0aeb1fb267
Model: fpadovani/urd-arab-10mb-after-ppt-shuff-dyck-10mb-ckpt500_seed10 Source: Original Platform
4079 lines
110 KiB
JSON
4079 lines
110 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 2.372479240806643,
|
|
"eval_steps": 500,
|
|
"global_step": 2000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 4.882989406585693,
|
|
"epoch": 0.005931198102016607,
|
|
"grad_norm": 22.5,
|
|
"learning_rate": 2e-06,
|
|
"loss": 14.5486,
|
|
"mean_token_accuracy": 0.00012430351926013828,
|
|
"num_tokens": 14276.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 4.891684865951538,
|
|
"epoch": 0.011862396204033215,
|
|
"grad_norm": 22.875,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 14.4388,
|
|
"mean_token_accuracy": 7.288630004040897e-05,
|
|
"num_tokens": 26454.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 4.908612489700317,
|
|
"epoch": 0.017793594306049824,
|
|
"grad_norm": 25.75,
|
|
"learning_rate": 7e-06,
|
|
"loss": 14.3151,
|
|
"mean_token_accuracy": 5.268703680485487e-05,
|
|
"num_tokens": 38987.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 4.985533952713013,
|
|
"epoch": 0.02372479240806643,
|
|
"grad_norm": 31.375,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 14.0964,
|
|
"mean_token_accuracy": 4.755111876875162e-05,
|
|
"num_tokens": 53908.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 5.351950407028198,
|
|
"epoch": 0.02965599051008304,
|
|
"grad_norm": 36.5,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 13.2683,
|
|
"mean_token_accuracy": 0.00010129763977602124,
|
|
"num_tokens": 66630.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 6.508245420455933,
|
|
"epoch": 0.03558718861209965,
|
|
"grad_norm": 38.0,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 12.2847,
|
|
"mean_token_accuracy": 0.00019785047043114902,
|
|
"num_tokens": 80227.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 9.8948805809021,
|
|
"epoch": 0.04151838671411625,
|
|
"grad_norm": 3.265625,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.9874,
|
|
"mean_token_accuracy": 0.00011806375114247203,
|
|
"num_tokens": 93260.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.6931321144104,
|
|
"epoch": 0.04744958481613286,
|
|
"grad_norm": 2.734375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.7316,
|
|
"mean_token_accuracy": 0.0010395277058705688,
|
|
"num_tokens": 107368.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.705201148986816,
|
|
"epoch": 0.05338078291814947,
|
|
"grad_norm": 2.515625,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.6199,
|
|
"mean_token_accuracy": 0.002701799588976428,
|
|
"num_tokens": 121910.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.703446388244629,
|
|
"epoch": 0.05931198102016608,
|
|
"grad_norm": 2.125,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 10.4434,
|
|
"mean_token_accuracy": 0.012030716100707651,
|
|
"num_tokens": 136756.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.688198661804199,
|
|
"epoch": 0.06524317912218268,
|
|
"grad_norm": 2.015625,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 10.3119,
|
|
"mean_token_accuracy": 0.009988007182255388,
|
|
"num_tokens": 152921.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.69539442062378,
|
|
"epoch": 0.0711743772241993,
|
|
"grad_norm": 1.9140625,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 10.1697,
|
|
"mean_token_accuracy": 0.012235769163817167,
|
|
"num_tokens": 167253.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.70052261352539,
|
|
"epoch": 0.0771055753262159,
|
|
"grad_norm": 1.8828125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 10.0223,
|
|
"mean_token_accuracy": 0.028588037192821502,
|
|
"num_tokens": 181088.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.69765625,
|
|
"epoch": 0.0830367734282325,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.9461,
|
|
"mean_token_accuracy": 0.030834931693971158,
|
|
"num_tokens": 193906.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.691731262207032,
|
|
"epoch": 0.08896797153024912,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.9097,
|
|
"mean_token_accuracy": 0.03433742746710777,
|
|
"num_tokens": 209520.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.694703388214112,
|
|
"epoch": 0.09489916963226572,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.8355,
|
|
"mean_token_accuracy": 0.03231241405010223,
|
|
"num_tokens": 223967.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.692516708374024,
|
|
"epoch": 0.10083036773428232,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.7472,
|
|
"mean_token_accuracy": 0.031197282299399377,
|
|
"num_tokens": 238991.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.681371212005615,
|
|
"epoch": 0.10676156583629894,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.635,
|
|
"mean_token_accuracy": 0.034548941254615786,
|
|
"num_tokens": 254281.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.67520875930786,
|
|
"epoch": 0.11269276393831554,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.4991,
|
|
"mean_token_accuracy": 0.03750690445303917,
|
|
"num_tokens": 268697.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.662280464172364,
|
|
"epoch": 0.11862396204033215,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.4058,
|
|
"mean_token_accuracy": 0.03157996293157339,
|
|
"num_tokens": 283026.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.645224285125732,
|
|
"epoch": 0.12455516014234876,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.2969,
|
|
"mean_token_accuracy": 0.033328772895038127,
|
|
"num_tokens": 298802.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.620039367675782,
|
|
"epoch": 0.13048635824436536,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.1491,
|
|
"mean_token_accuracy": 0.031768145971000195,
|
|
"num_tokens": 312390.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.58546199798584,
|
|
"epoch": 0.13641755634638197,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 9.0007,
|
|
"mean_token_accuracy": 0.033303617313504216,
|
|
"num_tokens": 325525.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.541824913024902,
|
|
"epoch": 0.1423487544483986,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.8754,
|
|
"mean_token_accuracy": 0.031658035889267924,
|
|
"num_tokens": 338168.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.4882963180542,
|
|
"epoch": 0.14827995255041518,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.7439,
|
|
"mean_token_accuracy": 0.034061219170689584,
|
|
"num_tokens": 351065.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.421833515167236,
|
|
"epoch": 0.1542111506524318,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.6351,
|
|
"mean_token_accuracy": 0.03405258022248745,
|
|
"num_tokens": 366974.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.340956115722657,
|
|
"epoch": 0.1601423487544484,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.4547,
|
|
"mean_token_accuracy": 0.03421926330775023,
|
|
"num_tokens": 380570.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.244134426116943,
|
|
"epoch": 0.166073546856465,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.3673,
|
|
"mean_token_accuracy": 0.03386142123490572,
|
|
"num_tokens": 395653.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.126684474945069,
|
|
"epoch": 0.17200474495848161,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.2663,
|
|
"mean_token_accuracy": 0.03231785073876381,
|
|
"num_tokens": 412138.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 9.98361644744873,
|
|
"epoch": 0.17793594306049823,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.1045,
|
|
"mean_token_accuracy": 0.03140157163143158,
|
|
"num_tokens": 425588.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.823878002166747,
|
|
"epoch": 0.18386714116251482,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.9299,
|
|
"mean_token_accuracy": 0.03208461999893188,
|
|
"num_tokens": 438907.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.627764892578124,
|
|
"epoch": 0.18979833926453143,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.8438,
|
|
"mean_token_accuracy": 0.032532901503145695,
|
|
"num_tokens": 454179.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.385260581970215,
|
|
"epoch": 0.19572953736654805,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.7521,
|
|
"mean_token_accuracy": 0.034452256932854655,
|
|
"num_tokens": 468367.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.124578285217286,
|
|
"epoch": 0.20166073546856464,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.7039,
|
|
"mean_token_accuracy": 0.03381109684705734,
|
|
"num_tokens": 483127.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.871719264984131,
|
|
"epoch": 0.20759193357058126,
|
|
"grad_norm": 0.76171875,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.6055,
|
|
"mean_token_accuracy": 0.031578476540744305,
|
|
"num_tokens": 498918.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.65745210647583,
|
|
"epoch": 0.21352313167259787,
|
|
"grad_norm": 0.74609375,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.5238,
|
|
"mean_token_accuracy": 0.0336442444473505,
|
|
"num_tokens": 512890.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.445704460144043,
|
|
"epoch": 0.21945432977461446,
|
|
"grad_norm": 0.59375,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.4868,
|
|
"mean_token_accuracy": 0.03619574885815382,
|
|
"num_tokens": 528517.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.257825183868409,
|
|
"epoch": 0.22538552787663108,
|
|
"grad_norm": 0.66796875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.4523,
|
|
"mean_token_accuracy": 0.03298983443528414,
|
|
"num_tokens": 542528.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.134656715393067,
|
|
"epoch": 0.2313167259786477,
|
|
"grad_norm": 0.59765625,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.4896,
|
|
"mean_token_accuracy": 0.0337029367685318,
|
|
"num_tokens": 557801.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.057815742492675,
|
|
"epoch": 0.2372479240806643,
|
|
"grad_norm": 0.52734375,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.466,
|
|
"mean_token_accuracy": 0.03353662136942148,
|
|
"num_tokens": 575318.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.983192443847656,
|
|
"epoch": 0.2431791221826809,
|
|
"grad_norm": 0.58984375,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.5065,
|
|
"mean_token_accuracy": 0.03289526663720608,
|
|
"num_tokens": 589283.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.915411520004272,
|
|
"epoch": 0.2491103202846975,
|
|
"grad_norm": 0.60546875,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.3919,
|
|
"mean_token_accuracy": 0.03649586420506239,
|
|
"num_tokens": 603112.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.879695129394531,
|
|
"epoch": 0.25504151838671413,
|
|
"grad_norm": 0.6015625,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.4059,
|
|
"mean_token_accuracy": 0.03437527269124985,
|
|
"num_tokens": 616993.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.861191368103027,
|
|
"epoch": 0.2609727164887307,
|
|
"grad_norm": 0.55859375,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.415,
|
|
"mean_token_accuracy": 0.032618867419660094,
|
|
"num_tokens": 630717.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.825675582885742,
|
|
"epoch": 0.2669039145907473,
|
|
"grad_norm": 0.55859375,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.4089,
|
|
"mean_token_accuracy": 0.03296402785927057,
|
|
"num_tokens": 646826.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.793451166152954,
|
|
"epoch": 0.27283511269276395,
|
|
"grad_norm": 0.54296875,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.4121,
|
|
"mean_token_accuracy": 0.03432573340833187,
|
|
"num_tokens": 662277.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.786031770706177,
|
|
"epoch": 0.27876631079478054,
|
|
"grad_norm": 0.60546875,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.4215,
|
|
"mean_token_accuracy": 0.03696324098855257,
|
|
"num_tokens": 676522.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.828532743453979,
|
|
"epoch": 0.2846975088967972,
|
|
"grad_norm": 0.59375,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.5228,
|
|
"mean_token_accuracy": 0.033165539428591725,
|
|
"num_tokens": 690425.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.870665788650513,
|
|
"epoch": 0.29062870699881377,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.5003,
|
|
"mean_token_accuracy": 0.037998564168810846,
|
|
"num_tokens": 704629.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.887278270721436,
|
|
"epoch": 0.29655990510083036,
|
|
"grad_norm": 0.609375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.4081,
|
|
"mean_token_accuracy": 0.0402442242950201,
|
|
"num_tokens": 721777.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.88170485496521,
|
|
"epoch": 0.302491103202847,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.3309,
|
|
"mean_token_accuracy": 0.04053100124001503,
|
|
"num_tokens": 734712.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.850059461593628,
|
|
"epoch": 0.3084223013048636,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.2525,
|
|
"mean_token_accuracy": 0.044293657690286634,
|
|
"num_tokens": 750222.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.8487138748168945,
|
|
"epoch": 0.3143534994068802,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.4068,
|
|
"mean_token_accuracy": 0.038819624856114385,
|
|
"num_tokens": 764699.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.8781726360321045,
|
|
"epoch": 0.3202846975088968,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.3138,
|
|
"mean_token_accuracy": 0.04715476129204035,
|
|
"num_tokens": 778738.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.895059490203858,
|
|
"epoch": 0.3262158956109134,
|
|
"grad_norm": 0.75,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.3011,
|
|
"mean_token_accuracy": 0.04708148837089539,
|
|
"num_tokens": 793904.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.921955537796021,
|
|
"epoch": 0.33214709371293,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.3829,
|
|
"mean_token_accuracy": 0.05229228027164936,
|
|
"num_tokens": 808310.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.833415269851685,
|
|
"epoch": 0.33807829181494664,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.235,
|
|
"mean_token_accuracy": 0.05603795982897282,
|
|
"num_tokens": 826253.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.822495317459106,
|
|
"epoch": 0.34400948991696323,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.1784,
|
|
"mean_token_accuracy": 0.05520997829735279,
|
|
"num_tokens": 839635.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.788468694686889,
|
|
"epoch": 0.3499406880189798,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.2809,
|
|
"mean_token_accuracy": 0.05334235914051533,
|
|
"num_tokens": 853870.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.822190046310425,
|
|
"epoch": 0.35587188612099646,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.2341,
|
|
"mean_token_accuracy": 0.05465950295329094,
|
|
"num_tokens": 868316.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.7704771041870115,
|
|
"epoch": 0.36180308422301305,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.1512,
|
|
"mean_token_accuracy": 0.05797473154962063,
|
|
"num_tokens": 881412.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.757186412811279,
|
|
"epoch": 0.36773428232502964,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.1676,
|
|
"mean_token_accuracy": 0.06134983114898205,
|
|
"num_tokens": 895656.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.723504638671875,
|
|
"epoch": 0.3736654804270463,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.0343,
|
|
"mean_token_accuracy": 0.06331710368394852,
|
|
"num_tokens": 909339.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.670499658584594,
|
|
"epoch": 0.37959667852906287,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.136,
|
|
"mean_token_accuracy": 0.05801789090037346,
|
|
"num_tokens": 922838.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.7397764205932615,
|
|
"epoch": 0.38552787663107946,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.1359,
|
|
"mean_token_accuracy": 0.06155713014304638,
|
|
"num_tokens": 936767.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.730058193206787,
|
|
"epoch": 0.3914590747330961,
|
|
"grad_norm": 2.03125,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.1752,
|
|
"mean_token_accuracy": 0.05990114398300648,
|
|
"num_tokens": 954078.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.619611692428589,
|
|
"epoch": 0.3973902728351127,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.0933,
|
|
"mean_token_accuracy": 0.06247694194316864,
|
|
"num_tokens": 968714.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.707681131362915,
|
|
"epoch": 0.4033214709371293,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.0391,
|
|
"mean_token_accuracy": 0.07253243401646614,
|
|
"num_tokens": 983001.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.655651664733886,
|
|
"epoch": 0.4092526690391459,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.0613,
|
|
"mean_token_accuracy": 0.06838090978562832,
|
|
"num_tokens": 997501.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.6562416553497314,
|
|
"epoch": 0.4151838671411625,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.0229,
|
|
"mean_token_accuracy": 0.07049233056604862,
|
|
"num_tokens": 1011683.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.5962440967559814,
|
|
"epoch": 0.4211150652431791,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.9731,
|
|
"mean_token_accuracy": 0.0693504311144352,
|
|
"num_tokens": 1024871.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.58204460144043,
|
|
"epoch": 0.42704626334519574,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.9777,
|
|
"mean_token_accuracy": 0.07172644883394241,
|
|
"num_tokens": 1038776.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.63423752784729,
|
|
"epoch": 0.43297746144721233,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.9462,
|
|
"mean_token_accuracy": 0.07613400146365165,
|
|
"num_tokens": 1052457.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.596765565872192,
|
|
"epoch": 0.4389086595492289,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.985,
|
|
"mean_token_accuracy": 0.06885807923972606,
|
|
"num_tokens": 1067566.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.509483337402344,
|
|
"epoch": 0.44483985765124556,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.8577,
|
|
"mean_token_accuracy": 0.0751548632979393,
|
|
"num_tokens": 1081285.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.488412714004516,
|
|
"epoch": 0.45077105575326215,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.8741,
|
|
"mean_token_accuracy": 0.08314598426222801,
|
|
"num_tokens": 1093944.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.485768795013428,
|
|
"epoch": 0.45670225385527874,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.8729,
|
|
"mean_token_accuracy": 0.08215565755963325,
|
|
"num_tokens": 1107210.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.5015223026275635,
|
|
"epoch": 0.4626334519572954,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.9484,
|
|
"mean_token_accuracy": 0.07512684762477875,
|
|
"num_tokens": 1120035.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.528394508361816,
|
|
"epoch": 0.46856465005931197,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.9877,
|
|
"mean_token_accuracy": 0.07711658254265785,
|
|
"num_tokens": 1132735.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.506278848648071,
|
|
"epoch": 0.4744958481613286,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.8883,
|
|
"mean_token_accuracy": 0.07609864845871925,
|
|
"num_tokens": 1146964.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.526645755767822,
|
|
"epoch": 0.4804270462633452,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.8871,
|
|
"mean_token_accuracy": 0.07854933217167855,
|
|
"num_tokens": 1161174.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.447538709640503,
|
|
"epoch": 0.4863582443653618,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.7717,
|
|
"mean_token_accuracy": 0.080623859167099,
|
|
"num_tokens": 1176833.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.474844169616699,
|
|
"epoch": 0.49228944246737844,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.9384,
|
|
"mean_token_accuracy": 0.07880584299564361,
|
|
"num_tokens": 1190845.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.3909095287322994,
|
|
"epoch": 0.498220640569395,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.8394,
|
|
"mean_token_accuracy": 0.07916948273777961,
|
|
"num_tokens": 1203878.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.414308786392212,
|
|
"epoch": 0.5041518386714117,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.8609,
|
|
"mean_token_accuracy": 0.07901249378919602,
|
|
"num_tokens": 1219135.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.502074193954468,
|
|
"epoch": 0.5100830367734283,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.8924,
|
|
"mean_token_accuracy": 0.08341019675135612,
|
|
"num_tokens": 1233167.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.4084759712219235,
|
|
"epoch": 0.5160142348754448,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.7389,
|
|
"mean_token_accuracy": 0.08747640326619148,
|
|
"num_tokens": 1246221.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.389473342895508,
|
|
"epoch": 0.5219454329774614,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.732,
|
|
"mean_token_accuracy": 0.08541311249136925,
|
|
"num_tokens": 1260856.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.329696989059448,
|
|
"epoch": 0.527876631079478,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.8368,
|
|
"mean_token_accuracy": 0.08163531869649887,
|
|
"num_tokens": 1276688.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.337572288513184,
|
|
"epoch": 0.5338078291814946,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.6987,
|
|
"mean_token_accuracy": 0.08712242096662522,
|
|
"num_tokens": 1291525.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.3344158172607425,
|
|
"epoch": 0.5397390272835113,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.7266,
|
|
"mean_token_accuracy": 0.08258413746953011,
|
|
"num_tokens": 1307769.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.2570515155792235,
|
|
"epoch": 0.5456702253855279,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.7115,
|
|
"mean_token_accuracy": 0.08097989335656167,
|
|
"num_tokens": 1322462.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.2473875999450685,
|
|
"epoch": 0.5516014234875445,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.6807,
|
|
"mean_token_accuracy": 0.08938496559858322,
|
|
"num_tokens": 1338063.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.260567998886108,
|
|
"epoch": 0.5575326215895611,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.6452,
|
|
"mean_token_accuracy": 0.09352756068110465,
|
|
"num_tokens": 1351384.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.236065340042114,
|
|
"epoch": 0.5634638196915777,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.6144,
|
|
"mean_token_accuracy": 0.08987314701080322,
|
|
"num_tokens": 1363781.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.16572265625,
|
|
"epoch": 0.5693950177935944,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.6055,
|
|
"mean_token_accuracy": 0.08821886777877808,
|
|
"num_tokens": 1379463.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.195259714126587,
|
|
"epoch": 0.575326215895611,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.6073,
|
|
"mean_token_accuracy": 0.09338206499814987,
|
|
"num_tokens": 1393631.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.307022905349731,
|
|
"epoch": 0.5812574139976275,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.6937,
|
|
"mean_token_accuracy": 0.08968389555811881,
|
|
"num_tokens": 1407401.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.144880056381226,
|
|
"epoch": 0.5871886120996441,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.5565,
|
|
"mean_token_accuracy": 0.09463186785578728,
|
|
"num_tokens": 1422167.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.16906509399414,
|
|
"epoch": 0.5931198102016607,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.6659,
|
|
"mean_token_accuracy": 0.08580891266465188,
|
|
"num_tokens": 1438662.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.5931198102016607,
|
|
"eval_entropy": 7.088789626864568,
|
|
"eval_loss": 6.709393501281738,
|
|
"eval_mean_token_accuracy": 0.09428181704362294,
|
|
"eval_num_tokens": 1438662.0,
|
|
"eval_runtime": 1.9148,
|
|
"eval_samples_per_second": 1362.036,
|
|
"eval_steps_per_second": 170.254,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.0674371242523195,
|
|
"epoch": 0.5990510083036773,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.4938,
|
|
"mean_token_accuracy": 0.09932567328214645,
|
|
"num_tokens": 1452224.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.124333143234253,
|
|
"epoch": 0.604982206405694,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.476,
|
|
"mean_token_accuracy": 0.09715204015374183,
|
|
"num_tokens": 1465853.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.134642124176025,
|
|
"epoch": 0.6109134045077106,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.6224,
|
|
"mean_token_accuracy": 0.09282981902360916,
|
|
"num_tokens": 1480125.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.0999956130981445,
|
|
"epoch": 0.6168446026097272,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.649,
|
|
"mean_token_accuracy": 0.09250865951180458,
|
|
"num_tokens": 1495040.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.164534378051758,
|
|
"epoch": 0.6227758007117438,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.5171,
|
|
"mean_token_accuracy": 0.0987244538962841,
|
|
"num_tokens": 1509672.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.013086462020874,
|
|
"epoch": 0.6287069988137604,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.5082,
|
|
"mean_token_accuracy": 0.09528515487909317,
|
|
"num_tokens": 1521921.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.022832727432251,
|
|
"epoch": 0.6346381969157769,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.4685,
|
|
"mean_token_accuracy": 0.09804724678397178,
|
|
"num_tokens": 1535993.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.1515604019165036,
|
|
"epoch": 0.6405693950177936,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.5696,
|
|
"mean_token_accuracy": 0.09566482156515121,
|
|
"num_tokens": 1548292.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.995516347885132,
|
|
"epoch": 0.6465005931198102,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.5626,
|
|
"mean_token_accuracy": 0.09593151062726975,
|
|
"num_tokens": 1564004.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.1294818878173825,
|
|
"epoch": 0.6524317912218268,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.4788,
|
|
"mean_token_accuracy": 0.10118974596261979,
|
|
"num_tokens": 1577543.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.000808572769165,
|
|
"epoch": 0.6583629893238434,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.4619,
|
|
"mean_token_accuracy": 0.10349823236465454,
|
|
"num_tokens": 1590529.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.059898233413696,
|
|
"epoch": 0.66429418742586,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.5415,
|
|
"mean_token_accuracy": 0.09783142134547233,
|
|
"num_tokens": 1605584.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.057692337036133,
|
|
"epoch": 0.6702253855278766,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.5684,
|
|
"mean_token_accuracy": 0.1004520446062088,
|
|
"num_tokens": 1618905.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.014936447143555,
|
|
"epoch": 0.6761565836298933,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.545,
|
|
"mean_token_accuracy": 0.09201160371303559,
|
|
"num_tokens": 1632538.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.974603891372681,
|
|
"epoch": 0.6820877817319099,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.3815,
|
|
"mean_token_accuracy": 0.10909996330738067,
|
|
"num_tokens": 1646327.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.008380365371704,
|
|
"epoch": 0.6880189798339265,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.5448,
|
|
"mean_token_accuracy": 0.10221554338932037,
|
|
"num_tokens": 1659913.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.0425440788269045,
|
|
"epoch": 0.693950177935943,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.4521,
|
|
"mean_token_accuracy": 0.10268813446164131,
|
|
"num_tokens": 1673625.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.9817821979522705,
|
|
"epoch": 0.6998813760379596,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.466,
|
|
"mean_token_accuracy": 0.09939212873578071,
|
|
"num_tokens": 1688495.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.9125754833221436,
|
|
"epoch": 0.7058125741399762,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.3759,
|
|
"mean_token_accuracy": 0.10623608976602554,
|
|
"num_tokens": 1702989.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.892080068588257,
|
|
"epoch": 0.7117437722419929,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.3151,
|
|
"mean_token_accuracy": 0.10541760846972466,
|
|
"num_tokens": 1717331.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.9624193668365475,
|
|
"epoch": 0.7176749703440095,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.4368,
|
|
"mean_token_accuracy": 0.10582135245203972,
|
|
"num_tokens": 1733012.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.895502758026123,
|
|
"epoch": 0.7236061684460261,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.446,
|
|
"mean_token_accuracy": 0.10651310458779335,
|
|
"num_tokens": 1747058.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.954757976531982,
|
|
"epoch": 0.7295373665480427,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.4605,
|
|
"mean_token_accuracy": 0.10364130288362502,
|
|
"num_tokens": 1761983.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.902099752426148,
|
|
"epoch": 0.7354685646500593,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.477,
|
|
"mean_token_accuracy": 0.09844475984573364,
|
|
"num_tokens": 1777220.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.894039106369019,
|
|
"epoch": 0.741399762752076,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.3396,
|
|
"mean_token_accuracy": 0.10894461125135421,
|
|
"num_tokens": 1791770.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.9334691047668455,
|
|
"epoch": 0.7473309608540926,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.4141,
|
|
"mean_token_accuracy": 0.11063586622476577,
|
|
"num_tokens": 1805717.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.90194263458252,
|
|
"epoch": 0.7532621589561092,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.3982,
|
|
"mean_token_accuracy": 0.10740711987018585,
|
|
"num_tokens": 1819943.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.841003894805908,
|
|
"epoch": 0.7591933570581257,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.3223,
|
|
"mean_token_accuracy": 0.10887554213404656,
|
|
"num_tokens": 1835219.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.813411712646484,
|
|
"epoch": 0.7651245551601423,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.2411,
|
|
"mean_token_accuracy": 0.11231007426977158,
|
|
"num_tokens": 1849467.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.809050035476685,
|
|
"epoch": 0.7710557532621589,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.2833,
|
|
"mean_token_accuracy": 0.11221997663378716,
|
|
"num_tokens": 1864160.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.759777450561524,
|
|
"epoch": 0.7769869513641756,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.2426,
|
|
"mean_token_accuracy": 0.11081567630171776,
|
|
"num_tokens": 1879324.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.799328899383545,
|
|
"epoch": 0.7829181494661922,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.314,
|
|
"mean_token_accuracy": 0.11415420845150948,
|
|
"num_tokens": 1893902.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.700550222396851,
|
|
"epoch": 0.7888493475682088,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.1519,
|
|
"mean_token_accuracy": 0.1172316811978817,
|
|
"num_tokens": 1908216.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.70758090019226,
|
|
"epoch": 0.7947805456702254,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.2537,
|
|
"mean_token_accuracy": 0.11493151113390923,
|
|
"num_tokens": 1923408.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.809796285629273,
|
|
"epoch": 0.800711743772242,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.3333,
|
|
"mean_token_accuracy": 0.10704677328467369,
|
|
"num_tokens": 1938307.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.800277757644653,
|
|
"epoch": 0.8066429418742586,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.3288,
|
|
"mean_token_accuracy": 0.10877522677183152,
|
|
"num_tokens": 1953173.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.750416469573975,
|
|
"epoch": 0.8125741399762753,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.2148,
|
|
"mean_token_accuracy": 0.11826780289411545,
|
|
"num_tokens": 1966198.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.749280786514282,
|
|
"epoch": 0.8185053380782918,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.3035,
|
|
"mean_token_accuracy": 0.11757627055048943,
|
|
"num_tokens": 1979608.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.750239419937134,
|
|
"epoch": 0.8244365361803084,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.2271,
|
|
"mean_token_accuracy": 0.11730416044592858,
|
|
"num_tokens": 1992949.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.718313598632813,
|
|
"epoch": 0.830367734282325,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.3097,
|
|
"mean_token_accuracy": 0.11092614904046058,
|
|
"num_tokens": 2007149.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.6663001537322994,
|
|
"epoch": 0.8362989323843416,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.2049,
|
|
"mean_token_accuracy": 0.11998757123947143,
|
|
"num_tokens": 2019444.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.776406240463257,
|
|
"epoch": 0.8422301304863582,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.2367,
|
|
"mean_token_accuracy": 0.1130843311548233,
|
|
"num_tokens": 2034499.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.747935152053833,
|
|
"epoch": 0.8481613285883749,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.2989,
|
|
"mean_token_accuracy": 0.11696135327219963,
|
|
"num_tokens": 2048293.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.70717420578003,
|
|
"epoch": 0.8540925266903915,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.263,
|
|
"mean_token_accuracy": 0.11379396319389343,
|
|
"num_tokens": 2062350.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.708211469650268,
|
|
"epoch": 0.8600237247924081,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.2456,
|
|
"mean_token_accuracy": 0.11864681616425514,
|
|
"num_tokens": 2076379.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.748144721984863,
|
|
"epoch": 0.8659549228944247,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.2164,
|
|
"mean_token_accuracy": 0.11808046251535416,
|
|
"num_tokens": 2090071.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.681804513931274,
|
|
"epoch": 0.8718861209964412,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.2306,
|
|
"mean_token_accuracy": 0.11399208307266236,
|
|
"num_tokens": 2102585.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.720635604858399,
|
|
"epoch": 0.8778173190984578,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.2257,
|
|
"mean_token_accuracy": 0.11735560595989228,
|
|
"num_tokens": 2116559.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.577832317352295,
|
|
"epoch": 0.8837485172004745,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.1975,
|
|
"mean_token_accuracy": 0.1139634482562542,
|
|
"num_tokens": 2130048.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.723380517959595,
|
|
"epoch": 0.8896797153024911,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.258,
|
|
"mean_token_accuracy": 0.11037925630807877,
|
|
"num_tokens": 2145176.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.668367147445679,
|
|
"epoch": 0.8956109134045077,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.174,
|
|
"mean_token_accuracy": 0.12165194749832153,
|
|
"num_tokens": 2160615.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.569440317153931,
|
|
"epoch": 0.9015421115065243,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.1306,
|
|
"mean_token_accuracy": 0.11893835440278053,
|
|
"num_tokens": 2174131.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.665890264511108,
|
|
"epoch": 0.9074733096085409,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.1707,
|
|
"mean_token_accuracy": 0.12153652384877205,
|
|
"num_tokens": 2188906.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.573056364059449,
|
|
"epoch": 0.9134045077105575,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.2009,
|
|
"mean_token_accuracy": 0.11878187209367752,
|
|
"num_tokens": 2203620.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.587138843536377,
|
|
"epoch": 0.9193357058125742,
|
|
"grad_norm": 1.9765625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.1415,
|
|
"mean_token_accuracy": 0.12185289338231087,
|
|
"num_tokens": 2218133.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.585116243362426,
|
|
"epoch": 0.9252669039145908,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.2137,
|
|
"mean_token_accuracy": 0.11551980301737785,
|
|
"num_tokens": 2233958.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.772628355026245,
|
|
"epoch": 0.9311981020166074,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.2822,
|
|
"mean_token_accuracy": 0.1171758972108364,
|
|
"num_tokens": 2250430.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.646478462219238,
|
|
"epoch": 0.9371293001186239,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.1562,
|
|
"mean_token_accuracy": 0.1206930473446846,
|
|
"num_tokens": 2266098.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.496437501907349,
|
|
"epoch": 0.9430604982206405,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.0743,
|
|
"mean_token_accuracy": 0.12797637432813644,
|
|
"num_tokens": 2279446.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.623060703277588,
|
|
"epoch": 0.9489916963226572,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.1254,
|
|
"mean_token_accuracy": 0.121622234582901,
|
|
"num_tokens": 2293714.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.50568151473999,
|
|
"epoch": 0.9549228944246738,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.1666,
|
|
"mean_token_accuracy": 0.11893192976713181,
|
|
"num_tokens": 2309619.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.559406423568726,
|
|
"epoch": 0.9608540925266904,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.0854,
|
|
"mean_token_accuracy": 0.12828661650419235,
|
|
"num_tokens": 2324091.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.625924682617187,
|
|
"epoch": 0.966785290628707,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.2051,
|
|
"mean_token_accuracy": 0.12474795132875442,
|
|
"num_tokens": 2338066.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.44111385345459,
|
|
"epoch": 0.9727164887307236,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.0949,
|
|
"mean_token_accuracy": 0.12794609144330024,
|
|
"num_tokens": 2354283.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.648028707504272,
|
|
"epoch": 0.9786476868327402,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.1346,
|
|
"mean_token_accuracy": 0.11840463951230049,
|
|
"num_tokens": 2368788.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.441373825073242,
|
|
"epoch": 0.9845788849347569,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.1425,
|
|
"mean_token_accuracy": 0.12159557864069939,
|
|
"num_tokens": 2383267.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.50308928489685,
|
|
"epoch": 0.9905100830367735,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.0508,
|
|
"mean_token_accuracy": 0.13076636716723442,
|
|
"num_tokens": 2396156.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.507014608383178,
|
|
"epoch": 0.99644128113879,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.1286,
|
|
"mean_token_accuracy": 0.11710731312632561,
|
|
"num_tokens": 2411839.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.392842102050781,
|
|
"epoch": 1.0023724792408066,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000422,
|
|
"loss": 5.9181,
|
|
"mean_token_accuracy": 0.12669287249445915,
|
|
"num_tokens": 2425541.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.5243079662323,
|
|
"epoch": 1.0083036773428233,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.1084,
|
|
"mean_token_accuracy": 0.12104234322905541,
|
|
"num_tokens": 2440445.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.520119667053223,
|
|
"epoch": 1.0142348754448398,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.0277,
|
|
"mean_token_accuracy": 0.1227425530552864,
|
|
"num_tokens": 2454894.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.527532386779785,
|
|
"epoch": 1.0201660735468565,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.019,
|
|
"mean_token_accuracy": 0.1259557895362377,
|
|
"num_tokens": 2470706.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.457006645202637,
|
|
"epoch": 1.026097271648873,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.035,
|
|
"mean_token_accuracy": 0.1288458950817585,
|
|
"num_tokens": 2484134.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.425363540649414,
|
|
"epoch": 1.0320284697508897,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 5.8931,
|
|
"mean_token_accuracy": 0.13428079336881638,
|
|
"num_tokens": 2498744.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.434927558898925,
|
|
"epoch": 1.0379596678529064,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.0079,
|
|
"mean_token_accuracy": 0.12984821051359177,
|
|
"num_tokens": 2513456.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.387080335617066,
|
|
"epoch": 1.0438908659549229,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 5.9021,
|
|
"mean_token_accuracy": 0.1296766184270382,
|
|
"num_tokens": 2528895.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.425879383087159,
|
|
"epoch": 1.0498220640569396,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000442,
|
|
"loss": 5.9424,
|
|
"mean_token_accuracy": 0.14008270800113679,
|
|
"num_tokens": 2541415.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.41387038230896,
|
|
"epoch": 1.055753262158956,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 5.927,
|
|
"mean_token_accuracy": 0.13471830263733864,
|
|
"num_tokens": 2554822.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.359044361114502,
|
|
"epoch": 1.0616844602609727,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000447,
|
|
"loss": 5.9672,
|
|
"mean_token_accuracy": 0.13547785803675652,
|
|
"num_tokens": 2567676.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.4110674381256105,
|
|
"epoch": 1.0676156583629894,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.0008,
|
|
"mean_token_accuracy": 0.13326788023114206,
|
|
"num_tokens": 2580828.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.451390743255615,
|
|
"epoch": 1.073546856465006,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.0095,
|
|
"mean_token_accuracy": 0.12793489992618562,
|
|
"num_tokens": 2595402.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.346421480178833,
|
|
"epoch": 1.0794780545670226,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 5.9495,
|
|
"mean_token_accuracy": 0.1260056674480438,
|
|
"num_tokens": 2612266.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.373950624465943,
|
|
"epoch": 1.085409252669039,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 5.9615,
|
|
"mean_token_accuracy": 0.13041894584894181,
|
|
"num_tokens": 2626064.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.378324556350708,
|
|
"epoch": 1.0913404507710558,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.0058,
|
|
"mean_token_accuracy": 0.12815107107162477,
|
|
"num_tokens": 2642134.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.3164959907531735,
|
|
"epoch": 1.0972716488730723,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 5.925,
|
|
"mean_token_accuracy": 0.13265205621719361,
|
|
"num_tokens": 2656009.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.399133729934692,
|
|
"epoch": 1.103202846975089,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 5.9817,
|
|
"mean_token_accuracy": 0.13066953867673875,
|
|
"num_tokens": 2671353.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.354629421234131,
|
|
"epoch": 1.1091340450771057,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000467,
|
|
"loss": 5.9245,
|
|
"mean_token_accuracy": 0.13608077466487883,
|
|
"num_tokens": 2685488.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.361505746841431,
|
|
"epoch": 1.1150652431791221,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 5.9299,
|
|
"mean_token_accuracy": 0.13026007637381554,
|
|
"num_tokens": 2700988.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.421771621704101,
|
|
"epoch": 1.1209964412811388,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000472,
|
|
"loss": 5.9483,
|
|
"mean_token_accuracy": 0.13384367898106575,
|
|
"num_tokens": 2717490.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.251776599884034,
|
|
"epoch": 1.1269276393831553,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 5.8468,
|
|
"mean_token_accuracy": 0.13876772224903106,
|
|
"num_tokens": 2730463.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.280179023742676,
|
|
"epoch": 1.132858837485172,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000477,
|
|
"loss": 5.8257,
|
|
"mean_token_accuracy": 0.13821708783507347,
|
|
"num_tokens": 2743458.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.371533966064453,
|
|
"epoch": 1.1387900355871885,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 5.9696,
|
|
"mean_token_accuracy": 0.13559629395604134,
|
|
"num_tokens": 2756064.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.283103799819946,
|
|
"epoch": 1.1447212336892052,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.000482,
|
|
"loss": 5.9031,
|
|
"mean_token_accuracy": 0.13596163243055343,
|
|
"num_tokens": 2770046.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.244359016418457,
|
|
"epoch": 1.150652431791222,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 5.9025,
|
|
"mean_token_accuracy": 0.13229298815131188,
|
|
"num_tokens": 2784308.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.287460231781006,
|
|
"epoch": 1.1565836298932384,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.000487,
|
|
"loss": 5.8552,
|
|
"mean_token_accuracy": 0.13137058764696122,
|
|
"num_tokens": 2799315.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.288997983932495,
|
|
"epoch": 1.162514827995255,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 5.9336,
|
|
"mean_token_accuracy": 0.1334098532795906,
|
|
"num_tokens": 2814561.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.325786352157593,
|
|
"epoch": 1.1684460260972718,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000492,
|
|
"loss": 5.916,
|
|
"mean_token_accuracy": 0.1331045538187027,
|
|
"num_tokens": 2829823.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.34386134147644,
|
|
"epoch": 1.1743772241992882,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 5.9348,
|
|
"mean_token_accuracy": 0.12784818038344384,
|
|
"num_tokens": 2843555.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.350662803649902,
|
|
"epoch": 1.180308422301305,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000497,
|
|
"loss": 5.9539,
|
|
"mean_token_accuracy": 0.1361467495560646,
|
|
"num_tokens": 2857071.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.224388265609742,
|
|
"epoch": 1.1862396204033214,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 5.8728,
|
|
"mean_token_accuracy": 0.1347593829035759,
|
|
"num_tokens": 2873640.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 1.1862396204033214,
|
|
"eval_entropy": 6.175969420766538,
|
|
"eval_loss": 6.0720014572143555,
|
|
"eval_mean_token_accuracy": 0.13131910120279877,
|
|
"eval_num_tokens": 2873640.0,
|
|
"eval_runtime": 1.8498,
|
|
"eval_samples_per_second": 1409.862,
|
|
"eval_steps_per_second": 176.233,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.261200714111328,
|
|
"epoch": 1.1921708185053381,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00049999967732573,
|
|
"loss": 5.8506,
|
|
"mean_token_accuracy": 0.14092798456549643,
|
|
"num_tokens": 2887453.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.166975975036621,
|
|
"epoch": 1.1981020166073546,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999983664630945,
|
|
"loss": 5.8298,
|
|
"mean_token_accuracy": 0.13897372782230377,
|
|
"num_tokens": 2904225.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.26166615486145,
|
|
"epoch": 1.2040332147093713,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004999960472508219,
|
|
"loss": 5.8504,
|
|
"mean_token_accuracy": 0.13864508718252183,
|
|
"num_tokens": 2918551.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.1890528202056885,
|
|
"epoch": 1.209964412811388,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004999927196993059,
|
|
"loss": 5.8572,
|
|
"mean_token_accuracy": 0.13447567522525788,
|
|
"num_tokens": 2932219.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.280620288848877,
|
|
"epoch": 1.2158956109134045,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0004999883838234592,
|
|
"loss": 5.8728,
|
|
"mean_token_accuracy": 0.13982290029525757,
|
|
"num_tokens": 2947053.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.307330179214477,
|
|
"epoch": 1.2218268090154212,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004999830396427133,
|
|
"loss": 5.9092,
|
|
"mean_token_accuracy": 0.13675478994846343,
|
|
"num_tokens": 2961901.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.2650683403015135,
|
|
"epoch": 1.2277580071174377,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999766871810188,
|
|
"loss": 5.9234,
|
|
"mean_token_accuracy": 0.12962723821401595,
|
|
"num_tokens": 2976789.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.266324710845947,
|
|
"epoch": 1.2336892052194544,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004999693264668446,
|
|
"loss": 5.9277,
|
|
"mean_token_accuracy": 0.1318037800490856,
|
|
"num_tokens": 2992061.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.30645899772644,
|
|
"epoch": 1.2396204033214708,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999609575331786,
|
|
"loss": 5.8895,
|
|
"mean_token_accuracy": 0.1368556059896946,
|
|
"num_tokens": 3007628.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.219874858856201,
|
|
"epoch": 1.2455516014234875,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004999515804175268,
|
|
"loss": 5.8365,
|
|
"mean_token_accuracy": 0.13651281595230103,
|
|
"num_tokens": 3022422.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.183523321151734,
|
|
"epoch": 1.2514827995255042,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004999411951619134,
|
|
"loss": 5.8362,
|
|
"mean_token_accuracy": 0.13714614510536194,
|
|
"num_tokens": 3036060.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.293585252761841,
|
|
"epoch": 1.2574139976275207,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000499929801812881,
|
|
"loss": 5.8794,
|
|
"mean_token_accuracy": 0.14127954691648484,
|
|
"num_tokens": 3050531.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.1592449188232425,
|
|
"epoch": 1.2633451957295374,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999174004214899,
|
|
"loss": 5.7805,
|
|
"mean_token_accuracy": 0.14164935201406478,
|
|
"num_tokens": 3065038.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.235931015014648,
|
|
"epoch": 1.269276393831554,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999039910433179,
|
|
"loss": 5.8957,
|
|
"mean_token_accuracy": 0.13519704565405846,
|
|
"num_tokens": 3080501.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.14642448425293,
|
|
"epoch": 1.2752075919335706,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004998895737384605,
|
|
"loss": 5.7994,
|
|
"mean_token_accuracy": 0.14483253732323648,
|
|
"num_tokens": 3095070.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.188129091262818,
|
|
"epoch": 1.281138790035587,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004998741485715297,
|
|
"loss": 5.8213,
|
|
"mean_token_accuracy": 0.138034887611866,
|
|
"num_tokens": 3109827.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.17890419960022,
|
|
"epoch": 1.2870699881376038,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004998577156116551,
|
|
"loss": 5.7828,
|
|
"mean_token_accuracy": 0.14100482687354088,
|
|
"num_tokens": 3124176.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.258465957641602,
|
|
"epoch": 1.2930011862396205,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004998402749324823,
|
|
"loss": 5.8919,
|
|
"mean_token_accuracy": 0.13448047116398812,
|
|
"num_tokens": 3137696.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.150644254684448,
|
|
"epoch": 1.298932384341637,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004998218266121731,
|
|
"loss": 5.7421,
|
|
"mean_token_accuracy": 0.1446476884186268,
|
|
"num_tokens": 3151245.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.12621955871582,
|
|
"epoch": 1.3048635824436536,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998023707334055,
|
|
"loss": 5.7242,
|
|
"mean_token_accuracy": 0.14792525842785836,
|
|
"num_tokens": 3165414.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.181356716156006,
|
|
"epoch": 1.3107947805456703,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004997819073833724,
|
|
"loss": 5.8237,
|
|
"mean_token_accuracy": 0.13567239567637443,
|
|
"num_tokens": 3179998.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.103460025787354,
|
|
"epoch": 1.3167259786476868,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004997604366537823,
|
|
"loss": 5.9501,
|
|
"mean_token_accuracy": 0.13358369320631028,
|
|
"num_tokens": 3194998.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.129420804977417,
|
|
"epoch": 1.3226571767497035,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000499737958640858,
|
|
"loss": 5.7875,
|
|
"mean_token_accuracy": 0.1355852596461773,
|
|
"num_tokens": 3211626.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.1442975997924805,
|
|
"epoch": 1.32858837485172,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004997144734453367,
|
|
"loss": 5.7596,
|
|
"mean_token_accuracy": 0.1420934334397316,
|
|
"num_tokens": 3225681.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.0875319480896,
|
|
"epoch": 1.3345195729537367,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004996899811724693,
|
|
"loss": 5.7775,
|
|
"mean_token_accuracy": 0.1395690955221653,
|
|
"num_tokens": 3240603.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.186561536788941,
|
|
"epoch": 1.3404507710557532,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004996644819320201,
|
|
"loss": 5.8046,
|
|
"mean_token_accuracy": 0.13793422132730485,
|
|
"num_tokens": 3254261.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.1490716457366945,
|
|
"epoch": 1.3463819691577699,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004996379758382662,
|
|
"loss": 5.8794,
|
|
"mean_token_accuracy": 0.13867162466049193,
|
|
"num_tokens": 3270602.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.176916170120239,
|
|
"epoch": 1.3523131672597866,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004996104630099969,
|
|
"loss": 5.7939,
|
|
"mean_token_accuracy": 0.13939716890454293,
|
|
"num_tokens": 3283226.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.120092248916626,
|
|
"epoch": 1.358244365361803,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004995819435705133,
|
|
"loss": 5.7043,
|
|
"mean_token_accuracy": 0.15347797572612762,
|
|
"num_tokens": 3296296.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.150380182266235,
|
|
"epoch": 1.3641755634638197,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004995524176476277,
|
|
"loss": 5.7537,
|
|
"mean_token_accuracy": 0.14649274945259094,
|
|
"num_tokens": 3310817.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.15944538116455,
|
|
"epoch": 1.3701067615658362,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004995218853736631,
|
|
"loss": 5.7455,
|
|
"mean_token_accuracy": 0.14735049456357957,
|
|
"num_tokens": 3326319.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.005965852737427,
|
|
"epoch": 1.376037959667853,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004994903468854527,
|
|
"loss": 5.7052,
|
|
"mean_token_accuracy": 0.14868801534175874,
|
|
"num_tokens": 3340015.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.097800588607788,
|
|
"epoch": 1.3819691577698694,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0004994578023243386,
|
|
"loss": 5.7548,
|
|
"mean_token_accuracy": 0.1444389432668686,
|
|
"num_tokens": 3354246.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.061732959747315,
|
|
"epoch": 1.387900355871886,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004994242518361723,
|
|
"loss": 5.7807,
|
|
"mean_token_accuracy": 0.14473409205675125,
|
|
"num_tokens": 3368857.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.090450572967529,
|
|
"epoch": 1.3938315539739028,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004993896955713133,
|
|
"loss": 5.7257,
|
|
"mean_token_accuracy": 0.14366455227136612,
|
|
"num_tokens": 3380801.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.117128992080689,
|
|
"epoch": 1.3997627520759193,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004993541336846283,
|
|
"loss": 5.6886,
|
|
"mean_token_accuracy": 0.1452789694070816,
|
|
"num_tokens": 3394435.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 5.991869878768921,
|
|
"epoch": 1.405693950177936,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000499317566335491,
|
|
"loss": 5.6642,
|
|
"mean_token_accuracy": 0.1542855590581894,
|
|
"num_tokens": 3408420.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.142762470245361,
|
|
"epoch": 1.4116251482799527,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004992799936877813,
|
|
"loss": 5.8448,
|
|
"mean_token_accuracy": 0.14580541998147964,
|
|
"num_tokens": 3420491.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.057984399795532,
|
|
"epoch": 1.4175563463819691,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004992414159098841,
|
|
"loss": 5.6982,
|
|
"mean_token_accuracy": 0.15121331959962844,
|
|
"num_tokens": 3434037.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.031634330749512,
|
|
"epoch": 1.4234875444839858,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004992018331746891,
|
|
"loss": 5.6921,
|
|
"mean_token_accuracy": 0.15349351465702057,
|
|
"num_tokens": 3448127.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.030265045166016,
|
|
"epoch": 1.4294187425860023,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004991612456595896,
|
|
"loss": 5.7577,
|
|
"mean_token_accuracy": 0.14504036903381348,
|
|
"num_tokens": 3462144.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.0907364845275875,
|
|
"epoch": 1.435349940688019,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004991196535464824,
|
|
"loss": 5.7082,
|
|
"mean_token_accuracy": 0.14401856660842896,
|
|
"num_tokens": 3477362.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.056795883178711,
|
|
"epoch": 1.4412811387900355,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004990770570217659,
|
|
"loss": 5.712,
|
|
"mean_token_accuracy": 0.14216629415750504,
|
|
"num_tokens": 3490211.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.025155782699585,
|
|
"epoch": 1.4472123368920522,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004990334562763401,
|
|
"loss": 5.6299,
|
|
"mean_token_accuracy": 0.15666819214820862,
|
|
"num_tokens": 3502553.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.0306840419769285,
|
|
"epoch": 1.453143534994069,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004989888515056057,
|
|
"loss": 5.7362,
|
|
"mean_token_accuracy": 0.14873925745487213,
|
|
"num_tokens": 3517758.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.040766811370849,
|
|
"epoch": 1.4590747330960854,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004989432429094627,
|
|
"loss": 5.6727,
|
|
"mean_token_accuracy": 0.14901096224784852,
|
|
"num_tokens": 3533303.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.05291166305542,
|
|
"epoch": 1.465005931198102,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004988966306923099,
|
|
"loss": 5.7784,
|
|
"mean_token_accuracy": 0.14436503797769545,
|
|
"num_tokens": 3547401.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.071095895767212,
|
|
"epoch": 1.4709371293001186,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004988490150630441,
|
|
"loss": 5.7429,
|
|
"mean_token_accuracy": 0.14811725914478302,
|
|
"num_tokens": 3561854.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.093766927719116,
|
|
"epoch": 1.4768683274021353,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004988003962350589,
|
|
"loss": 5.8124,
|
|
"mean_token_accuracy": 0.13454216867685317,
|
|
"num_tokens": 3578461.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.064093065261841,
|
|
"epoch": 1.4827995255041517,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004987507744262436,
|
|
"loss": 5.7559,
|
|
"mean_token_accuracy": 0.138331101834774,
|
|
"num_tokens": 3592660.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.066448545455932,
|
|
"epoch": 1.4887307236061684,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004987001498589828,
|
|
"loss": 5.7088,
|
|
"mean_token_accuracy": 0.14475469440221786,
|
|
"num_tokens": 3606050.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.1090662479400635,
|
|
"epoch": 1.4946619217081851,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004986485227601548,
|
|
"loss": 5.7276,
|
|
"mean_token_accuracy": 0.14912586510181428,
|
|
"num_tokens": 3619446.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.049404096603394,
|
|
"epoch": 1.5005931198102016,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000498595893361131,
|
|
"loss": 5.6724,
|
|
"mean_token_accuracy": 0.15446974337100983,
|
|
"num_tokens": 3632359.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 5.9836235523223875,
|
|
"epoch": 1.5065243179122183,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004985422618977744,
|
|
"loss": 5.6234,
|
|
"mean_token_accuracy": 0.1495246633887291,
|
|
"num_tokens": 3646728.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.04389181137085,
|
|
"epoch": 1.512455516014235,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004984876286104392,
|
|
"loss": 5.6728,
|
|
"mean_token_accuracy": 0.15024697855114938,
|
|
"num_tokens": 3660021.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.029508495330811,
|
|
"epoch": 1.5183867141162515,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004984319937439693,
|
|
"loss": 5.6785,
|
|
"mean_token_accuracy": 0.14460248425602912,
|
|
"num_tokens": 3675487.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 5.980725574493408,
|
|
"epoch": 1.524317912218268,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0004983753575476969,
|
|
"loss": 5.652,
|
|
"mean_token_accuracy": 0.14840786904096603,
|
|
"num_tokens": 3689095.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 5.969368553161621,
|
|
"epoch": 1.5302491103202847,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004983177202754421,
|
|
"loss": 5.5954,
|
|
"mean_token_accuracy": 0.15209269672632217,
|
|
"num_tokens": 3702737.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 5.925890970230102,
|
|
"epoch": 1.5361803084223014,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004982590821855115,
|
|
"loss": 5.7067,
|
|
"mean_token_accuracy": 0.14797620624303817,
|
|
"num_tokens": 3717844.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 5.99383578300476,
|
|
"epoch": 1.5421115065243178,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004981994435406965,
|
|
"loss": 5.6651,
|
|
"mean_token_accuracy": 0.1470349095761776,
|
|
"num_tokens": 3733228.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 5.96784405708313,
|
|
"epoch": 1.5480427046263345,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004981388046082731,
|
|
"loss": 5.6357,
|
|
"mean_token_accuracy": 0.14797436743974685,
|
|
"num_tokens": 3748297.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 5.904844045639038,
|
|
"epoch": 1.5539739027283512,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004980771656599998,
|
|
"loss": 5.5686,
|
|
"mean_token_accuracy": 0.1505537673830986,
|
|
"num_tokens": 3762530.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 5.974849843978882,
|
|
"epoch": 1.5599051008303677,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004980145269721168,
|
|
"loss": 5.7,
|
|
"mean_token_accuracy": 0.14675320386886598,
|
|
"num_tokens": 3776113.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.0750159740448,
|
|
"epoch": 1.5658362989323842,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000497950888825345,
|
|
"loss": 5.7374,
|
|
"mean_token_accuracy": 0.14834038615226747,
|
|
"num_tokens": 3790888.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 5.993779802322388,
|
|
"epoch": 1.571767497034401,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000497886251504884,
|
|
"loss": 5.6816,
|
|
"mean_token_accuracy": 0.14576735198497773,
|
|
"num_tokens": 3803684.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.01502423286438,
|
|
"epoch": 1.5776986951364176,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004978206153004117,
|
|
"loss": 5.6725,
|
|
"mean_token_accuracy": 0.151952113956213,
|
|
"num_tokens": 3817418.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 5.989662456512451,
|
|
"epoch": 1.583629893238434,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004977539805060824,
|
|
"loss": 5.6764,
|
|
"mean_token_accuracy": 0.14674202278256415,
|
|
"num_tokens": 3831131.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 5.9614674091339115,
|
|
"epoch": 1.5895610913404508,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004976863474205257,
|
|
"loss": 5.602,
|
|
"mean_token_accuracy": 0.15287365317344664,
|
|
"num_tokens": 3843149.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.027898216247559,
|
|
"epoch": 1.5954922894424675,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004976177163468451,
|
|
"loss": 5.7621,
|
|
"mean_token_accuracy": 0.14517709985375404,
|
|
"num_tokens": 3857920.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 5.954496240615844,
|
|
"epoch": 1.601423487544484,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004975480875926168,
|
|
"loss": 5.5358,
|
|
"mean_token_accuracy": 0.1539545789361,
|
|
"num_tokens": 3871233.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 5.976214933395386,
|
|
"epoch": 1.6073546856465006,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000497477461469888,
|
|
"loss": 5.6625,
|
|
"mean_token_accuracy": 0.14626852720975875,
|
|
"num_tokens": 3886211.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 5.968742895126343,
|
|
"epoch": 1.6132858837485173,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004974058382951757,
|
|
"loss": 5.6239,
|
|
"mean_token_accuracy": 0.14970412850379944,
|
|
"num_tokens": 3901432.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 5.925413274765015,
|
|
"epoch": 1.6192170818505338,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004973332183894653,
|
|
"loss": 5.6312,
|
|
"mean_token_accuracy": 0.15616006404161453,
|
|
"num_tokens": 3915871.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 5.889155912399292,
|
|
"epoch": 1.6251482799525503,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004972596020782096,
|
|
"loss": 5.5756,
|
|
"mean_token_accuracy": 0.15282159447669982,
|
|
"num_tokens": 3929176.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 5.917853927612304,
|
|
"epoch": 1.631079478054567,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004971849896913259,
|
|
"loss": 5.5837,
|
|
"mean_token_accuracy": 0.1516695201396942,
|
|
"num_tokens": 3943346.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 5.8619325160980225,
|
|
"epoch": 1.6370106761565837,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004971093815631965,
|
|
"loss": 5.5182,
|
|
"mean_token_accuracy": 0.15619884431362152,
|
|
"num_tokens": 3958029.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 5.9396899223327635,
|
|
"epoch": 1.6429418742586002,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004970327780326655,
|
|
"loss": 5.6141,
|
|
"mean_token_accuracy": 0.14542030096054076,
|
|
"num_tokens": 3973165.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 5.9169244289398195,
|
|
"epoch": 1.6488730723606169,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004969551794430383,
|
|
"loss": 5.5832,
|
|
"mean_token_accuracy": 0.15040257573127747,
|
|
"num_tokens": 3988141.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 5.900255250930786,
|
|
"epoch": 1.6548042704626336,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004968765861420799,
|
|
"loss": 5.5859,
|
|
"mean_token_accuracy": 0.15164702236652375,
|
|
"num_tokens": 4003034.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 5.976015520095825,
|
|
"epoch": 1.66073546856465,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004967969984820128,
|
|
"loss": 5.6204,
|
|
"mean_token_accuracy": 0.14684486538171768,
|
|
"num_tokens": 4016748.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 5.931428146362305,
|
|
"epoch": 1.6666666666666665,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004967164168195159,
|
|
"loss": 5.6268,
|
|
"mean_token_accuracy": 0.15338827520608903,
|
|
"num_tokens": 4030731.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 5.901250648498535,
|
|
"epoch": 1.6725978647686834,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000496634841515723,
|
|
"loss": 5.697,
|
|
"mean_token_accuracy": 0.1430331736803055,
|
|
"num_tokens": 4045315.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 5.882366561889649,
|
|
"epoch": 1.6785290628707,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004965522729362209,
|
|
"loss": 5.5977,
|
|
"mean_token_accuracy": 0.1583336055278778,
|
|
"num_tokens": 4060884.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 5.909760904312134,
|
|
"epoch": 1.6844602609727164,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004964687114510478,
|
|
"loss": 5.636,
|
|
"mean_token_accuracy": 0.15020709335803986,
|
|
"num_tokens": 4075582.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 5.906455612182617,
|
|
"epoch": 1.690391459074733,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004963841574346917,
|
|
"loss": 5.5648,
|
|
"mean_token_accuracy": 0.15351162999868392,
|
|
"num_tokens": 4091006.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 5.905172061920166,
|
|
"epoch": 1.6963226571767498,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004962986112660887,
|
|
"loss": 5.5828,
|
|
"mean_token_accuracy": 0.15652085244655609,
|
|
"num_tokens": 4104182.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 5.782209491729736,
|
|
"epoch": 1.7022538552787663,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004962120733286215,
|
|
"loss": 5.5387,
|
|
"mean_token_accuracy": 0.1527836561203003,
|
|
"num_tokens": 4118406.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.0054114818572994,
|
|
"epoch": 1.708185053380783,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004961245440101171,
|
|
"loss": 5.6006,
|
|
"mean_token_accuracy": 0.15566613972187043,
|
|
"num_tokens": 4132105.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 5.880414438247681,
|
|
"epoch": 1.7141162514827997,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000496036023702846,
|
|
"loss": 5.5791,
|
|
"mean_token_accuracy": 0.14796797037124634,
|
|
"num_tokens": 4148165.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 5.833616399765015,
|
|
"epoch": 1.7200474495848161,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004959465128035195,
|
|
"loss": 5.5562,
|
|
"mean_token_accuracy": 0.15645333975553513,
|
|
"num_tokens": 4163611.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 5.837632656097412,
|
|
"epoch": 1.7259786476868326,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004958560117132884,
|
|
"loss": 5.4278,
|
|
"mean_token_accuracy": 0.16689967811107637,
|
|
"num_tokens": 4177140.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 5.825499486923218,
|
|
"epoch": 1.7319098457888493,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004957645208377411,
|
|
"loss": 5.5569,
|
|
"mean_token_accuracy": 0.1538987524807453,
|
|
"num_tokens": 4189531.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 5.915851497650147,
|
|
"epoch": 1.737841043890866,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000495672040586902,
|
|
"loss": 5.588,
|
|
"mean_token_accuracy": 0.15057967826724053,
|
|
"num_tokens": 4203583.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 5.873620128631591,
|
|
"epoch": 1.7437722419928825,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004955785713752292,
|
|
"loss": 5.5409,
|
|
"mean_token_accuracy": 0.15645942687988282,
|
|
"num_tokens": 4216568.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 5.826131200790405,
|
|
"epoch": 1.7497034400948992,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004954841136216132,
|
|
"loss": 5.5305,
|
|
"mean_token_accuracy": 0.15381396561861038,
|
|
"num_tokens": 4230107.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 5.878738212585449,
|
|
"epoch": 1.755634638196916,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004953886677493746,
|
|
"loss": 5.5768,
|
|
"mean_token_accuracy": 0.15454063564538956,
|
|
"num_tokens": 4242695.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 5.887821245193481,
|
|
"epoch": 1.7615658362989324,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004952922341862622,
|
|
"loss": 5.6451,
|
|
"mean_token_accuracy": 0.14898051470518112,
|
|
"num_tokens": 4257509.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 5.973342370986939,
|
|
"epoch": 1.7674970344009489,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004951948133644514,
|
|
"loss": 5.6117,
|
|
"mean_token_accuracy": 0.1451162502169609,
|
|
"num_tokens": 4272888.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 5.800426387786866,
|
|
"epoch": 1.7734282325029656,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004950964057205421,
|
|
"loss": 5.5295,
|
|
"mean_token_accuracy": 0.15365246683359146,
|
|
"num_tokens": 4286452.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 5.801223611831665,
|
|
"epoch": 1.7793594306049823,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004949970116955566,
|
|
"loss": 5.5372,
|
|
"mean_token_accuracy": 0.16073279082775116,
|
|
"num_tokens": 4301928.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.7793594306049823,
|
|
"eval_entropy": 5.786939268463229,
|
|
"eval_loss": 5.7374067306518555,
|
|
"eval_mean_token_accuracy": 0.15165954421451486,
|
|
"eval_num_tokens": 4301928.0,
|
|
"eval_runtime": 1.8412,
|
|
"eval_samples_per_second": 1416.448,
|
|
"eval_steps_per_second": 177.056,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 5.9356677532196045,
|
|
"epoch": 1.7852906287069987,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004948966317349379,
|
|
"loss": 5.551,
|
|
"mean_token_accuracy": 0.14964706301689149,
|
|
"num_tokens": 4315416.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 5.806493616104126,
|
|
"epoch": 1.7912218268090154,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004947952662885473,
|
|
"loss": 5.5623,
|
|
"mean_token_accuracy": 0.15524133592844008,
|
|
"num_tokens": 4329666.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 5.863862752914429,
|
|
"epoch": 1.7971530249110321,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000494692915810663,
|
|
"loss": 5.6011,
|
|
"mean_token_accuracy": 0.15324967205524445,
|
|
"num_tokens": 4344934.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 5.8631922721862795,
|
|
"epoch": 1.8030842230130486,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004945895807599775,
|
|
"loss": 5.5668,
|
|
"mean_token_accuracy": 0.15294425189495087,
|
|
"num_tokens": 4358753.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 5.891944885253906,
|
|
"epoch": 1.809015421115065,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004944852615995956,
|
|
"loss": 5.559,
|
|
"mean_token_accuracy": 0.15507696568965912,
|
|
"num_tokens": 4372831.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 5.830299997329712,
|
|
"epoch": 1.814946619217082,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004943799587970328,
|
|
"loss": 5.5527,
|
|
"mean_token_accuracy": 0.15730138272047042,
|
|
"num_tokens": 4385491.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 5.871082401275634,
|
|
"epoch": 1.8208778173190985,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004942736728242127,
|
|
"loss": 5.5607,
|
|
"mean_token_accuracy": 0.14975678473711013,
|
|
"num_tokens": 4398847.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 5.822240734100342,
|
|
"epoch": 1.826809015421115,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004941664041574651,
|
|
"loss": 5.5539,
|
|
"mean_token_accuracy": 0.15635693669319153,
|
|
"num_tokens": 4413643.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 5.8459906578063965,
|
|
"epoch": 1.8327402135231317,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004940581532775237,
|
|
"loss": 5.5423,
|
|
"mean_token_accuracy": 0.15766398310661317,
|
|
"num_tokens": 4429450.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 5.824446487426758,
|
|
"epoch": 1.8386714116251484,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004939489206695241,
|
|
"loss": 5.502,
|
|
"mean_token_accuracy": 0.15289542078971863,
|
|
"num_tokens": 4446345.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 5.775029802322388,
|
|
"epoch": 1.8446026097271648,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004938387068230021,
|
|
"loss": 5.5173,
|
|
"mean_token_accuracy": 0.16020766794681549,
|
|
"num_tokens": 4463114.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 5.879529714584351,
|
|
"epoch": 1.8505338078291815,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004937275122318902,
|
|
"loss": 5.5359,
|
|
"mean_token_accuracy": 0.15493431687355042,
|
|
"num_tokens": 4477182.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 5.83151159286499,
|
|
"epoch": 1.8564650059311982,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004936153373945165,
|
|
"loss": 5.5607,
|
|
"mean_token_accuracy": 0.15771184861660004,
|
|
"num_tokens": 4492150.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 5.786841678619385,
|
|
"epoch": 1.8623962040332147,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004935021828136025,
|
|
"loss": 5.4989,
|
|
"mean_token_accuracy": 0.16028912961483002,
|
|
"num_tokens": 4505554.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 5.8885918140411375,
|
|
"epoch": 1.8683274021352312,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004933880489962602,
|
|
"loss": 5.5574,
|
|
"mean_token_accuracy": 0.15621060729026795,
|
|
"num_tokens": 4520531.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 5.77139630317688,
|
|
"epoch": 1.8742586002372479,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004932729364539901,
|
|
"loss": 5.5399,
|
|
"mean_token_accuracy": 0.15808913707733155,
|
|
"num_tokens": 4535275.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 5.834245920181274,
|
|
"epoch": 1.8801897983392646,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004931568457026789,
|
|
"loss": 5.5046,
|
|
"mean_token_accuracy": 0.16141868382692337,
|
|
"num_tokens": 4549760.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 5.812886381149292,
|
|
"epoch": 1.886120996441281,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004930397772625976,
|
|
"loss": 5.4636,
|
|
"mean_token_accuracy": 0.1592896267771721,
|
|
"num_tokens": 4565285.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 5.776328659057617,
|
|
"epoch": 1.8920521945432978,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004929217316583985,
|
|
"loss": 5.4441,
|
|
"mean_token_accuracy": 0.15965035259723664,
|
|
"num_tokens": 4581072.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 5.7703155994415285,
|
|
"epoch": 1.8979833926453145,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000492802709419113,
|
|
"loss": 5.4619,
|
|
"mean_token_accuracy": 0.15460458248853684,
|
|
"num_tokens": 4594351.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 5.757806015014649,
|
|
"epoch": 1.903914590747331,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004926827110781496,
|
|
"loss": 5.5066,
|
|
"mean_token_accuracy": 0.15677811503410338,
|
|
"num_tokens": 4608851.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 5.895300960540771,
|
|
"epoch": 1.9098457888493474,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004925617371732914,
|
|
"loss": 5.5946,
|
|
"mean_token_accuracy": 0.15780434310436248,
|
|
"num_tokens": 4622809.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 5.799103403091431,
|
|
"epoch": 1.9157769869513643,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004924397882466933,
|
|
"loss": 5.572,
|
|
"mean_token_accuracy": 0.15562279522418976,
|
|
"num_tokens": 4638205.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 5.818878030776977,
|
|
"epoch": 1.9217081850533808,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004923168648448799,
|
|
"loss": 5.5022,
|
|
"mean_token_accuracy": 0.16238696724176407,
|
|
"num_tokens": 4652847.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 5.816390609741211,
|
|
"epoch": 1.9276393831553973,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004921929675187433,
|
|
"loss": 5.5572,
|
|
"mean_token_accuracy": 0.15800073593854905,
|
|
"num_tokens": 4665875.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 5.856627559661865,
|
|
"epoch": 1.933570581257414,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004920680968235397,
|
|
"loss": 5.578,
|
|
"mean_token_accuracy": 0.15377815514802934,
|
|
"num_tokens": 4682604.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 5.845769596099854,
|
|
"epoch": 1.9395017793594307,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004919422533188882,
|
|
"loss": 5.5417,
|
|
"mean_token_accuracy": 0.15837402045726776,
|
|
"num_tokens": 4696070.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 5.896516275405884,
|
|
"epoch": 1.9454329774614472,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004918154375687673,
|
|
"loss": 5.6164,
|
|
"mean_token_accuracy": 0.15318400859832765,
|
|
"num_tokens": 4710692.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 5.816887331008911,
|
|
"epoch": 1.9513641755634639,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004916876501415129,
|
|
"loss": 5.4553,
|
|
"mean_token_accuracy": 0.16236065775156022,
|
|
"num_tokens": 4725830.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 5.770136833190918,
|
|
"epoch": 1.9572953736654806,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004915588916098151,
|
|
"loss": 5.5446,
|
|
"mean_token_accuracy": 0.154208692163229,
|
|
"num_tokens": 4739167.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 5.781821155548096,
|
|
"epoch": 1.963226571767497,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004914291625507164,
|
|
"loss": 5.446,
|
|
"mean_token_accuracy": 0.16111351996660234,
|
|
"num_tokens": 4753588.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 5.730515575408935,
|
|
"epoch": 1.9691577698695135,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000491298463545609,
|
|
"loss": 5.3918,
|
|
"mean_token_accuracy": 0.16754134446382524,
|
|
"num_tokens": 4764757.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 5.805826234817505,
|
|
"epoch": 1.9750889679715302,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004911667951802318,
|
|
"loss": 5.4842,
|
|
"mean_token_accuracy": 0.15701884925365447,
|
|
"num_tokens": 4779880.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 5.751394748687744,
|
|
"epoch": 1.981020166073547,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000491034158044668,
|
|
"loss": 5.533,
|
|
"mean_token_accuracy": 0.16572649627923966,
|
|
"num_tokens": 4792126.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 5.931813764572143,
|
|
"epoch": 1.9869513641755634,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004909005527333421,
|
|
"loss": 5.5678,
|
|
"mean_token_accuracy": 0.153676076233387,
|
|
"num_tokens": 4806147.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 5.865698957443238,
|
|
"epoch": 1.99288256227758,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000490765979845018,
|
|
"loss": 5.5725,
|
|
"mean_token_accuracy": 0.1579641193151474,
|
|
"num_tokens": 4822151.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 5.702477741241455,
|
|
"epoch": 1.9988137603795968,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000490630439982796,
|
|
"loss": 5.5069,
|
|
"mean_token_accuracy": 0.16614492237567902,
|
|
"num_tokens": 4835952.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 5.866048336029053,
|
|
"epoch": 2.0047449584816133,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004904939337541093,
|
|
"loss": 5.3389,
|
|
"mean_token_accuracy": 0.1676741734147072,
|
|
"num_tokens": 4849830.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 5.736663961410523,
|
|
"epoch": 2.0106761565836297,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004903564617707227,
|
|
"loss": 5.2251,
|
|
"mean_token_accuracy": 0.1749313712120056,
|
|
"num_tokens": 4863187.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 5.674867725372314,
|
|
"epoch": 2.0166073546856467,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004902180246487285,
|
|
"loss": 5.2664,
|
|
"mean_token_accuracy": 0.17031993269920348,
|
|
"num_tokens": 4878006.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 5.729181289672852,
|
|
"epoch": 2.022538552787663,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000490078623008545,
|
|
"loss": 5.1827,
|
|
"mean_token_accuracy": 0.17386887222528458,
|
|
"num_tokens": 4890624.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 5.630424547195434,
|
|
"epoch": 2.0284697508896796,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004899382574749124,
|
|
"loss": 5.226,
|
|
"mean_token_accuracy": 0.16898248344659805,
|
|
"num_tokens": 4904318.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 5.700745296478272,
|
|
"epoch": 2.0344009489916965,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000489796928676891,
|
|
"loss": 5.2258,
|
|
"mean_token_accuracy": 0.17715904116630554,
|
|
"num_tokens": 4920075.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 5.718631172180176,
|
|
"epoch": 2.040332147093713,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004896546372478581,
|
|
"loss": 5.2822,
|
|
"mean_token_accuracy": 0.17024712413549423,
|
|
"num_tokens": 4936488.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 5.720815515518188,
|
|
"epoch": 2.0462633451957295,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004895113838255049,
|
|
"loss": 5.214,
|
|
"mean_token_accuracy": 0.1729263797402382,
|
|
"num_tokens": 4952192.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 5.70762071609497,
|
|
"epoch": 2.052194543297746,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004893671690518343,
|
|
"loss": 5.3076,
|
|
"mean_token_accuracy": 0.1653238743543625,
|
|
"num_tokens": 4967736.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 5.761155271530152,
|
|
"epoch": 2.058125741399763,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004892219935731571,
|
|
"loss": 5.2909,
|
|
"mean_token_accuracy": 0.17624197006225586,
|
|
"num_tokens": 4980754.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 5.681296873092651,
|
|
"epoch": 2.0640569395017794,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004890758580400898,
|
|
"loss": 5.2314,
|
|
"mean_token_accuracy": 0.17614012509584426,
|
|
"num_tokens": 4995272.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 5.665003442764283,
|
|
"epoch": 2.069988137603796,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004889287631075513,
|
|
"loss": 5.2742,
|
|
"mean_token_accuracy": 0.17507000565528869,
|
|
"num_tokens": 5009149.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 5.748960971832275,
|
|
"epoch": 2.0759193357058128,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004887807094347606,
|
|
"loss": 5.3056,
|
|
"mean_token_accuracy": 0.16657923460006713,
|
|
"num_tokens": 5024106.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 5.673666334152221,
|
|
"epoch": 2.0818505338078293,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004886316976852326,
|
|
"loss": 5.2585,
|
|
"mean_token_accuracy": 0.1768916815519333,
|
|
"num_tokens": 5036956.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 5.692249059677124,
|
|
"epoch": 2.0877817319098457,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004884817285267766,
|
|
"loss": 5.2678,
|
|
"mean_token_accuracy": 0.1770847797393799,
|
|
"num_tokens": 5051439.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.657824516296387,
|
|
"epoch": 2.093712930011862,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004883308026314923,
|
|
"loss": 5.2389,
|
|
"mean_token_accuracy": 0.17888718843460083,
|
|
"num_tokens": 5066073.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 5.700278234481812,
|
|
"epoch": 2.099644128113879,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00048817892067576726,
|
|
"loss": 5.1927,
|
|
"mean_token_accuracy": 0.17897644639015198,
|
|
"num_tokens": 5078573.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 5.616072320938111,
|
|
"epoch": 2.1055753262158956,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004880260833402734,
|
|
"loss": 5.2432,
|
|
"mean_token_accuracy": 0.17429422289133073,
|
|
"num_tokens": 5092780.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 5.628483629226684,
|
|
"epoch": 2.111506524317912,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004878722913099645,
|
|
"loss": 5.1356,
|
|
"mean_token_accuracy": 0.17683007568120956,
|
|
"num_tokens": 5107199.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 5.603928327560425,
|
|
"epoch": 2.117437722419929,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000487717545274073,
|
|
"loss": 5.1822,
|
|
"mean_token_accuracy": 0.1796892687678337,
|
|
"num_tokens": 5121758.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.623524761199951,
|
|
"epoch": 2.1233689205219455,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004875618459261065,
|
|
"loss": 5.2684,
|
|
"mean_token_accuracy": 0.17580897361040115,
|
|
"num_tokens": 5136882.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 5.693287706375122,
|
|
"epoch": 2.129300118623962,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00048740519396384515,
|
|
"loss": 5.245,
|
|
"mean_token_accuracy": 0.17159996926784515,
|
|
"num_tokens": 5154795.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 5.703963279724121,
|
|
"epoch": 2.135231316725979,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004872475900893382,
|
|
"loss": 5.2809,
|
|
"mean_token_accuracy": 0.17368003576993943,
|
|
"num_tokens": 5168979.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.687782526016235,
|
|
"epoch": 2.1411625148279954,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004870890350089011,
|
|
"loss": 5.2747,
|
|
"mean_token_accuracy": 0.17227655202150344,
|
|
"num_tokens": 5183502.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.58332462310791,
|
|
"epoch": 2.147093712930012,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00048692952943311214,
|
|
"loss": 5.192,
|
|
"mean_token_accuracy": 0.17441501021385192,
|
|
"num_tokens": 5198600.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.6405181884765625,
|
|
"epoch": 2.1530249110320283,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004867690740768094,
|
|
"loss": 5.1461,
|
|
"mean_token_accuracy": 0.18086205720901488,
|
|
"num_tokens": 5212189.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 5.66403341293335,
|
|
"epoch": 2.1589561091340452,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00048660766965908736,
|
|
"loss": 5.3672,
|
|
"mean_token_accuracy": 0.16767664402723312,
|
|
"num_tokens": 5226666.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.637210655212402,
|
|
"epoch": 2.1648873072360617,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000486445316903294,
|
|
"loss": 5.2065,
|
|
"mean_token_accuracy": 0.1795545995235443,
|
|
"num_tokens": 5241592.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.617842102050782,
|
|
"epoch": 2.170818505338078,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000486282016537027,
|
|
"loss": 5.2765,
|
|
"mean_token_accuracy": 0.1726476162672043,
|
|
"num_tokens": 5257949.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.698722076416016,
|
|
"epoch": 2.176749703440095,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004861177692921313,
|
|
"loss": 5.2722,
|
|
"mean_token_accuracy": 0.16946842521429062,
|
|
"num_tokens": 5271545.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.653340530395508,
|
|
"epoch": 2.1826809015421116,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000485952575904695,
|
|
"loss": 5.2879,
|
|
"mean_token_accuracy": 0.17474988102912903,
|
|
"num_tokens": 5285615.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 5.669216823577881,
|
|
"epoch": 2.188612099644128,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004857864371150467,
|
|
"loss": 5.285,
|
|
"mean_token_accuracy": 0.1740844339132309,
|
|
"num_tokens": 5298755.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 5.645978355407715,
|
|
"epoch": 2.1945432977461445,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004856193536677515,
|
|
"loss": 5.2386,
|
|
"mean_token_accuracy": 0.18111101388931275,
|
|
"num_tokens": 5311225.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.6569397926330565,
|
|
"epoch": 2.2004744958481615,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048545132631160856,
|
|
"loss": 5.2583,
|
|
"mean_token_accuracy": 0.17769812494516374,
|
|
"num_tokens": 5326020.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.6589381217956545,
|
|
"epoch": 2.206405693950178,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000485282355799647,
|
|
"loss": 5.2692,
|
|
"mean_token_accuracy": 0.17553827911615372,
|
|
"num_tokens": 5339280.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.615429306030274,
|
|
"epoch": 2.2123368920521944,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004851124428891226,
|
|
"loss": 5.2616,
|
|
"mean_token_accuracy": 0.17267712354660034,
|
|
"num_tokens": 5354606.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 5.64393458366394,
|
|
"epoch": 2.2182680901542113,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048494158834151485,
|
|
"loss": 5.2569,
|
|
"mean_token_accuracy": 0.17619943767786025,
|
|
"num_tokens": 5371217.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.572863245010376,
|
|
"epoch": 2.224199288256228,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00048476979292252323,
|
|
"loss": 5.1907,
|
|
"mean_token_accuracy": 0.1723210945725441,
|
|
"num_tokens": 5385040.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.686497974395752,
|
|
"epoch": 2.2301304863582443,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004845970574020637,
|
|
"loss": 5.1913,
|
|
"mean_token_accuracy": 0.17279084026813507,
|
|
"num_tokens": 5398418.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.549385643005371,
|
|
"epoch": 2.2360616844602608,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048442338255426555,
|
|
"loss": 5.2355,
|
|
"mean_token_accuracy": 0.17002784162759782,
|
|
"num_tokens": 5413517.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.66954607963562,
|
|
"epoch": 2.2419928825622777,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004842487691574675,
|
|
"loss": 5.2031,
|
|
"mean_token_accuracy": 0.17829595357179642,
|
|
"num_tokens": 5428663.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.559753322601319,
|
|
"epoch": 2.247924080664294,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004840732179942146,
|
|
"loss": 5.2539,
|
|
"mean_token_accuracy": 0.17741060256958008,
|
|
"num_tokens": 5441903.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.6504683017730715,
|
|
"epoch": 2.2538552787663106,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048389672985125467,
|
|
"loss": 5.2436,
|
|
"mean_token_accuracy": 0.1759451746940613,
|
|
"num_tokens": 5458824.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.548000478744507,
|
|
"epoch": 2.2597864768683276,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004837193055195344,
|
|
"loss": 5.206,
|
|
"mean_token_accuracy": 0.17390945851802825,
|
|
"num_tokens": 5474174.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.6555421352386475,
|
|
"epoch": 2.265717674970344,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00048354094579419656,
|
|
"loss": 5.1861,
|
|
"mean_token_accuracy": 0.17711658626794816,
|
|
"num_tokens": 5487544.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.5846880912780765,
|
|
"epoch": 2.2716488730723605,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00048336165147457537,
|
|
"loss": 5.1992,
|
|
"mean_token_accuracy": 0.17726835906505584,
|
|
"num_tokens": 5503297.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.586446714401245,
|
|
"epoch": 2.277580071174377,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00048318142336419413,
|
|
"loss": 5.1867,
|
|
"mean_token_accuracy": 0.17828324437141418,
|
|
"num_tokens": 5517861.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.553643751144409,
|
|
"epoch": 2.283511269276394,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004830002622707604,
|
|
"loss": 5.1966,
|
|
"mean_token_accuracy": 0.1780408576130867,
|
|
"num_tokens": 5531530.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.545027494430542,
|
|
"epoch": 2.2894424673784104,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048281816900616354,
|
|
"loss": 5.1195,
|
|
"mean_token_accuracy": 0.18244230896234512,
|
|
"num_tokens": 5544744.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.594576406478882,
|
|
"epoch": 2.295373665480427,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00048263514438647015,
|
|
"loss": 5.2177,
|
|
"mean_token_accuracy": 0.17840056568384172,
|
|
"num_tokens": 5556810.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.572685575485229,
|
|
"epoch": 2.301304863582444,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004824511892319209,
|
|
"loss": 5.1837,
|
|
"mean_token_accuracy": 0.18053331077098847,
|
|
"num_tokens": 5570597.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.609456014633179,
|
|
"epoch": 2.3072360616844603,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004822663043669267,
|
|
"loss": 5.2089,
|
|
"mean_token_accuracy": 0.1729913905262947,
|
|
"num_tokens": 5583685.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.6058872699737545,
|
|
"epoch": 2.3131672597864767,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048208049062006505,
|
|
"loss": 5.2755,
|
|
"mean_token_accuracy": 0.17355445474386216,
|
|
"num_tokens": 5598597.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.604036808013916,
|
|
"epoch": 2.3190984578884937,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004818937488240764,
|
|
"loss": 5.2397,
|
|
"mean_token_accuracy": 0.1725240781903267,
|
|
"num_tokens": 5613274.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.4700079441070555,
|
|
"epoch": 2.32502965599051,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048170607981586023,
|
|
"loss": 5.2301,
|
|
"mean_token_accuracy": 0.17317135483026505,
|
|
"num_tokens": 5626253.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.650290155410767,
|
|
"epoch": 2.3309608540925266,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004815174844364714,
|
|
"loss": 5.2502,
|
|
"mean_token_accuracy": 0.17107724398374557,
|
|
"num_tokens": 5639200.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.535002565383911,
|
|
"epoch": 2.3368920521945435,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004813279635311165,
|
|
"loss": 5.2551,
|
|
"mean_token_accuracy": 0.16849012970924376,
|
|
"num_tokens": 5653603.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.611714267730713,
|
|
"epoch": 2.34282325029656,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00048113751794914974,
|
|
"loss": 5.1286,
|
|
"mean_token_accuracy": 0.18730779737234116,
|
|
"num_tokens": 5666083.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.55033802986145,
|
|
"epoch": 2.3487544483985765,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004809461485440694,
|
|
"loss": 5.2675,
|
|
"mean_token_accuracy": 0.1771186113357544,
|
|
"num_tokens": 5681369.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.6295370101928714,
|
|
"epoch": 2.354685646500593,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004807538561735143,
|
|
"loss": 5.193,
|
|
"mean_token_accuracy": 0.17823005467653275,
|
|
"num_tokens": 5696620.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.624615335464478,
|
|
"epoch": 2.36061684460261,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00048056064169925904,
|
|
"loss": 5.259,
|
|
"mean_token_accuracy": 0.16968683004379273,
|
|
"num_tokens": 5710022.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.682260417938233,
|
|
"epoch": 2.3665480427046264,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048036650598721125,
|
|
"loss": 5.2462,
|
|
"mean_token_accuracy": 0.1781255468726158,
|
|
"num_tokens": 5724503.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.542639303207397,
|
|
"epoch": 2.372479240806643,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00048017144990740673,
|
|
"loss": 5.2046,
|
|
"mean_token_accuracy": 0.17966311872005464,
|
|
"num_tokens": 5738365.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 2.372479240806643,
|
|
"eval_entropy": 5.439236042689692,
|
|
"eval_loss": 5.5663251876831055,
|
|
"eval_mean_token_accuracy": 0.16570363479432154,
|
|
"eval_num_tokens": 5738365.0,
|
|
"eval_runtime": 1.846,
|
|
"eval_samples_per_second": 1412.782,
|
|
"eval_steps_per_second": 176.598,
|
|
"step": 2000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 8420,
|
|
"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": 1491403969474560.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|