442bd82e67
Model: fpadovani/eus-latn-100mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407 Source: Original Platform
3068 lines
82 KiB
JSON
3068 lines
82 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.2887838418564677,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 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
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 11630,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 4136966316288000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|