"num_tokens": 1441907.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.072665357589722,
|
|
"epoch": 0.6875805758487322,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.9795,
|
|
"mean_token_accuracy": 0.11090076863765716,
|
|
"num_tokens": 1451062.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.22030816078186,
|
|
"epoch": 0.6918779544477869,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000402,
|
|
"loss": 7.0569,
|
|
"mean_token_accuracy": 0.10773894265294075,
|
|
"num_tokens": 1460132.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.0464826107025145,
|
|
"epoch": 0.6961753330468414,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.95,
|
|
"mean_token_accuracy": 0.11331479921936989,
|
|
"num_tokens": 1469582.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.150194263458252,
|
|
"epoch": 0.700472711645896,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 7.0273,
|
|
"mean_token_accuracy": 0.10949807912111283,
|
|
"num_tokens": 1479053.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.140777254104615,
|
|
"epoch": 0.7047700902449506,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 7.0444,
|
|
"mean_token_accuracy": 0.10801303759217262,
|
|
"num_tokens": 1488189.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.048080348968506,
|
|
"epoch": 0.7090674688440052,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.909,
|
|
"mean_token_accuracy": 0.11225773021578789,
|
|
"num_tokens": 1497324.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.119872570037842,
|
|
"epoch": 0.7133648474430597,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 7.0142,
|
|
"mean_token_accuracy": 0.11574618220329284,
|
|
"num_tokens": 1506543.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.156819486618042,
|
|
"epoch": 0.7176622260421143,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.99,
|
|
"mean_token_accuracy": 0.11488222405314445,
|
|
"num_tokens": 1516737.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.952379846572876,
|
|
"epoch": 0.7219596046411689,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.8673,
|
|
"mean_token_accuracy": 0.11356691420078277,
|
|
"num_tokens": 1525561.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.108857202529907,
|
|
"epoch": 0.7262569832402235,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.8969,
|
|
"mean_token_accuracy": 0.11771438643336296,
|
|
"num_tokens": 1533323.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.133508110046387,
|
|
"epoch": 0.730554361839278,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.9472,
|
|
"mean_token_accuracy": 0.11780929937958717,
|
|
"num_tokens": 1542632.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.068312883377075,
|
|
"epoch": 0.7348517404383326,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.9049,
|
|
"mean_token_accuracy": 0.11859052330255508,
|
|
"num_tokens": 1551236.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.930421543121338,
|
|
"epoch": 0.7391491190373872,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.9216,
|
|
"mean_token_accuracy": 0.11472166553139687,
|
|
"num_tokens": 1559674.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.060778665542602,
|
|
"epoch": 0.7434464976364418,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.9908,
|
|
"mean_token_accuracy": 0.11511028632521629,
|
|
"num_tokens": 1569481.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.130613136291504,
|
|
"epoch": 0.7477438762354963,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 7.0149,
|
|
"mean_token_accuracy": 0.11255836263298988,
|
|
"num_tokens": 1578488.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.021413326263428,
|
|
"epoch": 0.752041254834551,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.9125,
|
|
"mean_token_accuracy": 0.11169529184699059,
|
|
"num_tokens": 1586675.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.099101734161377,
|
|
"epoch": 0.7563386334336055,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 7.0387,
|
|
"mean_token_accuracy": 0.10673936307430268,
|
|
"num_tokens": 1595411.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.1430614471435545,
|
|
"epoch": 0.7606360120326601,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000442,
|
|
"loss": 7.0139,
|
|
"mean_token_accuracy": 0.1113392397761345,
|
|
"num_tokens": 1604046.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.069469738006592,
|
|
"epoch": 0.7649333906317146,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.9978,
|
|
"mean_token_accuracy": 0.10678019747138023,
|
|
"num_tokens": 1613759.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.96642370223999,
|
|
"epoch": 0.7692307692307693,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.9579,
|
|
"mean_token_accuracy": 0.11460690051317216,
|
|
"num_tokens": 1623323.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.08248085975647,
|
|
"epoch": 0.7735281478298238,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.9397,
|
|
"mean_token_accuracy": 0.11283179298043251,
|
|
"num_tokens": 1631727.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.034572553634644,
|
|
"epoch": 0.7778255264288784,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.8681,
|
|
"mean_token_accuracy": 0.11410197094082833,
|
|
"num_tokens": 1639544.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.053156805038452,
|
|
"epoch": 0.7821229050279329,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.9805,
|
|
"mean_token_accuracy": 0.11028798148036004,
|
|
"num_tokens": 1648931.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.9188769340515135,
|
|
"epoch": 0.7864202836269876,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.8809,
|
|
"mean_token_accuracy": 0.11601683646440505,
|
|
"num_tokens": 1657688.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.098619890213013,
|
|
"epoch": 0.7907176622260421,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.9528,
|
|
"mean_token_accuracy": 0.11290125101804734,
|
|
"num_tokens": 1666879.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.971384572982788,
|
|
"epoch": 0.7950150408250967,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.9249,
|
|
"mean_token_accuracy": 0.11611186563968659,
|
|
"num_tokens": 1676773.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.025462627410889,
|
|
"epoch": 0.7993124194241513,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 7.0319,
|
|
"mean_token_accuracy": 0.10499885082244872,
|
|
"num_tokens": 1686144.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.038275098800659,
|
|
"epoch": 0.8036097980232059,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000467,
|
|
"loss": 7.0061,
|
|
"mean_token_accuracy": 0.10581193268299102,
|
|
"num_tokens": 1695476.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.974601221084595,
|
|
"epoch": 0.8079071766222604,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.8898,
|
|
"mean_token_accuracy": 0.11397958695888519,
|
|
"num_tokens": 1704907.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.0045435428619385,
|
|
"epoch": 0.812204555221315,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.918,
|
|
"mean_token_accuracy": 0.11368482336401939,
|
|
"num_tokens": 1714564.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.094973802566528,
|
|
"epoch": 0.8165019338203696,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.956,
|
|
"mean_token_accuracy": 0.110804034024477,
|
|
"num_tokens": 1725285.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.9100405216217045,
|
|
"epoch": 0.8207993124194242,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.9344,
|
|
"mean_token_accuracy": 0.11342336311936378,
|
|
"num_tokens": 1734331.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.086512899398803,
|
|
"epoch": 0.8250966910184787,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.8794,
|
|
"mean_token_accuracy": 0.11003802865743637,
|
|
"num_tokens": 1742340.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.859744882583618,
|
|
"epoch": 0.8293940696175333,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.8616,
|
|
"mean_token_accuracy": 0.11626880392432212,
|
|
"num_tokens": 1751725.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.893418073654175,
|
|
"epoch": 0.8336914482165879,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.7937,
|
|
"mean_token_accuracy": 0.12318155318498611,
|
|
"num_tokens": 1760294.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.883913803100586,
|
|
"epoch": 0.8379888268156425,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.8749,
|
|
"mean_token_accuracy": 0.11642721146345139,
|
|
"num_tokens": 1768912.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.098875427246094,
|
|
"epoch": 0.842286205414697,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 7.0701,
|
|
"mean_token_accuracy": 0.11430519297719002,
|
|
"num_tokens": 1778633.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.9482495307922365,
|
|
"epoch": 0.8465835840137517,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.9373,
|
|
"mean_token_accuracy": 0.11983695551753044,
|
|
"num_tokens": 1787275.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.047261524200439,
|
|
"epoch": 0.8508809626128062,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.9066,
|
|
"mean_token_accuracy": 0.11325752139091491,
|
|
"num_tokens": 1795994.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.881836032867431,
|
|
"epoch": 0.8551783412118608,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.8176,
|
|
"mean_token_accuracy": 0.1229711078107357,
|
|
"num_tokens": 1806379.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.983829498291016,
|
|
"epoch": 0.8594757198109153,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.8635,
|
|
"mean_token_accuracy": 0.12418088242411614,
|
|
"num_tokens": 1816135.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.8594757198109153,
|
|
"eval_entropy": 6.5237359571027325,
|
|
"eval_loss": 6.886626243591309,
|
|
"eval_mean_token_accuracy": 0.12066693746627451,
|
|
"eval_num_tokens": 1816135.0,
|
|
"eval_runtime": 2.6304,
|
|
"eval_samples_per_second": 1349.239,
|
|
"eval_steps_per_second": 168.797,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.860518598556519,
|
|
"epoch": 0.86377309840997,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004999998427807679,
|
|
"loss": 6.8687,
|
|
"mean_token_accuracy": 0.11553931757807731,
|
|
"num_tokens": 1824777.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.905518817901611,
|
|
"epoch": 0.8680704770090245,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004999992040780138,
|
|
"loss": 6.9043,
|
|
"mean_token_accuracy": 0.11664655804634094,
|
|
"num_tokens": 1833807.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 7.013458967208862,
|
|
"epoch": 0.8723678556080791,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004999980740669294,
|
|
"loss": 6.866,
|
|
"mean_token_accuracy": 0.11891232207417488,
|
|
"num_tokens": 1843375.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.952661991119385,
|
|
"epoch": 0.8766652342071336,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999964527499823,
|
|
"loss": 6.9331,
|
|
"mean_token_accuracy": 0.10813112482428551,
|
|
"num_tokens": 1853036.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.9601478099823,
|
|
"epoch": 0.8809626128061883,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999943401307127,
|
|
"loss": 6.9041,
|
|
"mean_token_accuracy": 0.10686428099870682,
|
|
"num_tokens": 1862041.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.8647785663604735,
|
|
"epoch": 0.8852599914052428,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999917362137337,
|
|
"loss": 6.7971,
|
|
"mean_token_accuracy": 0.11803380921483039,
|
|
"num_tokens": 1870707.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.97495756149292,
|
|
"epoch": 0.8895573700042974,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999886410047312,
|
|
"loss": 6.8129,
|
|
"mean_token_accuracy": 0.12053435295820236,
|
|
"num_tokens": 1879787.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.89826807975769,
|
|
"epoch": 0.8938547486033519,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999850545104638,
|
|
"loss": 6.8449,
|
|
"mean_token_accuracy": 0.11536444425582885,
|
|
"num_tokens": 1889413.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.934769964218139,
|
|
"epoch": 0.8981521272024066,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004999809767387633,
|
|
"loss": 6.8562,
|
|
"mean_token_accuracy": 0.11443335115909577,
|
|
"num_tokens": 1898283.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.856943893432617,
|
|
"epoch": 0.9024495058014611,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999764076985337,
|
|
"loss": 6.8634,
|
|
"mean_token_accuracy": 0.11876284033060074,
|
|
"num_tokens": 1907175.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.854787921905517,
|
|
"epoch": 0.9067468844005157,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999713473997519,
|
|
"loss": 6.9125,
|
|
"mean_token_accuracy": 0.11652960851788521,
|
|
"num_tokens": 1918223.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.947815942764282,
|
|
"epoch": 0.9110442629995703,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004999657958534677,
|
|
"loss": 6.8447,
|
|
"mean_token_accuracy": 0.11502521187067032,
|
|
"num_tokens": 1928801.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.816051054000854,
|
|
"epoch": 0.9153416415986249,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999597530718034,
|
|
"loss": 6.8097,
|
|
"mean_token_accuracy": 0.12080107033252716,
|
|
"num_tokens": 1937406.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.9242844581604,
|
|
"epoch": 0.9196390201976794,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000499953219067954,
|
|
"loss": 6.7973,
|
|
"mean_token_accuracy": 0.1198230266571045,
|
|
"num_tokens": 1947184.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.883242225646972,
|
|
"epoch": 0.923936398796734,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999461938561873,
|
|
"loss": 6.834,
|
|
"mean_token_accuracy": 0.1205545112490654,
|
|
"num_tokens": 1956293.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.884571552276611,
|
|
"epoch": 0.9282337773957886,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999386774518432,
|
|
"loss": 6.8074,
|
|
"mean_token_accuracy": 0.11995837539434433,
|
|
"num_tokens": 1964791.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.769898319244385,
|
|
"epoch": 0.9325311559948432,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999306698713349,
|
|
"loss": 6.7141,
|
|
"mean_token_accuracy": 0.12293745428323746,
|
|
"num_tokens": 1973754.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.906006813049316,
|
|
"epoch": 0.9368285345938977,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999221711321477,
|
|
"loss": 6.7984,
|
|
"mean_token_accuracy": 0.12007073611021042,
|
|
"num_tokens": 1983035.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.752451467514038,
|
|
"epoch": 0.9411259131929522,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999131812528393,
|
|
"loss": 6.8138,
|
|
"mean_token_accuracy": 0.12244802862405776,
|
|
"num_tokens": 1992584.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.889402627944946,
|
|
"epoch": 0.9454232917920069,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00049990370025304,
|
|
"loss": 6.8571,
|
|
"mean_token_accuracy": 0.11934988871216774,
|
|
"num_tokens": 2001876.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.8836750984191895,
|
|
"epoch": 0.9497206703910615,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004998937281534526,
|
|
"loss": 6.7468,
|
|
"mean_token_accuracy": 0.12761817276477813,
|
|
"num_tokens": 2011067.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.86335597038269,
|
|
"epoch": 0.954018048990116,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004998832649758521,
|
|
"loss": 6.8229,
|
|
"mean_token_accuracy": 0.11846679598093032,
|
|
"num_tokens": 2020763.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.765384292602539,
|
|
"epoch": 0.9583154275891707,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004998723107430862,
|
|
"loss": 6.8129,
|
|
"mean_token_accuracy": 0.1182372123003006,
|
|
"num_tokens": 2029534.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.886447429656982,
|
|
"epoch": 0.9626128061882252,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004998608654790741,
|
|
"loss": 6.7573,
|
|
"mean_token_accuracy": 0.11676643267273903,
|
|
"num_tokens": 2039143.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.7921895503997805,
|
|
"epoch": 0.9669101847872797,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000499848929208808,
|
|
"loss": 6.7191,
|
|
"mean_token_accuracy": 0.12415711209177971,
|
|
"num_tokens": 2048253.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.730324459075928,
|
|
"epoch": 0.9712075633863343,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004998365019583519,
|
|
"loss": 6.7645,
|
|
"mean_token_accuracy": 0.1259891837835312,
|
|
"num_tokens": 2057234.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.922126626968383,
|
|
"epoch": 0.975504941985389,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004998235837548417,
|
|
"loss": 6.8125,
|
|
"mean_token_accuracy": 0.12313916236162185,
|
|
"num_tokens": 2065431.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.802910661697387,
|
|
"epoch": 0.9798023205844435,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000499810174626486,
|
|
"loss": 6.8096,
|
|
"mean_token_accuracy": 0.12332509085536003,
|
|
"num_tokens": 2074723.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.796639680862427,
|
|
"epoch": 0.984099699183498,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004997962746025646,
|
|
"loss": 6.7022,
|
|
"mean_token_accuracy": 0.12409603446722031,
|
|
"num_tokens": 2084509.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.916227102279663,
|
|
"epoch": 0.9883970777825526,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004997818837134298,
|
|
"loss": 6.8258,
|
|
"mean_token_accuracy": 0.12093790173530579,
|
|
"num_tokens": 2093110.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.6372894763946535,
|
|
"epoch": 0.9926944563816072,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004997670019905057,
|
|
"loss": 6.6684,
|
|
"mean_token_accuracy": 0.12281760796904564,
|
|
"num_tokens": 2102355.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.856326389312744,
|
|
"epoch": 0.9969918349806618,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004997516294662876,
|
|
"loss": 6.7315,
|
|
"mean_token_accuracy": 0.12142538279294968,
|
|
"num_tokens": 2110418.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.7563766903347435,
|
|
"epoch": 1.0008594757198108,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004997357661743433,
|
|
"loss": 6.7106,
|
|
"mean_token_accuracy": 0.12642807099554274,
|
|
"num_tokens": 2117866.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.752506065368652,
|
|
"epoch": 1.0051568543188656,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004997194121493118,
|
|
"loss": 6.5302,
|
|
"mean_token_accuracy": 0.12761134281754494,
|
|
"num_tokens": 2126082.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.704809331893921,
|
|
"epoch": 1.0094542329179201,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004997025674269037,
|
|
"loss": 6.5085,
|
|
"mean_token_accuracy": 0.13516685590147973,
|
|
"num_tokens": 2134042.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.653689575195313,
|
|
"epoch": 1.0137516115169747,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004996852320439013,
|
|
"loss": 6.591,
|
|
"mean_token_accuracy": 0.12557103708386422,
|
|
"num_tokens": 2142570.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.779953145980835,
|
|
"epoch": 1.0180489901160292,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004996674060381578,
|
|
"loss": 6.5451,
|
|
"mean_token_accuracy": 0.1291184291243553,
|
|
"num_tokens": 2151310.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.66243085861206,
|
|
"epoch": 1.0223463687150838,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004996490894485985,
|
|
"loss": 6.5506,
|
|
"mean_token_accuracy": 0.12935295850038528,
|
|
"num_tokens": 2160662.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.731634521484375,
|
|
"epoch": 1.0266437473141383,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004996302823152193,
|
|
"loss": 6.5437,
|
|
"mean_token_accuracy": 0.12944689616560937,
|
|
"num_tokens": 2170067.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.652099084854126,
|
|
"epoch": 1.0309411259131929,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004996109846790873,
|
|
"loss": 6.4931,
|
|
"mean_token_accuracy": 0.134644240885973,
|
|
"num_tokens": 2178850.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.650523138046265,
|
|
"epoch": 1.0352385045122476,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004995911965823412,
|
|
"loss": 6.5253,
|
|
"mean_token_accuracy": 0.13209154531359674,
|
|
"num_tokens": 2188307.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.697449731826782,
|
|
"epoch": 1.0395358831113022,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004995709180681899,
|
|
"loss": 6.5163,
|
|
"mean_token_accuracy": 0.13310288339853288,
|
|
"num_tokens": 2197026.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.669093418121338,
|
|
"epoch": 1.0438332617103567,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000499550149180914,
|
|
"loss": 6.4938,
|
|
"mean_token_accuracy": 0.1298208139836788,
|
|
"num_tokens": 2205537.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.640572357177734,
|
|
"epoch": 1.0481306403094113,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004995288899658641,
|
|
"loss": 6.5253,
|
|
"mean_token_accuracy": 0.13521900922060012,
|
|
"num_tokens": 2214508.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.663976192474365,
|
|
"epoch": 1.0524280189084658,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004995071404694619,
|
|
"loss": 6.6448,
|
|
"mean_token_accuracy": 0.12222216129302979,
|
|
"num_tokens": 2223084.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.801617097854614,
|
|
"epoch": 1.0567253975075204,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004994849007391996,
|
|
"loss": 6.5559,
|
|
"mean_token_accuracy": 0.12804892435669898,
|
|
"num_tokens": 2231406.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.55214557647705,
|
|
"epoch": 1.061022776106575,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004994621708236401,
|
|
"loss": 6.4763,
|
|
"mean_token_accuracy": 0.1329495020210743,
|
|
"num_tokens": 2239867.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.705474472045898,
|
|
"epoch": 1.0653201547056295,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000499438950772416,
|
|
"loss": 6.5366,
|
|
"mean_token_accuracy": 0.12770846113562584,
|
|
"num_tokens": 2248844.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.590055131912232,
|
|
"epoch": 1.0696175333046842,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004994152406362311,
|
|
"loss": 6.4625,
|
|
"mean_token_accuracy": 0.13395937606692315,
|
|
"num_tokens": 2257599.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.684833335876465,
|
|
"epoch": 1.0739149119037388,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004993910404668586,
|
|
"loss": 6.5105,
|
|
"mean_token_accuracy": 0.1265903040766716,
|
|
"num_tokens": 2266510.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.589158630371093,
|
|
"epoch": 1.0782122905027933,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000499366350317142,
|
|
"loss": 6.5074,
|
|
"mean_token_accuracy": 0.13423952087759972,
|
|
"num_tokens": 2275462.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.62285099029541,
|
|
"epoch": 1.0825096691018479,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004993411702409948,
|
|
"loss": 6.4747,
|
|
"mean_token_accuracy": 0.13087429031729697,
|
|
"num_tokens": 2283826.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.651145029067993,
|
|
"epoch": 1.0868070477009024,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004993155002934002,
|
|
"loss": 6.4916,
|
|
"mean_token_accuracy": 0.1317196421325207,
|
|
"num_tokens": 2292967.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.677734279632569,
|
|
"epoch": 1.091104426299957,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0004992893405304111,
|
|
"loss": 6.6155,
|
|
"mean_token_accuracy": 0.13076187074184417,
|
|
"num_tokens": 2302336.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.616421175003052,
|
|
"epoch": 1.0954018048990115,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00049926269100915,
|
|
"loss": 6.526,
|
|
"mean_token_accuracy": 0.13683564439415932,
|
|
"num_tokens": 2311465.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.670866680145264,
|
|
"epoch": 1.0996991834980663,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004992355517878087,
|
|
"loss": 6.6418,
|
|
"mean_token_accuracy": 0.125611037760973,
|
|
"num_tokens": 2320281.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.660199451446533,
|
|
"epoch": 1.1039965620971208,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004992079229256484,
|
|
"loss": 6.5372,
|
|
"mean_token_accuracy": 0.12755262702703477,
|
|
"num_tokens": 2329755.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.533046245574951,
|
|
"epoch": 1.1082939406961754,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004991798044829996,
|
|
"loss": 6.4921,
|
|
"mean_token_accuracy": 0.1316668689250946,
|
|
"num_tokens": 2338807.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.721467113494873,
|
|
"epoch": 1.11259131929523,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004991511965212618,
|
|
"loss": 6.5841,
|
|
"mean_token_accuracy": 0.13009289875626565,
|
|
"num_tokens": 2348056.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.537866449356079,
|
|
"epoch": 1.1168886978942845,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004991220991029032,
|
|
"loss": 6.5754,
|
|
"mean_token_accuracy": 0.12793516367673874,
|
|
"num_tokens": 2357780.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.719857215881348,
|
|
"epoch": 1.121186076493339,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000499092512291461,
|
|
"loss": 6.5582,
|
|
"mean_token_accuracy": 0.1265191540122032,
|
|
"num_tokens": 2367060.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.613448905944824,
|
|
"epoch": 1.1254834550923936,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000499062436151541,
|
|
"loss": 6.5412,
|
|
"mean_token_accuracy": 0.13018715083599092,
|
|
"num_tokens": 2375751.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.668878126144409,
|
|
"epoch": 1.129780833691448,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004990318707488173,
|
|
"loss": 6.5926,
|
|
"mean_token_accuracy": 0.1261170446872711,
|
|
"num_tokens": 2385013.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.6744931697845455,
|
|
"epoch": 1.1340782122905029,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004990008161500327,
|
|
"loss": 6.4939,
|
|
"mean_token_accuracy": 0.13619888350367546,
|
|
"num_tokens": 2392935.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.598589897155762,
|
|
"epoch": 1.1383755908895574,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000498969272422998,
|
|
"loss": 6.5151,
|
|
"mean_token_accuracy": 0.1341960161924362,
|
|
"num_tokens": 2401560.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.5921900272369385,
|
|
"epoch": 1.142672969488612,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004989372396365921,
|
|
"loss": 6.4283,
|
|
"mean_token_accuracy": 0.13502056449651717,
|
|
"num_tokens": 2410050.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.658480310440064,
|
|
"epoch": 1.1469703480876665,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004989047178607618,
|
|
"loss": 6.5281,
|
|
"mean_token_accuracy": 0.13080739006400108,
|
|
"num_tokens": 2418980.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.614177894592285,
|
|
"epoch": 1.151267726686721,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004988717071665215,
|
|
"loss": 6.531,
|
|
"mean_token_accuracy": 0.12711252719163896,
|
|
"num_tokens": 2427992.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.590984678268432,
|
|
"epoch": 1.1555651052857756,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004988382076259537,
|
|
"loss": 6.4322,
|
|
"mean_token_accuracy": 0.13690400421619414,
|
|
"num_tokens": 2436368.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.564122438430786,
|
|
"epoch": 1.1598624838848304,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004988042193122077,
|
|
"loss": 6.431,
|
|
"mean_token_accuracy": 0.1385386474430561,
|
|
"num_tokens": 2445499.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.659243583679199,
|
|
"epoch": 1.164159862483885,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004987697422995005,
|
|
"loss": 6.4754,
|
|
"mean_token_accuracy": 0.1324051983654499,
|
|
"num_tokens": 2454312.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.5486212253570555,
|
|
"epoch": 1.1684572410829395,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004987347766631161,
|
|
"loss": 6.5325,
|
|
"mean_token_accuracy": 0.13133806958794594,
|
|
"num_tokens": 2462922.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.615437030792236,
|
|
"epoch": 1.172754619681994,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004986993224794055,
|
|
"loss": 6.583,
|
|
"mean_token_accuracy": 0.12835746854543686,
|
|
"num_tokens": 2472195.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.589010429382324,
|
|
"epoch": 1.1770519982810486,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004986633798257865,
|
|
"loss": 6.4796,
|
|
"mean_token_accuracy": 0.13408638834953307,
|
|
"num_tokens": 2481021.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.539653873443603,
|
|
"epoch": 1.181349376880103,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004986269487807434,
|
|
"loss": 6.4946,
|
|
"mean_token_accuracy": 0.13308287411928177,
|
|
"num_tokens": 2490250.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.634622812271118,
|
|
"epoch": 1.1856467554791577,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000498590029423827,
|
|
"loss": 6.5705,
|
|
"mean_token_accuracy": 0.132661934196949,
|
|
"num_tokens": 2499122.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.700552654266358,
|
|
"epoch": 1.1899441340782122,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004985526218356546,
|
|
"loss": 6.5204,
|
|
"mean_token_accuracy": 0.12895534336566924,
|
|
"num_tokens": 2508454.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.527865076065064,
|
|
"epoch": 1.1942415126772667,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004985147260979093,
|
|
"loss": 6.4634,
|
|
"mean_token_accuracy": 0.14121531173586846,
|
|
"num_tokens": 2517353.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.636118793487549,
|
|
"epoch": 1.1985388912763215,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004984763422933402,
|
|
"loss": 6.4827,
|
|
"mean_token_accuracy": 0.13766878992319107,
|
|
"num_tokens": 2526321.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.5368650436401365,
|
|
"epoch": 1.202836269875376,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004984374705057623,
|
|
"loss": 6.4993,
|
|
"mean_token_accuracy": 0.1347013756632805,
|
|
"num_tokens": 2535924.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.610053443908692,
|
|
"epoch": 1.2071336484744306,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004983981108200561,
|
|
"loss": 6.5027,
|
|
"mean_token_accuracy": 0.13288647755980493,
|
|
"num_tokens": 2545606.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.4733387470245365,
|
|
"epoch": 1.2114310270734852,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004983582633221672,
|
|
"loss": 6.477,
|
|
"mean_token_accuracy": 0.13318859785795212,
|
|
"num_tokens": 2554947.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.720026731491089,
|
|
"epoch": 1.2157284056725397,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004983179280991068,
|
|
"loss": 6.6352,
|
|
"mean_token_accuracy": 0.12735798433423043,
|
|
"num_tokens": 2564462.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.607101345062256,
|
|
"epoch": 1.2200257842715942,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004982771052389508,
|
|
"loss": 6.4639,
|
|
"mean_token_accuracy": 0.13422766998410224,
|
|
"num_tokens": 2573124.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.607088661193847,
|
|
"epoch": 1.224323162870649,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004982357948308401,
|
|
"loss": 6.56,
|
|
"mean_token_accuracy": 0.12319548428058624,
|
|
"num_tokens": 2581829.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.636404609680175,
|
|
"epoch": 1.2286205414697036,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004981939969649799,
|
|
"loss": 6.427,
|
|
"mean_token_accuracy": 0.1378232181072235,
|
|
"num_tokens": 2590631.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.567480611801147,
|
|
"epoch": 1.232917920068758,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004981517117326404,
|
|
"loss": 6.5469,
|
|
"mean_token_accuracy": 0.13434875309467315,
|
|
"num_tokens": 2600684.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.560835504531861,
|
|
"epoch": 1.2372152986678127,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004981089392261553,
|
|
"loss": 6.4626,
|
|
"mean_token_accuracy": 0.13627152964472772,
|
|
"num_tokens": 2609667.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.532538223266601,
|
|
"epoch": 1.2415126772668672,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000498065679538923,
|
|
"loss": 6.5324,
|
|
"mean_token_accuracy": 0.13216254264116287,
|
|
"num_tokens": 2620025.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.637110471725464,
|
|
"epoch": 1.2458100558659218,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004980219327654049,
|
|
"loss": 6.4511,
|
|
"mean_token_accuracy": 0.1365908809006214,
|
|
"num_tokens": 2629032.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.5011430263519285,
|
|
"epoch": 1.2501074344649763,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000497977699001127,
|
|
"loss": 6.4337,
|
|
"mean_token_accuracy": 0.14005313366651534,
|
|
"num_tokens": 2638303.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.583467960357666,
|
|
"epoch": 1.2544048130640308,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004979329783426778,
|
|
"loss": 6.473,
|
|
"mean_token_accuracy": 0.1399681955575943,
|
|
"num_tokens": 2647902.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.504309749603271,
|
|
"epoch": 1.2587021916630854,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004978877708877094,
|
|
"loss": 6.5023,
|
|
"mean_token_accuracy": 0.1329456977546215,
|
|
"num_tokens": 2657902.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.5875834941864015,
|
|
"epoch": 1.2629995702621402,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004978420767349368,
|
|
"loss": 6.4405,
|
|
"mean_token_accuracy": 0.13250284716486932,
|
|
"num_tokens": 2667082.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.548440313339233,
|
|
"epoch": 1.2672969488611947,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004977958959841379,
|
|
"loss": 6.5259,
|
|
"mean_token_accuracy": 0.13341723382472992,
|
|
"num_tokens": 2676855.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.60371618270874,
|
|
"epoch": 1.2715943274602493,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000497749228736153,
|
|
"loss": 6.4516,
|
|
"mean_token_accuracy": 0.1368909828364849,
|
|
"num_tokens": 2685750.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.530312252044678,
|
|
"epoch": 1.2758917060593038,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004977020750928845,
|
|
"loss": 6.5123,
|
|
"mean_token_accuracy": 0.14152853339910507,
|
|
"num_tokens": 2695272.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.559909868240356,
|
|
"epoch": 1.2801890846583583,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004976544351572973,
|
|
"loss": 6.4417,
|
|
"mean_token_accuracy": 0.13969839066267015,
|
|
"num_tokens": 2704806.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.4414268970489506,
|
|
"epoch": 1.2844864632574131,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004976063090334179,
|
|
"loss": 6.5131,
|
|
"mean_token_accuracy": 0.13546389266848563,
|
|
"num_tokens": 2713521.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.612242555618286,
|
|
"epoch": 1.2887838418564677,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004975576968263346,
|
|
"loss": 6.4793,
|
|
"mean_token_accuracy": 0.12537865340709686,
|
|
"num_tokens": 2721848.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.2887838418564677,
|
|
"eval_entropy": 6.396234018308622,
|
|
"eval_loss": 6.579471111297607,
|
|
"eval_mean_token_accuracy": 0.13534304930887245,
|
|
"eval_num_tokens": 2721848.0,
|
|
"eval_runtime": 2.8309,
|
|
"eval_samples_per_second": 1253.663,
|
|
"eval_steps_per_second": 156.84,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.469302129745484,
|
|
"epoch": 1.2930812204555222,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000497508598642197,
|
|
"loss": 6.4775,
|
|
"mean_token_accuracy": 0.13551390171051025,
|
|
"num_tokens": 2731473.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.6442944526672365,
|
|
"epoch": 1.2973785990545768,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000497459014588216,
|
|
"loss": 6.5218,
|
|
"mean_token_accuracy": 0.12938549742102623,
|
|
"num_tokens": 2739867.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.515505123138428,
|
|
"epoch": 1.3016759776536313,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000497408944772663,
|
|
"loss": 6.447,
|
|
"mean_token_accuracy": 0.13564593568444253,
|
|
"num_tokens": 2748903.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.512873935699463,
|
|
"epoch": 1.3059733562526858,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004973583893048707,
|
|
"loss": 6.4229,
|
|
"mean_token_accuracy": 0.13668794333934783,
|
|
"num_tokens": 2757711.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.633020448684692,
|
|
"epoch": 1.3102707348517404,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004973073482952321,
|
|
"loss": 6.4576,
|
|
"mean_token_accuracy": 0.13406115919351577,
|
|
"num_tokens": 2765633.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.488648509979248,
|
|
"epoch": 1.314568113450795,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004972558218552004,
|
|
"loss": 6.4817,
|
|
"mean_token_accuracy": 0.1359037533402443,
|
|
"num_tokens": 2774495.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.504700994491577,
|
|
"epoch": 1.3188654920498495,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004972038100972885,
|
|
"loss": 6.5048,
|
|
"mean_token_accuracy": 0.1317047119140625,
|
|
"num_tokens": 2782665.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.534765195846558,
|
|
"epoch": 1.323162870648904,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004971513131350697,
|
|
"loss": 6.4356,
|
|
"mean_token_accuracy": 0.1329593062400818,
|
|
"num_tokens": 2791394.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.490444326400757,
|
|
"epoch": 1.3274602492479588,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004970983310831759,
|
|
"loss": 6.4197,
|
|
"mean_token_accuracy": 0.13485073372721673,
|
|
"num_tokens": 2800488.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.544950771331787,
|
|
"epoch": 1.3317576278470133,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004970448640572989,
|
|
"loss": 6.5464,
|
|
"mean_token_accuracy": 0.13310329020023345,
|
|
"num_tokens": 2810116.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.5609852313995365,
|
|
"epoch": 1.336055006446068,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004969909121741895,
|
|
"loss": 6.3531,
|
|
"mean_token_accuracy": 0.13995248079299927,
|
|
"num_tokens": 2819205.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.486909627914429,
|
|
"epoch": 1.3403523850451224,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004969364755516569,
|
|
"loss": 6.428,
|
|
"mean_token_accuracy": 0.13530739918351173,
|
|
"num_tokens": 2828017.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.5232460498809814,
|
|
"epoch": 1.344649763644177,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004968815543085689,
|
|
"loss": 6.4764,
|
|
"mean_token_accuracy": 0.13402760103344918,
|
|
"num_tokens": 2837125.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.532066297531128,
|
|
"epoch": 1.3489471422432318,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004968261485648516,
|
|
"loss": 6.4831,
|
|
"mean_token_accuracy": 0.1357971176505089,
|
|
"num_tokens": 2845438.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.543150091171265,
|
|
"epoch": 1.3532445208422863,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000496770258441489,
|
|
"loss": 6.4545,
|
|
"mean_token_accuracy": 0.14239902943372726,
|
|
"num_tokens": 2854389.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.534445333480835,
|
|
"epoch": 1.3575418994413408,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004967138840605228,
|
|
"loss": 6.4277,
|
|
"mean_token_accuracy": 0.13663119375705718,
|
|
"num_tokens": 2863654.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.371426105499268,
|
|
"epoch": 1.3618392780403954,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000496657025545052,
|
|
"loss": 6.3269,
|
|
"mean_token_accuracy": 0.13967968374490738,
|
|
"num_tokens": 2872871.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.499322366714478,
|
|
"epoch": 1.36613665663945,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000496599683019233,
|
|
"loss": 6.4312,
|
|
"mean_token_accuracy": 0.13665958493947983,
|
|
"num_tokens": 2881140.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.529193878173828,
|
|
"epoch": 1.3704340352385045,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000496541856608279,
|
|
"loss": 6.4037,
|
|
"mean_token_accuracy": 0.13579225316643714,
|
|
"num_tokens": 2889945.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.398252201080322,
|
|
"epoch": 1.374731413837559,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004964835464384595,
|
|
"loss": 6.3424,
|
|
"mean_token_accuracy": 0.14328474253416063,
|
|
"num_tokens": 2898897.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.480287218093872,
|
|
"epoch": 1.3790287924366136,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000496424752637101,
|
|
"loss": 6.3699,
|
|
"mean_token_accuracy": 0.14322275519371033,
|
|
"num_tokens": 2907717.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.459333705902099,
|
|
"epoch": 1.3833261710356681,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004963654753325853,
|
|
"loss": 6.3616,
|
|
"mean_token_accuracy": 0.14588853269815444,
|
|
"num_tokens": 2916213.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.520181608200073,
|
|
"epoch": 1.387623549634723,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004963057146543505,
|
|
"loss": 6.5216,
|
|
"mean_token_accuracy": 0.13179171159863473,
|
|
"num_tokens": 2925706.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.476054716110229,
|
|
"epoch": 1.3919209282337774,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00049624547073289,
|
|
"loss": 6.4326,
|
|
"mean_token_accuracy": 0.13477319702506066,
|
|
"num_tokens": 2934464.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.495903539657593,
|
|
"epoch": 1.396218306832832,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004961847436997526,
|
|
"loss": 6.3254,
|
|
"mean_token_accuracy": 0.14422428011894226,
|
|
"num_tokens": 2944095.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.374484729766846,
|
|
"epoch": 1.4005156854318865,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004961235336875416,
|
|
"loss": 6.358,
|
|
"mean_token_accuracy": 0.14231155291199685,
|
|
"num_tokens": 2953357.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.491524600982666,
|
|
"epoch": 1.404813064030941,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004960618408299154,
|
|
"loss": 6.4899,
|
|
"mean_token_accuracy": 0.1311335600912571,
|
|
"num_tokens": 2963020.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.494795560836792,
|
|
"epoch": 1.4091104426299956,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004959996652615865,
|
|
"loss": 6.3217,
|
|
"mean_token_accuracy": 0.1407495141029358,
|
|
"num_tokens": 2971955.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.465664577484131,
|
|
"epoch": 1.4134078212290504,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004959370071183216,
|
|
"loss": 6.3813,
|
|
"mean_token_accuracy": 0.1411087453365326,
|
|
"num_tokens": 2980662.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.45834002494812,
|
|
"epoch": 1.417705199828105,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004958738665369407,
|
|
"loss": 6.5253,
|
|
"mean_token_accuracy": 0.12874177545309068,
|
|
"num_tokens": 2990038.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.516742801666259,
|
|
"epoch": 1.4220025784271595,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004958102436553179,
|
|
"loss": 6.4256,
|
|
"mean_token_accuracy": 0.1377176247537136,
|
|
"num_tokens": 2999835.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.502577257156372,
|
|
"epoch": 1.426299957026214,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00049574613861238,
|
|
"loss": 6.4241,
|
|
"mean_token_accuracy": 0.13374193757772446,
|
|
"num_tokens": 3009593.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.572561645507813,
|
|
"epoch": 1.4305973356252686,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004956815515481069,
|
|
"loss": 6.4608,
|
|
"mean_token_accuracy": 0.14016298577189445,
|
|
"num_tokens": 3019187.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.403193092346191,
|
|
"epoch": 1.4348947142243231,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004956164826035309,
|
|
"loss": 6.357,
|
|
"mean_token_accuracy": 0.13920670598745347,
|
|
"num_tokens": 3027875.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.453193140029907,
|
|
"epoch": 1.4391920928233777,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004955509319207363,
|
|
"loss": 6.3935,
|
|
"mean_token_accuracy": 0.14017415642738343,
|
|
"num_tokens": 3036902.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.406733131408691,
|
|
"epoch": 1.4434894714224322,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004954848996428601,
|
|
"loss": 6.3749,
|
|
"mean_token_accuracy": 0.14780990928411483,
|
|
"num_tokens": 3046653.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.486447191238403,
|
|
"epoch": 1.4477868500214868,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00049541838591409,
|
|
"loss": 6.4719,
|
|
"mean_token_accuracy": 0.13637785688042642,
|
|
"num_tokens": 3056273.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.465806913375855,
|
|
"epoch": 1.4520842286205415,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004953513908796657,
|
|
"loss": 6.3789,
|
|
"mean_token_accuracy": 0.137953469902277,
|
|
"num_tokens": 3065662.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.500221633911133,
|
|
"epoch": 1.456381607219596,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004952839146858773,
|
|
"loss": 6.4145,
|
|
"mean_token_accuracy": 0.14149838909506798,
|
|
"num_tokens": 3073970.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.40098557472229,
|
|
"epoch": 1.4606789858186506,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004952159574800658,
|
|
"loss": 6.409,
|
|
"mean_token_accuracy": 0.1353250853717327,
|
|
"num_tokens": 3082500.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.496115350723267,
|
|
"epoch": 1.4649763644177052,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004951475194106229,
|
|
"loss": 6.3622,
|
|
"mean_token_accuracy": 0.14535047560930253,
|
|
"num_tokens": 3091574.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.4026820182800295,
|
|
"epoch": 1.4692737430167597,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004950786006269898,
|
|
"loss": 6.4757,
|
|
"mean_token_accuracy": 0.13191901370882989,
|
|
"num_tokens": 3102402.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.5063896656036375,
|
|
"epoch": 1.4735711216158143,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004950092012796576,
|
|
"loss": 6.2814,
|
|
"mean_token_accuracy": 0.143421570956707,
|
|
"num_tokens": 3111347.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.346796178817749,
|
|
"epoch": 1.477868500214869,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004949393215201666,
|
|
"loss": 6.3502,
|
|
"mean_token_accuracy": 0.1382603771984577,
|
|
"num_tokens": 3120018.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.488202047348023,
|
|
"epoch": 1.4821658788139236,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004948689615011065,
|
|
"loss": 6.4354,
|
|
"mean_token_accuracy": 0.13543768227100372,
|
|
"num_tokens": 3129669.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.442359161376953,
|
|
"epoch": 1.4864632574129781,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004947981213761154,
|
|
"loss": 6.3728,
|
|
"mean_token_accuracy": 0.13687124252319335,
|
|
"num_tokens": 3139112.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.462133312225342,
|
|
"epoch": 1.4907606360120327,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004947268012998797,
|
|
"loss": 6.3215,
|
|
"mean_token_accuracy": 0.14909568279981614,
|
|
"num_tokens": 3148437.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.405123472213745,
|
|
"epoch": 1.4950580146110872,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000494655001428134,
|
|
"loss": 6.3124,
|
|
"mean_token_accuracy": 0.14245474264025687,
|
|
"num_tokens": 3158165.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.4186896800994875,
|
|
"epoch": 1.4993553932101418,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004945827219176604,
|
|
"loss": 6.4,
|
|
"mean_token_accuracy": 0.14140995144844054,
|
|
"num_tokens": 3167262.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.396775150299073,
|
|
"epoch": 1.5036527718091963,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004945099629262888,
|
|
"loss": 6.3753,
|
|
"mean_token_accuracy": 0.14143237695097924,
|
|
"num_tokens": 3176696.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.565005731582642,
|
|
"epoch": 1.5079501504082509,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004944367246128954,
|
|
"loss": 6.4545,
|
|
"mean_token_accuracy": 0.13297815918922423,
|
|
"num_tokens": 3185857.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.3231383800506595,
|
|
"epoch": 1.5122475290073054,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004943630071374036,
|
|
"loss": 6.2933,
|
|
"mean_token_accuracy": 0.14562440365552903,
|
|
"num_tokens": 3194687.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.474056005477905,
|
|
"epoch": 1.51654490760636,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004942888106607828,
|
|
"loss": 6.3535,
|
|
"mean_token_accuracy": 0.1362925410270691,
|
|
"num_tokens": 3204913.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.452696895599365,
|
|
"epoch": 1.5208422862054147,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004942141353450486,
|
|
"loss": 6.3414,
|
|
"mean_token_accuracy": 0.14594482108950615,
|
|
"num_tokens": 3213312.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.347502994537353,
|
|
"epoch": 1.5251396648044693,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004941389813532619,
|
|
"loss": 6.2698,
|
|
"mean_token_accuracy": 0.15359112471342087,
|
|
"num_tokens": 3222992.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.322586011886597,
|
|
"epoch": 1.5294370434035238,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000494063348849529,
|
|
"loss": 6.3051,
|
|
"mean_token_accuracy": 0.1480158232152462,
|
|
"num_tokens": 3232836.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.5091650009155275,
|
|
"epoch": 1.5337344220025786,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004939872379990011,
|
|
"loss": 6.4637,
|
|
"mean_token_accuracy": 0.13492583408951758,
|
|
"num_tokens": 3243171.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.4711463928222654,
|
|
"epoch": 1.5380318006016331,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004939106489678739,
|
|
"loss": 6.3839,
|
|
"mean_token_accuracy": 0.14319418519735336,
|
|
"num_tokens": 3251995.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.373999691009521,
|
|
"epoch": 1.5423291792006877,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000493833581923387,
|
|
"loss": 6.326,
|
|
"mean_token_accuracy": 0.14339777454733849,
|
|
"num_tokens": 3260841.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.5138959884643555,
|
|
"epoch": 1.5466265577997422,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004937560370338244,
|
|
"loss": 6.4805,
|
|
"mean_token_accuracy": 0.13039591386914254,
|
|
"num_tokens": 3270979.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.466180372238159,
|
|
"epoch": 1.5509239363987968,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000493678014468513,
|
|
"loss": 6.4021,
|
|
"mean_token_accuracy": 0.143599297106266,
|
|
"num_tokens": 3279848.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.3830766677856445,
|
|
"epoch": 1.5552213149978513,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004935995143978227,
|
|
"loss": 6.3995,
|
|
"mean_token_accuracy": 0.13850793689489366,
|
|
"num_tokens": 3289172.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.427796506881714,
|
|
"epoch": 1.5595186935969059,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004935205369931664,
|
|
"loss": 6.275,
|
|
"mean_token_accuracy": 0.14568763375282287,
|
|
"num_tokens": 3297432.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.4932654857635494,
|
|
"epoch": 1.5638160721959604,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004934410824269992,
|
|
"loss": 6.3083,
|
|
"mean_token_accuracy": 0.14067512452602388,
|
|
"num_tokens": 3307486.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.30113582611084,
|
|
"epoch": 1.568113450795015,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004933611508728182,
|
|
"loss": 6.2921,
|
|
"mean_token_accuracy": 0.1483084164559841,
|
|
"num_tokens": 3316296.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.427493619918823,
|
|
"epoch": 1.5724108293940695,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000493280742505162,
|
|
"loss": 6.3306,
|
|
"mean_token_accuracy": 0.1380011484026909,
|
|
"num_tokens": 3326080.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.297157955169678,
|
|
"epoch": 1.576708207993124,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004931998574996102,
|
|
"loss": 6.2414,
|
|
"mean_token_accuracy": 0.14382967874407768,
|
|
"num_tokens": 3334826.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.3207234859466555,
|
|
"epoch": 1.5810055865921788,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004931184960327832,
|
|
"loss": 6.2526,
|
|
"mean_token_accuracy": 0.1502455472946167,
|
|
"num_tokens": 3343261.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.337889814376831,
|
|
"epoch": 1.5853029651912334,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.0004930366582823421,
|
|
"loss": 6.3009,
|
|
"mean_token_accuracy": 0.140691427141428,
|
|
"num_tokens": 3352513.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.36025972366333,
|
|
"epoch": 1.589600343790288,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004929543444269879,
|
|
"loss": 6.319,
|
|
"mean_token_accuracy": 0.1484783448278904,
|
|
"num_tokens": 3361577.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.440754079818726,
|
|
"epoch": 1.5938977223893425,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000492871554646461,
|
|
"loss": 6.4157,
|
|
"mean_token_accuracy": 0.13772669062018394,
|
|
"num_tokens": 3370591.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.370141792297363,
|
|
"epoch": 1.5981951009883972,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004927882891215413,
|
|
"loss": 6.3196,
|
|
"mean_token_accuracy": 0.1388826124370098,
|
|
"num_tokens": 3379761.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.4343994617462155,
|
|
"epoch": 1.6024924795874518,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004927045480340475,
|
|
"loss": 6.4095,
|
|
"mean_token_accuracy": 0.13787330240011214,
|
|
"num_tokens": 3388974.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.3982750415802006,
|
|
"epoch": 1.6067898581865063,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004926203315668363,
|
|
"loss": 6.3234,
|
|
"mean_token_accuracy": 0.14442344531416892,
|
|
"num_tokens": 3398339.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.38454155921936,
|
|
"epoch": 1.6110872367855609,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004925356399038032,
|
|
"loss": 6.2867,
|
|
"mean_token_accuracy": 0.13828113302588463,
|
|
"num_tokens": 3408292.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.368298768997192,
|
|
"epoch": 1.6153846153846154,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004924504732298808,
|
|
"loss": 6.2622,
|
|
"mean_token_accuracy": 0.14790038838982583,
|
|
"num_tokens": 3417057.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.436639022827149,
|
|
"epoch": 1.61968199398367,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004923648317310391,
|
|
"loss": 6.381,
|
|
"mean_token_accuracy": 0.14184708818793296,
|
|
"num_tokens": 3425830.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.369312191009522,
|
|
"epoch": 1.6239793725827245,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004922787155942849,
|
|
"loss": 6.411,
|
|
"mean_token_accuracy": 0.13301676735281945,
|
|
"num_tokens": 3435513.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.379889011383057,
|
|
"epoch": 1.628276751181779,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004921921250076611,
|
|
"loss": 6.3129,
|
|
"mean_token_accuracy": 0.14551183134317397,
|
|
"num_tokens": 3444684.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.3351727485656735,
|
|
"epoch": 1.6325741297808336,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004921050601602475,
|
|
"loss": 6.3717,
|
|
"mean_token_accuracy": 0.14514745026826859,
|
|
"num_tokens": 3453454.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.364632034301758,
|
|
"epoch": 1.6368715083798882,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004920175212421587,
|
|
"loss": 6.3024,
|
|
"mean_token_accuracy": 0.1404518447816372,
|
|
"num_tokens": 3463228.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.370801830291748,
|
|
"epoch": 1.6411688869789427,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004919295084445445,
|
|
"loss": 6.2039,
|
|
"mean_token_accuracy": 0.15301429331302643,
|
|
"num_tokens": 3472131.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.300609064102173,
|
|
"epoch": 1.6454662655779975,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004918410219595899,
|
|
"loss": 6.2529,
|
|
"mean_token_accuracy": 0.15238626301288605,
|
|
"num_tokens": 3480642.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.482364797592163,
|
|
"epoch": 1.649763644177052,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000491752061980514,
|
|
"loss": 6.2502,
|
|
"mean_token_accuracy": 0.14723040014505387,
|
|
"num_tokens": 3489346.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.245130395889282,
|
|
"epoch": 1.6540610227761066,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004916626287015697,
|
|
"loss": 6.2956,
|
|
"mean_token_accuracy": 0.14454371482133865,
|
|
"num_tokens": 3498473.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.452545595169068,
|
|
"epoch": 1.658358401375161,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004915727223180436,
|
|
"loss": 6.3046,
|
|
"mean_token_accuracy": 0.14169235825538634,
|
|
"num_tokens": 3507415.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.395853805541992,
|
|
"epoch": 1.6626557799742159,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004914823430262554,
|
|
"loss": 6.4133,
|
|
"mean_token_accuracy": 0.13193509206175805,
|
|
"num_tokens": 3516873.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.325657653808594,
|
|
"epoch": 1.6669531585732704,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004913914910235573,
|
|
"loss": 6.2705,
|
|
"mean_token_accuracy": 0.14801762700080873,
|
|
"num_tokens": 3525047.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.363438558578491,
|
|
"epoch": 1.671250537172325,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004913001665083337,
|
|
"loss": 6.2905,
|
|
"mean_token_accuracy": 0.14471901580691338,
|
|
"num_tokens": 3534354.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.308793354034424,
|
|
"epoch": 1.6755479157713795,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004912083696800008,
|
|
"loss": 6.3216,
|
|
"mean_token_accuracy": 0.1346129782497883,
|
|
"num_tokens": 3543830.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.393164157867432,
|
|
"epoch": 1.679845294370434,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004911161007390063,
|
|
"loss": 6.2271,
|
|
"mean_token_accuracy": 0.14336094558238982,
|
|
"num_tokens": 3552314.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.246313667297363,
|
|
"epoch": 1.6841426729694886,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004910233598868287,
|
|
"loss": 6.3015,
|
|
"mean_token_accuracy": 0.14284233823418618,
|
|
"num_tokens": 3561656.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.327203702926636,
|
|
"epoch": 1.6884400515685432,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004909301473259769,
|
|
"loss": 6.2888,
|
|
"mean_token_accuracy": 0.14548132047057152,
|
|
"num_tokens": 3571784.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.413244581222534,
|
|
"epoch": 1.6927374301675977,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004908364632599899,
|
|
"loss": 6.2576,
|
|
"mean_token_accuracy": 0.1450193226337433,
|
|
"num_tokens": 3580626.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.291885328292847,
|
|
"epoch": 1.6970348087666522,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004907423078934362,
|
|
"loss": 6.2453,
|
|
"mean_token_accuracy": 0.148175410926342,
|
|
"num_tokens": 3589916.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.36896767616272,
|
|
"epoch": 1.7013321873657068,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004906476814319134,
|
|
"loss": 6.284,
|
|
"mean_token_accuracy": 0.14784513786435127,
|
|
"num_tokens": 3599128.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.303310966491699,
|
|
"epoch": 1.7056295659647613,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004905525840820481,
|
|
"loss": 6.2935,
|
|
"mean_token_accuracy": 0.14577944353222846,
|
|
"num_tokens": 3608764.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.450990772247314,
|
|
"epoch": 1.709926944563816,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004904570160514948,
|
|
"loss": 6.3469,
|
|
"mean_token_accuracy": 0.13999350219964982,
|
|
"num_tokens": 3619082.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.30384087562561,
|
|
"epoch": 1.7142243231628707,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004903609775489358,
|
|
"loss": 6.3176,
|
|
"mean_token_accuracy": 0.1439313054084778,
|
|
"num_tokens": 3628695.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.3533117294311525,
|
|
"epoch": 1.7185217017619252,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004902644687840809,
|
|
"loss": 6.2817,
|
|
"mean_token_accuracy": 0.14347996339201927,
|
|
"num_tokens": 3637599.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.7185217017619252,
|
|
"eval_entropy": 6.091574696807174,
|
|
"eval_loss": 6.368002414703369,
|
|
"eval_mean_token_accuracy": 0.1463287656807953,
|
|
"eval_num_tokens": 3637599.0,
|
|
"eval_runtime": 2.6343,
|
|
"eval_samples_per_second": 1347.23,
|
|
"eval_steps_per_second": 168.546,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.318877649307251,
|
|
"epoch": 1.7228190803609797,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004901674899676667,
|
|
"loss": 6.2771,
|
|
"mean_token_accuracy": 0.14787266850471498,
|
|
"num_tokens": 3647406.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.291737365722656,
|
|
"epoch": 1.7271164589600345,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004900700413114561,
|
|
"loss": 6.1647,
|
|
"mean_token_accuracy": 0.15139409750699998,
|
|
"num_tokens": 3656531.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.29096736907959,
|
|
"epoch": 1.731413837559089,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000489972123028238,
|
|
"loss": 6.2599,
|
|
"mean_token_accuracy": 0.14592444598674775,
|
|
"num_tokens": 3664922.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.352511978149414,
|
|
"epoch": 1.7357112161581436,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004898737353318268,
|
|
"loss": 6.1849,
|
|
"mean_token_accuracy": 0.1494704321026802,
|
|
"num_tokens": 3673283.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.282111120223999,
|
|
"epoch": 1.7400085947571982,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000489774878437062,
|
|
"loss": 6.3298,
|
|
"mean_token_accuracy": 0.14433063864707946,
|
|
"num_tokens": 3681760.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.375459289550781,
|
|
"epoch": 1.7443059733562527,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004896755525598074,
|
|
"loss": 6.144,
|
|
"mean_token_accuracy": 0.15006723552942275,
|
|
"num_tokens": 3689408.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.231076049804687,
|
|
"epoch": 1.7486033519553073,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004895757579169511,
|
|
"loss": 6.2602,
|
|
"mean_token_accuracy": 0.1471289351582527,
|
|
"num_tokens": 3697904.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.453372764587402,
|
|
"epoch": 1.7529007305543618,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004894754947264047,
|
|
"loss": 6.2916,
|
|
"mean_token_accuracy": 0.1484273798763752,
|
|
"num_tokens": 3706704.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.315323305130005,
|
|
"epoch": 1.7571981091534163,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000489374763207103,
|
|
"loss": 6.3484,
|
|
"mean_token_accuracy": 0.14002273380756378,
|
|
"num_tokens": 3715690.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.346850919723511,
|
|
"epoch": 1.761495487752471,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004892735635790033,
|
|
"loss": 6.1477,
|
|
"mean_token_accuracy": 0.15207798033952713,
|
|
"num_tokens": 3724835.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 6.233658790588379,
|
|
"epoch": 1.7657928663515254,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000489171896063085,
|
|
"loss": 6.1926,
|
|
"mean_token_accuracy": 0.14881692826747894,
|
|
"num_tokens": 3733977.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.367916107177734,
|
|
"epoch": 1.77009024495058,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004890697608813495,
|
|
"loss": 6.3022,
|
|
"mean_token_accuracy": 0.13871972635388374,
|
|
"num_tokens": 3742665.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.434072065353393,
|
|
"epoch": 1.7743876235496348,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004889671582568193,
|
|
"loss": 6.3778,
|
|
"mean_token_accuracy": 0.1384774811565876,
|
|
"num_tokens": 3751647.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.281269264221192,
|
|
"epoch": 1.7786850021486893,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004888640884135374,
|
|
"loss": 6.2769,
|
|
"mean_token_accuracy": 0.14352479204535484,
|
|
"num_tokens": 3760852.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 6.328350830078125,
|
|
"epoch": 1.7829823807477438,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004887605515765671,
|
|
"loss": 6.2042,
|
|
"mean_token_accuracy": 0.15113454610109328,
|
|
"num_tokens": 3768640.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.2967079162597654,
|
|
"epoch": 1.7872797593467986,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004886565479719914,
|
|
"loss": 6.2419,
|
|
"mean_token_accuracy": 0.1427251249551773,
|
|
"num_tokens": 3776504.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.399069786071777,
|
|
"epoch": 1.7915771379458532,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004885520778269128,
|
|
"loss": 6.2894,
|
|
"mean_token_accuracy": 0.14642380475997924,
|
|
"num_tokens": 3786353.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.313576030731201,
|
|
"epoch": 1.7958745165449077,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004884471413694523,
|
|
"loss": 6.3197,
|
|
"mean_token_accuracy": 0.14385596364736558,
|
|
"num_tokens": 3795902.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 6.347762823104858,
|
|
"epoch": 1.8001718951439623,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004883417388287491,
|
|
"loss": 6.2388,
|
|
"mean_token_accuracy": 0.1416109524667263,
|
|
"num_tokens": 3805986.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.318545532226563,
|
|
"epoch": 1.8044692737430168,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004882358704349603,
|
|
"loss": 6.3591,
|
|
"mean_token_accuracy": 0.1477951005101204,
|
|
"num_tokens": 3814915.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.37258472442627,
|
|
"epoch": 1.8087666523420713,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004881295364192601,
|
|
"loss": 6.2406,
|
|
"mean_token_accuracy": 0.1547567516565323,
|
|
"num_tokens": 3823966.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 6.420681047439575,
|
|
"epoch": 1.813064030941126,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004880227370138394,
|
|
"loss": 6.2861,
|
|
"mean_token_accuracy": 0.14362874031066894,
|
|
"num_tokens": 3832775.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.228458261489868,
|
|
"epoch": 1.8173614095401804,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004879154724519057,
|
|
"loss": 6.2121,
|
|
"mean_token_accuracy": 0.1502902552485466,
|
|
"num_tokens": 3842808.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 6.378505086898803,
|
|
"epoch": 1.821658788139235,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004878077429676816,
|
|
"loss": 6.3469,
|
|
"mean_token_accuracy": 0.1420361042022705,
|
|
"num_tokens": 3853303.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.311634635925293,
|
|
"epoch": 1.8259561667382895,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004876995487964054,
|
|
"loss": 6.25,
|
|
"mean_token_accuracy": 0.13848483711481094,
|
|
"num_tokens": 3862462.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.302492427825928,
|
|
"epoch": 1.830253545337344,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048759089017432996,
|
|
"loss": 6.3741,
|
|
"mean_token_accuracy": 0.14119420051574708,
|
|
"num_tokens": 3871596.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.385835075378418,
|
|
"epoch": 1.8345509239363988,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004874817673387222,
|
|
"loss": 6.2773,
|
|
"mean_token_accuracy": 0.1481023609638214,
|
|
"num_tokens": 3881276.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.259650325775146,
|
|
"epoch": 1.8388483025354534,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00048737218052786275,
|
|
"loss": 6.3435,
|
|
"mean_token_accuracy": 0.13825829476118087,
|
|
"num_tokens": 3891610.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.423104095458984,
|
|
"epoch": 1.843145681134508,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048726212998104554,
|
|
"loss": 6.288,
|
|
"mean_token_accuracy": 0.1408744402229786,
|
|
"num_tokens": 3900584.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.276949644088745,
|
|
"epoch": 1.8474430597335625,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004871516159385768,
|
|
"loss": 6.2314,
|
|
"mean_token_accuracy": 0.1494441844522953,
|
|
"num_tokens": 3910208.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.17384672164917,
|
|
"epoch": 1.8517404383326173,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004870406386417752,
|
|
"loss": 6.1413,
|
|
"mean_token_accuracy": 0.1555544376373291,
|
|
"num_tokens": 3918424.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.253190803527832,
|
|
"epoch": 1.8560378169316718,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004869291983329707,
|
|
"loss": 6.0808,
|
|
"mean_token_accuracy": 0.1601550966501236,
|
|
"num_tokens": 3926206.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 6.334660243988037,
|
|
"epoch": 1.8603351955307263,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004868172952555044,
|
|
"loss": 6.182,
|
|
"mean_token_accuracy": 0.13937367051839827,
|
|
"num_tokens": 3935769.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.230923652648926,
|
|
"epoch": 1.864632574129781,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004867049296537278,
|
|
"loss": 6.1661,
|
|
"mean_token_accuracy": 0.1482834830880165,
|
|
"num_tokens": 3945118.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.328389406204224,
|
|
"epoch": 1.8689299527288354,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004865921017730027,
|
|
"loss": 6.2661,
|
|
"mean_token_accuracy": 0.1476903736591339,
|
|
"num_tokens": 3954012.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 6.291385698318481,
|
|
"epoch": 1.87322733132789,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00048647881185969995,
|
|
"loss": 6.2742,
|
|
"mean_token_accuracy": 0.14401593804359436,
|
|
"num_tokens": 3964239.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 6.288094854354858,
|
|
"epoch": 1.8775247099269445,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004863650601611994,
|
|
"loss": 6.1621,
|
|
"mean_token_accuracy": 0.15318569391965867,
|
|
"num_tokens": 3973694.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.297281074523926,
|
|
"epoch": 1.881822088525999,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048625084692588937,
|
|
"loss": 6.2134,
|
|
"mean_token_accuracy": 0.15158866047859193,
|
|
"num_tokens": 3982706.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.197356653213501,
|
|
"epoch": 1.8861194671250536,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048613617240316593,
|
|
"loss": 6.174,
|
|
"mean_token_accuracy": 0.15385224521160126,
|
|
"num_tokens": 3990934.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.360083818435669,
|
|
"epoch": 1.8904168457241082,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004860210368434323,
|
|
"loss": 6.2316,
|
|
"mean_token_accuracy": 0.15272461101412774,
|
|
"num_tokens": 3999864.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 6.193952560424805,
|
|
"epoch": 1.8947142243231627,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00048590544049809857,
|
|
"loss": 6.2365,
|
|
"mean_token_accuracy": 0.1530345231294632,
|
|
"num_tokens": 4008273.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 6.370530462265014,
|
|
"epoch": 1.8990116029222175,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000485789383619581,
|
|
"loss": 6.249,
|
|
"mean_token_accuracy": 0.15136862844228743,
|
|
"num_tokens": 4017697.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.214098501205444,
|
|
"epoch": 1.903308981521272,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004856728664613015,
|
|
"loss": 6.2656,
|
|
"mean_token_accuracy": 0.1445502944290638,
|
|
"num_tokens": 4026775.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.3401977062225345,
|
|
"epoch": 1.9076063601203266,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00048555588927768674,
|
|
"loss": 6.2475,
|
|
"mean_token_accuracy": 0.14767130538821222,
|
|
"num_tokens": 4036476.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.289152050018311,
|
|
"epoch": 1.9119037387193811,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004854384523241683,
|
|
"loss": 6.2226,
|
|
"mean_token_accuracy": 0.14828546568751336,
|
|
"num_tokens": 4045221.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 6.2168541431427,
|
|
"epoch": 1.916201117318436,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048532055585718143,
|
|
"loss": 6.1471,
|
|
"mean_token_accuracy": 0.1514877662062645,
|
|
"num_tokens": 4053754.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 6.268859577178955,
|
|
"epoch": 1.9204984959174904,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048520220013416505,
|
|
"loss": 6.1911,
|
|
"mean_token_accuracy": 0.15194396525621415,
|
|
"num_tokens": 4061730.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 6.280244159698486,
|
|
"epoch": 1.924795874516545,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004850833854135607,
|
|
"loss": 6.2515,
|
|
"mean_token_accuracy": 0.1469673454761505,
|
|
"num_tokens": 4070501.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.289591121673584,
|
|
"epoch": 1.9290932531155995,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004849641119548122,
|
|
"loss": 6.3447,
|
|
"mean_token_accuracy": 0.1468745768070221,
|
|
"num_tokens": 4079621.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.453436183929443,
|
|
"epoch": 1.933390631714654,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000484844380018365,
|
|
"loss": 6.3098,
|
|
"mean_token_accuracy": 0.1452730752527714,
|
|
"num_tokens": 4090106.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.334768152236938,
|
|
"epoch": 1.9376880103137086,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000484724189865666,
|
|
"loss": 6.2396,
|
|
"mean_token_accuracy": 0.14894529432058334,
|
|
"num_tokens": 4099269.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 6.102892255783081,
|
|
"epoch": 1.9419853889127632,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004846035417591624,
|
|
"loss": 6.1739,
|
|
"mean_token_accuracy": 0.15417957305908203,
|
|
"num_tokens": 4108414.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 6.362226295471191,
|
|
"epoch": 1.9462827675118177,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004844824359623014,
|
|
"loss": 6.2842,
|
|
"mean_token_accuracy": 0.14404889196157455,
|
|
"num_tokens": 4117731.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.320440101623535,
|
|
"epoch": 1.9505801461108723,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048436087273952966,
|
|
"loss": 6.2651,
|
|
"mean_token_accuracy": 0.1450080007314682,
|
|
"num_tokens": 4127194.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 6.2928619384765625,
|
|
"epoch": 1.9548775247099268,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048423885235629265,
|
|
"loss": 6.2306,
|
|
"mean_token_accuracy": 0.15602879226207733,
|
|
"num_tokens": 4135594.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 6.247224378585815,
|
|
"epoch": 1.9591749033089814,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004841163750790342,
|
|
"loss": 6.2688,
|
|
"mean_token_accuracy": 0.14728586077690126,
|
|
"num_tokens": 4145027.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.255398988723755,
|
|
"epoch": 1.9634722819080361,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048399344117519555,
|
|
"loss": 6.1311,
|
|
"mean_token_accuracy": 0.152347894012928,
|
|
"num_tokens": 4153754.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 6.192139434814453,
|
|
"epoch": 1.9677696605070907,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00048387005091321544,
|
|
"loss": 6.1927,
|
|
"mean_token_accuracy": 0.15041765347123146,
|
|
"num_tokens": 4162765.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.372733211517334,
|
|
"epoch": 1.9720670391061452,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048374620456252877,
|
|
"loss": 6.2137,
|
|
"mean_token_accuracy": 0.14727154970169068,
|
|
"num_tokens": 4171589.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 6.218329286575317,
|
|
"epoch": 1.9763644177052,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048362190239356644,
|
|
"loss": 6.2076,
|
|
"mean_token_accuracy": 0.14930796027183532,
|
|
"num_tokens": 4181817.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 6.213005685806275,
|
|
"epoch": 1.9806617963042545,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00048349714467775474,
|
|
"loss": 6.1767,
|
|
"mean_token_accuracy": 0.14748584479093552,
|
|
"num_tokens": 4191350.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 6.370962333679199,
|
|
"epoch": 1.984959174903309,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048337193168751464,
|
|
"loss": 6.2288,
|
|
"mean_token_accuracy": 0.1438884899020195,
|
|
"num_tokens": 4199888.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.2559257507324215,
|
|
"epoch": 1.9892565535023636,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004832462636962613,
|
|
"loss": 6.1964,
|
|
"mean_token_accuracy": 0.14573389440774917,
|
|
"num_tokens": 4209509.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 6.252245616912842,
|
|
"epoch": 1.9935539321014182,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004831201409784034,
|
|
"loss": 6.1734,
|
|
"mean_token_accuracy": 0.15078805685043334,
|
|
"num_tokens": 4218496.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 6.245168590545655,
|
|
"epoch": 1.9978513107004727,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004829935638093424,
|
|
"loss": 6.1689,
|
|
"mean_token_accuracy": 0.15246346592903137,
|
|
"num_tokens": 4227504.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 6.250787470075819,
|
|
"epoch": 2.0017189514396216,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004828665324654724,
|
|
"loss": 6.0796,
|
|
"mean_token_accuracy": 0.1526774697833591,
|
|
"num_tokens": 4235338.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 6.299807786941528,
|
|
"epoch": 2.006016330038676,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004827390472241791,
|
|
"loss": 5.8319,
|
|
"mean_token_accuracy": 0.1594364270567894,
|
|
"num_tokens": 4244905.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 6.186063194274903,
|
|
"epoch": 2.010313708637731,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004826111083638392,
|
|
"loss": 5.9311,
|
|
"mean_token_accuracy": 0.15949834063649176,
|
|
"num_tokens": 4254533.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 6.186539793014527,
|
|
"epoch": 2.0146110872367857,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048248271616382,
|
|
"loss": 5.9071,
|
|
"mean_token_accuracy": 0.1573925197124481,
|
|
"num_tokens": 4264023.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 6.196358966827392,
|
|
"epoch": 2.0189084658358403,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048235387090447894,
|
|
"loss": 5.9306,
|
|
"mean_token_accuracy": 0.15745894163846968,
|
|
"num_tokens": 4273298.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 6.200879001617432,
|
|
"epoch": 2.023205844434895,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048222457286716235,
|
|
"loss": 5.8942,
|
|
"mean_token_accuracy": 0.16266611367464065,
|
|
"num_tokens": 4283244.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 6.158419275283814,
|
|
"epoch": 2.0275032230339494,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00048209482233420564,
|
|
"loss": 5.8262,
|
|
"mean_token_accuracy": 0.16634698510169982,
|
|
"num_tokens": 4291677.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 6.176835346221924,
|
|
"epoch": 2.031800601633004,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000481964619588932,
|
|
"loss": 5.9018,
|
|
"mean_token_accuracy": 0.16285468339920045,
|
|
"num_tokens": 4300822.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 6.165877914428711,
|
|
"epoch": 2.0360979802320585,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004818339649156523,
|
|
"loss": 5.9042,
|
|
"mean_token_accuracy": 0.1610295593738556,
|
|
"num_tokens": 4310149.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 6.042674922943116,
|
|
"epoch": 2.040395358831113,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048170285859966395,
|
|
"loss": 5.8132,
|
|
"mean_token_accuracy": 0.17229840457439421,
|
|
"num_tokens": 4319109.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 6.18833737373352,
|
|
"epoch": 2.0446927374301676,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048157130092725087,
|
|
"loss": 5.7992,
|
|
"mean_token_accuracy": 0.16624504029750825,
|
|
"num_tokens": 4327921.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 6.128254032135009,
|
|
"epoch": 2.048990116029222,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004814392921856824,
|
|
"loss": 5.9541,
|
|
"mean_token_accuracy": 0.16383159458637236,
|
|
"num_tokens": 4338026.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 6.095228052139282,
|
|
"epoch": 2.0532874946282766,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004813068326632128,
|
|
"loss": 5.7884,
|
|
"mean_token_accuracy": 0.168255914747715,
|
|
"num_tokens": 4347794.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 6.24313006401062,
|
|
"epoch": 2.057584873227331,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004811739226490809,
|
|
"loss": 5.9717,
|
|
"mean_token_accuracy": 0.15837548822164535,
|
|
"num_tokens": 4357249.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 6.085185909271241,
|
|
"epoch": 2.0618822518263857,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048104056243350896,
|
|
"loss": 5.9248,
|
|
"mean_token_accuracy": 0.1550598293542862,
|
|
"num_tokens": 4366053.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.192628812789917,
|
|
"epoch": 2.0661796304254403,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004809067523077023,
|
|
"loss": 5.9337,
|
|
"mean_token_accuracy": 0.16046450436115264,
|
|
"num_tokens": 4375543.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.059787702560425,
|
|
"epoch": 2.0704770090244953,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048077249256384884,
|
|
"loss": 5.8025,
|
|
"mean_token_accuracy": 0.16983114629983903,
|
|
"num_tokens": 4384332.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 6.084091567993164,
|
|
"epoch": 2.07477438762355,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004806377834951182,
|
|
"loss": 5.9162,
|
|
"mean_token_accuracy": 0.15598006695508956,
|
|
"num_tokens": 4393670.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 6.219714832305908,
|
|
"epoch": 2.0790717662226044,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048050262539566104,
|
|
"loss": 5.9082,
|
|
"mean_token_accuracy": 0.16180012375116348,
|
|
"num_tokens": 4402763.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 6.109860849380493,
|
|
"epoch": 2.083369144821659,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004803670185606087,
|
|
"loss": 5.8379,
|
|
"mean_token_accuracy": 0.16459062099456787,
|
|
"num_tokens": 4411863.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 6.097035360336304,
|
|
"epoch": 2.0876665234207135,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004802309632860724,
|
|
"loss": 5.9035,
|
|
"mean_token_accuracy": 0.16392969191074372,
|
|
"num_tokens": 4421110.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 6.110820770263672,
|
|
"epoch": 2.091963902019768,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048009445986914236,
|
|
"loss": 5.9159,
|
|
"mean_token_accuracy": 0.15644611269235612,
|
|
"num_tokens": 4430249.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.130997562408448,
|
|
"epoch": 2.0962612806188226,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047995750860788756,
|
|
"loss": 5.8785,
|
|
"mean_token_accuracy": 0.15592641681432723,
|
|
"num_tokens": 4439686.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 6.098659610748291,
|
|
"epoch": 2.100558659217877,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004798201098013547,
|
|
"loss": 5.8752,
|
|
"mean_token_accuracy": 0.1653648167848587,
|
|
"num_tokens": 4448645.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 6.098980140686035,
|
|
"epoch": 2.1048560378169316,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047968226374956797,
|
|
"loss": 5.8269,
|
|
"mean_token_accuracy": 0.1635419562458992,
|
|
"num_tokens": 4456870.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 6.087638330459595,
|
|
"epoch": 2.109153416415986,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047954397075352794,
|
|
"loss": 5.8799,
|
|
"mean_token_accuracy": 0.16591449379920958,
|
|
"num_tokens": 4466287.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 6.110076379776001,
|
|
"epoch": 2.1134507950150407,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047940523111521136,
|
|
"loss": 5.7739,
|
|
"mean_token_accuracy": 0.16818707287311555,
|
|
"num_tokens": 4474461.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 6.054969310760498,
|
|
"epoch": 2.1177481736140953,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004792660451375701,
|
|
"loss": 5.8418,
|
|
"mean_token_accuracy": 0.16488105356693267,
|
|
"num_tokens": 4483002.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 6.0769671440124515,
|
|
"epoch": 2.12204555221315,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047912641312453064,
|
|
"loss": 5.8564,
|
|
"mean_token_accuracy": 0.16275329440832137,
|
|
"num_tokens": 4492061.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 6.130937194824218,
|
|
"epoch": 2.1263429308122044,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00047898633538099363,
|
|
"loss": 5.9163,
|
|
"mean_token_accuracy": 0.15475746542215346,
|
|
"num_tokens": 4501829.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 6.041531896591186,
|
|
"epoch": 2.130640309411259,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004788458122128327,
|
|
"loss": 5.9373,
|
|
"mean_token_accuracy": 0.1593800500035286,
|
|
"num_tokens": 4511539.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 6.092942094802856,
|
|
"epoch": 2.134937688010314,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047870484392689434,
|
|
"loss": 5.806,
|
|
"mean_token_accuracy": 0.1652669280767441,
|
|
"num_tokens": 4520425.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 6.10265851020813,
|
|
"epoch": 2.1392350666093685,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000478563430830997,
|
|
"loss": 5.8856,
|
|
"mean_token_accuracy": 0.15846171379089355,
|
|
"num_tokens": 4529474.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 6.099829435348511,
|
|
"epoch": 2.143532445208423,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047842157323393035,
|
|
"loss": 5.84,
|
|
"mean_token_accuracy": 0.1621888354420662,
|
|
"num_tokens": 4538082.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 6.131585359573364,
|
|
"epoch": 2.1478298238074776,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004782792714454547,
|
|
"loss": 6.0238,
|
|
"mean_token_accuracy": 0.15583711564540864,
|
|
"num_tokens": 4547340.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.1478298238074776,
|
|
"eval_entropy": 6.039603040025041,
|
|
"eval_loss": 6.27166223526001,
|
|
"eval_mean_token_accuracy": 0.1527276904666209,
|
|
"eval_num_tokens": 4547340.0,
|
|
"eval_runtime": 2.6239,
|
|
"eval_samples_per_second": 1352.543,
|
|
"eval_steps_per_second": 169.211,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 6.112734031677246,
|
|
"epoch": 2.152127202406532,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004781365257763002,
|
|
"loss": 5.8688,
|
|
"mean_token_accuracy": 0.16214950904250144,
|
|
"num_tokens": 4556415.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 6.020659351348877,
|
|
"epoch": 2.1564245810055866,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00047799333653816633,
|
|
"loss": 5.7514,
|
|
"mean_token_accuracy": 0.17052749693393707,
|
|
"num_tokens": 4565156.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 6.076334428787232,
|
|
"epoch": 2.160721959604641,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047784970404372124,
|
|
"loss": 5.8509,
|
|
"mean_token_accuracy": 0.1637386590242386,
|
|
"num_tokens": 4574678.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 6.012679958343506,
|
|
"epoch": 2.1650193382036957,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047770562860660083,
|
|
"loss": 5.8575,
|
|
"mean_token_accuracy": 0.15799981504678726,
|
|
"num_tokens": 4583253.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 6.172503328323364,
|
|
"epoch": 2.1693167168027503,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004775611105414083,
|
|
"loss": 5.9464,
|
|
"mean_token_accuracy": 0.15288824141025542,
|
|
"num_tokens": 4594042.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 6.031742715835572,
|
|
"epoch": 2.173614095401805,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004774161501637133,
|
|
"loss": 5.9014,
|
|
"mean_token_accuracy": 0.16449007391929626,
|
|
"num_tokens": 4603128.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 6.089631795883179,
|
|
"epoch": 2.1779114740008594,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004772707477900514,
|
|
"loss": 5.8645,
|
|
"mean_token_accuracy": 0.1612997442483902,
|
|
"num_tokens": 4611537.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 6.1750177383422855,
|
|
"epoch": 2.182208852599914,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004771249037379232,
|
|
"loss": 5.9786,
|
|
"mean_token_accuracy": 0.1547987088561058,
|
|
"num_tokens": 4622481.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 6.0418188095092775,
|
|
"epoch": 2.1865062311989685,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004769786183257939,
|
|
"loss": 5.8752,
|
|
"mean_token_accuracy": 0.16891982555389404,
|
|
"num_tokens": 4631259.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 6.038510131835937,
|
|
"epoch": 2.190803609798023,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004768318918730924,
|
|
"loss": 5.8046,
|
|
"mean_token_accuracy": 0.1702958658337593,
|
|
"num_tokens": 4640266.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 6.085853242874146,
|
|
"epoch": 2.195100988397078,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047668472470021044,
|
|
"loss": 5.8902,
|
|
"mean_token_accuracy": 0.15410513281822205,
|
|
"num_tokens": 4649520.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 6.106590795516968,
|
|
"epoch": 2.1993983669961326,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004765371171285025,
|
|
"loss": 5.8194,
|
|
"mean_token_accuracy": 0.16876625418663024,
|
|
"num_tokens": 4658501.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.971577548980713,
|
|
"epoch": 2.203695745595187,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047638906948028445,
|
|
"loss": 5.8927,
|
|
"mean_token_accuracy": 0.15966321527957916,
|
|
"num_tokens": 4667567.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 6.086361265182495,
|
|
"epoch": 2.2079931241942417,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047624058207883317,
|
|
"loss": 5.8803,
|
|
"mean_token_accuracy": 0.1661355510354042,
|
|
"num_tokens": 4676618.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 6.191226530075073,
|
|
"epoch": 2.212290502793296,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047609165524838576,
|
|
"loss": 5.9549,
|
|
"mean_token_accuracy": 0.1582304894924164,
|
|
"num_tokens": 4685967.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.970425224304199,
|
|
"epoch": 2.2165878813923507,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004759422893141389,
|
|
"loss": 5.8483,
|
|
"mean_token_accuracy": 0.1639987826347351,
|
|
"num_tokens": 4694568.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 6.077778053283692,
|
|
"epoch": 2.2208852599914053,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004757924846022482,
|
|
"loss": 5.9149,
|
|
"mean_token_accuracy": 0.16335686296224594,
|
|
"num_tokens": 4703648.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 6.03687014579773,
|
|
"epoch": 2.22518263859046,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047564224143982714,
|
|
"loss": 5.7617,
|
|
"mean_token_accuracy": 0.16921258717775345,
|
|
"num_tokens": 4712444.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 6.101342964172363,
|
|
"epoch": 2.2294800171895144,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047549156015494676,
|
|
"loss": 5.9223,
|
|
"mean_token_accuracy": 0.1618524432182312,
|
|
"num_tokens": 4722034.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.988636541366577,
|
|
"epoch": 2.233777395788569,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047534044107663484,
|
|
"loss": 5.9315,
|
|
"mean_token_accuracy": 0.16420270651578903,
|
|
"num_tokens": 4731344.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 6.179015493392944,
|
|
"epoch": 2.2380747743876235,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047518888453487496,
|
|
"loss": 5.8325,
|
|
"mean_token_accuracy": 0.16798994094133377,
|
|
"num_tokens": 4739302.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.9993438720703125,
|
|
"epoch": 2.242372152986678,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004750368908606061,
|
|
"loss": 5.9712,
|
|
"mean_token_accuracy": 0.1615530237555504,
|
|
"num_tokens": 4748848.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 6.1699446678161625,
|
|
"epoch": 2.2466695315857326,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047488446038572164,
|
|
"loss": 6.0177,
|
|
"mean_token_accuracy": 0.1542590841650963,
|
|
"num_tokens": 4758194.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 6.096167087554932,
|
|
"epoch": 2.250966910184787,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004747315934430688,
|
|
"loss": 5.9141,
|
|
"mean_token_accuracy": 0.1600179821252823,
|
|
"num_tokens": 4768081.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 6.005567502975464,
|
|
"epoch": 2.2552642887838417,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000474578290366448,
|
|
"loss": 5.8429,
|
|
"mean_token_accuracy": 0.16374872177839278,
|
|
"num_tokens": 4776471.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 6.133287858963013,
|
|
"epoch": 2.259561667382896,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004744245514906117,
|
|
"loss": 5.8587,
|
|
"mean_token_accuracy": 0.15926120728254317,
|
|
"num_tokens": 4784403.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.980273675918579,
|
|
"epoch": 2.263859045981951,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047427037715126426,
|
|
"loss": 5.816,
|
|
"mean_token_accuracy": 0.16032798886299132,
|
|
"num_tokens": 4792779.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 6.001334095001221,
|
|
"epoch": 2.2681564245810057,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004741157676850608,
|
|
"loss": 5.8348,
|
|
"mean_token_accuracy": 0.1639099195599556,
|
|
"num_tokens": 4801426.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 6.066247367858887,
|
|
"epoch": 2.2724538031800603,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00047396072342960663,
|
|
"loss": 5.8534,
|
|
"mean_token_accuracy": 0.162247434258461,
|
|
"num_tokens": 4810329.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 6.035758686065674,
|
|
"epoch": 2.276751181779115,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047380524472345645,
|
|
"loss": 5.9372,
|
|
"mean_token_accuracy": 0.1603607341647148,
|
|
"num_tokens": 4819544.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 6.040545797348022,
|
|
"epoch": 2.2810485603781694,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004736493319061134,
|
|
"loss": 5.9218,
|
|
"mean_token_accuracy": 0.16008897721767426,
|
|
"num_tokens": 4828113.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 6.024325180053711,
|
|
"epoch": 2.285345938977224,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004734929853180291,
|
|
"loss": 5.9024,
|
|
"mean_token_accuracy": 0.1628168672323227,
|
|
"num_tokens": 4836989.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 6.152593612670898,
|
|
"epoch": 2.2896433175762785,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047333620530060175,
|
|
"loss": 5.9454,
|
|
"mean_token_accuracy": 0.1576353132724762,
|
|
"num_tokens": 4847103.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 6.021273136138916,
|
|
"epoch": 2.293940696175333,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004731789921961764,
|
|
"loss": 5.9643,
|
|
"mean_token_accuracy": 0.15701263695955275,
|
|
"num_tokens": 4856238.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 6.13275842666626,
|
|
"epoch": 2.2982380747743876,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004730213463480434,
|
|
"loss": 5.8596,
|
|
"mean_token_accuracy": 0.1687108978629112,
|
|
"num_tokens": 4864608.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 6.022077465057373,
|
|
"epoch": 2.302535453373442,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047286326810043857,
|
|
"loss": 5.8102,
|
|
"mean_token_accuracy": 0.16301658600568772,
|
|
"num_tokens": 4873889.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 6.018551015853882,
|
|
"epoch": 2.3068328319724967,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047270475779854137,
|
|
"loss": 5.8724,
|
|
"mean_token_accuracy": 0.16501303017139435,
|
|
"num_tokens": 4882902.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 6.1423060417175295,
|
|
"epoch": 2.311130210571551,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047254581578847507,
|
|
"loss": 5.9,
|
|
"mean_token_accuracy": 0.16175711005926133,
|
|
"num_tokens": 4892390.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.996535396575927,
|
|
"epoch": 2.3154275891706058,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004723864424173055,
|
|
"loss": 6.0089,
|
|
"mean_token_accuracy": 0.15623133033514022,
|
|
"num_tokens": 4901625.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 6.0932330131530765,
|
|
"epoch": 2.3197249677696608,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004722266380330403,
|
|
"loss": 5.7842,
|
|
"mean_token_accuracy": 0.17170698791742325,
|
|
"num_tokens": 4910804.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 6.0394035339355465,
|
|
"epoch": 2.3240223463687153,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047206640298462857,
|
|
"loss": 5.8615,
|
|
"mean_token_accuracy": 0.16718332320451737,
|
|
"num_tokens": 4920441.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 6.0346497058868405,
|
|
"epoch": 2.32831972496777,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047190573762195945,
|
|
"loss": 5.9206,
|
|
"mean_token_accuracy": 0.1583698570728302,
|
|
"num_tokens": 4930204.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 6.057749891281128,
|
|
"epoch": 2.3326171035668244,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00047174464229586186,
|
|
"loss": 6.0217,
|
|
"mean_token_accuracy": 0.15289230942726134,
|
|
"num_tokens": 4941191.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 6.169630098342895,
|
|
"epoch": 2.336914482165879,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004715831173581036,
|
|
"loss": 5.9768,
|
|
"mean_token_accuracy": 0.15604398101568223,
|
|
"num_tokens": 4951825.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 6.009410858154297,
|
|
"epoch": 2.3412118607649335,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047142116316139073,
|
|
"loss": 5.9168,
|
|
"mean_token_accuracy": 0.1644278198480606,
|
|
"num_tokens": 4960632.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 6.152362537384033,
|
|
"epoch": 2.345509239363988,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004712587800593663,
|
|
"loss": 5.932,
|
|
"mean_token_accuracy": 0.16140228062868117,
|
|
"num_tokens": 4969455.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 6.026926755905151,
|
|
"epoch": 2.3498066179630426,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004710959684066102,
|
|
"loss": 5.8494,
|
|
"mean_token_accuracy": 0.16802738159894942,
|
|
"num_tokens": 4978997.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 6.05658655166626,
|
|
"epoch": 2.354103996562097,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047093272855863803,
|
|
"loss": 5.9097,
|
|
"mean_token_accuracy": 0.16449163407087325,
|
|
"num_tokens": 4988305.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.999591207504272,
|
|
"epoch": 2.3584013751611517,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004707690608719003,
|
|
"loss": 5.8536,
|
|
"mean_token_accuracy": 0.17064749151468278,
|
|
"num_tokens": 4997022.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 6.097990083694458,
|
|
"epoch": 2.362698753760206,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004706049657037818,
|
|
"loss": 5.9126,
|
|
"mean_token_accuracy": 0.1574753388762474,
|
|
"num_tokens": 5005664.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 6.010551643371582,
|
|
"epoch": 2.3669961323592608,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004704404434126009,
|
|
"loss": 5.8714,
|
|
"mean_token_accuracy": 0.15295194536447526,
|
|
"num_tokens": 5014769.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 6.082769918441772,
|
|
"epoch": 2.3712935109583153,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047027549435760843,
|
|
"loss": 5.9393,
|
|
"mean_token_accuracy": 0.1611029714345932,
|
|
"num_tokens": 5024060.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 6.1582804203033445,
|
|
"epoch": 2.37559088955737,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004701101188989872,
|
|
"loss": 5.9804,
|
|
"mean_token_accuracy": 0.155602391064167,
|
|
"num_tokens": 5033046.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.99857497215271,
|
|
"epoch": 2.3798882681564244,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00046994431739785114,
|
|
"loss": 5.8236,
|
|
"mean_token_accuracy": 0.1750484824180603,
|
|
"num_tokens": 5040894.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 6.079819822311402,
|
|
"epoch": 2.384185646755479,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046977809021624454,
|
|
"loss": 5.9986,
|
|
"mean_token_accuracy": 0.159964120388031,
|
|
"num_tokens": 5050961.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 6.148093128204346,
|
|
"epoch": 2.3884830253545335,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004696114377171409,
|
|
"loss": 5.923,
|
|
"mean_token_accuracy": 0.15658502727746965,
|
|
"num_tokens": 5060226.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 6.005710458755493,
|
|
"epoch": 2.3927804039535885,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004694443602644429,
|
|
"loss": 5.891,
|
|
"mean_token_accuracy": 0.15860626548528672,
|
|
"num_tokens": 5069225.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 6.099031782150268,
|
|
"epoch": 2.397077782552643,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004692768582229808,
|
|
"loss": 5.872,
|
|
"mean_token_accuracy": 0.1625506117939949,
|
|
"num_tokens": 5078386.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.995739078521728,
|
|
"epoch": 2.4013751611516976,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046910893195851213,
|
|
"loss": 5.8067,
|
|
"mean_token_accuracy": 0.1610507681965828,
|
|
"num_tokens": 5087161.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 6.081733751296997,
|
|
"epoch": 2.405672539750752,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046894058183772074,
|
|
"loss": 5.9655,
|
|
"mean_token_accuracy": 0.16117437928915024,
|
|
"num_tokens": 5096613.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 6.038573598861694,
|
|
"epoch": 2.4099699183498067,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000468771808228216,
|
|
"loss": 5.9261,
|
|
"mean_token_accuracy": 0.1605885148048401,
|
|
"num_tokens": 5106534.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 6.009059953689575,
|
|
"epoch": 2.414267296948861,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00046860261149853197,
|
|
"loss": 5.9486,
|
|
"mean_token_accuracy": 0.15776654332876205,
|
|
"num_tokens": 5115975.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 6.048645448684693,
|
|
"epoch": 2.4185646755479158,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004684329920181268,
|
|
"loss": 5.8462,
|
|
"mean_token_accuracy": 0.16311540752649306,
|
|
"num_tokens": 5124635.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 6.0368053913116455,
|
|
"epoch": 2.4228620541469703,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00046826295015738154,
|
|
"loss": 5.8119,
|
|
"mean_token_accuracy": 0.17295191586017608,
|
|
"num_tokens": 5133226.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.987456226348877,
|
|
"epoch": 2.427159432746025,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004680924862875996,
|
|
"loss": 5.8974,
|
|
"mean_token_accuracy": 0.16351867467164993,
|
|
"num_tokens": 5142257.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 6.084688806533814,
|
|
"epoch": 2.4314568113450794,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046792160078100605,
|
|
"loss": 5.9038,
|
|
"mean_token_accuracy": 0.16701326072216033,
|
|
"num_tokens": 5150752.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 6.010568571090698,
|
|
"epoch": 2.435754189944134,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046775029401074653,
|
|
"loss": 5.7857,
|
|
"mean_token_accuracy": 0.17085398733615875,
|
|
"num_tokens": 5160237.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 6.090199661254883,
|
|
"epoch": 2.4400515685431885,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00046757856635088645,
|
|
"loss": 5.8873,
|
|
"mean_token_accuracy": 0.1663772866129875,
|
|
"num_tokens": 5169752.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.990356922149658,
|
|
"epoch": 2.444348947142243,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004674064181764105,
|
|
"loss": 5.9324,
|
|
"mean_token_accuracy": 0.16507768481969834,
|
|
"num_tokens": 5178892.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 6.05471134185791,
|
|
"epoch": 2.448646325741298,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046723384986322147,
|
|
"loss": 5.9014,
|
|
"mean_token_accuracy": 0.16048227846622468,
|
|
"num_tokens": 5188468.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.975317525863647,
|
|
"epoch": 2.4529437043403526,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004670608617881395,
|
|
"loss": 5.8258,
|
|
"mean_token_accuracy": 0.16768420934677125,
|
|
"num_tokens": 5197565.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.986767673492432,
|
|
"epoch": 2.457241082939407,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004668874543289014,
|
|
"loss": 5.8118,
|
|
"mean_token_accuracy": 0.17526791095733643,
|
|
"num_tokens": 5205791.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.987915706634522,
|
|
"epoch": 2.4615384615384617,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046671362786415986,
|
|
"loss": 5.8336,
|
|
"mean_token_accuracy": 0.1755102038383484,
|
|
"num_tokens": 5214773.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.960792493820191,
|
|
"epoch": 2.465835840137516,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046653938277348237,
|
|
"loss": 5.8568,
|
|
"mean_token_accuracy": 0.17312622368335723,
|
|
"num_tokens": 5223734.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 6.149091148376465,
|
|
"epoch": 2.4701332187365708,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004663647194373505,
|
|
"loss": 5.9354,
|
|
"mean_token_accuracy": 0.16000303328037263,
|
|
"num_tokens": 5231742.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.962928104400635,
|
|
"epoch": 2.4744305973356253,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046618963823715913,
|
|
"loss": 5.8985,
|
|
"mean_token_accuracy": 0.16055782586336137,
|
|
"num_tokens": 5241673.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 6.057924270629883,
|
|
"epoch": 2.47872797593468,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00046601413955521575,
|
|
"loss": 5.8658,
|
|
"mean_token_accuracy": 0.16624459624290466,
|
|
"num_tokens": 5250082.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 6.010685443878174,
|
|
"epoch": 2.4830253545337344,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004658382237747393,
|
|
"loss": 5.9297,
|
|
"mean_token_accuracy": 0.16147278249263763,
|
|
"num_tokens": 5259680.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 6.041595315933227,
|
|
"epoch": 2.487322733132789,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046566189127985946,
|
|
"loss": 5.8996,
|
|
"mean_token_accuracy": 0.16485785096883773,
|
|
"num_tokens": 5269561.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 6.085468482971192,
|
|
"epoch": 2.4916201117318435,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000465485142455616,
|
|
"loss": 5.867,
|
|
"mean_token_accuracy": 0.16827538460493088,
|
|
"num_tokens": 5278659.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.905921554565429,
|
|
"epoch": 2.495917490330898,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046530797768795765,
|
|
"loss": 5.8331,
|
|
"mean_token_accuracy": 0.17265328168869018,
|
|
"num_tokens": 5287619.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 6.0317473888397215,
|
|
"epoch": 2.5002148689299526,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046513039736374153,
|
|
"loss": 5.9676,
|
|
"mean_token_accuracy": 0.1589129939675331,
|
|
"num_tokens": 5297334.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 6.07217264175415,
|
|
"epoch": 2.504512247529007,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004649524018707319,
|
|
"loss": 5.8606,
|
|
"mean_token_accuracy": 0.16698037683963776,
|
|
"num_tokens": 5306208.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.981895017623901,
|
|
"epoch": 2.5088096261280617,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00046477399159759996,
|
|
"loss": 5.8081,
|
|
"mean_token_accuracy": 0.16871434301137925,
|
|
"num_tokens": 5314754.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.913084983825684,
|
|
"epoch": 2.5131070047271162,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046459516693392246,
|
|
"loss": 5.8211,
|
|
"mean_token_accuracy": 0.16699100136756898,
|
|
"num_tokens": 5324000.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 6.0547689437866214,
|
|
"epoch": 2.517404383326171,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004644159282701808,
|
|
"loss": 5.8884,
|
|
"mean_token_accuracy": 0.16294697523117066,
|
|
"num_tokens": 5332478.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 6.078831005096435,
|
|
"epoch": 2.5217017619252258,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046423627599776076,
|
|
"loss": 5.9459,
|
|
"mean_token_accuracy": 0.15489169508218764,
|
|
"num_tokens": 5341635.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 6.072408723831177,
|
|
"epoch": 2.5259991405242803,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000464056210508951,
|
|
"loss": 5.957,
|
|
"mean_token_accuracy": 0.15589358359575273,
|
|
"num_tokens": 5350144.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.971312141418457,
|
|
"epoch": 2.530296519123335,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004638757321969426,
|
|
"loss": 5.8671,
|
|
"mean_token_accuracy": 0.16721404045820237,
|
|
"num_tokens": 5358788.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 6.0780237197875975,
|
|
"epoch": 2.5345938977223894,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00046369484145582815,
|
|
"loss": 5.9583,
|
|
"mean_token_accuracy": 0.16172538995742797,
|
|
"num_tokens": 5368057.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.921319103240966,
|
|
"epoch": 2.538891276321444,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046351353868060054,
|
|
"loss": 5.8036,
|
|
"mean_token_accuracy": 0.16979097872972487,
|
|
"num_tokens": 5376739.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 6.045459365844726,
|
|
"epoch": 2.5431886549204985,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046333182426715273,
|
|
"loss": 5.9206,
|
|
"mean_token_accuracy": 0.16544167399406434,
|
|
"num_tokens": 5385967.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 6.095268678665161,
|
|
"epoch": 2.547486033519553,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046314969861227626,
|
|
"loss": 5.9374,
|
|
"mean_token_accuracy": 0.15496064722537994,
|
|
"num_tokens": 5395192.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.993819046020508,
|
|
"epoch": 2.5517834121186076,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004629671621136608,
|
|
"loss": 5.9091,
|
|
"mean_token_accuracy": 0.16285971105098723,
|
|
"num_tokens": 5404694.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 6.065847873687744,
|
|
"epoch": 2.556080790717662,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004627842151698931,
|
|
"loss": 5.9263,
|
|
"mean_token_accuracy": 0.16298119574785233,
|
|
"num_tokens": 5413102.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 6.028053331375122,
|
|
"epoch": 2.5603781693167167,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046260085818045625,
|
|
"loss": 5.9072,
|
|
"mean_token_accuracy": 0.1584775522351265,
|
|
"num_tokens": 5423339.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 6.078606796264649,
|
|
"epoch": 2.5646755479157712,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004624170915457284,
|
|
"loss": 5.8959,
|
|
"mean_token_accuracy": 0.16520549952983857,
|
|
"num_tokens": 5432377.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.980776977539063,
|
|
"epoch": 2.5689729265148262,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046223291566698264,
|
|
"loss": 5.8141,
|
|
"mean_token_accuracy": 0.1644345499575138,
|
|
"num_tokens": 5441038.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 6.00663743019104,
|
|
"epoch": 2.5732703051138808,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004620483309463855,
|
|
"loss": 5.8539,
|
|
"mean_token_accuracy": 0.16262196451425553,
|
|
"num_tokens": 5449557.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 6.075806379318237,
|
|
"epoch": 2.5775676837129353,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004618633377869961,
|
|
"loss": 5.969,
|
|
"mean_token_accuracy": 0.1565746933221817,
|
|
"num_tokens": 5458931.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.5775676837129353,
|
|
"eval_entropy": 5.9535146479134085,
|
|
"eval_loss": 6.191389560699463,
|
|
"eval_mean_token_accuracy": 0.15702831731723235,
|
|
"eval_num_tokens": 5458931.0,
|
|
"eval_runtime": 2.6385,
|
|
"eval_samples_per_second": 1345.059,
|
|
"eval_steps_per_second": 168.275,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 6.015251874923706,
|
|
"epoch": 2.58186506231199,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004616779365927656,
|
|
"loss": 5.7933,
|
|
"mean_token_accuracy": 0.17440636605024337,
|
|
"num_tokens": 5468539.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.870741844177246,
|
|
"epoch": 2.5861624409110444,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004614921277685361,
|
|
"loss": 5.7313,
|
|
"mean_token_accuracy": 0.1767333447933197,
|
|
"num_tokens": 5475710.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.986069059371948,
|
|
"epoch": 2.590459819510099,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046130591172003976,
|
|
"loss": 5.876,
|
|
"mean_token_accuracy": 0.16390150636434556,
|
|
"num_tokens": 5484597.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 6.0947503566741945,
|
|
"epoch": 2.5947571981091535,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004611192888538981,
|
|
"loss": 5.9718,
|
|
"mean_token_accuracy": 0.15505678951740265,
|
|
"num_tokens": 5493213.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 6.071368312835693,
|
|
"epoch": 2.599054576708208,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046093225957762084,
|
|
"loss": 5.9669,
|
|
"mean_token_accuracy": 0.15736154466867447,
|
|
"num_tokens": 5502556.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 6.0553131103515625,
|
|
"epoch": 2.6033519553072626,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004607448242996051,
|
|
"loss": 5.8744,
|
|
"mean_token_accuracy": 0.16855462789535522,
|
|
"num_tokens": 5511779.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 6.043158531188965,
|
|
"epoch": 2.607649333906317,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004605569834291347,
|
|
"loss": 5.8432,
|
|
"mean_token_accuracy": 0.1685672163963318,
|
|
"num_tokens": 5520836.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.961936092376709,
|
|
"epoch": 2.6119467125053717,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00046036873737637904,
|
|
"loss": 5.8813,
|
|
"mean_token_accuracy": 0.1675259366631508,
|
|
"num_tokens": 5529285.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 6.0443174839019775,
|
|
"epoch": 2.6162440911044262,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004601800865523921,
|
|
"loss": 5.883,
|
|
"mean_token_accuracy": 0.16329028606414794,
|
|
"num_tokens": 5538160.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.952384805679321,
|
|
"epoch": 2.620541469703481,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045999103136911204,
|
|
"loss": 5.8701,
|
|
"mean_token_accuracy": 0.1625372052192688,
|
|
"num_tokens": 5547355.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 6.0581084251403805,
|
|
"epoch": 2.6248388483025353,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045980157223935965,
|
|
"loss": 5.8848,
|
|
"mean_token_accuracy": 0.1610598996281624,
|
|
"num_tokens": 5557299.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.965495777130127,
|
|
"epoch": 2.62913622690159,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045961170957683806,
|
|
"loss": 5.8156,
|
|
"mean_token_accuracy": 0.1695963054895401,
|
|
"num_tokens": 5565469.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.990341711044311,
|
|
"epoch": 2.6334336055006444,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045942144379613147,
|
|
"loss": 5.935,
|
|
"mean_token_accuracy": 0.15782787054777145,
|
|
"num_tokens": 5574740.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 6.084517765045166,
|
|
"epoch": 2.637730984099699,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045923077531270426,
|
|
"loss": 5.9137,
|
|
"mean_token_accuracy": 0.16793479323387145,
|
|
"num_tokens": 5583438.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.985931777954102,
|
|
"epoch": 2.6420283626987535,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004590397045429001,
|
|
"loss": 5.8725,
|
|
"mean_token_accuracy": 0.16909938603639602,
|
|
"num_tokens": 5592389.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.992589473724365,
|
|
"epoch": 2.646325741297808,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00045884823190394134,
|
|
"loss": 5.7798,
|
|
"mean_token_accuracy": 0.17591131776571273,
|
|
"num_tokens": 5601598.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.99040412902832,
|
|
"epoch": 2.650623119896863,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004586563578139275,
|
|
"loss": 5.8858,
|
|
"mean_token_accuracy": 0.1604371339082718,
|
|
"num_tokens": 5610498.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.977429723739624,
|
|
"epoch": 2.6549204984959176,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00045846408269183505,
|
|
"loss": 5.7831,
|
|
"mean_token_accuracy": 0.17481062114238738,
|
|
"num_tokens": 5620082.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 6.044168376922608,
|
|
"epoch": 2.659217877094972,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045827140695751603,
|
|
"loss": 5.8733,
|
|
"mean_token_accuracy": 0.16334586590528488,
|
|
"num_tokens": 5630291.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.9464648246765135,
|
|
"epoch": 2.6635152556940267,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004580783310316971,
|
|
"loss": 5.874,
|
|
"mean_token_accuracy": 0.16866674721240998,
|
|
"num_tokens": 5638784.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.949322175979614,
|
|
"epoch": 2.6678126342930812,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045788485533597895,
|
|
"loss": 5.7265,
|
|
"mean_token_accuracy": 0.17600842714309692,
|
|
"num_tokens": 5647968.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 6.025244140625,
|
|
"epoch": 2.672110012892136,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045769098029283526,
|
|
"loss": 5.9623,
|
|
"mean_token_accuracy": 0.1540097177028656,
|
|
"num_tokens": 5657543.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 6.02379412651062,
|
|
"epoch": 2.6764073914911903,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004574967063256115,
|
|
"loss": 5.8573,
|
|
"mean_token_accuracy": 0.16828878223896027,
|
|
"num_tokens": 5666535.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.992211484909058,
|
|
"epoch": 2.680704770090245,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045730203385852447,
|
|
"loss": 5.9371,
|
|
"mean_token_accuracy": 0.16220323592424393,
|
|
"num_tokens": 5676273.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.9227536678314205,
|
|
"epoch": 2.6850021486892994,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000457106963316661,
|
|
"loss": 5.8682,
|
|
"mean_token_accuracy": 0.16720343232154847,
|
|
"num_tokens": 5684888.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 6.045551300048828,
|
|
"epoch": 2.689299527288354,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045691149512597717,
|
|
"loss": 5.9047,
|
|
"mean_token_accuracy": 0.16129783689975738,
|
|
"num_tokens": 5693626.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 6.0508557796478275,
|
|
"epoch": 2.6935969058874085,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 0.00045671562971329736,
|
|
"loss": 5.8291,
|
|
"mean_token_accuracy": 0.16866300702095033,
|
|
"num_tokens": 5702542.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.952105188369751,
|
|
"epoch": 2.6978942844864635,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00045651936750631337,
|
|
"loss": 5.8452,
|
|
"mean_token_accuracy": 0.16578937768936158,
|
|
"num_tokens": 5711440.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 6.055763578414917,
|
|
"epoch": 2.702191663085518,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045632270893358333,
|
|
"loss": 5.9322,
|
|
"mean_token_accuracy": 0.16412092000246048,
|
|
"num_tokens": 5721495.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 6.03286509513855,
|
|
"epoch": 2.7064890416845726,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004561256544245312,
|
|
"loss": 5.9443,
|
|
"mean_token_accuracy": 0.15768667608499526,
|
|
"num_tokens": 5730664.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.962044620513916,
|
|
"epoch": 2.710786420283627,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000455928204409445,
|
|
"loss": 5.8253,
|
|
"mean_token_accuracy": 0.17572951316833496,
|
|
"num_tokens": 5740229.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.970143127441406,
|
|
"epoch": 2.7150837988826817,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00045573035931947684,
|
|
"loss": 5.8324,
|
|
"mean_token_accuracy": 0.1705564573407173,
|
|
"num_tokens": 5748549.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 6.028087854385376,
|
|
"epoch": 2.7193811774817362,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004555321195866411,
|
|
"loss": 5.7649,
|
|
"mean_token_accuracy": 0.1665789157152176,
|
|
"num_tokens": 5757603.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.972912311553955,
|
|
"epoch": 2.723678556080791,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004553334856438143,
|
|
"loss": 5.9561,
|
|
"mean_token_accuracy": 0.15752369910478592,
|
|
"num_tokens": 5767520.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 6.066450452804565,
|
|
"epoch": 2.7279759346798453,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045513445792473356,
|
|
"loss": 5.9355,
|
|
"mean_token_accuracy": 0.15854838043451308,
|
|
"num_tokens": 5776778.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 6.077041244506836,
|
|
"epoch": 2.7322733132789,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004549350368639958,
|
|
"loss": 5.9885,
|
|
"mean_token_accuracy": 0.15888068675994874,
|
|
"num_tokens": 5785652.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 6.099004125595092,
|
|
"epoch": 2.7365706918779544,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045473522289705693,
|
|
"loss": 5.9167,
|
|
"mean_token_accuracy": 0.16159161776304246,
|
|
"num_tokens": 5795766.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.919166326522827,
|
|
"epoch": 2.740868070477009,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00045453501646023085,
|
|
"loss": 5.9589,
|
|
"mean_token_accuracy": 0.16025417894124985,
|
|
"num_tokens": 5804504.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 6.0491148948669435,
|
|
"epoch": 2.7451654490760635,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00045433441799068837,
|
|
"loss": 5.8771,
|
|
"mean_token_accuracy": 0.16158021241426468,
|
|
"num_tokens": 5814161.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.993768501281738,
|
|
"epoch": 2.749462827675118,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004541334279264562,
|
|
"loss": 5.7725,
|
|
"mean_token_accuracy": 0.17674892395734787,
|
|
"num_tokens": 5822235.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.940905618667602,
|
|
"epoch": 2.7537602062741726,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00045393204670641656,
|
|
"loss": 5.8042,
|
|
"mean_token_accuracy": 0.16777520775794982,
|
|
"num_tokens": 5831572.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.931118297576904,
|
|
"epoch": 2.758057584873227,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004537302747703055,
|
|
"loss": 5.813,
|
|
"mean_token_accuracy": 0.17369199246168138,
|
|
"num_tokens": 5839694.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 6.1083742618560795,
|
|
"epoch": 2.7623549634722817,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00045352811255871216,
|
|
"loss": 5.9466,
|
|
"mean_token_accuracy": 0.15793339610099794,
|
|
"num_tokens": 5849131.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 6.030238151550293,
|
|
"epoch": 2.7666523420713363,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00045332556051307804,
|
|
"loss": 5.8486,
|
|
"mean_token_accuracy": 0.16480160504579544,
|
|
"num_tokens": 5858861.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 6.031745100021363,
|
|
"epoch": 2.770949720670391,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045312261907569585,
|
|
"loss": 5.8528,
|
|
"mean_token_accuracy": 0.1642756223678589,
|
|
"num_tokens": 5867585.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.9777271270751955,
|
|
"epoch": 2.775247099269446,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045291928868970867,
|
|
"loss": 5.8738,
|
|
"mean_token_accuracy": 0.16010442823171617,
|
|
"num_tokens": 5876256.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.9892865180969235,
|
|
"epoch": 2.7795444778685003,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004527155697991087,
|
|
"loss": 5.9467,
|
|
"mean_token_accuracy": 0.15326642394065856,
|
|
"num_tokens": 5885302.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 6.024959230422974,
|
|
"epoch": 2.783841856467555,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004525114628487365,
|
|
"loss": 5.9677,
|
|
"mean_token_accuracy": 0.15789304226636885,
|
|
"num_tokens": 5895066.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 6.062315177917481,
|
|
"epoch": 2.7881392350666094,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045230696828428026,
|
|
"loss": 5.9293,
|
|
"mean_token_accuracy": 0.16333307176828385,
|
|
"num_tokens": 5903258.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.992421674728393,
|
|
"epoch": 2.792436613665664,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004521020865522742,
|
|
"loss": 5.8251,
|
|
"mean_token_accuracy": 0.16470370143651963,
|
|
"num_tokens": 5911714.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.975275611877441,
|
|
"epoch": 2.7967339922647185,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045189681810009827,
|
|
"loss": 5.8962,
|
|
"mean_token_accuracy": 0.15860241800546646,
|
|
"num_tokens": 5920432.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 6.068870687484742,
|
|
"epoch": 2.801031370863773,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00045169116337597653,
|
|
"loss": 5.9161,
|
|
"mean_token_accuracy": 0.15756052285432814,
|
|
"num_tokens": 5929202.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 6.080472087860107,
|
|
"epoch": 2.8053287494628276,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000451485122828977,
|
|
"loss": 5.9589,
|
|
"mean_token_accuracy": 0.15693847239017486,
|
|
"num_tokens": 5938034.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.944226551055908,
|
|
"epoch": 2.809626128061882,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045127869690900956,
|
|
"loss": 5.7934,
|
|
"mean_token_accuracy": 0.17031458169221877,
|
|
"num_tokens": 5946944.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.920538330078125,
|
|
"epoch": 2.8139235066609367,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00045107188606682613,
|
|
"loss": 5.892,
|
|
"mean_token_accuracy": 0.16842835694551467,
|
|
"num_tokens": 5956475.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 6.080301523208618,
|
|
"epoch": 2.8182208852599913,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004508646907540188,
|
|
"loss": 5.8798,
|
|
"mean_token_accuracy": 0.15972289741039275,
|
|
"num_tokens": 5965814.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.996534824371338,
|
|
"epoch": 2.8225182638590463,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004506571114230195,
|
|
"loss": 5.9075,
|
|
"mean_token_accuracy": 0.16237858533859253,
|
|
"num_tokens": 5973850.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.90730504989624,
|
|
"epoch": 2.826815642458101,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045044914852709824,
|
|
"loss": 5.8495,
|
|
"mean_token_accuracy": 0.16205417662858962,
|
|
"num_tokens": 5982987.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 6.092641592025757,
|
|
"epoch": 2.8311130210571553,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004502408025203631,
|
|
"loss": 5.8606,
|
|
"mean_token_accuracy": 0.1686910644173622,
|
|
"num_tokens": 5992227.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.963792896270752,
|
|
"epoch": 2.83541039965621,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004500320738577584,
|
|
"loss": 5.8338,
|
|
"mean_token_accuracy": 0.16970308721065522,
|
|
"num_tokens": 6000243.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.981186723709106,
|
|
"epoch": 2.8397077782552644,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00044982296299506407,
|
|
"loss": 5.8322,
|
|
"mean_token_accuracy": 0.1722040668129921,
|
|
"num_tokens": 6009771.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 6.045083284378052,
|
|
"epoch": 2.844005156854319,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004496134703888948,
|
|
"loss": 5.905,
|
|
"mean_token_accuracy": 0.1612884059548378,
|
|
"num_tokens": 6018683.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.9778515815734865,
|
|
"epoch": 2.8483025354533735,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00044940359649669846,
|
|
"loss": 5.7719,
|
|
"mean_token_accuracy": 0.17384362816810608,
|
|
"num_tokens": 6027422.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.9929955959320065,
|
|
"epoch": 2.852599914052428,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00044919334177675595,
|
|
"loss": 5.8633,
|
|
"mean_token_accuracy": 0.16016940772533417,
|
|
"num_tokens": 6035670.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.966201400756836,
|
|
"epoch": 2.8568972926514826,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00044898270668817955,
|
|
"loss": 5.7873,
|
|
"mean_token_accuracy": 0.17351211607456207,
|
|
"num_tokens": 6044092.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.956835079193115,
|
|
"epoch": 2.861194671250537,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000448771691690912,
|
|
"loss": 5.8567,
|
|
"mean_token_accuracy": 0.1599970579147339,
|
|
"num_tokens": 6053970.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.994864130020142,
|
|
"epoch": 2.8654920498495917,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004485602972457257,
|
|
"loss": 5.8204,
|
|
"mean_token_accuracy": 0.16462148129940032,
|
|
"num_tokens": 6062965.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.992863368988037,
|
|
"epoch": 2.8697894284486463,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00044834852381422165,
|
|
"loss": 5.9008,
|
|
"mean_token_accuracy": 0.1637433260679245,
|
|
"num_tokens": 6072420.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.955929565429687,
|
|
"epoch": 2.874086807047701,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00044813637185882836,
|
|
"loss": 5.7987,
|
|
"mean_token_accuracy": 0.16683831959962844,
|
|
"num_tokens": 6080915.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 6.066890859603882,
|
|
"epoch": 2.8783841856467554,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00044792384184280106,
|
|
"loss": 5.9239,
|
|
"mean_token_accuracy": 0.16229262053966523,
|
|
"num_tokens": 6090453.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.943010091781616,
|
|
"epoch": 2.88268156424581,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00044771093423022013,
|
|
"loss": 5.9791,
|
|
"mean_token_accuracy": 0.15519970804452896,
|
|
"num_tokens": 6099390.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.986198377609253,
|
|
"epoch": 2.8869789428448644,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004474976494859909,
|
|
"loss": 5.8692,
|
|
"mean_token_accuracy": 0.17025581151247024,
|
|
"num_tokens": 6108677.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 6.007058477401733,
|
|
"epoch": 2.891276321443919,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004472839880758419,
|
|
"loss": 5.8018,
|
|
"mean_token_accuracy": 0.16929676234722138,
|
|
"num_tokens": 6117151.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 6.097532176971436,
|
|
"epoch": 2.8955737000429735,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004470699504663242,
|
|
"loss": 5.9198,
|
|
"mean_token_accuracy": 0.1557392880320549,
|
|
"num_tokens": 6127167.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.9331482410430905,
|
|
"epoch": 2.899871078642028,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004468555371248104,
|
|
"loss": 5.811,
|
|
"mean_token_accuracy": 0.16359427124261855,
|
|
"num_tokens": 6136487.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 6.0199809074401855,
|
|
"epoch": 2.904168457241083,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004466407485194937,
|
|
"loss": 5.9189,
|
|
"mean_token_accuracy": 0.16040531545877457,
|
|
"num_tokens": 6145334.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 6.026473665237427,
|
|
"epoch": 2.9084658358401376,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004464255851193864,
|
|
"loss": 5.8414,
|
|
"mean_token_accuracy": 0.1662326857447624,
|
|
"num_tokens": 6155062.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.972359037399292,
|
|
"epoch": 2.912763214439192,
|
|
"grad_norm": 1.90625,
|
|
"learning_rate": 0.0004462100473943194,
|
|
"loss": 5.8159,
|
|
"mean_token_accuracy": 0.17152023315429688,
|
|
"num_tokens": 6164313.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.973877334594727,
|
|
"epoch": 2.9170605930382467,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000445994135814941,
|
|
"loss": 5.8574,
|
|
"mean_token_accuracy": 0.16147627383470536,
|
|
"num_tokens": 6173513.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.96266450881958,
|
|
"epoch": 2.9213579716373013,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00044577785085271566,
|
|
"loss": 5.87,
|
|
"mean_token_accuracy": 0.16429724544286728,
|
|
"num_tokens": 6182000.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 6.0837915420532225,
|
|
"epoch": 2.925655350236356,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004455611929799235,
|
|
"loss": 5.8837,
|
|
"mean_token_accuracy": 0.15638401508331298,
|
|
"num_tokens": 6191887.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.879487800598144,
|
|
"epoch": 2.9299527288354104,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004453441626696585,
|
|
"loss": 5.9357,
|
|
"mean_token_accuracy": 0.15588561296463013,
|
|
"num_tokens": 6202897.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 6.037405920028687,
|
|
"epoch": 2.934250107434465,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00044512676039582823,
|
|
"loss": 5.8192,
|
|
"mean_token_accuracy": 0.1726475328207016,
|
|
"num_tokens": 6211811.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 6.025434732437134,
|
|
"epoch": 2.9385474860335195,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004449089866331524,
|
|
"loss": 5.83,
|
|
"mean_token_accuracy": 0.17189965546131133,
|
|
"num_tokens": 6219896.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.89447660446167,
|
|
"epoch": 2.942844864632574,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004446908418571617,
|
|
"loss": 5.8145,
|
|
"mean_token_accuracy": 0.171225206553936,
|
|
"num_tokens": 6228212.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.967305994033813,
|
|
"epoch": 2.9471422432316285,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004444723265441973,
|
|
"loss": 5.9781,
|
|
"mean_token_accuracy": 0.15706060230731964,
|
|
"num_tokens": 6238133.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.97793173789978,
|
|
"epoch": 2.9514396218306835,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004442534411714092,
|
|
"loss": 5.8927,
|
|
"mean_token_accuracy": 0.15664541125297546,
|
|
"num_tokens": 6247331.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 6.087286186218262,
|
|
"epoch": 2.955737000429738,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00044403418621675555,
|
|
"loss": 5.8851,
|
|
"mean_token_accuracy": 0.1622688055038452,
|
|
"num_tokens": 6255280.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.984484243392944,
|
|
"epoch": 2.9600343790287926,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004438145621590017,
|
|
"loss": 5.8304,
|
|
"mean_token_accuracy": 0.16720051914453507,
|
|
"num_tokens": 6264752.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.910861301422119,
|
|
"epoch": 2.964331757627847,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00044359456947771857,
|
|
"loss": 5.7809,
|
|
"mean_token_accuracy": 0.16352355778217315,
|
|
"num_tokens": 6273258.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.880699491500854,
|
|
"epoch": 2.9686291362269017,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004433742086532824,
|
|
"loss": 5.6995,
|
|
"mean_token_accuracy": 0.18464418053627013,
|
|
"num_tokens": 6281584.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.95316252708435,
|
|
"epoch": 2.9729265148259563,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00044315348016687317,
|
|
"loss": 5.8224,
|
|
"mean_token_accuracy": 0.1641302987933159,
|
|
"num_tokens": 6290016.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.891777420043946,
|
|
"epoch": 2.977223893425011,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004429323845004736,
|
|
"loss": 5.734,
|
|
"mean_token_accuracy": 0.17175184190273285,
|
|
"num_tokens": 6298569.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.903365993499756,
|
|
"epoch": 2.9815212720240654,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00044271092213686824,
|
|
"loss": 5.7661,
|
|
"mean_token_accuracy": 0.17182787805795668,
|
|
"num_tokens": 6307684.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 6.026407480239868,
|
|
"epoch": 2.98581865062312,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00044248909355964247,
|
|
"loss": 5.9057,
|
|
"mean_token_accuracy": 0.15918205678462982,
|
|
"num_tokens": 6317767.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 6.092149639129639,
|
|
"epoch": 2.9901160292221745,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00044226689925318117,
|
|
"loss": 5.9344,
|
|
"mean_token_accuracy": 0.1575716406106949,
|
|
"num_tokens": 6327457.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.980013179779053,
|
|
"epoch": 2.994413407821229,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00044204433970266785,
|
|
"loss": 5.7377,
|
|
"mean_token_accuracy": 0.18050013333559037,
|
|
"num_tokens": 6335747.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.940433025360107,
|
|
"epoch": 2.9987107864202835,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004418214153940837,
|
|
"loss": 5.8196,
|
|
"mean_token_accuracy": 0.1658599093556404,
|
|
"num_tokens": 6344750.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.953445116678874,
|
|
"epoch": 3.002578427159433,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00044159812681420624,
|
|
"loss": 5.7485,
|
|
"mean_token_accuracy": 0.16718086269166735,
|
|
"num_tokens": 6354779.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.959359550476075,
|
|
"epoch": 3.0068758057584875,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004413744744506086,
|
|
"loss": 5.5163,
|
|
"mean_token_accuracy": 0.17318416386842728,
|
|
"num_tokens": 6363809.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 3.0068758057584875,
|
|
"eval_entropy": 5.808071028004895,
|
|
"eval_loss": 6.120314598083496,
|
|
"eval_mean_token_accuracy": 0.16263622614378864,
|
|
"eval_num_tokens": 6363809.0,
|
|
"eval_runtime": 2.6272,
|
|
"eval_samples_per_second": 1350.889,
|
|
"eval_steps_per_second": 169.004,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.933870029449463,
|
|
"epoch": 3.011173184357542,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00044115045879165806,
|
|
"loss": 5.5839,
|
|
"mean_token_accuracy": 0.175676928460598,
|
|
"num_tokens": 6373082.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.9161820888519285,
|
|
"epoch": 3.0154705629565965,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00044092608032651515,
|
|
"loss": 5.5454,
|
|
"mean_token_accuracy": 0.17348759919404982,
|
|
"num_tokens": 6381286.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.909457683563232,
|
|
"epoch": 3.019767941555651,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044070133954513305,
|
|
"loss": 5.5092,
|
|
"mean_token_accuracy": 0.18552801311016082,
|
|
"num_tokens": 6390217.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.969851350784301,
|
|
"epoch": 3.0240653201547056,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004404762369382555,
|
|
"loss": 5.5226,
|
|
"mean_token_accuracy": 0.17632414549589157,
|
|
"num_tokens": 6399276.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.897236680984497,
|
|
"epoch": 3.02836269875376,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00044025077299741683,
|
|
"loss": 5.525,
|
|
"mean_token_accuracy": 0.18381931334733964,
|
|
"num_tokens": 6407981.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.844799900054932,
|
|
"epoch": 3.0326600773528147,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00044002494821494007,
|
|
"loss": 5.5114,
|
|
"mean_token_accuracy": 0.18218347877264024,
|
|
"num_tokens": 6416159.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.806324815750122,
|
|
"epoch": 3.0369574559518693,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00043979876308393635,
|
|
"loss": 5.5665,
|
|
"mean_token_accuracy": 0.17806299924850463,
|
|
"num_tokens": 6424564.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.99275279045105,
|
|
"epoch": 3.041254834550924,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004395722180983036,
|
|
"loss": 5.589,
|
|
"mean_token_accuracy": 0.17690781652927398,
|
|
"num_tokens": 6434163.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.7671958923339846,
|
|
"epoch": 3.0455522131499784,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00043934531375272535,
|
|
"loss": 5.417,
|
|
"mean_token_accuracy": 0.19137026965618134,
|
|
"num_tokens": 6443372.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.868448972702026,
|
|
"epoch": 3.049849591749033,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00043911805054267015,
|
|
"loss": 5.5048,
|
|
"mean_token_accuracy": 0.18185081630945205,
|
|
"num_tokens": 6452638.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.985547065734863,
|
|
"epoch": 3.0541469703480875,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00043889042896439004,
|
|
"loss": 5.5345,
|
|
"mean_token_accuracy": 0.17991046011447906,
|
|
"num_tokens": 6461319.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.893108177185058,
|
|
"epoch": 3.0584443489471425,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00043866244951491946,
|
|
"loss": 5.4336,
|
|
"mean_token_accuracy": 0.19298922568559645,
|
|
"num_tokens": 6469506.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.863250494003296,
|
|
"epoch": 3.062741727546197,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00043843411269207445,
|
|
"loss": 5.5134,
|
|
"mean_token_accuracy": 0.18835846781730653,
|
|
"num_tokens": 6478404.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.778197145462036,
|
|
"epoch": 3.0670391061452515,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004382054189944514,
|
|
"loss": 5.44,
|
|
"mean_token_accuracy": 0.18331104516983032,
|
|
"num_tokens": 6487447.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.819038820266724,
|
|
"epoch": 3.071336484744306,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004379763689214259,
|
|
"loss": 5.4877,
|
|
"mean_token_accuracy": 0.17835415601730348,
|
|
"num_tokens": 6496738.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.899236726760864,
|
|
"epoch": 3.0756338633433606,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004377469629731518,
|
|
"loss": 5.5141,
|
|
"mean_token_accuracy": 0.17553481161594392,
|
|
"num_tokens": 6505848.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.843332052230835,
|
|
"epoch": 3.079931241942415,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004375172016505599,
|
|
"loss": 5.4782,
|
|
"mean_token_accuracy": 0.17995121628046035,
|
|
"num_tokens": 6515731.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.891195487976074,
|
|
"epoch": 3.0842286205414697,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004372870854553572,
|
|
"loss": 5.543,
|
|
"mean_token_accuracy": 0.1805647909641266,
|
|
"num_tokens": 6524914.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.884959983825683,
|
|
"epoch": 3.0885259991405243,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004370566148900255,
|
|
"loss": 5.5398,
|
|
"mean_token_accuracy": 0.1865120217204094,
|
|
"num_tokens": 6533712.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.859037780761719,
|
|
"epoch": 3.092823377739579,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00043682579045782024,
|
|
"loss": 5.6039,
|
|
"mean_token_accuracy": 0.17290965765714644,
|
|
"num_tokens": 6543313.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.807743120193481,
|
|
"epoch": 3.0971207563386334,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004365946126627699,
|
|
"loss": 5.4985,
|
|
"mean_token_accuracy": 0.18919602334499358,
|
|
"num_tokens": 6551634.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.85457592010498,
|
|
"epoch": 3.101418134937688,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00043636308200967433,
|
|
"loss": 5.4938,
|
|
"mean_token_accuracy": 0.18733223974704744,
|
|
"num_tokens": 6560695.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.76024374961853,
|
|
"epoch": 3.1057155135367425,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004361311990041039,
|
|
"loss": 5.3974,
|
|
"mean_token_accuracy": 0.18700604140758514,
|
|
"num_tokens": 6569086.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.808054399490357,
|
|
"epoch": 3.110012892135797,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00043589896415239843,
|
|
"loss": 5.4793,
|
|
"mean_token_accuracy": 0.18851352334022523,
|
|
"num_tokens": 6578287.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.828329801559448,
|
|
"epoch": 3.1143102707348516,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00043566637796166595,
|
|
"loss": 5.5526,
|
|
"mean_token_accuracy": 0.17525139153003694,
|
|
"num_tokens": 6587015.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.867431020736694,
|
|
"epoch": 3.118607649333906,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00043543344093978186,
|
|
"loss": 5.601,
|
|
"mean_token_accuracy": 0.1718316853046417,
|
|
"num_tokens": 6596187.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.864822340011597,
|
|
"epoch": 3.122905027932961,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00043520015359538745,
|
|
"loss": 5.4591,
|
|
"mean_token_accuracy": 0.18958659023046492,
|
|
"num_tokens": 6605226.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.794321203231812,
|
|
"epoch": 3.1272024065320156,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004349665164378891,
|
|
"loss": 5.5008,
|
|
"mean_token_accuracy": 0.18200621455907823,
|
|
"num_tokens": 6613232.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.789428567886352,
|
|
"epoch": 3.13149978513107,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00043473252997745684,
|
|
"loss": 5.5113,
|
|
"mean_token_accuracy": 0.18076512068510056,
|
|
"num_tokens": 6622247.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.893785858154297,
|
|
"epoch": 3.1357971637301247,
|
|
"grad_norm": 1.984375,
|
|
"learning_rate": 0.00043449819472502366,
|
|
"loss": 5.4675,
|
|
"mean_token_accuracy": 0.18034528195858002,
|
|
"num_tokens": 6630883.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.818866157531739,
|
|
"epoch": 3.1400945423291793,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004342635111922841,
|
|
"loss": 5.5915,
|
|
"mean_token_accuracy": 0.17699178606271743,
|
|
"num_tokens": 6639399.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.8470982074737545,
|
|
"epoch": 3.144391920928234,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004340284798916931,
|
|
"loss": 5.5204,
|
|
"mean_token_accuracy": 0.17858008593320845,
|
|
"num_tokens": 6649288.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.844093990325928,
|
|
"epoch": 3.1486892995272884,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004337931013364653,
|
|
"loss": 5.4288,
|
|
"mean_token_accuracy": 0.18451717346906663,
|
|
"num_tokens": 6658670.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.8302186012268065,
|
|
"epoch": 3.152986678126343,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000433557376040573,
|
|
"loss": 5.5096,
|
|
"mean_token_accuracy": 0.1836533859372139,
|
|
"num_tokens": 6667302.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.885258197784424,
|
|
"epoch": 3.1572840567253975,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00043332130451874645,
|
|
"loss": 5.5686,
|
|
"mean_token_accuracy": 0.18251876533031464,
|
|
"num_tokens": 6677393.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.848063898086548,
|
|
"epoch": 3.161581435324452,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00043308488728647127,
|
|
"loss": 5.539,
|
|
"mean_token_accuracy": 0.17765716165304185,
|
|
"num_tokens": 6686727.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.796309280395508,
|
|
"epoch": 3.1658788139235066,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004328481248599882,
|
|
"loss": 5.4597,
|
|
"mean_token_accuracy": 0.1872706487774849,
|
|
"num_tokens": 6696116.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.8640275478363035,
|
|
"epoch": 3.170176192522561,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004326110177562918,
|
|
"loss": 5.5685,
|
|
"mean_token_accuracy": 0.17389472126960753,
|
|
"num_tokens": 6704640.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.78633599281311,
|
|
"epoch": 3.1744735711216157,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00043237356649312926,
|
|
"loss": 5.4412,
|
|
"mean_token_accuracy": 0.18975964039564133,
|
|
"num_tokens": 6713663.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.780803060531616,
|
|
"epoch": 3.17877094972067,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004321357715889991,
|
|
"loss": 5.5799,
|
|
"mean_token_accuracy": 0.17476617842912673,
|
|
"num_tokens": 6722965.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.826023006439209,
|
|
"epoch": 3.1830683283197247,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004318976335631505,
|
|
"loss": 5.5202,
|
|
"mean_token_accuracy": 0.18648175597190858,
|
|
"num_tokens": 6732776.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.850239086151123,
|
|
"epoch": 3.1873657069187797,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00043165915293558155,
|
|
"loss": 5.51,
|
|
"mean_token_accuracy": 0.17653508335351945,
|
|
"num_tokens": 6741309.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.831938314437866,
|
|
"epoch": 3.1916630855178343,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004314203302270388,
|
|
"loss": 5.5491,
|
|
"mean_token_accuracy": 0.17970845699310303,
|
|
"num_tokens": 6750584.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.865281438827514,
|
|
"epoch": 3.195960464116889,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004311811659590154,
|
|
"loss": 5.5361,
|
|
"mean_token_accuracy": 0.1788480058312416,
|
|
"num_tokens": 6759344.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.934193706512451,
|
|
"epoch": 3.2002578427159434,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004309416606537507,
|
|
"loss": 5.6777,
|
|
"mean_token_accuracy": 0.17327607274055482,
|
|
"num_tokens": 6770345.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.842852067947388,
|
|
"epoch": 3.204555221314998,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00043070181483422843,
|
|
"loss": 5.5815,
|
|
"mean_token_accuracy": 0.17351661771535873,
|
|
"num_tokens": 6779991.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.782969617843628,
|
|
"epoch": 3.2088525999140525,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000430461629024176,
|
|
"loss": 5.5133,
|
|
"mean_token_accuracy": 0.1805244117975235,
|
|
"num_tokens": 6788972.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.787181186676025,
|
|
"epoch": 3.213149978513107,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004302211037480634,
|
|
"loss": 5.4534,
|
|
"mean_token_accuracy": 0.18245744109153747,
|
|
"num_tokens": 6796967.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.786061096191406,
|
|
"epoch": 3.2174473571121616,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004299802395311015,
|
|
"loss": 5.5499,
|
|
"mean_token_accuracy": 0.18429655432701111,
|
|
"num_tokens": 6805961.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.716186952590943,
|
|
"epoch": 3.221744735711216,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004297390368992414,
|
|
"loss": 5.4619,
|
|
"mean_token_accuracy": 0.18117133677005767,
|
|
"num_tokens": 6814657.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.7997174739837645,
|
|
"epoch": 3.2260421143102707,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00042949749637917353,
|
|
"loss": 5.487,
|
|
"mean_token_accuracy": 0.18614423125982285,
|
|
"num_tokens": 6823095.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.865487480163575,
|
|
"epoch": 3.230339492909325,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004292556184983256,
|
|
"loss": 5.5216,
|
|
"mean_token_accuracy": 0.18153394907712936,
|
|
"num_tokens": 6832195.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.8879210472106935,
|
|
"epoch": 3.2346368715083798,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004290134037848623,
|
|
"loss": 5.6524,
|
|
"mean_token_accuracy": 0.17485911697149276,
|
|
"num_tokens": 6840922.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.814740896224976,
|
|
"epoch": 3.2389342501074343,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00042877085276768386,
|
|
"loss": 5.496,
|
|
"mean_token_accuracy": 0.18620195537805556,
|
|
"num_tokens": 6849182.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.8426494121551515,
|
|
"epoch": 3.243231628706489,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00042852796597642455,
|
|
"loss": 5.4881,
|
|
"mean_token_accuracy": 0.18392587453126907,
|
|
"num_tokens": 6857932.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.930465126037598,
|
|
"epoch": 3.247529007305544,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004282847439414522,
|
|
"loss": 5.6207,
|
|
"mean_token_accuracy": 0.16870806813240052,
|
|
"num_tokens": 6867283.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.864709043502808,
|
|
"epoch": 3.2518263859045984,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004280411871938664,
|
|
"loss": 5.608,
|
|
"mean_token_accuracy": 0.17931383848190308,
|
|
"num_tokens": 6876123.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.877500772476196,
|
|
"epoch": 3.256123764503653,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004277972962654979,
|
|
"loss": 5.5472,
|
|
"mean_token_accuracy": 0.17602882534265518,
|
|
"num_tokens": 6885239.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.858795881271362,
|
|
"epoch": 3.2604211431027075,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004275530716889069,
|
|
"loss": 5.5937,
|
|
"mean_token_accuracy": 0.171337129175663,
|
|
"num_tokens": 6895061.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.898361396789551,
|
|
"epoch": 3.264718521701762,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004273085139973822,
|
|
"loss": 5.6602,
|
|
"mean_token_accuracy": 0.16613161861896514,
|
|
"num_tokens": 6903828.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.880637073516846,
|
|
"epoch": 3.2690159003008166,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004270636237249401,
|
|
"loss": 5.5461,
|
|
"mean_token_accuracy": 0.17980466932058334,
|
|
"num_tokens": 6912805.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.835609292984008,
|
|
"epoch": 3.273313278899871,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00042681840140632314,
|
|
"loss": 5.6146,
|
|
"mean_token_accuracy": 0.17179694026708603,
|
|
"num_tokens": 6922165.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.856044054031372,
|
|
"epoch": 3.2776106574989257,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004265728475769989,
|
|
"loss": 5.5718,
|
|
"mean_token_accuracy": 0.17612663954496383,
|
|
"num_tokens": 6931677.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.847190523147583,
|
|
"epoch": 3.28190803609798,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004263269627731586,
|
|
"loss": 5.5376,
|
|
"mean_token_accuracy": 0.1813596546649933,
|
|
"num_tokens": 6940486.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.766942977905273,
|
|
"epoch": 3.2862054146970348,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004260807475317164,
|
|
"loss": 5.5387,
|
|
"mean_token_accuracy": 0.1778130680322647,
|
|
"num_tokens": 6948990.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.867075777053833,
|
|
"epoch": 3.2905027932960893,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004258342023903081,
|
|
"loss": 5.6853,
|
|
"mean_token_accuracy": 0.1678419291973114,
|
|
"num_tokens": 6959311.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.880772256851197,
|
|
"epoch": 3.294800171895144,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00042558732788728975,
|
|
"loss": 5.458,
|
|
"mean_token_accuracy": 0.18143573701381682,
|
|
"num_tokens": 6968619.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.764143800735473,
|
|
"epoch": 3.2990975504941984,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00042534012456173643,
|
|
"loss": 5.524,
|
|
"mean_token_accuracy": 0.17863315343856812,
|
|
"num_tokens": 6977469.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.848656415939331,
|
|
"epoch": 3.303394929093253,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00042509259295344157,
|
|
"loss": 5.4816,
|
|
"mean_token_accuracy": 0.17849756330251693,
|
|
"num_tokens": 6986772.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.847779750823975,
|
|
"epoch": 3.3076923076923075,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00042484473360291514,
|
|
"loss": 5.5227,
|
|
"mean_token_accuracy": 0.1756258174777031,
|
|
"num_tokens": 6993937.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.738813400268555,
|
|
"epoch": 3.311989686291362,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00042459654705138294,
|
|
"loss": 5.5679,
|
|
"mean_token_accuracy": 0.1823994942009449,
|
|
"num_tokens": 7003222.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.774629783630371,
|
|
"epoch": 3.316287064890417,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004243480338407853,
|
|
"loss": 5.525,
|
|
"mean_token_accuracy": 0.18862827271223068,
|
|
"num_tokens": 7012055.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.843663358688355,
|
|
"epoch": 3.3205844434894716,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004240991945137755,
|
|
"loss": 5.5162,
|
|
"mean_token_accuracy": 0.18519170731306075,
|
|
"num_tokens": 7021036.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.8147810935974125,
|
|
"epoch": 3.324881822088526,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00042385002961371944,
|
|
"loss": 5.5325,
|
|
"mean_token_accuracy": 0.18135716915130615,
|
|
"num_tokens": 7030450.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.889339017868042,
|
|
"epoch": 3.3291792006875807,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004236005396846935,
|
|
"loss": 5.6222,
|
|
"mean_token_accuracy": 0.17458055168390274,
|
|
"num_tokens": 7039740.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.856438732147216,
|
|
"epoch": 3.333476579286635,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00042335072527148406,
|
|
"loss": 5.6147,
|
|
"mean_token_accuracy": 0.17771665155887603,
|
|
"num_tokens": 7050430.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.803697156906128,
|
|
"epoch": 3.3377739578856898,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004231005869195859,
|
|
"loss": 5.5804,
|
|
"mean_token_accuracy": 0.17390600442886353,
|
|
"num_tokens": 7059477.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.7714502811431885,
|
|
"epoch": 3.3420713364847443,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004228501251752011,
|
|
"loss": 5.5353,
|
|
"mean_token_accuracy": 0.18197794258594513,
|
|
"num_tokens": 7067805.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.799749040603638,
|
|
"epoch": 3.346368715083799,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00042259934058523814,
|
|
"loss": 5.5407,
|
|
"mean_token_accuracy": 0.17531825453042985,
|
|
"num_tokens": 7077606.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.843122863769532,
|
|
"epoch": 3.3506660936828534,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00042234823369731027,
|
|
"loss": 5.5208,
|
|
"mean_token_accuracy": 0.1831410214304924,
|
|
"num_tokens": 7085647.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.75871639251709,
|
|
"epoch": 3.354963472281908,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00042209680505973465,
|
|
"loss": 5.5217,
|
|
"mean_token_accuracy": 0.18487536758184434,
|
|
"num_tokens": 7095298.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.7882904529571535,
|
|
"epoch": 3.3592608508809625,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004218450552215308,
|
|
"loss": 5.5837,
|
|
"mean_token_accuracy": 0.1783793792128563,
|
|
"num_tokens": 7105207.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.808549976348877,
|
|
"epoch": 3.363558229480017,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004215929847324199,
|
|
"loss": 5.6568,
|
|
"mean_token_accuracy": 0.17325044721364974,
|
|
"num_tokens": 7114833.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.890059137344361,
|
|
"epoch": 3.3678556080790716,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000421340594142823,
|
|
"loss": 5.447,
|
|
"mean_token_accuracy": 0.1935821980237961,
|
|
"num_tokens": 7123608.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.772169685363769,
|
|
"epoch": 3.3721529866781266,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00042108788400386035,
|
|
"loss": 5.5202,
|
|
"mean_token_accuracy": 0.18366121798753737,
|
|
"num_tokens": 7132250.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.777109003067016,
|
|
"epoch": 3.376450365277181,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004208348548673498,
|
|
"loss": 5.6247,
|
|
"mean_token_accuracy": 0.17855982035398482,
|
|
"num_tokens": 7142086.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.884216690063477,
|
|
"epoch": 3.3807477438762357,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000420581507285806,
|
|
"loss": 5.571,
|
|
"mean_token_accuracy": 0.16913529187440873,
|
|
"num_tokens": 7152434.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.7801055908203125,
|
|
"epoch": 3.38504512247529,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004203278418124386,
|
|
"loss": 5.5064,
|
|
"mean_token_accuracy": 0.18806122690439225,
|
|
"num_tokens": 7163041.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.74062147140503,
|
|
"epoch": 3.3893425010743448,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004200738590011518,
|
|
"loss": 5.4675,
|
|
"mean_token_accuracy": 0.1898912951350212,
|
|
"num_tokens": 7171875.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.762895250320435,
|
|
"epoch": 3.3936398796733993,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00041981955940654245,
|
|
"loss": 5.6127,
|
|
"mean_token_accuracy": 0.17519034296274186,
|
|
"num_tokens": 7180803.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.843291091918945,
|
|
"epoch": 3.397937258272454,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004195649435838992,
|
|
"loss": 5.6222,
|
|
"mean_token_accuracy": 0.16873589307069778,
|
|
"num_tokens": 7190661.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.777584600448608,
|
|
"epoch": 3.4022346368715084,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004193100120892013,
|
|
"loss": 5.4614,
|
|
"mean_token_accuracy": 0.1874678000807762,
|
|
"num_tokens": 7199357.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.876493167877197,
|
|
"epoch": 3.406532015470563,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004190547654791172,
|
|
"loss": 5.6328,
|
|
"mean_token_accuracy": 0.1726130172610283,
|
|
"num_tokens": 7209856.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.870272159576416,
|
|
"epoch": 3.4108293940696175,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00041879920431100347,
|
|
"loss": 5.6018,
|
|
"mean_token_accuracy": 0.17143317461013793,
|
|
"num_tokens": 7218778.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.822156190872192,
|
|
"epoch": 3.415126772668672,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004185433291429036,
|
|
"loss": 5.6106,
|
|
"mean_token_accuracy": 0.18186013400554657,
|
|
"num_tokens": 7228442.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.845083904266358,
|
|
"epoch": 3.4194241512677266,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00041828714053354665,
|
|
"loss": 5.5896,
|
|
"mean_token_accuracy": 0.17571255266666413,
|
|
"num_tokens": 7238724.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.775502252578735,
|
|
"epoch": 3.423721529866781,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004180306390423462,
|
|
"loss": 5.5535,
|
|
"mean_token_accuracy": 0.1877201244235039,
|
|
"num_tokens": 7247844.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.8070330142974855,
|
|
"epoch": 3.4280189084658357,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00041777382522939884,
|
|
"loss": 5.6153,
|
|
"mean_token_accuracy": 0.1745489463210106,
|
|
"num_tokens": 7257260.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.832814073562622,
|
|
"epoch": 3.4323162870648902,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00041751669965548344,
|
|
"loss": 5.6241,
|
|
"mean_token_accuracy": 0.17142881453037262,
|
|
"num_tokens": 7266890.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.893997573852539,
|
|
"epoch": 3.4366136656639448,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00041725926288205945,
|
|
"loss": 5.645,
|
|
"mean_token_accuracy": 0.17508225589990617,
|
|
"num_tokens": 7276114.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 3.4366136656639448,
|
|
"eval_entropy": 5.724426681930955,
|
|
"eval_loss": 6.096716403961182,
|
|
"eval_mean_token_accuracy": 0.16408700819823657,
|
|
"eval_num_tokens": 7276114.0,
|
|
"eval_runtime": 2.82,
|
|
"eval_samples_per_second": 1258.515,
|
|
"eval_steps_per_second": 157.447,
|
|
"step": 4000
|
|
}
|
|
],
|
|
"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": 1.1067323582592e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|