758236a625
Model: fpadovani/tur-latn-10mb-after-ppt-Dp-10mb-ckpt500_seed10 Source: Original Platform
10090 lines
274 KiB
JSON
10090 lines
274 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 4.970661362506216,
|
|
"eval_steps": 1000,
|
|
"global_step": 5000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 8.294325065612792,
|
|
"epoch": 0.004972650422675286,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 2e-06,
|
|
"loss": 7.9407,
|
|
"mean_token_accuracy": 0.08834519535303116,
|
|
"num_tokens": 9311.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 8.259415340423583,
|
|
"epoch": 0.009945300845350571,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 8.0084,
|
|
"mean_token_accuracy": 0.07713913097977639,
|
|
"num_tokens": 20562.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 8.223557472229004,
|
|
"epoch": 0.014917951268025857,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 7e-06,
|
|
"loss": 7.9707,
|
|
"mean_token_accuracy": 0.08674330189824105,
|
|
"num_tokens": 31490.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 8.246721458435058,
|
|
"epoch": 0.019890601690701143,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 7.89,
|
|
"mean_token_accuracy": 0.08834159150719642,
|
|
"num_tokens": 40741.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 8.267284965515136,
|
|
"epoch": 0.02486325211337643,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 7.865,
|
|
"mean_token_accuracy": 0.09002188965678215,
|
|
"num_tokens": 50749.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 8.258776950836182,
|
|
"epoch": 0.029835902536051714,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 7.9304,
|
|
"mean_token_accuracy": 0.09212828278541565,
|
|
"num_tokens": 60296.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 8.257113647460937,
|
|
"epoch": 0.034808552958727,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 7.9095,
|
|
"mean_token_accuracy": 0.09328181892633439,
|
|
"num_tokens": 70797.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 8.294310569763184,
|
|
"epoch": 0.039781203381402286,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 7.9458,
|
|
"mean_token_accuracy": 0.0856083333492279,
|
|
"num_tokens": 80815.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 8.263056659698487,
|
|
"epoch": 0.04475385380407757,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 7.897,
|
|
"mean_token_accuracy": 0.09171751812100411,
|
|
"num_tokens": 90326.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 8.305741024017333,
|
|
"epoch": 0.04972650422675286,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 7.9054,
|
|
"mean_token_accuracy": 0.08634501695632935,
|
|
"num_tokens": 99800.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 8.256485271453858,
|
|
"epoch": 0.05469915464942814,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 8.0127,
|
|
"mean_token_accuracy": 0.09032509922981262,
|
|
"num_tokens": 110636.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 8.283671665191651,
|
|
"epoch": 0.05967180507210343,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 7.918,
|
|
"mean_token_accuracy": 0.08834953606128693,
|
|
"num_tokens": 120269.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 8.314575004577637,
|
|
"epoch": 0.06464445549477872,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 7.9141,
|
|
"mean_token_accuracy": 0.08136967048048974,
|
|
"num_tokens": 130176.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 8.258645057678223,
|
|
"epoch": 0.069617105917454,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 7.8765,
|
|
"mean_token_accuracy": 0.0901374228298664,
|
|
"num_tokens": 140020.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 8.252040672302247,
|
|
"epoch": 0.07458975634012929,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 7.9541,
|
|
"mean_token_accuracy": 0.08882354721426963,
|
|
"num_tokens": 150713.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 8.258432102203368,
|
|
"epoch": 0.07956240676280457,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 7.8864,
|
|
"mean_token_accuracy": 0.08713632896542549,
|
|
"num_tokens": 159828.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 8.263443851470948,
|
|
"epoch": 0.08453505718547986,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 7.9056,
|
|
"mean_token_accuracy": 0.09245348498225212,
|
|
"num_tokens": 169909.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 8.316218376159668,
|
|
"epoch": 0.08950770760815514,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 7.9613,
|
|
"mean_token_accuracy": 0.08627018332481384,
|
|
"num_tokens": 179462.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 8.290662479400634,
|
|
"epoch": 0.09448035803083044,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 7.9602,
|
|
"mean_token_accuracy": 0.0799452781677246,
|
|
"num_tokens": 188965.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 8.284315204620361,
|
|
"epoch": 0.09945300845350571,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 7.9587,
|
|
"mean_token_accuracy": 0.0864613514393568,
|
|
"num_tokens": 200382.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 8.303834819793702,
|
|
"epoch": 0.10442565887618101,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 7.8477,
|
|
"mean_token_accuracy": 0.09324368312954903,
|
|
"num_tokens": 209718.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 8.257446384429931,
|
|
"epoch": 0.10939830929885629,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 7.8336,
|
|
"mean_token_accuracy": 0.09145947843790055,
|
|
"num_tokens": 219216.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 8.195090293884277,
|
|
"epoch": 0.11437095972153158,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 7.9026,
|
|
"mean_token_accuracy": 0.09406888335943223,
|
|
"num_tokens": 228219.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 8.23755178451538,
|
|
"epoch": 0.11934361014420686,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 7.8584,
|
|
"mean_token_accuracy": 0.0880578763782978,
|
|
"num_tokens": 237776.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 8.235713577270507,
|
|
"epoch": 0.12431626056688215,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 7.8502,
|
|
"mean_token_accuracy": 0.09442176297307014,
|
|
"num_tokens": 248411.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 8.223017120361328,
|
|
"epoch": 0.12928891098955744,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 7.9307,
|
|
"mean_token_accuracy": 0.08167004808783532,
|
|
"num_tokens": 258797.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 8.265431118011474,
|
|
"epoch": 0.13426156141223272,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.8642,
|
|
"mean_token_accuracy": 0.08920244127511978,
|
|
"num_tokens": 269228.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 8.231118106842041,
|
|
"epoch": 0.139234211834908,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.9405,
|
|
"mean_token_accuracy": 0.08824159651994705,
|
|
"num_tokens": 279191.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 8.255726528167724,
|
|
"epoch": 0.14420686225758328,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.8564,
|
|
"mean_token_accuracy": 0.08566267415881157,
|
|
"num_tokens": 289382.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.270699310302735,
|
|
"epoch": 0.14917951268025859,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.8058,
|
|
"mean_token_accuracy": 0.08951546400785446,
|
|
"num_tokens": 299163.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.2356463432312,
|
|
"epoch": 0.15415216310293386,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.8727,
|
|
"mean_token_accuracy": 0.08964284360408784,
|
|
"num_tokens": 308551.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.247602558135986,
|
|
"epoch": 0.15912481352560914,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.8447,
|
|
"mean_token_accuracy": 0.08380615264177323,
|
|
"num_tokens": 318427.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.207133197784424,
|
|
"epoch": 0.16409746394828442,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.9229,
|
|
"mean_token_accuracy": 0.08464949056506157,
|
|
"num_tokens": 327539.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.285111618041991,
|
|
"epoch": 0.16907011437095973,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.8549,
|
|
"mean_token_accuracy": 0.09108748883008957,
|
|
"num_tokens": 336676.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.27844171524048,
|
|
"epoch": 0.174042764793635,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.8547,
|
|
"mean_token_accuracy": 0.09992765560746193,
|
|
"num_tokens": 345914.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.213016605377197,
|
|
"epoch": 0.1790154152163103,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.8325,
|
|
"mean_token_accuracy": 0.0939701370894909,
|
|
"num_tokens": 355982.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.270163917541504,
|
|
"epoch": 0.1839880656389856,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.8643,
|
|
"mean_token_accuracy": 0.08404637724161149,
|
|
"num_tokens": 365398.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.256326866149902,
|
|
"epoch": 0.18896071606166087,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.863,
|
|
"mean_token_accuracy": 0.09230407699942589,
|
|
"num_tokens": 374519.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.19491024017334,
|
|
"epoch": 0.19393336648433615,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.8077,
|
|
"mean_token_accuracy": 0.09135130345821381,
|
|
"num_tokens": 383536.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.321134948730469,
|
|
"epoch": 0.19890601690701143,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.8841,
|
|
"mean_token_accuracy": 0.09590880051255227,
|
|
"num_tokens": 393703.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.27638511657715,
|
|
"epoch": 0.20387866732968674,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.9667,
|
|
"mean_token_accuracy": 0.08295246548950672,
|
|
"num_tokens": 404065.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.142377758026123,
|
|
"epoch": 0.20885131775236201,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.8811,
|
|
"mean_token_accuracy": 0.08889581225812435,
|
|
"num_tokens": 413869.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.355494785308839,
|
|
"epoch": 0.2138239681750373,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.9051,
|
|
"mean_token_accuracy": 0.0895668026059866,
|
|
"num_tokens": 424251.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.171804428100586,
|
|
"epoch": 0.21879661859771257,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.8335,
|
|
"mean_token_accuracy": 0.09142067879438401,
|
|
"num_tokens": 432530.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.221377086639404,
|
|
"epoch": 0.22376926902038788,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.8111,
|
|
"mean_token_accuracy": 0.09241886958479881,
|
|
"num_tokens": 442335.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.299753284454345,
|
|
"epoch": 0.22874191944306316,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.8422,
|
|
"mean_token_accuracy": 0.09034919887781143,
|
|
"num_tokens": 452048.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.125889539718628,
|
|
"epoch": 0.23371456986573844,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.7678,
|
|
"mean_token_accuracy": 0.09646162688732147,
|
|
"num_tokens": 460840.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.212695217132568,
|
|
"epoch": 0.23868722028841372,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.8627,
|
|
"mean_token_accuracy": 0.08924896270036697,
|
|
"num_tokens": 470198.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.287675857543945,
|
|
"epoch": 0.24365987071108902,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.8171,
|
|
"mean_token_accuracy": 0.0940744325518608,
|
|
"num_tokens": 481030.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.143581008911132,
|
|
"epoch": 0.2486325211337643,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.7735,
|
|
"mean_token_accuracy": 0.09191242530941963,
|
|
"num_tokens": 490411.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.310920333862304,
|
|
"epoch": 0.2536051715564396,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.8102,
|
|
"mean_token_accuracy": 0.09700686782598496,
|
|
"num_tokens": 499551.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.202103424072266,
|
|
"epoch": 0.2585778219791149,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.8364,
|
|
"mean_token_accuracy": 0.08541068732738495,
|
|
"num_tokens": 508885.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.195527076721191,
|
|
"epoch": 0.26355047240179014,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.8451,
|
|
"mean_token_accuracy": 0.08657787889242172,
|
|
"num_tokens": 518291.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.21373691558838,
|
|
"epoch": 0.26852312282446544,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.7504,
|
|
"mean_token_accuracy": 0.09659530520439148,
|
|
"num_tokens": 528035.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 8.114533519744873,
|
|
"epoch": 0.27349577324714075,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.8156,
|
|
"mean_token_accuracy": 0.09262970834970474,
|
|
"num_tokens": 538410.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.237305450439454,
|
|
"epoch": 0.278468423669816,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.7497,
|
|
"mean_token_accuracy": 0.09546326473355293,
|
|
"num_tokens": 549697.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 8.143435287475587,
|
|
"epoch": 0.2834410740924913,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.6651,
|
|
"mean_token_accuracy": 0.09623958095908165,
|
|
"num_tokens": 558932.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 8.087209558486938,
|
|
"epoch": 0.28841372451516656,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.7688,
|
|
"mean_token_accuracy": 0.08862602412700653,
|
|
"num_tokens": 568441.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 8.289060688018798,
|
|
"epoch": 0.29338637493784187,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.8673,
|
|
"mean_token_accuracy": 0.0926568903028965,
|
|
"num_tokens": 577861.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 8.105233335494995,
|
|
"epoch": 0.29835902536051717,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.8223,
|
|
"mean_token_accuracy": 0.08772151693701744,
|
|
"num_tokens": 587858.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 8.266038990020752,
|
|
"epoch": 0.3033316757831924,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.8405,
|
|
"mean_token_accuracy": 0.09540271908044815,
|
|
"num_tokens": 597806.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 8.187761306762695,
|
|
"epoch": 0.30830432620586773,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.7743,
|
|
"mean_token_accuracy": 0.10029298216104507,
|
|
"num_tokens": 607408.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 8.16814022064209,
|
|
"epoch": 0.31327697662854304,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.7147,
|
|
"mean_token_accuracy": 0.0964276671409607,
|
|
"num_tokens": 617862.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 8.192989444732666,
|
|
"epoch": 0.3182496270512183,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.7305,
|
|
"mean_token_accuracy": 0.09274147972464561,
|
|
"num_tokens": 627128.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 8.032805109024048,
|
|
"epoch": 0.3232222774738936,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.7453,
|
|
"mean_token_accuracy": 0.09363794773817062,
|
|
"num_tokens": 636775.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 8.435432052612304,
|
|
"epoch": 0.32819492789656884,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.7993,
|
|
"mean_token_accuracy": 0.08975734375417233,
|
|
"num_tokens": 645906.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 8.087645673751831,
|
|
"epoch": 0.33316757831924415,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.719,
|
|
"mean_token_accuracy": 0.09149112850427628,
|
|
"num_tokens": 655601.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 8.23168420791626,
|
|
"epoch": 0.33814022874191946,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.7826,
|
|
"mean_token_accuracy": 0.09446617141366005,
|
|
"num_tokens": 665682.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 8.065727710723877,
|
|
"epoch": 0.3431128791645947,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.7366,
|
|
"mean_token_accuracy": 0.09515505582094193,
|
|
"num_tokens": 675200.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 8.17827262878418,
|
|
"epoch": 0.34808552958727,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.7539,
|
|
"mean_token_accuracy": 0.09401869773864746,
|
|
"num_tokens": 685590.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 8.199443244934082,
|
|
"epoch": 0.3530581800099453,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.7327,
|
|
"mean_token_accuracy": 0.09991735145449639,
|
|
"num_tokens": 695665.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 8.18546781539917,
|
|
"epoch": 0.3580308304326206,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.7672,
|
|
"mean_token_accuracy": 0.09240800514817238,
|
|
"num_tokens": 705850.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 8.245514583587646,
|
|
"epoch": 0.3630034808552959,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.7806,
|
|
"mean_token_accuracy": 0.09033955484628678,
|
|
"num_tokens": 716010.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 8.129885625839233,
|
|
"epoch": 0.3679761312779712,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.7279,
|
|
"mean_token_accuracy": 0.09400108084082603,
|
|
"num_tokens": 724908.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 8.065651798248291,
|
|
"epoch": 0.37294878170064644,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.6965,
|
|
"mean_token_accuracy": 0.09816657826304435,
|
|
"num_tokens": 733479.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 8.14595718383789,
|
|
"epoch": 0.37792143212332174,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.7217,
|
|
"mean_token_accuracy": 0.08874044567346573,
|
|
"num_tokens": 743365.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 8.16227297782898,
|
|
"epoch": 0.382894082545997,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.7618,
|
|
"mean_token_accuracy": 0.09624904319643975,
|
|
"num_tokens": 753720.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 8.151291179656983,
|
|
"epoch": 0.3878667329686723,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.6438,
|
|
"mean_token_accuracy": 0.09636546224355698,
|
|
"num_tokens": 762524.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 8.13357491493225,
|
|
"epoch": 0.3928393833913476,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.7387,
|
|
"mean_token_accuracy": 0.09395921602845192,
|
|
"num_tokens": 772549.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 8.074169540405274,
|
|
"epoch": 0.39781203381402286,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.7146,
|
|
"mean_token_accuracy": 0.09670355543494225,
|
|
"num_tokens": 782304.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 8.18697247505188,
|
|
"epoch": 0.40278468423669817,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.8206,
|
|
"mean_token_accuracy": 0.0916778601706028,
|
|
"num_tokens": 792010.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 8.05876727104187,
|
|
"epoch": 0.40775733465937347,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.6559,
|
|
"mean_token_accuracy": 0.10227638110518456,
|
|
"num_tokens": 801667.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 8.145513725280761,
|
|
"epoch": 0.4127299850820487,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.7361,
|
|
"mean_token_accuracy": 0.09709529057145119,
|
|
"num_tokens": 812385.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 8.071045780181885,
|
|
"epoch": 0.41770263550472403,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.6508,
|
|
"mean_token_accuracy": 0.10053839311003684,
|
|
"num_tokens": 822177.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 8.176684093475341,
|
|
"epoch": 0.4226752859273993,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.7252,
|
|
"mean_token_accuracy": 0.09335672929883003,
|
|
"num_tokens": 832576.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.983503007888794,
|
|
"epoch": 0.4276479363500746,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.6683,
|
|
"mean_token_accuracy": 0.10134745985269547,
|
|
"num_tokens": 842290.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 8.12824535369873,
|
|
"epoch": 0.4326205867727499,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.684,
|
|
"mean_token_accuracy": 0.10076010972261429,
|
|
"num_tokens": 852500.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 8.012931871414185,
|
|
"epoch": 0.43759323719542514,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.6047,
|
|
"mean_token_accuracy": 0.10121534466743469,
|
|
"num_tokens": 861401.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 8.086879587173462,
|
|
"epoch": 0.44256588761810045,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.6227,
|
|
"mean_token_accuracy": 0.09979584887623787,
|
|
"num_tokens": 871523.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 8.090540599822997,
|
|
"epoch": 0.44753853804077576,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.7061,
|
|
"mean_token_accuracy": 0.09213669747114181,
|
|
"num_tokens": 881384.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 8.126852178573609,
|
|
"epoch": 0.452511188463451,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.734,
|
|
"mean_token_accuracy": 0.09309234991669654,
|
|
"num_tokens": 890511.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 8.062766981124877,
|
|
"epoch": 0.4574838388861263,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.6788,
|
|
"mean_token_accuracy": 0.09279735162854194,
|
|
"num_tokens": 900064.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 8.13046703338623,
|
|
"epoch": 0.46245648930880157,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.7215,
|
|
"mean_token_accuracy": 0.09545360058546067,
|
|
"num_tokens": 910892.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 8.09913067817688,
|
|
"epoch": 0.4674291397314769,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.7194,
|
|
"mean_token_accuracy": 0.09483447819948196,
|
|
"num_tokens": 920566.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 8.045233345031738,
|
|
"epoch": 0.4724017901541522,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.6206,
|
|
"mean_token_accuracy": 0.09445327222347259,
|
|
"num_tokens": 930609.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 8.060771894454955,
|
|
"epoch": 0.47737444057682743,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.66,
|
|
"mean_token_accuracy": 0.09503448456525802,
|
|
"num_tokens": 940383.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 8.15862307548523,
|
|
"epoch": 0.48234709099950274,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.6285,
|
|
"mean_token_accuracy": 0.09771560356020928,
|
|
"num_tokens": 951302.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 8.081371116638184,
|
|
"epoch": 0.48731974142217804,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.5969,
|
|
"mean_token_accuracy": 0.09744079262018204,
|
|
"num_tokens": 961126.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 8.032482051849366,
|
|
"epoch": 0.4922923918448533,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.5935,
|
|
"mean_token_accuracy": 0.09629232585430145,
|
|
"num_tokens": 970845.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.959290885925293,
|
|
"epoch": 0.4972650422675286,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.5338,
|
|
"mean_token_accuracy": 0.10219172462821006,
|
|
"num_tokens": 980919.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.940049409866333,
|
|
"epoch": 0.5022376926902039,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.5322,
|
|
"mean_token_accuracy": 0.10311833545565605,
|
|
"num_tokens": 990659.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 8.02933964729309,
|
|
"epoch": 0.5072103431128792,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.6205,
|
|
"mean_token_accuracy": 0.09970935881137848,
|
|
"num_tokens": 999540.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 8.019573259353638,
|
|
"epoch": 0.5121829935355544,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.6061,
|
|
"mean_token_accuracy": 0.09358469545841216,
|
|
"num_tokens": 1008688.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 8.0077889919281,
|
|
"epoch": 0.5171556439582298,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.5144,
|
|
"mean_token_accuracy": 0.11092083007097245,
|
|
"num_tokens": 1018310.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.852058458328247,
|
|
"epoch": 0.522128294380905,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.6069,
|
|
"mean_token_accuracy": 0.0994601771235466,
|
|
"num_tokens": 1027973.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 8.132920122146606,
|
|
"epoch": 0.5271009448035803,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.5264,
|
|
"mean_token_accuracy": 0.10466317981481552,
|
|
"num_tokens": 1038141.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.880703258514404,
|
|
"epoch": 0.5320735952262556,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.626,
|
|
"mean_token_accuracy": 0.09480768665671349,
|
|
"num_tokens": 1047932.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.882956504821777,
|
|
"epoch": 0.5370462456489309,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.4267,
|
|
"mean_token_accuracy": 0.11068027839064598,
|
|
"num_tokens": 1056826.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 8.010830736160278,
|
|
"epoch": 0.5420188960716061,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.5807,
|
|
"mean_token_accuracy": 0.10220433995127678,
|
|
"num_tokens": 1066025.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.966916275024414,
|
|
"epoch": 0.5469915464942815,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.5869,
|
|
"mean_token_accuracy": 0.09801997393369674,
|
|
"num_tokens": 1075452.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 8.02659068107605,
|
|
"epoch": 0.5519641969169568,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.5895,
|
|
"mean_token_accuracy": 0.10260884463787079,
|
|
"num_tokens": 1085300.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.869441032409668,
|
|
"epoch": 0.556936847339632,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.4643,
|
|
"mean_token_accuracy": 0.10986730009317398,
|
|
"num_tokens": 1094548.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 8.092569923400879,
|
|
"epoch": 0.5619094977623074,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.5082,
|
|
"mean_token_accuracy": 0.10508983358740806,
|
|
"num_tokens": 1104321.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.939152812957763,
|
|
"epoch": 0.5668821481849826,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.6251,
|
|
"mean_token_accuracy": 0.09839587658643723,
|
|
"num_tokens": 1115047.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 8.003477239608765,
|
|
"epoch": 0.5718547986076579,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.585,
|
|
"mean_token_accuracy": 0.10319317504763603,
|
|
"num_tokens": 1124757.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.930534553527832,
|
|
"epoch": 0.5768274490303331,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.5566,
|
|
"mean_token_accuracy": 0.09697693884372711,
|
|
"num_tokens": 1134255.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 8.075233602523804,
|
|
"epoch": 0.5818000994530085,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.6043,
|
|
"mean_token_accuracy": 0.10006313472986221,
|
|
"num_tokens": 1143117.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.938198089599609,
|
|
"epoch": 0.5867727498756837,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.5148,
|
|
"mean_token_accuracy": 0.10014821812510491,
|
|
"num_tokens": 1153203.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.920211458206177,
|
|
"epoch": 0.591745400298359,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.5,
|
|
"mean_token_accuracy": 0.10478221327066421,
|
|
"num_tokens": 1162839.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.8557953357696535,
|
|
"epoch": 0.5967180507210343,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.4608,
|
|
"mean_token_accuracy": 0.10512098148465157,
|
|
"num_tokens": 1172580.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 8.031906652450562,
|
|
"epoch": 0.6016907011437096,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.473,
|
|
"mean_token_accuracy": 0.11015030816197395,
|
|
"num_tokens": 1183715.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.937970399856567,
|
|
"epoch": 0.6066633515663848,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.6279,
|
|
"mean_token_accuracy": 0.09497415050864219,
|
|
"num_tokens": 1194217.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.901769924163818,
|
|
"epoch": 0.6116360019890602,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.4873,
|
|
"mean_token_accuracy": 0.10889495462179184,
|
|
"num_tokens": 1203297.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 8.01526050567627,
|
|
"epoch": 0.6166086524117355,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.5916,
|
|
"mean_token_accuracy": 0.09512657150626183,
|
|
"num_tokens": 1213394.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.85530571937561,
|
|
"epoch": 0.6215813028344107,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.5358,
|
|
"mean_token_accuracy": 0.10607103481888772,
|
|
"num_tokens": 1222771.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.95513277053833,
|
|
"epoch": 0.6265539532570861,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.5204,
|
|
"mean_token_accuracy": 0.0972771666944027,
|
|
"num_tokens": 1232171.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.86173448562622,
|
|
"epoch": 0.6315266036797613,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.4505,
|
|
"mean_token_accuracy": 0.10430286675691605,
|
|
"num_tokens": 1241640.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.854068088531494,
|
|
"epoch": 0.6364992541024366,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.4687,
|
|
"mean_token_accuracy": 0.10413866713643075,
|
|
"num_tokens": 1251785.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.883517074584961,
|
|
"epoch": 0.6414719045251119,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.4631,
|
|
"mean_token_accuracy": 0.10138216093182564,
|
|
"num_tokens": 1262327.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.836060333251953,
|
|
"epoch": 0.6464445549477872,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.4038,
|
|
"mean_token_accuracy": 0.10368820801377296,
|
|
"num_tokens": 1270871.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.805431079864502,
|
|
"epoch": 0.6514172053704624,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.4444,
|
|
"mean_token_accuracy": 0.10368885323405266,
|
|
"num_tokens": 1281563.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.854015874862671,
|
|
"epoch": 0.6563898557931377,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.4545,
|
|
"mean_token_accuracy": 0.10161851048469543,
|
|
"num_tokens": 1290867.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.927043771743774,
|
|
"epoch": 0.661362506215813,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.4668,
|
|
"mean_token_accuracy": 0.10288965106010436,
|
|
"num_tokens": 1300384.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.893484973907471,
|
|
"epoch": 0.6663351566384883,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.3865,
|
|
"mean_token_accuracy": 0.10967770665884018,
|
|
"num_tokens": 1310443.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.822868919372558,
|
|
"epoch": 0.6713078070611636,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.3685,
|
|
"mean_token_accuracy": 0.1109682209789753,
|
|
"num_tokens": 1320345.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.8312232971191404,
|
|
"epoch": 0.6762804574838389,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.53,
|
|
"mean_token_accuracy": 0.10490024462342262,
|
|
"num_tokens": 1330868.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.818981647491455,
|
|
"epoch": 0.6812531079065142,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.4442,
|
|
"mean_token_accuracy": 0.10243335589766503,
|
|
"num_tokens": 1340605.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.875829458236694,
|
|
"epoch": 0.6862257583291894,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.4607,
|
|
"mean_token_accuracy": 0.10384754240512847,
|
|
"num_tokens": 1350248.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.650998640060425,
|
|
"epoch": 0.6911984087518648,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.4145,
|
|
"mean_token_accuracy": 0.10926363319158554,
|
|
"num_tokens": 1360251.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.897835302352905,
|
|
"epoch": 0.69617105917454,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.3653,
|
|
"mean_token_accuracy": 0.11317518725991249,
|
|
"num_tokens": 1370156.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.83317084312439,
|
|
"epoch": 0.7011437095972153,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.5418,
|
|
"mean_token_accuracy": 0.09714026600122452,
|
|
"num_tokens": 1380346.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.78965392112732,
|
|
"epoch": 0.7061163600198906,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.4048,
|
|
"mean_token_accuracy": 0.10425653830170631,
|
|
"num_tokens": 1390958.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.844746923446655,
|
|
"epoch": 0.7110890104425659,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.4823,
|
|
"mean_token_accuracy": 0.10559975057840347,
|
|
"num_tokens": 1400324.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.866388559341431,
|
|
"epoch": 0.7160616608652411,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.4002,
|
|
"mean_token_accuracy": 0.10767291709780694,
|
|
"num_tokens": 1410272.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.707793235778809,
|
|
"epoch": 0.7210343112879165,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.4192,
|
|
"mean_token_accuracy": 0.10780517980456353,
|
|
"num_tokens": 1420061.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.7463947296142575,
|
|
"epoch": 0.7260069617105918,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.3613,
|
|
"mean_token_accuracy": 0.10962173566222191,
|
|
"num_tokens": 1428771.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.700264024734497,
|
|
"epoch": 0.730979612133267,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.3579,
|
|
"mean_token_accuracy": 0.10923694595694541,
|
|
"num_tokens": 1438154.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.6981738090515135,
|
|
"epoch": 0.7359522625559424,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.3366,
|
|
"mean_token_accuracy": 0.10916702821850777,
|
|
"num_tokens": 1447403.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.838420009613037,
|
|
"epoch": 0.7409249129786176,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.4225,
|
|
"mean_token_accuracy": 0.10422661527991295,
|
|
"num_tokens": 1457443.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.775690364837646,
|
|
"epoch": 0.7458975634012929,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.4171,
|
|
"mean_token_accuracy": 0.1078345961868763,
|
|
"num_tokens": 1466783.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.842795324325562,
|
|
"epoch": 0.7508702138239681,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.4309,
|
|
"mean_token_accuracy": 0.10510188713669777,
|
|
"num_tokens": 1476320.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.6635977268219,
|
|
"epoch": 0.7558428642466435,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.3422,
|
|
"mean_token_accuracy": 0.10771172270178794,
|
|
"num_tokens": 1487581.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.852895784378052,
|
|
"epoch": 0.7608155146693187,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000382,
|
|
"loss": 7.4748,
|
|
"mean_token_accuracy": 0.10103904530405998,
|
|
"num_tokens": 1496455.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.726494026184082,
|
|
"epoch": 0.765788165091994,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 7.4027,
|
|
"mean_token_accuracy": 0.1115493893623352,
|
|
"num_tokens": 1506422.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.752800941467285,
|
|
"epoch": 0.7707608155146694,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.3697,
|
|
"mean_token_accuracy": 0.10680537670850754,
|
|
"num_tokens": 1516593.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.726449012756348,
|
|
"epoch": 0.7757334659373446,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.4068,
|
|
"mean_token_accuracy": 0.10679919570684433,
|
|
"num_tokens": 1526011.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.742998838424683,
|
|
"epoch": 0.7807061163600199,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.3367,
|
|
"mean_token_accuracy": 0.10538461208343505,
|
|
"num_tokens": 1536018.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.617919445037842,
|
|
"epoch": 0.7856787667826952,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.2812,
|
|
"mean_token_accuracy": 0.11040928065776826,
|
|
"num_tokens": 1545568.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.749273729324341,
|
|
"epoch": 0.7906514172053705,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.2682,
|
|
"mean_token_accuracy": 0.11172266751527786,
|
|
"num_tokens": 1556389.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.814680528640747,
|
|
"epoch": 0.7956240676280457,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.3847,
|
|
"mean_token_accuracy": 0.10775300115346909,
|
|
"num_tokens": 1565587.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.6317754745483395,
|
|
"epoch": 0.8005967180507211,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000402,
|
|
"loss": 7.3027,
|
|
"mean_token_accuracy": 0.10983041971921921,
|
|
"num_tokens": 1574911.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.558075761795044,
|
|
"epoch": 0.8055693684733963,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 7.264,
|
|
"mean_token_accuracy": 0.11495908200740815,
|
|
"num_tokens": 1584477.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.660688591003418,
|
|
"epoch": 0.8105420188960716,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 7.3449,
|
|
"mean_token_accuracy": 0.11345159783959388,
|
|
"num_tokens": 1594408.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.662034893035889,
|
|
"epoch": 0.8155146693187469,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 7.2873,
|
|
"mean_token_accuracy": 0.11071835607290267,
|
|
"num_tokens": 1603494.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.699939775466919,
|
|
"epoch": 0.8204873197414222,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000412,
|
|
"loss": 7.3896,
|
|
"mean_token_accuracy": 0.10532766357064247,
|
|
"num_tokens": 1612994.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.750362157821655,
|
|
"epoch": 0.8254599701640974,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 7.2704,
|
|
"mean_token_accuracy": 0.10584538355469704,
|
|
"num_tokens": 1622150.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.638842678070068,
|
|
"epoch": 0.8304326205867727,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.000417,
|
|
"loss": 7.2926,
|
|
"mean_token_accuracy": 0.11065409779548645,
|
|
"num_tokens": 1631040.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.656509637832642,
|
|
"epoch": 0.8354052710094481,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 7.2518,
|
|
"mean_token_accuracy": 0.11168953701853752,
|
|
"num_tokens": 1641519.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.702224683761597,
|
|
"epoch": 0.8403779214321233,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000422,
|
|
"loss": 7.3027,
|
|
"mean_token_accuracy": 0.1080830842256546,
|
|
"num_tokens": 1652218.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.77684383392334,
|
|
"epoch": 0.8453505718547986,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 7.3732,
|
|
"mean_token_accuracy": 0.10448917225003243,
|
|
"num_tokens": 1663763.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.692452812194825,
|
|
"epoch": 0.8503232222774739,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000427,
|
|
"loss": 7.3174,
|
|
"mean_token_accuracy": 0.10640130266547203,
|
|
"num_tokens": 1674160.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.675528860092163,
|
|
"epoch": 0.8552958727001492,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 7.4251,
|
|
"mean_token_accuracy": 0.1123006097972393,
|
|
"num_tokens": 1685138.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.7082442283630375,
|
|
"epoch": 0.8602685231228244,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000432,
|
|
"loss": 7.3299,
|
|
"mean_token_accuracy": 0.10580715984106064,
|
|
"num_tokens": 1694847.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.687242841720581,
|
|
"epoch": 0.8652411735454998,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 7.3152,
|
|
"mean_token_accuracy": 0.10903403162956238,
|
|
"num_tokens": 1704503.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.559535503387451,
|
|
"epoch": 0.870213823968175,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.000437,
|
|
"loss": 7.2871,
|
|
"mean_token_accuracy": 0.11060690507292747,
|
|
"num_tokens": 1714011.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.73149561882019,
|
|
"epoch": 0.8751864743908503,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 7.2414,
|
|
"mean_token_accuracy": 0.11003299355506897,
|
|
"num_tokens": 1724254.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.559349107742309,
|
|
"epoch": 0.8801591248135257,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000442,
|
|
"loss": 7.3015,
|
|
"mean_token_accuracy": 0.10640292093157769,
|
|
"num_tokens": 1734696.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.5397727489471436,
|
|
"epoch": 0.8851317752362009,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 7.2606,
|
|
"mean_token_accuracy": 0.11507974788546563,
|
|
"num_tokens": 1743650.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.599524927139282,
|
|
"epoch": 0.8901044256588762,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000447,
|
|
"loss": 7.2203,
|
|
"mean_token_accuracy": 0.11115283817052841,
|
|
"num_tokens": 1753449.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.655688762664795,
|
|
"epoch": 0.8950770760815515,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 7.3675,
|
|
"mean_token_accuracy": 0.11229334026575089,
|
|
"num_tokens": 1764104.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.566684341430664,
|
|
"epoch": 0.9000497265042268,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 7.1293,
|
|
"mean_token_accuracy": 0.12582117319107056,
|
|
"num_tokens": 1773172.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.634956216812133,
|
|
"epoch": 0.905022376926902,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 7.2032,
|
|
"mean_token_accuracy": 0.11464175954461098,
|
|
"num_tokens": 1783268.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.5021251201629635,
|
|
"epoch": 0.9099950273495774,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 7.2476,
|
|
"mean_token_accuracy": 0.10757804960012436,
|
|
"num_tokens": 1792169.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.66026554107666,
|
|
"epoch": 0.9149676777722526,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 7.2501,
|
|
"mean_token_accuracy": 0.11052607968449593,
|
|
"num_tokens": 1802280.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.5866508960723875,
|
|
"epoch": 0.9199403281949279,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000462,
|
|
"loss": 7.25,
|
|
"mean_token_accuracy": 0.11102810874581337,
|
|
"num_tokens": 1812086.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.470366525650024,
|
|
"epoch": 0.9249129786176031,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 7.1273,
|
|
"mean_token_accuracy": 0.1212368443608284,
|
|
"num_tokens": 1821554.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.644198083877564,
|
|
"epoch": 0.9298856290402785,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000467,
|
|
"loss": 7.2551,
|
|
"mean_token_accuracy": 0.11455676332116127,
|
|
"num_tokens": 1831088.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.505097246170044,
|
|
"epoch": 0.9348582794629537,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 7.2723,
|
|
"mean_token_accuracy": 0.1153840146958828,
|
|
"num_tokens": 1840606.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.486701536178589,
|
|
"epoch": 0.939830929885629,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000472,
|
|
"loss": 7.1843,
|
|
"mean_token_accuracy": 0.12234615907073021,
|
|
"num_tokens": 1850343.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.49072756767273,
|
|
"epoch": 0.9448035803083044,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 7.2992,
|
|
"mean_token_accuracy": 0.11201107949018478,
|
|
"num_tokens": 1860727.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.53571457862854,
|
|
"epoch": 0.9497762307309796,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000477,
|
|
"loss": 7.2114,
|
|
"mean_token_accuracy": 0.1107769139111042,
|
|
"num_tokens": 1870703.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.566974258422851,
|
|
"epoch": 0.9547488811536549,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 7.236,
|
|
"mean_token_accuracy": 0.11286089569330215,
|
|
"num_tokens": 1880694.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.500005054473877,
|
|
"epoch": 0.9597215315763302,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000482,
|
|
"loss": 7.1751,
|
|
"mean_token_accuracy": 0.11533893570303917,
|
|
"num_tokens": 1890195.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.554698896408081,
|
|
"epoch": 0.9646941819990055,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 7.1661,
|
|
"mean_token_accuracy": 0.11099867224693298,
|
|
"num_tokens": 1900221.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 7.39583911895752,
|
|
"epoch": 0.9696668324216807,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000487,
|
|
"loss": 7.2126,
|
|
"mean_token_accuracy": 0.11516022086143493,
|
|
"num_tokens": 1910122.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.621148586273193,
|
|
"epoch": 0.9746394828443561,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 7.1956,
|
|
"mean_token_accuracy": 0.11264971271157265,
|
|
"num_tokens": 1918814.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.444791698455811,
|
|
"epoch": 0.9796121332670313,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000492,
|
|
"loss": 7.1927,
|
|
"mean_token_accuracy": 0.11786668226122857,
|
|
"num_tokens": 1928446.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.4477911472320555,
|
|
"epoch": 0.9845847836897066,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 7.1764,
|
|
"mean_token_accuracy": 0.11415334343910218,
|
|
"num_tokens": 1939060.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 7.5346626281738285,
|
|
"epoch": 0.989557434112382,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000497,
|
|
"loss": 7.1698,
|
|
"mean_token_accuracy": 0.11574856638908386,
|
|
"num_tokens": 1948809.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.45610032081604,
|
|
"epoch": 0.9945300845350572,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 7.184,
|
|
"mean_token_accuracy": 0.111851417273283,
|
|
"num_tokens": 1959232.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.9945300845350572,
|
|
"eval_entropy": 7.415475522318194,
|
|
"eval_loss": 7.2727508544921875,
|
|
"eval_mean_token_accuracy": 0.10570303410772354,
|
|
"eval_num_tokens": 1959232.0,
|
|
"eval_runtime": 1.9499,
|
|
"eval_samples_per_second": 1777.046,
|
|
"eval_steps_per_second": 222.58,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 7.493302965164185,
|
|
"epoch": 0.9995027349577325,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004999997830922735,
|
|
"loss": 7.1029,
|
|
"mean_token_accuracy": 0.11844968572258949,
|
|
"num_tokens": 1969117.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 7.496229383680555,
|
|
"epoch": 1.0039781203381402,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004999989019053515,
|
|
"loss": 7.1509,
|
|
"mean_token_accuracy": 0.11668031331565645,
|
|
"num_tokens": 1978202.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 7.339719772338867,
|
|
"epoch": 1.0089507707608154,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004999973428851536,
|
|
"loss": 6.9271,
|
|
"mean_token_accuracy": 0.12847438231110572,
|
|
"num_tokens": 1988085.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 7.394905757904053,
|
|
"epoch": 1.0139234211834909,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999951060363766,
|
|
"loss": 6.9203,
|
|
"mean_token_accuracy": 0.12345911487936974,
|
|
"num_tokens": 1996977.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 7.557369518280029,
|
|
"epoch": 1.0188960716061661,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004999921913657592,
|
|
"loss": 7.0022,
|
|
"mean_token_accuracy": 0.12040132135152817,
|
|
"num_tokens": 2006622.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 7.476524496078492,
|
|
"epoch": 1.0238687220288414,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999885988820821,
|
|
"loss": 7.0937,
|
|
"mean_token_accuracy": 0.11633466556668282,
|
|
"num_tokens": 2016390.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 7.517478179931641,
|
|
"epoch": 1.0288413724515166,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999843285961683,
|
|
"loss": 6.9541,
|
|
"mean_token_accuracy": 0.11428895443677903,
|
|
"num_tokens": 2026439.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 7.35127477645874,
|
|
"epoch": 1.0338140228741919,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999793805208822,
|
|
"loss": 7.0121,
|
|
"mean_token_accuracy": 0.12344333082437516,
|
|
"num_tokens": 2035611.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 7.427029991149903,
|
|
"epoch": 1.0387866732968671,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999737546711305,
|
|
"loss": 7.032,
|
|
"mean_token_accuracy": 0.11545429825782776,
|
|
"num_tokens": 2047166.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 7.396841049194336,
|
|
"epoch": 1.0437593237195426,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999674510638618,
|
|
"loss": 6.9481,
|
|
"mean_token_accuracy": 0.1212002918124199,
|
|
"num_tokens": 2057216.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 7.379416799545288,
|
|
"epoch": 1.0487319741422179,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004999604697180661,
|
|
"loss": 7.0116,
|
|
"mean_token_accuracy": 0.1184284508228302,
|
|
"num_tokens": 2067339.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 7.3042182445526125,
|
|
"epoch": 1.053704624564893,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004999528106547759,
|
|
"loss": 6.8967,
|
|
"mean_token_accuracy": 0.12114272117614747,
|
|
"num_tokens": 2077815.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 7.3766937255859375,
|
|
"epoch": 1.0586772749875684,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004999444738970644,
|
|
"loss": 6.994,
|
|
"mean_token_accuracy": 0.11613365784287452,
|
|
"num_tokens": 2087738.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 7.388165855407715,
|
|
"epoch": 1.0636499254102436,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999354594700474,
|
|
"loss": 6.9292,
|
|
"mean_token_accuracy": 0.12312228381633758,
|
|
"num_tokens": 2097246.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 7.34119062423706,
|
|
"epoch": 1.0686225758329189,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004999257674008817,
|
|
"loss": 6.8507,
|
|
"mean_token_accuracy": 0.12643605396151542,
|
|
"num_tokens": 2106415.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 7.21716423034668,
|
|
"epoch": 1.0735952262555943,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004999153977187656,
|
|
"loss": 6.8473,
|
|
"mean_token_accuracy": 0.12365620136260987,
|
|
"num_tokens": 2115950.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 7.422256946563721,
|
|
"epoch": 1.0785678766782696,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000499904350454939,
|
|
"loss": 7.0261,
|
|
"mean_token_accuracy": 0.12062097415328026,
|
|
"num_tokens": 2127896.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 7.303533363342285,
|
|
"epoch": 1.0835405271009448,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004998926256426829,
|
|
"loss": 6.9646,
|
|
"mean_token_accuracy": 0.11712555587291718,
|
|
"num_tokens": 2137238.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 7.3995646953582765,
|
|
"epoch": 1.08851317752362,
|
|
"grad_norm": 4.5625,
|
|
"learning_rate": 0.0004998802233173196,
|
|
"loss": 7.0234,
|
|
"mean_token_accuracy": 0.11911827549338341,
|
|
"num_tokens": 2147107.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 7.400513315200806,
|
|
"epoch": 1.0934858279462953,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004998671435162123,
|
|
"loss": 7.0165,
|
|
"mean_token_accuracy": 0.12362412586808205,
|
|
"num_tokens": 2157128.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 7.199777364730835,
|
|
"epoch": 1.0984584783689706,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004998533862787657,
|
|
"loss": 6.9465,
|
|
"mean_token_accuracy": 0.11821585893630981,
|
|
"num_tokens": 2167981.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 7.38196325302124,
|
|
"epoch": 1.1034311287916458,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004998389516464244,
|
|
"loss": 6.9082,
|
|
"mean_token_accuracy": 0.12267233282327653,
|
|
"num_tokens": 2177659.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 7.270491695404052,
|
|
"epoch": 1.1084037792143213,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004998238396626746,
|
|
"loss": 6.9114,
|
|
"mean_token_accuracy": 0.12499475553631782,
|
|
"num_tokens": 2186936.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 7.389007472991944,
|
|
"epoch": 1.1133764296369966,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998080503730427,
|
|
"loss": 6.9282,
|
|
"mean_token_accuracy": 0.1268167279660702,
|
|
"num_tokens": 2197126.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 7.228035116195679,
|
|
"epoch": 1.1183490800596718,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004997915838250956,
|
|
"loss": 6.9711,
|
|
"mean_token_accuracy": 0.12116536647081375,
|
|
"num_tokens": 2206337.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 7.330891036987305,
|
|
"epoch": 1.123321730482347,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004997744400684406,
|
|
"loss": 6.9367,
|
|
"mean_token_accuracy": 0.12688032314181327,
|
|
"num_tokens": 2217112.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 7.372451734542847,
|
|
"epoch": 1.1282943809050223,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004997566191547251,
|
|
"loss": 7.0098,
|
|
"mean_token_accuracy": 0.11632521897554397,
|
|
"num_tokens": 2226239.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 7.359096622467041,
|
|
"epoch": 1.1332670313276976,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004997381211376364,
|
|
"loss": 7.0033,
|
|
"mean_token_accuracy": 0.1202960804104805,
|
|
"num_tokens": 2235702.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 7.3794042587280275,
|
|
"epoch": 1.138239681750373,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004997189460729019,
|
|
"loss": 6.9715,
|
|
"mean_token_accuracy": 0.1264696702361107,
|
|
"num_tokens": 2246005.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 7.160685777664185,
|
|
"epoch": 1.1432123321730483,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004996990940182884,
|
|
"loss": 6.9226,
|
|
"mean_token_accuracy": 0.1226792573928833,
|
|
"num_tokens": 2255534.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 7.451617813110351,
|
|
"epoch": 1.1481849825957235,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004996785650336024,
|
|
"loss": 6.995,
|
|
"mean_token_accuracy": 0.11782700344920158,
|
|
"num_tokens": 2266374.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 7.1924505710601805,
|
|
"epoch": 1.1531576330183988,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004996573591806898,
|
|
"loss": 6.8232,
|
|
"mean_token_accuracy": 0.1305961549282074,
|
|
"num_tokens": 2275614.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 7.263517904281616,
|
|
"epoch": 1.158130283441074,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004996354765234353,
|
|
"loss": 6.8793,
|
|
"mean_token_accuracy": 0.13028667643666267,
|
|
"num_tokens": 2284403.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 7.221520471572876,
|
|
"epoch": 1.1631029338637493,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004996129171277628,
|
|
"loss": 6.9386,
|
|
"mean_token_accuracy": 0.12240026593208313,
|
|
"num_tokens": 2294153.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 7.2880021095275875,
|
|
"epoch": 1.1680755842864246,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004995896810616349,
|
|
"loss": 6.914,
|
|
"mean_token_accuracy": 0.12336086258292198,
|
|
"num_tokens": 2302897.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 7.392106246948242,
|
|
"epoch": 1.1730482347091,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004995657683950528,
|
|
"loss": 7.0269,
|
|
"mean_token_accuracy": 0.11793071627616883,
|
|
"num_tokens": 2312483.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 7.128571462631226,
|
|
"epoch": 1.1780208851317753,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004995411792000559,
|
|
"loss": 6.7559,
|
|
"mean_token_accuracy": 0.1343972183763981,
|
|
"num_tokens": 2321911.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 7.327469968795777,
|
|
"epoch": 1.1829935355544505,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004995159135507218,
|
|
"loss": 7.0446,
|
|
"mean_token_accuracy": 0.12160438150167466,
|
|
"num_tokens": 2331932.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 7.108690023422241,
|
|
"epoch": 1.1879661859771258,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000499489971523166,
|
|
"loss": 6.824,
|
|
"mean_token_accuracy": 0.12957706823945045,
|
|
"num_tokens": 2342264.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 7.316192150115967,
|
|
"epoch": 1.192938836399801,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004994633531955415,
|
|
"loss": 6.9808,
|
|
"mean_token_accuracy": 0.12357689663767815,
|
|
"num_tokens": 2353068.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 7.2259362697601315,
|
|
"epoch": 1.1979114868224765,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004994360586480391,
|
|
"loss": 6.791,
|
|
"mean_token_accuracy": 0.12446972727775574,
|
|
"num_tokens": 2361765.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 7.270137977600098,
|
|
"epoch": 1.2028841372451518,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004994080879628864,
|
|
"loss": 6.8741,
|
|
"mean_token_accuracy": 0.12269744649529457,
|
|
"num_tokens": 2372721.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 7.188381958007812,
|
|
"epoch": 1.207856787667827,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004993794412243481,
|
|
"loss": 6.9218,
|
|
"mean_token_accuracy": 0.12584674954414368,
|
|
"num_tokens": 2382288.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 7.248117685317993,
|
|
"epoch": 1.2128294380905023,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004993501185187255,
|
|
"loss": 6.9454,
|
|
"mean_token_accuracy": 0.12722492441534997,
|
|
"num_tokens": 2391953.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 7.3238441944122314,
|
|
"epoch": 1.2178020885131775,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004993201199343565,
|
|
"loss": 6.9689,
|
|
"mean_token_accuracy": 0.11886250600218773,
|
|
"num_tokens": 2402705.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 7.246498060226441,
|
|
"epoch": 1.2227747389358528,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004992894455616149,
|
|
"loss": 6.9298,
|
|
"mean_token_accuracy": 0.1228230431675911,
|
|
"num_tokens": 2412654.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 7.308147144317627,
|
|
"epoch": 1.227747389358528,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004992580954929108,
|
|
"loss": 7.0107,
|
|
"mean_token_accuracy": 0.1260378710925579,
|
|
"num_tokens": 2422958.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 7.1362627983093265,
|
|
"epoch": 1.2327200397812033,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004992260698226894,
|
|
"loss": 6.8427,
|
|
"mean_token_accuracy": 0.13524628728628157,
|
|
"num_tokens": 2432274.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 7.269167900085449,
|
|
"epoch": 1.2376926902038787,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004991933686474315,
|
|
"loss": 7.0119,
|
|
"mean_token_accuracy": 0.11604324206709862,
|
|
"num_tokens": 2443818.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 7.267391920089722,
|
|
"epoch": 1.242665340626554,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004991599920656529,
|
|
"loss": 6.9135,
|
|
"mean_token_accuracy": 0.12462805435061455,
|
|
"num_tokens": 2453474.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 7.154501295089721,
|
|
"epoch": 1.2476379910492292,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004991259401779042,
|
|
"loss": 6.8632,
|
|
"mean_token_accuracy": 0.12783660665154456,
|
|
"num_tokens": 2462715.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 7.122018146514892,
|
|
"epoch": 1.2526106414719045,
|
|
"grad_norm": 1.8203125,
|
|
"learning_rate": 0.0004990912130867704,
|
|
"loss": 6.793,
|
|
"mean_token_accuracy": 0.1286759704351425,
|
|
"num_tokens": 2473256.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 7.270669984817505,
|
|
"epoch": 1.2575832918945797,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004990558108968703,
|
|
"loss": 6.8362,
|
|
"mean_token_accuracy": 0.12040838897228241,
|
|
"num_tokens": 2482475.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 7.157846975326538,
|
|
"epoch": 1.2625559423172552,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004990197337148571,
|
|
"loss": 6.8845,
|
|
"mean_token_accuracy": 0.12453731372952462,
|
|
"num_tokens": 2493163.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 7.287322330474853,
|
|
"epoch": 1.2675285927399305,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000498982981649417,
|
|
"loss": 6.9298,
|
|
"mean_token_accuracy": 0.12113498970866203,
|
|
"num_tokens": 2502062.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 7.226325988769531,
|
|
"epoch": 1.2725012431626057,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004989455548112695,
|
|
"loss": 6.9066,
|
|
"mean_token_accuracy": 0.12696527019143106,
|
|
"num_tokens": 2511196.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 7.268891716003418,
|
|
"epoch": 1.277473893585281,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004989074533131671,
|
|
"loss": 6.8317,
|
|
"mean_token_accuracy": 0.12131210193037986,
|
|
"num_tokens": 2520492.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 7.129314327239991,
|
|
"epoch": 1.2824465440079562,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004988686772698945,
|
|
"loss": 6.8393,
|
|
"mean_token_accuracy": 0.12803662717342376,
|
|
"num_tokens": 2530368.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 7.228492403030396,
|
|
"epoch": 1.2874191944306315,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004988292267982689,
|
|
"loss": 6.8896,
|
|
"mean_token_accuracy": 0.12222845032811165,
|
|
"num_tokens": 2540734.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 7.137642669677734,
|
|
"epoch": 1.2923918448533067,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004987891020171387,
|
|
"loss": 6.9013,
|
|
"mean_token_accuracy": 0.12264095768332481,
|
|
"num_tokens": 2550907.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 7.275863742828369,
|
|
"epoch": 1.297364495275982,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004987483030473845,
|
|
"loss": 6.9567,
|
|
"mean_token_accuracy": 0.1269907049834728,
|
|
"num_tokens": 2560062.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 7.269666481018066,
|
|
"epoch": 1.3023371456986574,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004987068300119173,
|
|
"loss": 6.8745,
|
|
"mean_token_accuracy": 0.12158575430512428,
|
|
"num_tokens": 2568840.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 7.255507898330689,
|
|
"epoch": 1.3073097961213327,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004986646830356792,
|
|
"loss": 6.9418,
|
|
"mean_token_accuracy": 0.12193020433187485,
|
|
"num_tokens": 2578566.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 7.251036262512207,
|
|
"epoch": 1.312282446544008,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004986218622456423,
|
|
"loss": 6.935,
|
|
"mean_token_accuracy": 0.12300549000501633,
|
|
"num_tokens": 2589853.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 7.159052133560181,
|
|
"epoch": 1.3172550969666832,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000498578367770809,
|
|
"loss": 6.8971,
|
|
"mean_token_accuracy": 0.1280972309410572,
|
|
"num_tokens": 2600848.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 7.162569952011109,
|
|
"epoch": 1.3222277473893587,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004985341997422111,
|
|
"loss": 6.8438,
|
|
"mean_token_accuracy": 0.1278272859752178,
|
|
"num_tokens": 2610730.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 7.152881383895874,
|
|
"epoch": 1.327200397812034,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004984893582929093,
|
|
"loss": 6.9235,
|
|
"mean_token_accuracy": 0.12276390418410302,
|
|
"num_tokens": 2621129.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 7.262564945220947,
|
|
"epoch": 1.3321730482347092,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004984438435579933,
|
|
"loss": 6.8666,
|
|
"mean_token_accuracy": 0.12768579870462418,
|
|
"num_tokens": 2630306.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 7.203206634521484,
|
|
"epoch": 1.3371456986573844,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004983976556745812,
|
|
"loss": 6.965,
|
|
"mean_token_accuracy": 0.11911260932683945,
|
|
"num_tokens": 2639874.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 7.157221889495849,
|
|
"epoch": 1.3421183490800597,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000498350794781819,
|
|
"loss": 6.9507,
|
|
"mean_token_accuracy": 0.12042490616440774,
|
|
"num_tokens": 2650340.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 7.173311042785644,
|
|
"epoch": 1.347090999502735,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004983032610208801,
|
|
"loss": 6.7057,
|
|
"mean_token_accuracy": 0.1385875567793846,
|
|
"num_tokens": 2659541.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 7.1430010318756105,
|
|
"epoch": 1.3520636499254102,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004982550545349649,
|
|
"loss": 6.9138,
|
|
"mean_token_accuracy": 0.12124743461608886,
|
|
"num_tokens": 2669151.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 7.152199602127075,
|
|
"epoch": 1.3570363003480854,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004982061754693008,
|
|
"loss": 6.7927,
|
|
"mean_token_accuracy": 0.1370914548635483,
|
|
"num_tokens": 2678867.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 7.108493709564209,
|
|
"epoch": 1.362008950770761,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004981566239711412,
|
|
"loss": 6.9067,
|
|
"mean_token_accuracy": 0.1217210240662098,
|
|
"num_tokens": 2687572.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 7.179310941696167,
|
|
"epoch": 1.3669816011934361,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004981064001897651,
|
|
"loss": 6.8258,
|
|
"mean_token_accuracy": 0.12409792467951775,
|
|
"num_tokens": 2696922.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 7.15827956199646,
|
|
"epoch": 1.3719542516161114,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004980555042764772,
|
|
"loss": 6.858,
|
|
"mean_token_accuracy": 0.12861115857958794,
|
|
"num_tokens": 2707465.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 7.167220401763916,
|
|
"epoch": 1.3769269020387866,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004980039363846069,
|
|
"loss": 6.7731,
|
|
"mean_token_accuracy": 0.13282391503453256,
|
|
"num_tokens": 2718708.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 7.1543560981750485,
|
|
"epoch": 1.381899552461462,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000497951696669508,
|
|
"loss": 6.8469,
|
|
"mean_token_accuracy": 0.12289262413978577,
|
|
"num_tokens": 2728862.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 7.187507057189942,
|
|
"epoch": 1.3868722028841374,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004978987852885579,
|
|
"loss": 6.8147,
|
|
"mean_token_accuracy": 0.12499945238232613,
|
|
"num_tokens": 2738546.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 7.02512788772583,
|
|
"epoch": 1.3918448533068126,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004978452024011582,
|
|
"loss": 6.7638,
|
|
"mean_token_accuracy": 0.13051125630736352,
|
|
"num_tokens": 2747885.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 7.115453433990479,
|
|
"epoch": 1.3968175037294879,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000497790948168733,
|
|
"loss": 6.806,
|
|
"mean_token_accuracy": 0.13135685175657272,
|
|
"num_tokens": 2756851.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 7.17847204208374,
|
|
"epoch": 1.4017901541521631,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004977360227547288,
|
|
"loss": 6.8366,
|
|
"mean_token_accuracy": 0.12639565840363504,
|
|
"num_tokens": 2766309.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 7.103021621704102,
|
|
"epoch": 1.4067628045748384,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004976804263246146,
|
|
"loss": 6.8197,
|
|
"mean_token_accuracy": 0.1250031240284443,
|
|
"num_tokens": 2775679.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 7.066811227798462,
|
|
"epoch": 1.4117354549975136,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004976241590458803,
|
|
"loss": 6.7916,
|
|
"mean_token_accuracy": 0.12554823979735374,
|
|
"num_tokens": 2784610.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 7.18024435043335,
|
|
"epoch": 1.4167081054201889,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004975672210880372,
|
|
"loss": 6.8794,
|
|
"mean_token_accuracy": 0.12827789261937142,
|
|
"num_tokens": 2793969.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 7.087760019302368,
|
|
"epoch": 1.4216807558428641,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004975096126226171,
|
|
"loss": 6.7225,
|
|
"mean_token_accuracy": 0.13135690838098527,
|
|
"num_tokens": 2803096.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 7.031573009490967,
|
|
"epoch": 1.4266534062655396,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004974513338231715,
|
|
"loss": 6.792,
|
|
"mean_token_accuracy": 0.13075321167707443,
|
|
"num_tokens": 2812320.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 7.197575330734253,
|
|
"epoch": 1.4316260566882149,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004973923848652715,
|
|
"loss": 6.8669,
|
|
"mean_token_accuracy": 0.12681687399744987,
|
|
"num_tokens": 2821393.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 7.0385096073150635,
|
|
"epoch": 1.43659870711089,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004973327659265073,
|
|
"loss": 6.8145,
|
|
"mean_token_accuracy": 0.12959662079811096,
|
|
"num_tokens": 2831182.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 7.135791969299317,
|
|
"epoch": 1.4415713575335654,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004972724771864874,
|
|
"loss": 6.7979,
|
|
"mean_token_accuracy": 0.13130285441875458,
|
|
"num_tokens": 2840429.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 7.0527090549469,
|
|
"epoch": 1.4465440079562406,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004972115188268378,
|
|
"loss": 6.7799,
|
|
"mean_token_accuracy": 0.13332838863134383,
|
|
"num_tokens": 2849443.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 7.091077661514282,
|
|
"epoch": 1.451516658378916,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004971498910312024,
|
|
"loss": 6.7554,
|
|
"mean_token_accuracy": 0.13105802685022355,
|
|
"num_tokens": 2858908.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 7.193029689788818,
|
|
"epoch": 1.4564893088015913,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004970875939852412,
|
|
"loss": 6.84,
|
|
"mean_token_accuracy": 0.12812084704637527,
|
|
"num_tokens": 2869615.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 7.086412715911865,
|
|
"epoch": 1.4614619592242666,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000497024627876631,
|
|
"loss": 6.7764,
|
|
"mean_token_accuracy": 0.1278965711593628,
|
|
"num_tokens": 2879007.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 7.047615957260132,
|
|
"epoch": 1.4664346096469418,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004969609928950639,
|
|
"loss": 6.6855,
|
|
"mean_token_accuracy": 0.13203833848237992,
|
|
"num_tokens": 2887527.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 7.076382970809936,
|
|
"epoch": 1.471407260069617,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004968966892322468,
|
|
"loss": 6.7361,
|
|
"mean_token_accuracy": 0.13048939257860184,
|
|
"num_tokens": 2897137.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 7.0052777290344235,
|
|
"epoch": 1.4763799104922923,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004968317170819018,
|
|
"loss": 6.8217,
|
|
"mean_token_accuracy": 0.13370618224143982,
|
|
"num_tokens": 2908346.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 7.136820507049561,
|
|
"epoch": 1.4813525609149676,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004967660766397642,
|
|
"loss": 6.8398,
|
|
"mean_token_accuracy": 0.12695724442601203,
|
|
"num_tokens": 2918438.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 7.135380601882934,
|
|
"epoch": 1.4863252113376428,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004966997681035829,
|
|
"loss": 6.9338,
|
|
"mean_token_accuracy": 0.12171319648623466,
|
|
"num_tokens": 2928096.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 7.139851999282837,
|
|
"epoch": 1.4912978617603183,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004966327916731195,
|
|
"loss": 6.788,
|
|
"mean_token_accuracy": 0.13225842788815498,
|
|
"num_tokens": 2938150.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.979417085647583,
|
|
"epoch": 1.4962705121829936,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004965651475501476,
|
|
"loss": 6.7514,
|
|
"mean_token_accuracy": 0.13687803894281386,
|
|
"num_tokens": 2949410.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 7.091439294815063,
|
|
"epoch": 1.5012431626056688,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004964968359384524,
|
|
"loss": 6.7151,
|
|
"mean_token_accuracy": 0.1393497034907341,
|
|
"num_tokens": 2958564.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 7.003509664535523,
|
|
"epoch": 1.506215813028344,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004964278570438299,
|
|
"loss": 6.7822,
|
|
"mean_token_accuracy": 0.1343317523598671,
|
|
"num_tokens": 2968750.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 7.036017942428589,
|
|
"epoch": 1.5111884634510195,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004963582110740865,
|
|
"loss": 6.7354,
|
|
"mean_token_accuracy": 0.13797036409378052,
|
|
"num_tokens": 2978397.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 7.089057397842407,
|
|
"epoch": 1.5161611138736948,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004962878982390381,
|
|
"loss": 6.8543,
|
|
"mean_token_accuracy": 0.12393493577837944,
|
|
"num_tokens": 2987625.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 7.029309558868408,
|
|
"epoch": 1.52113376429637,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004962169187505097,
|
|
"loss": 6.7979,
|
|
"mean_token_accuracy": 0.13233162388205527,
|
|
"num_tokens": 2997266.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 7.0997051239013675,
|
|
"epoch": 1.5261064147190453,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004961452728223343,
|
|
"loss": 6.8037,
|
|
"mean_token_accuracy": 0.12844678908586502,
|
|
"num_tokens": 3007474.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 7.142696809768677,
|
|
"epoch": 1.5310790651417205,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004960729606703535,
|
|
"loss": 6.7646,
|
|
"mean_token_accuracy": 0.1274226538836956,
|
|
"num_tokens": 3016270.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.99416036605835,
|
|
"epoch": 1.5360517155643958,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004959999825124147,
|
|
"loss": 6.7622,
|
|
"mean_token_accuracy": 0.13078449293971062,
|
|
"num_tokens": 3027397.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 7.1135049819946286,
|
|
"epoch": 1.541024365987071,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000495926338568373,
|
|
"loss": 6.7947,
|
|
"mean_token_accuracy": 0.13005578964948655,
|
|
"num_tokens": 3037324.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 7.059345626831055,
|
|
"epoch": 1.5459970164097463,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004958520290600884,
|
|
"loss": 6.7522,
|
|
"mean_token_accuracy": 0.1320761352777481,
|
|
"num_tokens": 3048315.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.960471200942993,
|
|
"epoch": 1.5509696668324215,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004957770542114262,
|
|
"loss": 6.6805,
|
|
"mean_token_accuracy": 0.13555806800723075,
|
|
"num_tokens": 3057875.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 7.1203056335449215,
|
|
"epoch": 1.555942317255097,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0004957014142482563,
|
|
"loss": 6.7681,
|
|
"mean_token_accuracy": 0.12818291038274765,
|
|
"num_tokens": 3068006.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 7.021181964874268,
|
|
"epoch": 1.5609149676777723,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004956251093984521,
|
|
"loss": 6.7909,
|
|
"mean_token_accuracy": 0.13547587469220163,
|
|
"num_tokens": 3079450.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.992327356338501,
|
|
"epoch": 1.5658876181004475,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004955481398918902,
|
|
"loss": 6.5877,
|
|
"mean_token_accuracy": 0.1435450591146946,
|
|
"num_tokens": 3088385.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 7.081035327911377,
|
|
"epoch": 1.5708602685231228,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004954705059604495,
|
|
"loss": 6.7267,
|
|
"mean_token_accuracy": 0.12697325944900512,
|
|
"num_tokens": 3097854.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 7.133671236038208,
|
|
"epoch": 1.5758329189457982,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004953922078380104,
|
|
"loss": 6.8513,
|
|
"mean_token_accuracy": 0.1233741968870163,
|
|
"num_tokens": 3107877.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.986430072784424,
|
|
"epoch": 1.5808055693684735,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004953132457604544,
|
|
"loss": 6.8032,
|
|
"mean_token_accuracy": 0.1288697488605976,
|
|
"num_tokens": 3117967.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.936651039123535,
|
|
"epoch": 1.5857782197911487,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004952336199656632,
|
|
"loss": 6.6866,
|
|
"mean_token_accuracy": 0.13533323779702186,
|
|
"num_tokens": 3129827.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 7.07323956489563,
|
|
"epoch": 1.590750870213824,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004951533306935181,
|
|
"loss": 6.7126,
|
|
"mean_token_accuracy": 0.1283967450261116,
|
|
"num_tokens": 3138779.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.95408182144165,
|
|
"epoch": 1.5957235206364992,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004950723781858991,
|
|
"loss": 6.7396,
|
|
"mean_token_accuracy": 0.13301298320293425,
|
|
"num_tokens": 3148741.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.978694200515747,
|
|
"epoch": 1.6006961710591745,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004949907626866844,
|
|
"loss": 6.7679,
|
|
"mean_token_accuracy": 0.1357366606593132,
|
|
"num_tokens": 3158964.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.939979505538941,
|
|
"epoch": 1.6056688214818498,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004949084844417493,
|
|
"loss": 6.5938,
|
|
"mean_token_accuracy": 0.14317897260189055,
|
|
"num_tokens": 3168347.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.981011819839478,
|
|
"epoch": 1.610641471904525,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004948255436989657,
|
|
"loss": 6.7921,
|
|
"mean_token_accuracy": 0.1312498465180397,
|
|
"num_tokens": 3178276.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.967410850524902,
|
|
"epoch": 1.6156141223272003,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004947419407082017,
|
|
"loss": 6.7236,
|
|
"mean_token_accuracy": 0.126032754778862,
|
|
"num_tokens": 3186975.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.970744943618774,
|
|
"epoch": 1.6205867727498757,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004946576757213202,
|
|
"loss": 6.5959,
|
|
"mean_token_accuracy": 0.1481788009405136,
|
|
"num_tokens": 3197311.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.923173618316651,
|
|
"epoch": 1.625559423172551,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004945727489921784,
|
|
"loss": 6.7274,
|
|
"mean_token_accuracy": 0.13012531250715256,
|
|
"num_tokens": 3208534.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 7.096609210968017,
|
|
"epoch": 1.6305320735952262,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004944871607766273,
|
|
"loss": 6.6792,
|
|
"mean_token_accuracy": 0.1359664298593998,
|
|
"num_tokens": 3217440.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.960383129119873,
|
|
"epoch": 1.6355047240179017,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004944009113325104,
|
|
"loss": 6.8064,
|
|
"mean_token_accuracy": 0.12968278676271439,
|
|
"num_tokens": 3225858.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.954638862609864,
|
|
"epoch": 1.640477374440577,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004943140009196635,
|
|
"loss": 6.6886,
|
|
"mean_token_accuracy": 0.14025676250457764,
|
|
"num_tokens": 3235921.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.973278760910034,
|
|
"epoch": 1.6454500248632522,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004942264297999135,
|
|
"loss": 6.6679,
|
|
"mean_token_accuracy": 0.1431015580892563,
|
|
"num_tokens": 3245123.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 7.002093887329101,
|
|
"epoch": 1.6504226752859275,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004941381982370778,
|
|
"loss": 6.6843,
|
|
"mean_token_accuracy": 0.13884196057915688,
|
|
"num_tokens": 3254246.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 7.006673192977905,
|
|
"epoch": 1.6553953257086027,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004940493064969633,
|
|
"loss": 6.6729,
|
|
"mean_token_accuracy": 0.13730634599924088,
|
|
"num_tokens": 3263602.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 7.0129955291748045,
|
|
"epoch": 1.660367976131278,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000493959754847366,
|
|
"loss": 6.8508,
|
|
"mean_token_accuracy": 0.1310214728116989,
|
|
"num_tokens": 3274430.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.980470037460327,
|
|
"epoch": 1.6653406265539532,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004938695435580698,
|
|
"loss": 6.5756,
|
|
"mean_token_accuracy": 0.13692976981401445,
|
|
"num_tokens": 3283587.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.932946586608887,
|
|
"epoch": 1.6703132769766285,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0004937786729008459,
|
|
"loss": 6.7551,
|
|
"mean_token_accuracy": 0.12248265445232391,
|
|
"num_tokens": 3294165.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.993210983276367,
|
|
"epoch": 1.6752859273993037,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004936871431494517,
|
|
"loss": 6.7254,
|
|
"mean_token_accuracy": 0.13728469908237456,
|
|
"num_tokens": 3302756.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.935109996795655,
|
|
"epoch": 1.680258577821979,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004935949545796307,
|
|
"loss": 6.6467,
|
|
"mean_token_accuracy": 0.13777003958821296,
|
|
"num_tokens": 3313108.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.986012554168701,
|
|
"epoch": 1.6852312282446544,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004935021074691106,
|
|
"loss": 6.7302,
|
|
"mean_token_accuracy": 0.13472678437829017,
|
|
"num_tokens": 3323008.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.94013557434082,
|
|
"epoch": 1.6902038786673297,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004934086020976032,
|
|
"loss": 6.6468,
|
|
"mean_token_accuracy": 0.1360482841730118,
|
|
"num_tokens": 3332490.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.893417453765869,
|
|
"epoch": 1.695176529090005,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004933144387468037,
|
|
"loss": 6.6904,
|
|
"mean_token_accuracy": 0.1323254182934761,
|
|
"num_tokens": 3343069.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.932749938964844,
|
|
"epoch": 1.7001491795126804,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004932196177003891,
|
|
"loss": 6.7685,
|
|
"mean_token_accuracy": 0.1359359875321388,
|
|
"num_tokens": 3351834.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 7.079486131668091,
|
|
"epoch": 1.7051218299353557,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000493124139244018,
|
|
"loss": 6.6814,
|
|
"mean_token_accuracy": 0.13092866986989976,
|
|
"num_tokens": 3360915.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.913963794708252,
|
|
"epoch": 1.710094480358031,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004930280036653294,
|
|
"loss": 6.6667,
|
|
"mean_token_accuracy": 0.1348275937139988,
|
|
"num_tokens": 3370184.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.893805408477784,
|
|
"epoch": 1.7150671307807062,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000492931211253942,
|
|
"loss": 6.7491,
|
|
"mean_token_accuracy": 0.12614040821790695,
|
|
"num_tokens": 3379545.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.9495923042297365,
|
|
"epoch": 1.7200397812033814,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004928337623014535,
|
|
"loss": 6.6975,
|
|
"mean_token_accuracy": 0.1386477082967758,
|
|
"num_tokens": 3390238.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.97960844039917,
|
|
"epoch": 1.7250124316260567,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000492735657101439,
|
|
"loss": 6.7018,
|
|
"mean_token_accuracy": 0.1332371175289154,
|
|
"num_tokens": 3400437.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.89375410079956,
|
|
"epoch": 1.729985082048732,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004926368959494513,
|
|
"loss": 6.6747,
|
|
"mean_token_accuracy": 0.1363765239715576,
|
|
"num_tokens": 3410982.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.999945878982544,
|
|
"epoch": 1.7349577324714072,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004925374791430186,
|
|
"loss": 6.7725,
|
|
"mean_token_accuracy": 0.12880600914359092,
|
|
"num_tokens": 3420898.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.967786884307861,
|
|
"epoch": 1.7399303828940824,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004924374069816448,
|
|
"loss": 6.6319,
|
|
"mean_token_accuracy": 0.13581027537584306,
|
|
"num_tokens": 3429144.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.884613561630249,
|
|
"epoch": 1.7449030333167577,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004923366797668081,
|
|
"loss": 6.671,
|
|
"mean_token_accuracy": 0.1395179197192192,
|
|
"num_tokens": 3439427.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 7.030870532989502,
|
|
"epoch": 1.7498756837394331,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004922352978019597,
|
|
"loss": 6.6663,
|
|
"mean_token_accuracy": 0.1389183908700943,
|
|
"num_tokens": 3448218.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.950385427474975,
|
|
"epoch": 1.7548483341621084,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004921332613925239,
|
|
"loss": 6.7189,
|
|
"mean_token_accuracy": 0.1360587939620018,
|
|
"num_tokens": 3457656.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.990746355056762,
|
|
"epoch": 1.7598209845847836,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004920305708458964,
|
|
"loss": 6.6902,
|
|
"mean_token_accuracy": 0.13703734427690506,
|
|
"num_tokens": 3467414.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.882623529434204,
|
|
"epoch": 1.7647936350074591,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004919272264714431,
|
|
"loss": 6.6361,
|
|
"mean_token_accuracy": 0.13325225710868835,
|
|
"num_tokens": 3477195.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.934677791595459,
|
|
"epoch": 1.7697662854301344,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004918232285805004,
|
|
"loss": 6.7096,
|
|
"mean_token_accuracy": 0.12795531526207923,
|
|
"num_tokens": 3487497.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 7.038682126998902,
|
|
"epoch": 1.7747389358528096,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004917185774863727,
|
|
"loss": 6.6873,
|
|
"mean_token_accuracy": 0.13072459921240806,
|
|
"num_tokens": 3497363.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.867512369155884,
|
|
"epoch": 1.7797115862754849,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004916132735043329,
|
|
"loss": 6.6908,
|
|
"mean_token_accuracy": 0.13604979440569878,
|
|
"num_tokens": 3506969.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.91690731048584,
|
|
"epoch": 1.7846842366981601,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004915073169516204,
|
|
"loss": 6.6354,
|
|
"mean_token_accuracy": 0.14025795385241507,
|
|
"num_tokens": 3516418.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.872815942764282,
|
|
"epoch": 1.7896568871208354,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0004914007081474406,
|
|
"loss": 6.5792,
|
|
"mean_token_accuracy": 0.1402997076511383,
|
|
"num_tokens": 3528023.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.960012292861938,
|
|
"epoch": 1.7946295375435106,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004912934474129641,
|
|
"loss": 6.6684,
|
|
"mean_token_accuracy": 0.1340332642197609,
|
|
"num_tokens": 3537946.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.9085088729858395,
|
|
"epoch": 1.7996021879661859,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004911855350713251,
|
|
"loss": 6.6505,
|
|
"mean_token_accuracy": 0.1358492098748684,
|
|
"num_tokens": 3547299.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.916775703430176,
|
|
"epoch": 1.8045748383888611,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004910769714476214,
|
|
"loss": 6.7306,
|
|
"mean_token_accuracy": 0.13443647176027299,
|
|
"num_tokens": 3556075.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.931710147857666,
|
|
"epoch": 1.8095474888115366,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004909677568689122,
|
|
"loss": 6.6871,
|
|
"mean_token_accuracy": 0.13875397965312003,
|
|
"num_tokens": 3566314.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.913867044448852,
|
|
"epoch": 1.8145201392342118,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004908578916642183,
|
|
"loss": 6.6015,
|
|
"mean_token_accuracy": 0.14025142565369605,
|
|
"num_tokens": 3575095.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.932841873168945,
|
|
"epoch": 1.819492789656887,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004907473761645205,
|
|
"loss": 6.6872,
|
|
"mean_token_accuracy": 0.13179008662700653,
|
|
"num_tokens": 3584375.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.851043272018432,
|
|
"epoch": 1.8244654400795626,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004906362107027583,
|
|
"loss": 6.5686,
|
|
"mean_token_accuracy": 0.14170765727758408,
|
|
"num_tokens": 3594504.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.88632984161377,
|
|
"epoch": 1.8294380905022378,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004905243956138298,
|
|
"loss": 6.6974,
|
|
"mean_token_accuracy": 0.13204221725463866,
|
|
"num_tokens": 3604050.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.96671690940857,
|
|
"epoch": 1.834410740924913,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004904119312345896,
|
|
"loss": 6.6078,
|
|
"mean_token_accuracy": 0.14322111383080482,
|
|
"num_tokens": 3613388.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.874255704879761,
|
|
"epoch": 1.8393833913475883,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004902988179038492,
|
|
"loss": 6.778,
|
|
"mean_token_accuracy": 0.12332275211811065,
|
|
"num_tokens": 3623758.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 7.028712940216065,
|
|
"epoch": 1.8443560417702636,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000490185055962374,
|
|
"loss": 6.643,
|
|
"mean_token_accuracy": 0.13714932426810264,
|
|
"num_tokens": 3633040.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.839939117431641,
|
|
"epoch": 1.8493286921929388,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 0.0004900706457528845,
|
|
"loss": 6.5725,
|
|
"mean_token_accuracy": 0.13864953815937042,
|
|
"num_tokens": 3642980.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.9457542419433596,
|
|
"epoch": 1.854301342615614,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004899555876200535,
|
|
"loss": 6.7248,
|
|
"mean_token_accuracy": 0.13515753149986268,
|
|
"num_tokens": 3653817.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.911198997497559,
|
|
"epoch": 1.8592739930382893,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004898398819105059,
|
|
"loss": 6.6605,
|
|
"mean_token_accuracy": 0.1311890348792076,
|
|
"num_tokens": 3664460.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.951674890518189,
|
|
"epoch": 1.8642466434609646,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004897235289728178,
|
|
"loss": 6.6822,
|
|
"mean_token_accuracy": 0.1262006141245365,
|
|
"num_tokens": 3674221.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.816601419448853,
|
|
"epoch": 1.8692192938836398,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004896065291575145,
|
|
"loss": 6.5641,
|
|
"mean_token_accuracy": 0.14460689574480057,
|
|
"num_tokens": 3683130.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.959670495986939,
|
|
"epoch": 1.8741919443063153,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004894888828170708,
|
|
"loss": 6.5195,
|
|
"mean_token_accuracy": 0.1417189985513687,
|
|
"num_tokens": 3694544.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.837589168548584,
|
|
"epoch": 1.8791645947289906,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004893705903059087,
|
|
"loss": 6.4985,
|
|
"mean_token_accuracy": 0.14005153775215148,
|
|
"num_tokens": 3704857.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.892556285858154,
|
|
"epoch": 1.8841372451516658,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004892516519803971,
|
|
"loss": 6.6087,
|
|
"mean_token_accuracy": 0.14154447615146637,
|
|
"num_tokens": 3714623.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.8302037715911865,
|
|
"epoch": 1.8891098955743413,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004891320681988504,
|
|
"loss": 6.5789,
|
|
"mean_token_accuracy": 0.14388583824038506,
|
|
"num_tokens": 3723349.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.875624370574951,
|
|
"epoch": 1.8940825459970165,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004890118393215277,
|
|
"loss": 6.5673,
|
|
"mean_token_accuracy": 0.1395200990140438,
|
|
"num_tokens": 3732990.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.831943941116333,
|
|
"epoch": 1.8990551964196918,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004888909657106311,
|
|
"loss": 6.5626,
|
|
"mean_token_accuracy": 0.14156354665756227,
|
|
"num_tokens": 3742343.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.791019105911255,
|
|
"epoch": 1.904027846842367,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004887694477303055,
|
|
"loss": 6.5747,
|
|
"mean_token_accuracy": 0.13516438007354736,
|
|
"num_tokens": 3752128.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.887428951263428,
|
|
"epoch": 1.9090004972650423,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004886472857466368,
|
|
"loss": 6.5083,
|
|
"mean_token_accuracy": 0.13921474814414977,
|
|
"num_tokens": 3760431.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.837686777114868,
|
|
"epoch": 1.9139731476877175,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004885244801276509,
|
|
"loss": 6.6186,
|
|
"mean_token_accuracy": 0.1353122554719448,
|
|
"num_tokens": 3769623.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.748737239837647,
|
|
"epoch": 1.9189457981103928,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004884010312433131,
|
|
"loss": 6.56,
|
|
"mean_token_accuracy": 0.13537675738334656,
|
|
"num_tokens": 3779383.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.937736177444458,
|
|
"epoch": 1.923918448533068,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048827693946552615,
|
|
"loss": 6.7036,
|
|
"mean_token_accuracy": 0.1315899521112442,
|
|
"num_tokens": 3788847.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.770579624176025,
|
|
"epoch": 1.9288910989557433,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00048815220516813004,
|
|
"loss": 6.5721,
|
|
"mean_token_accuracy": 0.14341088607907296,
|
|
"num_tokens": 3798711.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.862802124023437,
|
|
"epoch": 1.9338637493784185,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004880268287269,
|
|
"loss": 6.5292,
|
|
"mean_token_accuracy": 0.13987967222929001,
|
|
"num_tokens": 3808837.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.861500453948975,
|
|
"epoch": 1.938836399801094,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048790081051954605,
|
|
"loss": 6.5165,
|
|
"mean_token_accuracy": 0.1424727886915207,
|
|
"num_tokens": 3818261.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.892284822463989,
|
|
"epoch": 1.9438090502237693,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004877741509257116,
|
|
"loss": 6.5845,
|
|
"mean_token_accuracy": 0.1345815621316433,
|
|
"num_tokens": 3828146.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.942962217330932,
|
|
"epoch": 1.9487817006464445,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004876468503269721,
|
|
"loss": 6.8112,
|
|
"mean_token_accuracy": 0.12276574075222016,
|
|
"num_tokens": 3838658.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.920194530487061,
|
|
"epoch": 1.95375435106912,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048751890910683434,
|
|
"loss": 6.5607,
|
|
"mean_token_accuracy": 0.13691648170351983,
|
|
"num_tokens": 3849097.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.843836736679077,
|
|
"epoch": 1.9587270014917952,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048739032765073487,
|
|
"loss": 6.6902,
|
|
"mean_token_accuracy": 0.12868692278861998,
|
|
"num_tokens": 3858645.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.858337259292602,
|
|
"epoch": 1.9636996519144705,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048726110634603913,
|
|
"loss": 6.5588,
|
|
"mean_token_accuracy": 0.1408264495432377,
|
|
"num_tokens": 3869232.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.959062719345093,
|
|
"epoch": 1.9686723023371457,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00048713124558204013,
|
|
"loss": 6.6508,
|
|
"mean_token_accuracy": 0.13310638517141343,
|
|
"num_tokens": 3879453.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.837125778198242,
|
|
"epoch": 1.973644952759821,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004870007457499573,
|
|
"loss": 6.5694,
|
|
"mean_token_accuracy": 0.1347861558198929,
|
|
"num_tokens": 3888257.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.88454384803772,
|
|
"epoch": 1.9786176031824962,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048686960724293537,
|
|
"loss": 6.5899,
|
|
"mean_token_accuracy": 0.14540353715419768,
|
|
"num_tokens": 3897980.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.74349193572998,
|
|
"epoch": 1.9835902536051715,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.0004867378304560432,
|
|
"loss": 6.4244,
|
|
"mean_token_accuracy": 0.14290129393339157,
|
|
"num_tokens": 3907479.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.848113250732422,
|
|
"epoch": 1.9885629040278467,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004866054157862723,
|
|
"loss": 6.6451,
|
|
"mean_token_accuracy": 0.13774104043841362,
|
|
"num_tokens": 3917249.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.9885629040278467,
|
|
"eval_entropy": 6.649700598782658,
|
|
"eval_loss": 6.7694196701049805,
|
|
"eval_mean_token_accuracy": 0.12889868372939697,
|
|
"eval_num_tokens": 3917249.0,
|
|
"eval_runtime": 1.9283,
|
|
"eval_samples_per_second": 1796.927,
|
|
"eval_steps_per_second": 225.07,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.786346101760865,
|
|
"epoch": 1.993535554450522,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004864723636325361,
|
|
"loss": 6.5652,
|
|
"mean_token_accuracy": 0.13984513580799102,
|
|
"num_tokens": 3928388.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.837109565734863,
|
|
"epoch": 1.9985082048731975,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048633867439566855,
|
|
"loss": 6.596,
|
|
"mean_token_accuracy": 0.1373481847345829,
|
|
"num_tokens": 3937323.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.834231747521295,
|
|
"epoch": 2.002983590253605,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004862043484784227,
|
|
"loss": 6.4637,
|
|
"mean_token_accuracy": 0.14175044496854147,
|
|
"num_tokens": 3946815.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.854365348815918,
|
|
"epoch": 2.0079562406762803,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048606938628546975,
|
|
"loss": 6.2622,
|
|
"mean_token_accuracy": 0.1447312392294407,
|
|
"num_tokens": 3956125.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.850778341293335,
|
|
"epoch": 2.0129288910989556,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004859337882233978,
|
|
"loss": 6.2985,
|
|
"mean_token_accuracy": 0.14958048537373542,
|
|
"num_tokens": 3966543.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.765303611755371,
|
|
"epoch": 2.017901541521631,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004857975547007106,
|
|
"loss": 6.3062,
|
|
"mean_token_accuracy": 0.14360406473279,
|
|
"num_tokens": 3978200.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.765876579284668,
|
|
"epoch": 2.0228741919443065,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048566068612782595,
|
|
"loss": 6.1741,
|
|
"mean_token_accuracy": 0.15523267835378646,
|
|
"num_tokens": 3988331.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.83152813911438,
|
|
"epoch": 2.0278468423669818,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048552318291707533,
|
|
"loss": 6.2358,
|
|
"mean_token_accuracy": 0.14919300377368927,
|
|
"num_tokens": 3998976.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.69407811164856,
|
|
"epoch": 2.032819492789657,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00048538504548270177,
|
|
"loss": 6.2527,
|
|
"mean_token_accuracy": 0.15024952888488768,
|
|
"num_tokens": 4008354.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.780992650985718,
|
|
"epoch": 2.0377921432123323,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048524627424085913,
|
|
"loss": 6.1801,
|
|
"mean_token_accuracy": 0.15816001892089843,
|
|
"num_tokens": 4018045.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 6.677500295639038,
|
|
"epoch": 2.0427647936350075,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048510686960961057,
|
|
"loss": 6.2022,
|
|
"mean_token_accuracy": 0.15982799232006073,
|
|
"num_tokens": 4027884.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.735596656799316,
|
|
"epoch": 2.0477374440576828,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004849668320089276,
|
|
"loss": 6.2311,
|
|
"mean_token_accuracy": 0.1520911917090416,
|
|
"num_tokens": 4037866.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.80460262298584,
|
|
"epoch": 2.052710094480358,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00048482616186068826,
|
|
"loss": 6.2492,
|
|
"mean_token_accuracy": 0.1510556533932686,
|
|
"num_tokens": 4048126.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.760310173034668,
|
|
"epoch": 2.0576827449030333,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048468485958867664,
|
|
"loss": 6.2633,
|
|
"mean_token_accuracy": 0.1487021990120411,
|
|
"num_tokens": 4058179.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 6.716266441345215,
|
|
"epoch": 2.0626553953257085,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048454292561858083,
|
|
"loss": 6.1486,
|
|
"mean_token_accuracy": 0.15171891525387765,
|
|
"num_tokens": 4067479.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.7428733825683596,
|
|
"epoch": 2.0676280457483838,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004844003603779921,
|
|
"loss": 6.2617,
|
|
"mean_token_accuracy": 0.14702339172363282,
|
|
"num_tokens": 4076624.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.679452180862427,
|
|
"epoch": 2.072600696171059,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004842571642964036,
|
|
"loss": 6.209,
|
|
"mean_token_accuracy": 0.15349285006523133,
|
|
"num_tokens": 4085752.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.717970657348633,
|
|
"epoch": 2.0775733465937343,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048411333780520876,
|
|
"loss": 6.2313,
|
|
"mean_token_accuracy": 0.15353571027517318,
|
|
"num_tokens": 4095026.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 6.734153604507446,
|
|
"epoch": 2.0825459970164095,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048396888133770034,
|
|
"loss": 6.2344,
|
|
"mean_token_accuracy": 0.15159628987312318,
|
|
"num_tokens": 4104589.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.69730315208435,
|
|
"epoch": 2.087518647439085,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048382379532906897,
|
|
"loss": 6.1873,
|
|
"mean_token_accuracy": 0.1529018387198448,
|
|
"num_tokens": 4114550.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.781340885162353,
|
|
"epoch": 2.0924912978617605,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004836780802164016,
|
|
"loss": 6.2846,
|
|
"mean_token_accuracy": 0.1485740602016449,
|
|
"num_tokens": 4125491.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 6.686945152282715,
|
|
"epoch": 2.0974639482844357,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00048353173643868064,
|
|
"loss": 6.2123,
|
|
"mean_token_accuracy": 0.1518661692738533,
|
|
"num_tokens": 4134599.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.631251335144043,
|
|
"epoch": 2.102436598707111,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004833847644367826,
|
|
"loss": 6.1503,
|
|
"mean_token_accuracy": 0.15814711451530455,
|
|
"num_tokens": 4143640.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 6.729331207275391,
|
|
"epoch": 2.107409249129786,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00048323716465347615,
|
|
"loss": 6.2424,
|
|
"mean_token_accuracy": 0.15285539925098418,
|
|
"num_tokens": 4152558.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.579533481597901,
|
|
"epoch": 2.1123818995524615,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004830889375334215,
|
|
"loss": 6.0845,
|
|
"mean_token_accuracy": 0.17071812450885773,
|
|
"num_tokens": 4162238.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.705658960342407,
|
|
"epoch": 2.1173545499751367,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00048294008352316875,
|
|
"loss": 6.2123,
|
|
"mean_token_accuracy": 0.1533783420920372,
|
|
"num_tokens": 4171437.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.625134658813477,
|
|
"epoch": 2.122327200397812,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004827906030711564,
|
|
"loss": 6.2032,
|
|
"mean_token_accuracy": 0.15235737413167955,
|
|
"num_tokens": 4181584.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.571455287933349,
|
|
"epoch": 2.1272998508204872,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048264049662771053,
|
|
"loss": 6.0983,
|
|
"mean_token_accuracy": 0.16358734667301178,
|
|
"num_tokens": 4191756.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.571852302551269,
|
|
"epoch": 2.1322725012431625,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00048248976464504266,
|
|
"loss": 6.1405,
|
|
"mean_token_accuracy": 0.16373035609722136,
|
|
"num_tokens": 4202864.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.802804565429687,
|
|
"epoch": 2.1372451516658377,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048233840757724926,
|
|
"loss": 6.2308,
|
|
"mean_token_accuracy": 0.1441970646381378,
|
|
"num_tokens": 4212839.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.646758127212524,
|
|
"epoch": 2.142217802088513,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00048218642588030943,
|
|
"loss": 6.2435,
|
|
"mean_token_accuracy": 0.1501287817955017,
|
|
"num_tokens": 4223150.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.637703847885132,
|
|
"epoch": 2.1471904525111887,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00048203382001208453,
|
|
"loss": 6.2404,
|
|
"mean_token_accuracy": 0.15799741744995116,
|
|
"num_tokens": 4232206.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 6.726072549819946,
|
|
"epoch": 2.152163102933864,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00048188059043231585,
|
|
"loss": 6.2243,
|
|
"mean_token_accuracy": 0.15078145563602446,
|
|
"num_tokens": 4240992.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.647400331497193,
|
|
"epoch": 2.157135753356539,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00048172673760262414,
|
|
"loss": 6.217,
|
|
"mean_token_accuracy": 0.150467661768198,
|
|
"num_tokens": 4250371.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.669999122619629,
|
|
"epoch": 2.1621084037792144,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048157226198650735,
|
|
"loss": 6.2455,
|
|
"mean_token_accuracy": 0.15382370054721833,
|
|
"num_tokens": 4260606.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 6.700578451156616,
|
|
"epoch": 2.1670810542018897,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004814171640493398,
|
|
"loss": 6.2256,
|
|
"mean_token_accuracy": 0.15223996788263322,
|
|
"num_tokens": 4269753.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 6.557087182998657,
|
|
"epoch": 2.172053704624565,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004812614442583708,
|
|
"loss": 6.1563,
|
|
"mean_token_accuracy": 0.15995459109544755,
|
|
"num_tokens": 4279485.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.759015274047852,
|
|
"epoch": 2.17702635504724,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004811051030827227,
|
|
"loss": 6.2437,
|
|
"mean_token_accuracy": 0.1531550943851471,
|
|
"num_tokens": 4289491.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.675690078735352,
|
|
"epoch": 2.1819990054699154,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048094814099339006,
|
|
"loss": 6.2637,
|
|
"mean_token_accuracy": 0.1570417270064354,
|
|
"num_tokens": 4300536.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.752048015594482,
|
|
"epoch": 2.1869716558925907,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004807905584632381,
|
|
"loss": 6.3245,
|
|
"mean_token_accuracy": 0.14069870561361314,
|
|
"num_tokens": 4311804.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 6.821823930740356,
|
|
"epoch": 2.191944306315266,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048063235596700087,
|
|
"loss": 6.3016,
|
|
"mean_token_accuracy": 0.14367157444357873,
|
|
"num_tokens": 4320662.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 6.695243406295776,
|
|
"epoch": 2.196916956737941,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048047353398128056,
|
|
"loss": 6.2937,
|
|
"mean_token_accuracy": 0.14494634568691253,
|
|
"num_tokens": 4330414.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.608447122573852,
|
|
"epoch": 2.2018896071606164,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004803140929845452,
|
|
"loss": 6.251,
|
|
"mean_token_accuracy": 0.153412364423275,
|
|
"num_tokens": 4340478.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.637951469421386,
|
|
"epoch": 2.2068622575832917,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048015403345712796,
|
|
"loss": 6.2041,
|
|
"mean_token_accuracy": 0.1512879580259323,
|
|
"num_tokens": 4350688.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.590618228912353,
|
|
"epoch": 2.2118349080059674,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004799933558812252,
|
|
"loss": 6.1062,
|
|
"mean_token_accuracy": 0.16607880890369414,
|
|
"num_tokens": 4360554.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 6.603150033950806,
|
|
"epoch": 2.2168075584286426,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047983206074089546,
|
|
"loss": 6.1454,
|
|
"mean_token_accuracy": 0.1603748083114624,
|
|
"num_tokens": 4370723.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 6.6336830139160154,
|
|
"epoch": 2.221780208851318,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004796701485220576,
|
|
"loss": 6.2298,
|
|
"mean_token_accuracy": 0.15252456963062286,
|
|
"num_tokens": 4380383.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 6.695860147476196,
|
|
"epoch": 2.226752859273993,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004795076197124893,
|
|
"loss": 6.3543,
|
|
"mean_token_accuracy": 0.13700178861618043,
|
|
"num_tokens": 4390408.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.669200372695923,
|
|
"epoch": 2.2317255096966684,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047934447480182626,
|
|
"loss": 6.2209,
|
|
"mean_token_accuracy": 0.15164826661348343,
|
|
"num_tokens": 4400043.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.684255123138428,
|
|
"epoch": 2.2366981601193436,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004791807142815598,
|
|
"loss": 6.2802,
|
|
"mean_token_accuracy": 0.1460638202726841,
|
|
"num_tokens": 4409420.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.6182393550872805,
|
|
"epoch": 2.241670810542019,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047901633864503614,
|
|
"loss": 6.154,
|
|
"mean_token_accuracy": 0.1582044929265976,
|
|
"num_tokens": 4418742.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 6.669830656051635,
|
|
"epoch": 2.246643460964694,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047885134838745433,
|
|
"loss": 6.1942,
|
|
"mean_token_accuracy": 0.1549680531024933,
|
|
"num_tokens": 4427921.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 6.605356502532959,
|
|
"epoch": 2.2516161113873694,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004786857440058654,
|
|
"loss": 6.227,
|
|
"mean_token_accuracy": 0.15065814331173896,
|
|
"num_tokens": 4438029.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.599242210388184,
|
|
"epoch": 2.2565887618100446,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004785195259991701,
|
|
"loss": 6.2023,
|
|
"mean_token_accuracy": 0.15980394035577775,
|
|
"num_tokens": 4446967.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 6.687356567382812,
|
|
"epoch": 2.26156141223272,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000478352694868118,
|
|
"loss": 6.1413,
|
|
"mean_token_accuracy": 0.15389355272054672,
|
|
"num_tokens": 4456469.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 6.624385356903076,
|
|
"epoch": 2.266534062655395,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004781852511153058,
|
|
"loss": 6.2896,
|
|
"mean_token_accuracy": 0.1513574466109276,
|
|
"num_tokens": 4465949.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.633196783065796,
|
|
"epoch": 2.271506713078071,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00047801719524517573,
|
|
"loss": 6.0961,
|
|
"mean_token_accuracy": 0.15549132376909255,
|
|
"num_tokens": 4475460.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 6.649049139022827,
|
|
"epoch": 2.276479363500746,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047784852776401403,
|
|
"loss": 6.2228,
|
|
"mean_token_accuracy": 0.15281574726104735,
|
|
"num_tokens": 4485540.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.635147142410278,
|
|
"epoch": 2.2814520139234213,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004776792491799495,
|
|
"loss": 6.2084,
|
|
"mean_token_accuracy": 0.15474381297826767,
|
|
"num_tokens": 4495545.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 6.6261351108551025,
|
|
"epoch": 2.2864246643460966,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047750936000295213,
|
|
"loss": 6.1569,
|
|
"mean_token_accuracy": 0.15675314962863923,
|
|
"num_tokens": 4505527.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 6.606781053543091,
|
|
"epoch": 2.291397314768772,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047733886074483115,
|
|
"loss": 6.1583,
|
|
"mean_token_accuracy": 0.1597209393978119,
|
|
"num_tokens": 4515366.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 6.636414289474487,
|
|
"epoch": 2.296369965191447,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004771677519192339,
|
|
"loss": 6.2537,
|
|
"mean_token_accuracy": 0.15066712349653244,
|
|
"num_tokens": 4525273.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.618545055389404,
|
|
"epoch": 2.3013426156141223,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004769960340416439,
|
|
"loss": 6.2276,
|
|
"mean_token_accuracy": 0.15042769834399222,
|
|
"num_tokens": 4533568.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 6.635939359664917,
|
|
"epoch": 2.3063152660367976,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047682370762937976,
|
|
"loss": 6.3759,
|
|
"mean_token_accuracy": 0.1424851842224598,
|
|
"num_tokens": 4544508.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 6.718221950531006,
|
|
"epoch": 2.311287916459473,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00047665077320159315,
|
|
"loss": 6.1277,
|
|
"mean_token_accuracy": 0.15752690583467482,
|
|
"num_tokens": 4554586.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 6.56561427116394,
|
|
"epoch": 2.316260566882148,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00047647723127926765,
|
|
"loss": 6.163,
|
|
"mean_token_accuracy": 0.15369710326194763,
|
|
"num_tokens": 4563883.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 6.534911870956421,
|
|
"epoch": 2.3212332173048233,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004763030823852168,
|
|
"loss": 6.2094,
|
|
"mean_token_accuracy": 0.15631043761968613,
|
|
"num_tokens": 4573675.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 6.673135471343994,
|
|
"epoch": 2.3262058677274986,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004761283270440829,
|
|
"loss": 6.2008,
|
|
"mean_token_accuracy": 0.147063772380352,
|
|
"num_tokens": 4583617.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 6.624844121932983,
|
|
"epoch": 2.331178518150174,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004759529657823349,
|
|
"loss": 6.2377,
|
|
"mean_token_accuracy": 0.15442849099636077,
|
|
"num_tokens": 4593792.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 6.631576442718506,
|
|
"epoch": 2.336151168572849,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047577699912826755,
|
|
"loss": 6.2319,
|
|
"mean_token_accuracy": 0.14803201109170913,
|
|
"num_tokens": 4603324.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 6.6467750549316404,
|
|
"epoch": 2.341123818995525,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047560042761199915,
|
|
"loss": 6.189,
|
|
"mean_token_accuracy": 0.15746352225542068,
|
|
"num_tokens": 4613509.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 6.546010541915893,
|
|
"epoch": 2.3460964694182,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004754232517654704,
|
|
"loss": 6.1575,
|
|
"mean_token_accuracy": 0.15881496369838716,
|
|
"num_tokens": 4624282.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 6.515215110778809,
|
|
"epoch": 2.3510691198408753,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004752454721224424,
|
|
"loss": 6.1154,
|
|
"mean_token_accuracy": 0.15782492607831955,
|
|
"num_tokens": 4633217.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 6.521375703811645,
|
|
"epoch": 2.3560417702635505,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047506708921849535,
|
|
"loss": 6.1496,
|
|
"mean_token_accuracy": 0.15903475135564804,
|
|
"num_tokens": 4642824.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 6.567485666275024,
|
|
"epoch": 2.361014420686226,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004748881035910269,
|
|
"loss": 6.1948,
|
|
"mean_token_accuracy": 0.15451549589633942,
|
|
"num_tokens": 4651938.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 6.535520315170288,
|
|
"epoch": 2.365987071108901,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004747085157792504,
|
|
"loss": 6.1899,
|
|
"mean_token_accuracy": 0.15533617585897447,
|
|
"num_tokens": 4661695.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 6.660605573654175,
|
|
"epoch": 2.3709597215315763,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004745283263241933,
|
|
"loss": 6.2719,
|
|
"mean_token_accuracy": 0.15383488535881043,
|
|
"num_tokens": 4671354.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 6.569707155227661,
|
|
"epoch": 2.3759323719542516,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004743475357686957,
|
|
"loss": 6.1179,
|
|
"mean_token_accuracy": 0.17257502675056458,
|
|
"num_tokens": 4681795.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 6.576448965072632,
|
|
"epoch": 2.380905022376927,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047416614465740843,
|
|
"loss": 6.2053,
|
|
"mean_token_accuracy": 0.16284550875425338,
|
|
"num_tokens": 4691229.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 6.5818023681640625,
|
|
"epoch": 2.385877672799602,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047398415353679167,
|
|
"loss": 6.2956,
|
|
"mean_token_accuracy": 0.15139417573809624,
|
|
"num_tokens": 4701880.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.648142242431641,
|
|
"epoch": 2.3908503232222773,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004738015629551132,
|
|
"loss": 6.1957,
|
|
"mean_token_accuracy": 0.15121400207281113,
|
|
"num_tokens": 4712041.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.388457441329956,
|
|
"epoch": 2.395822973644953,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00047361837346244655,
|
|
"loss": 6.0013,
|
|
"mean_token_accuracy": 0.17091195285320282,
|
|
"num_tokens": 4722576.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 6.585583019256592,
|
|
"epoch": 2.4007956240676283,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004734345856106698,
|
|
"loss": 6.2275,
|
|
"mean_token_accuracy": 0.1544385224580765,
|
|
"num_tokens": 4732569.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 6.679563093185425,
|
|
"epoch": 2.4057682744903035,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00047325019995346355,
|
|
"loss": 6.2763,
|
|
"mean_token_accuracy": 0.15366682261228562,
|
|
"num_tokens": 4741973.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 6.709687852859497,
|
|
"epoch": 2.4107409249129788,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004730652170463091,
|
|
"loss": 6.2318,
|
|
"mean_token_accuracy": 0.15407460033893586,
|
|
"num_tokens": 4751550.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 6.556988286972046,
|
|
"epoch": 2.415713575335654,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004728796374464875,
|
|
"loss": 6.1722,
|
|
"mean_token_accuracy": 0.15943695977330208,
|
|
"num_tokens": 4761620.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 6.581428909301758,
|
|
"epoch": 2.4206862257583293,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047269346171307716,
|
|
"loss": 6.1569,
|
|
"mean_token_accuracy": 0.15552216917276382,
|
|
"num_tokens": 4771893.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.592879343032837,
|
|
"epoch": 2.4256588761810045,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004725066904069523,
|
|
"loss": 6.2301,
|
|
"mean_token_accuracy": 0.1504606083035469,
|
|
"num_tokens": 4781094.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 6.525106906890869,
|
|
"epoch": 2.4306315266036798,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004723193240907814,
|
|
"loss": 6.1943,
|
|
"mean_token_accuracy": 0.15088403299450875,
|
|
"num_tokens": 4790337.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 6.561110115051269,
|
|
"epoch": 2.435604177026355,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004721313633290257,
|
|
"loss": 6.227,
|
|
"mean_token_accuracy": 0.1542115777730942,
|
|
"num_tokens": 4801758.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 6.621707773208618,
|
|
"epoch": 2.4405768274490303,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000471942808687937,
|
|
"loss": 6.2985,
|
|
"mean_token_accuracy": 0.14450567066669465,
|
|
"num_tokens": 4811040.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 6.588228702545166,
|
|
"epoch": 2.4455494778717055,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047175366073555654,
|
|
"loss": 6.1466,
|
|
"mean_token_accuracy": 0.1542899042367935,
|
|
"num_tokens": 4821361.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 6.561674928665161,
|
|
"epoch": 2.4505221282943808,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004715639200417125,
|
|
"loss": 6.2059,
|
|
"mean_token_accuracy": 0.1495930217206478,
|
|
"num_tokens": 4830919.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 6.514748239517212,
|
|
"epoch": 2.455494778717056,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004713735871780192,
|
|
"loss": 6.0365,
|
|
"mean_token_accuracy": 0.16840022653341294,
|
|
"num_tokens": 4840364.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 6.575282430648803,
|
|
"epoch": 2.4604674291397313,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004711826627178747,
|
|
"loss": 6.209,
|
|
"mean_token_accuracy": 0.15822732895612718,
|
|
"num_tokens": 4849044.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 6.498941230773926,
|
|
"epoch": 2.4654400795624065,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047099114723645944,
|
|
"loss": 6.1913,
|
|
"mean_token_accuracy": 0.1515636995434761,
|
|
"num_tokens": 4859183.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 6.602888107299805,
|
|
"epoch": 2.470412729985082,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00047079904131073434,
|
|
"loss": 6.1665,
|
|
"mean_token_accuracy": 0.15605650171637536,
|
|
"num_tokens": 4869625.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 6.594534111022949,
|
|
"epoch": 2.4753853804077575,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004706063455194389,
|
|
"loss": 6.1743,
|
|
"mean_token_accuracy": 0.15970076620578766,
|
|
"num_tokens": 4880089.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 6.5573179721832275,
|
|
"epoch": 2.4803580308304327,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047041306044308986,
|
|
"loss": 6.2642,
|
|
"mean_token_accuracy": 0.15200921297073364,
|
|
"num_tokens": 4890673.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 6.547742938995361,
|
|
"epoch": 2.485330681253108,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004702191866639792,
|
|
"loss": 6.1272,
|
|
"mean_token_accuracy": 0.15250463038682938,
|
|
"num_tokens": 4899444.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 6.5435985088348385,
|
|
"epoch": 2.490303331675783,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004700247247661726,
|
|
"loss": 6.2677,
|
|
"mean_token_accuracy": 0.14674118012189866,
|
|
"num_tokens": 4909186.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 6.488081264495849,
|
|
"epoch": 2.4952759820984585,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004698296753355072,
|
|
"loss": 6.0518,
|
|
"mean_token_accuracy": 0.17193612605333328,
|
|
"num_tokens": 4918886.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 6.6091328144073485,
|
|
"epoch": 2.5002486325211337,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004696340389595903,
|
|
"loss": 6.2436,
|
|
"mean_token_accuracy": 0.15216250270605086,
|
|
"num_tokens": 4929157.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 6.532401132583618,
|
|
"epoch": 2.505221282943809,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046943781622779733,
|
|
"loss": 6.0798,
|
|
"mean_token_accuracy": 0.16211776733398436,
|
|
"num_tokens": 4938127.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 6.5577586650848385,
|
|
"epoch": 2.510193933366484,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046924100773127055,
|
|
"loss": 6.0978,
|
|
"mean_token_accuracy": 0.1667572245001793,
|
|
"num_tokens": 4947849.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 6.468730545043945,
|
|
"epoch": 2.5151665837891595,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046904361406291633,
|
|
"loss": 6.1613,
|
|
"mean_token_accuracy": 0.15513846427202224,
|
|
"num_tokens": 4957046.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 6.510133409500122,
|
|
"epoch": 2.520139234211835,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004688456358174044,
|
|
"loss": 6.1811,
|
|
"mean_token_accuracy": 0.15946740806102752,
|
|
"num_tokens": 4967128.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 6.611709308624268,
|
|
"epoch": 2.5251118846345104,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004686470735911653,
|
|
"loss": 6.2471,
|
|
"mean_token_accuracy": 0.15146666690707206,
|
|
"num_tokens": 4976817.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 6.586194801330566,
|
|
"epoch": 2.5300845350571857,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004684479279823891,
|
|
"loss": 6.2364,
|
|
"mean_token_accuracy": 0.15271722450852393,
|
|
"num_tokens": 4985908.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 6.584090709686279,
|
|
"epoch": 2.535057185479861,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004682481995910232,
|
|
"loss": 6.1299,
|
|
"mean_token_accuracy": 0.1592479467391968,
|
|
"num_tokens": 4995218.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 6.54912633895874,
|
|
"epoch": 2.540029835902536,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00046804788901877084,
|
|
"loss": 6.2733,
|
|
"mean_token_accuracy": 0.14976896345615387,
|
|
"num_tokens": 5004197.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 6.571247577667236,
|
|
"epoch": 2.5450024863252114,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00046784699686908886,
|
|
"loss": 6.2014,
|
|
"mean_token_accuracy": 0.14773830771446228,
|
|
"num_tokens": 5015408.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 6.5191038131713865,
|
|
"epoch": 2.5499751367478867,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046764552374718656,
|
|
"loss": 6.1946,
|
|
"mean_token_accuracy": 0.1547203317284584,
|
|
"num_tokens": 5025138.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 6.53830041885376,
|
|
"epoch": 2.554947787170562,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046744347026002315,
|
|
"loss": 6.2125,
|
|
"mean_token_accuracy": 0.15321909114718438,
|
|
"num_tokens": 5034638.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 6.625218248367309,
|
|
"epoch": 2.559920437593237,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004672408370163066,
|
|
"loss": 6.3232,
|
|
"mean_token_accuracy": 0.14830388128757477,
|
|
"num_tokens": 5044826.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 6.622767782211303,
|
|
"epoch": 2.5648930880159124,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000467037624626491,
|
|
"loss": 6.3628,
|
|
"mean_token_accuracy": 0.1464301362633705,
|
|
"num_tokens": 5054914.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 6.442788410186767,
|
|
"epoch": 2.5698657384385877,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046683383370277554,
|
|
"loss": 6.0994,
|
|
"mean_token_accuracy": 0.16794165521860122,
|
|
"num_tokens": 5066689.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 6.570945453643799,
|
|
"epoch": 2.574838388861263,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046662946485910226,
|
|
"loss": 6.2753,
|
|
"mean_token_accuracy": 0.15411486029624938,
|
|
"num_tokens": 5076170.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 6.588211011886597,
|
|
"epoch": 2.579811039283938,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000466424518711154,
|
|
"loss": 6.1203,
|
|
"mean_token_accuracy": 0.15667143762111663,
|
|
"num_tokens": 5085433.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 6.550396156311035,
|
|
"epoch": 2.5847836897066134,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004662189958763531,
|
|
"loss": 6.1254,
|
|
"mean_token_accuracy": 0.16099537014961243,
|
|
"num_tokens": 5095692.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 6.45124077796936,
|
|
"epoch": 2.5897563401292887,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004660128969738591,
|
|
"loss": 6.2046,
|
|
"mean_token_accuracy": 0.14970657527446746,
|
|
"num_tokens": 5105037.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 6.500788068771362,
|
|
"epoch": 2.594728990551964,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004658062226245668,
|
|
"loss": 6.157,
|
|
"mean_token_accuracy": 0.15702567845582963,
|
|
"num_tokens": 5113868.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 6.678019332885742,
|
|
"epoch": 2.5997016409746396,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004655989734511052,
|
|
"loss": 6.1979,
|
|
"mean_token_accuracy": 0.15999541953206062,
|
|
"num_tokens": 5123136.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 6.514936304092407,
|
|
"epoch": 2.604674291397315,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004653911500778343,
|
|
"loss": 6.0718,
|
|
"mean_token_accuracy": 0.1664543554186821,
|
|
"num_tokens": 5133869.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 6.52581787109375,
|
|
"epoch": 2.60964694181999,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004651827531308443,
|
|
"loss": 6.1738,
|
|
"mean_token_accuracy": 0.1572820320725441,
|
|
"num_tokens": 5143119.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 6.386371660232544,
|
|
"epoch": 2.6146195922426654,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004649737832379534,
|
|
"loss": 5.999,
|
|
"mean_token_accuracy": 0.17316098213195802,
|
|
"num_tokens": 5152167.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 6.5685680389404295,
|
|
"epoch": 2.6195922426653406,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004647642410287057,
|
|
"loss": 6.26,
|
|
"mean_token_accuracy": 0.154359133541584,
|
|
"num_tokens": 5161962.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 6.55338191986084,
|
|
"epoch": 2.624564893088016,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004645541271343696,
|
|
"loss": 6.0962,
|
|
"mean_token_accuracy": 0.16010626405477524,
|
|
"num_tokens": 5171637.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 6.447772645950318,
|
|
"epoch": 2.629537543510691,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00046434344218793566,
|
|
"loss": 6.074,
|
|
"mean_token_accuracy": 0.16469110697507858,
|
|
"num_tokens": 5181839.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 6.47948522567749,
|
|
"epoch": 2.6345101939333664,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00046413218682411485,
|
|
"loss": 6.1592,
|
|
"mean_token_accuracy": 0.15918211489915848,
|
|
"num_tokens": 5192237.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 6.6197569370269775,
|
|
"epoch": 2.6394828443560416,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004639203616793365,
|
|
"loss": 6.3135,
|
|
"mean_token_accuracy": 0.14707784727215767,
|
|
"num_tokens": 5203879.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 6.531035852432251,
|
|
"epoch": 2.6444554947787173,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046370796739174657,
|
|
"loss": 6.1748,
|
|
"mean_token_accuracy": 0.153842993080616,
|
|
"num_tokens": 5213456.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 6.463725233078003,
|
|
"epoch": 2.6494281452013926,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046349500460120556,
|
|
"loss": 6.1195,
|
|
"mean_token_accuracy": 0.1572451762855053,
|
|
"num_tokens": 5222234.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 6.520852851867676,
|
|
"epoch": 2.654400795624068,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00046328147394928684,
|
|
"loss": 6.1511,
|
|
"mean_token_accuracy": 0.15687175691127778,
|
|
"num_tokens": 5230989.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 6.555701732635498,
|
|
"epoch": 2.659373446046743,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004630673760792741,
|
|
"loss": 6.1807,
|
|
"mean_token_accuracy": 0.15620526373386384,
|
|
"num_tokens": 5240607.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 6.560945510864258,
|
|
"epoch": 2.6643460964694183,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004628527116361603,
|
|
"loss": 6.2841,
|
|
"mean_token_accuracy": 0.14436158984899522,
|
|
"num_tokens": 5250837.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 6.592908477783203,
|
|
"epoch": 2.6693187468920936,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004626374812666449,
|
|
"loss": 6.1824,
|
|
"mean_token_accuracy": 0.1583193361759186,
|
|
"num_tokens": 5260542.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 6.5208680629730225,
|
|
"epoch": 2.674291397314769,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00046242168561913256,
|
|
"loss": 6.0766,
|
|
"mean_token_accuracy": 0.1593346729874611,
|
|
"num_tokens": 5272327.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 6.5047039031982425,
|
|
"epoch": 2.679264047737444,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00046220532534373067,
|
|
"loss": 6.1937,
|
|
"mean_token_accuracy": 0.1573296770453453,
|
|
"num_tokens": 5281407.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 6.588146162033081,
|
|
"epoch": 2.6842366981601193,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004619884010922477,
|
|
"loss": 6.165,
|
|
"mean_token_accuracy": 0.14946887791156768,
|
|
"num_tokens": 5291617.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 6.57141146659851,
|
|
"epoch": 2.6892093485827946,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004617709135181913,
|
|
"loss": 6.2239,
|
|
"mean_token_accuracy": 0.15283752083778382,
|
|
"num_tokens": 5301174.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 6.544818353652954,
|
|
"epoch": 2.69418199900547,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00046155286327676594,
|
|
"loss": 6.2072,
|
|
"mean_token_accuracy": 0.15086385756731033,
|
|
"num_tokens": 5311211.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 6.4996692657470705,
|
|
"epoch": 2.699154649428145,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00046133425102487134,
|
|
"loss": 6.2101,
|
|
"mean_token_accuracy": 0.15241022258996964,
|
|
"num_tokens": 5321343.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 6.487451696395874,
|
|
"epoch": 2.7041272998508203,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046111507742110035,
|
|
"loss": 6.158,
|
|
"mean_token_accuracy": 0.16770548671483992,
|
|
"num_tokens": 5330849.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 6.549419212341308,
|
|
"epoch": 2.7090999502734956,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000460895343125737,
|
|
"loss": 6.2515,
|
|
"mean_token_accuracy": 0.15465343296527861,
|
|
"num_tokens": 5340836.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 6.551375055313111,
|
|
"epoch": 2.714072600696171,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00046067504880075425,
|
|
"loss": 6.1325,
|
|
"mean_token_accuracy": 0.1601417377591133,
|
|
"num_tokens": 5351093.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 6.489064788818359,
|
|
"epoch": 2.719045251118846,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004604541951098122,
|
|
"loss": 6.1826,
|
|
"mean_token_accuracy": 0.15869345515966415,
|
|
"num_tokens": 5360859.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 6.417691087722778,
|
|
"epoch": 2.724017901541522,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004602327827182565,
|
|
"loss": 5.9519,
|
|
"mean_token_accuracy": 0.16777553856372834,
|
|
"num_tokens": 5370591.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 6.446378374099732,
|
|
"epoch": 2.728990551964197,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004600108122931156,
|
|
"loss": 6.0711,
|
|
"mean_token_accuracy": 0.16241830289363862,
|
|
"num_tokens": 5380019.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 6.476145648956299,
|
|
"epoch": 2.7339632023868723,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00045978828450309907,
|
|
"loss": 6.163,
|
|
"mean_token_accuracy": 0.1511037155985832,
|
|
"num_tokens": 5389558.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 6.521407842636108,
|
|
"epoch": 2.7389358528095475,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004595652000185958,
|
|
"loss": 6.1939,
|
|
"mean_token_accuracy": 0.16077455431222915,
|
|
"num_tokens": 5398305.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 6.57568507194519,
|
|
"epoch": 2.743908503232223,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00045934155951167163,
|
|
"loss": 6.2703,
|
|
"mean_token_accuracy": 0.14973231852054597,
|
|
"num_tokens": 5406886.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 6.539137601852417,
|
|
"epoch": 2.748881153654898,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004591173636560675,
|
|
"loss": 6.2566,
|
|
"mean_token_accuracy": 0.15772698521614076,
|
|
"num_tokens": 5416404.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 6.531837272644043,
|
|
"epoch": 2.7538538040775733,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004588926131271974,
|
|
"loss": 6.142,
|
|
"mean_token_accuracy": 0.16053656488656998,
|
|
"num_tokens": 5425746.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 6.543280601501465,
|
|
"epoch": 2.7588264545002485,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045866730860214655,
|
|
"loss": 6.2102,
|
|
"mean_token_accuracy": 0.15724788457155228,
|
|
"num_tokens": 5436089.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 6.4683390140533445,
|
|
"epoch": 2.763799104922924,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004584414507596687,
|
|
"loss": 6.1381,
|
|
"mean_token_accuracy": 0.1579010307788849,
|
|
"num_tokens": 5446270.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 6.528982496261596,
|
|
"epoch": 2.768771755345599,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045821504028018494,
|
|
"loss": 6.1562,
|
|
"mean_token_accuracy": 0.1572088345885277,
|
|
"num_tokens": 5455476.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 6.540053510665894,
|
|
"epoch": 2.7737444057682747,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004579880778457811,
|
|
"loss": 6.1733,
|
|
"mean_token_accuracy": 0.15304938405752183,
|
|
"num_tokens": 5463813.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 6.500091075897217,
|
|
"epoch": 2.77871705619095,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004577605641402056,
|
|
"loss": 6.1627,
|
|
"mean_token_accuracy": 0.15459904372692107,
|
|
"num_tokens": 5473127.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 6.42025671005249,
|
|
"epoch": 2.7836897066136252,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004575324998488681,
|
|
"loss": 6.1254,
|
|
"mean_token_accuracy": 0.15737488716840745,
|
|
"num_tokens": 5481595.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 6.455853414535523,
|
|
"epoch": 2.7886623570363005,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004573038856588364,
|
|
"loss": 6.1726,
|
|
"mean_token_accuracy": 0.15274386778473853,
|
|
"num_tokens": 5491495.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 6.465582275390625,
|
|
"epoch": 2.7936350074589757,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004570747222588354,
|
|
"loss": 6.1428,
|
|
"mean_token_accuracy": 0.15832369029521942,
|
|
"num_tokens": 5500702.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 6.519288682937622,
|
|
"epoch": 2.798607657881651,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045684501033924443,
|
|
"loss": 6.0806,
|
|
"mean_token_accuracy": 0.16239927038550378,
|
|
"num_tokens": 5510223.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 6.539275407791138,
|
|
"epoch": 2.8035803083043263,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045661475059209503,
|
|
"loss": 6.2583,
|
|
"mean_token_accuracy": 0.1480315700173378,
|
|
"num_tokens": 5522063.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 6.506518888473511,
|
|
"epoch": 2.8085529587270015,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004563839437110695,
|
|
"loss": 6.1465,
|
|
"mean_token_accuracy": 0.16144824028015137,
|
|
"num_tokens": 5531323.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 6.538668441772461,
|
|
"epoch": 2.8135256091496768,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045615259039149826,
|
|
"loss": 6.1277,
|
|
"mean_token_accuracy": 0.16058994978666305,
|
|
"num_tokens": 5540637.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 6.4470171451568605,
|
|
"epoch": 2.818498259572352,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045592069133035783,
|
|
"loss": 6.0692,
|
|
"mean_token_accuracy": 0.16713782250881196,
|
|
"num_tokens": 5550992.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 6.438177967071534,
|
|
"epoch": 2.8234709099950273,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004556882472262691,
|
|
"loss": 6.116,
|
|
"mean_token_accuracy": 0.15509629845619202,
|
|
"num_tokens": 5561140.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 6.504058694839477,
|
|
"epoch": 2.8284435604177025,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004554552587794948,
|
|
"loss": 6.1436,
|
|
"mean_token_accuracy": 0.1571763977408409,
|
|
"num_tokens": 5570457.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 6.507036542892456,
|
|
"epoch": 2.8334162108403778,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004552217266919375,
|
|
"loss": 6.3151,
|
|
"mean_token_accuracy": 0.14259750321507453,
|
|
"num_tokens": 5579950.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 6.528378200531006,
|
|
"epoch": 2.838388861263053,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004549876516671376,
|
|
"loss": 6.1906,
|
|
"mean_token_accuracy": 0.15532433539628981,
|
|
"num_tokens": 5589441.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 6.481816625595092,
|
|
"epoch": 2.8433615116857283,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045475303441027114,
|
|
"loss": 6.0349,
|
|
"mean_token_accuracy": 0.16319974213838578,
|
|
"num_tokens": 5598583.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 6.461427354812622,
|
|
"epoch": 2.8483341621084035,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004545178756281479,
|
|
"loss": 6.1134,
|
|
"mean_token_accuracy": 0.16304348409175873,
|
|
"num_tokens": 5609198.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 6.508690309524536,
|
|
"epoch": 2.853306812531079,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00045428217602920876,
|
|
"loss": 6.1178,
|
|
"mean_token_accuracy": 0.15762704610824585,
|
|
"num_tokens": 5620425.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 6.431004238128662,
|
|
"epoch": 2.8582794629537545,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00045404593632352385,
|
|
"loss": 5.949,
|
|
"mean_token_accuracy": 0.17371091842651368,
|
|
"num_tokens": 5629578.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 6.507540988922119,
|
|
"epoch": 2.8632521133764297,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004538091572227909,
|
|
"loss": 6.164,
|
|
"mean_token_accuracy": 0.1495412528514862,
|
|
"num_tokens": 5639817.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 6.542991638183594,
|
|
"epoch": 2.868224763799105,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004535718394403319,
|
|
"loss": 6.2686,
|
|
"mean_token_accuracy": 0.1552718386054039,
|
|
"num_tokens": 5649130.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 6.511079549789429,
|
|
"epoch": 2.87319741422178,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004533339836910924,
|
|
"loss": 6.1887,
|
|
"mean_token_accuracy": 0.15932107865810394,
|
|
"num_tokens": 5659672.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 6.521791791915893,
|
|
"epoch": 2.8781700646444555,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004530955906916382,
|
|
"loss": 6.2316,
|
|
"mean_token_accuracy": 0.15385505557060242,
|
|
"num_tokens": 5669491.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 6.33027663230896,
|
|
"epoch": 2.8831427150671307,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004528566611601536,
|
|
"loss": 5.9379,
|
|
"mean_token_accuracy": 0.17895400673151016,
|
|
"num_tokens": 5679444.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 6.5365081310272215,
|
|
"epoch": 2.888115365489806,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004526171958164396,
|
|
"loss": 6.2679,
|
|
"mean_token_accuracy": 0.14566816091537477,
|
|
"num_tokens": 5690997.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 6.535708999633789,
|
|
"epoch": 2.893088015912481,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045237719538191107,
|
|
"loss": 6.176,
|
|
"mean_token_accuracy": 0.15118112415075302,
|
|
"num_tokens": 5700727.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 6.544662761688232,
|
|
"epoch": 2.898060666335157,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000452136660579595,
|
|
"loss": 6.1516,
|
|
"mean_token_accuracy": 0.15779981017112732,
|
|
"num_tokens": 5710809.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 6.334124994277954,
|
|
"epoch": 2.903033316757832,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045189559213412825,
|
|
"loss": 6.0542,
|
|
"mean_token_accuracy": 0.17352839410305024,
|
|
"num_tokens": 5720286.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 6.439293432235718,
|
|
"epoch": 2.9080059671805074,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004516539907717553,
|
|
"loss": 6.1249,
|
|
"mean_token_accuracy": 0.1603862538933754,
|
|
"num_tokens": 5730627.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 6.449020481109619,
|
|
"epoch": 2.9129786176031827,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045141185722032617,
|
|
"loss": 6.0337,
|
|
"mean_token_accuracy": 0.1665644347667694,
|
|
"num_tokens": 5739752.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 6.484323596954345,
|
|
"epoch": 2.917951268025858,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00045116919220929407,
|
|
"loss": 6.1221,
|
|
"mean_token_accuracy": 0.15985846668481826,
|
|
"num_tokens": 5748277.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 6.361673402786255,
|
|
"epoch": 2.922923918448533,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0004509259964697133,
|
|
"loss": 5.946,
|
|
"mean_token_accuracy": 0.17565756142139435,
|
|
"num_tokens": 5758632.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 6.4824365139007565,
|
|
"epoch": 2.9278965688712084,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000450682270734237,
|
|
"loss": 6.1104,
|
|
"mean_token_accuracy": 0.1635764867067337,
|
|
"num_tokens": 5769107.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 6.458119440078735,
|
|
"epoch": 2.9328692192938837,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004504380157371152,
|
|
"loss": 6.1332,
|
|
"mean_token_accuracy": 0.1544796958565712,
|
|
"num_tokens": 5779051.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 6.4672284603118895,
|
|
"epoch": 2.937841869716559,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00045019323221419204,
|
|
"loss": 6.1155,
|
|
"mean_token_accuracy": 0.15404862463474273,
|
|
"num_tokens": 5788826.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 6.439382600784302,
|
|
"epoch": 2.942814520139234,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00044994792090290425,
|
|
"loss": 6.1522,
|
|
"mean_token_accuracy": 0.15480929836630822,
|
|
"num_tokens": 5798003.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 6.443703985214233,
|
|
"epoch": 2.9477871705619094,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00044970208254227836,
|
|
"loss": 6.1159,
|
|
"mean_token_accuracy": 0.15876110196113585,
|
|
"num_tokens": 5807215.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 6.4770910263061525,
|
|
"epoch": 2.9527598209845847,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044945571787292866,
|
|
"loss": 6.1203,
|
|
"mean_token_accuracy": 0.16003868877887725,
|
|
"num_tokens": 5816629.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 6.463996458053589,
|
|
"epoch": 2.95773247140726,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004492088276370552,
|
|
"loss": 6.1088,
|
|
"mean_token_accuracy": 0.16100065037608147,
|
|
"num_tokens": 5825840.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 6.571723413467407,
|
|
"epoch": 2.962705121829935,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004489614125784411,
|
|
"loss": 6.2822,
|
|
"mean_token_accuracy": 0.14693968445062638,
|
|
"num_tokens": 5835754.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 6.421692276000977,
|
|
"epoch": 2.9676777722526104,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004487134734424509,
|
|
"loss": 6.1508,
|
|
"mean_token_accuracy": 0.15898758172988892,
|
|
"num_tokens": 5846062.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 6.513372230529785,
|
|
"epoch": 2.9726504226752857,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004484650109760276,
|
|
"loss": 6.2097,
|
|
"mean_token_accuracy": 0.15308655053377151,
|
|
"num_tokens": 5855589.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 6.508023929595947,
|
|
"epoch": 2.9776230730979614,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004482160259276911,
|
|
"loss": 6.0059,
|
|
"mean_token_accuracy": 0.17406210154294968,
|
|
"num_tokens": 5864917.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 6.530429077148438,
|
|
"epoch": 2.9825957235206366,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004479665190475355,
|
|
"loss": 6.1399,
|
|
"mean_token_accuracy": 0.16042668521404266,
|
|
"num_tokens": 5875530.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.9825957235206366,
|
|
"eval_entropy": 6.320197295483356,
|
|
"eval_loss": 6.556489944458008,
|
|
"eval_mean_token_accuracy": 0.14044848300756946,
|
|
"eval_num_tokens": 5875530.0,
|
|
"eval_runtime": 1.9162,
|
|
"eval_samples_per_second": 1808.235,
|
|
"eval_steps_per_second": 226.486,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 6.543712568283081,
|
|
"epoch": 2.987568373943312,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00044771649108722695,
|
|
"loss": 6.2039,
|
|
"mean_token_accuracy": 0.15643370449542998,
|
|
"num_tokens": 5885363.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 6.542303466796875,
|
|
"epoch": 2.992541024365987,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004474659428000014,
|
|
"loss": 6.2023,
|
|
"mean_token_accuracy": 0.15645717680454255,
|
|
"num_tokens": 5895318.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 6.462405014038086,
|
|
"epoch": 2.9975136747886624,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00044721487494066254,
|
|
"loss": 6.2252,
|
|
"mean_token_accuracy": 0.15860049799084663,
|
|
"num_tokens": 5905742.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 6.471013334062365,
|
|
"epoch": 3.00198906016907,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000446963288265579,
|
|
"loss": 5.9789,
|
|
"mean_token_accuracy": 0.1748280723889669,
|
|
"num_tokens": 5913844.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 6.478049230575562,
|
|
"epoch": 3.0069617105917454,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00044671118353268257,
|
|
"loss": 5.8438,
|
|
"mean_token_accuracy": 0.1641690358519554,
|
|
"num_tokens": 5923724.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 6.416875886917114,
|
|
"epoch": 3.0119343610144207,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00044645856150146593,
|
|
"loss": 5.8634,
|
|
"mean_token_accuracy": 0.16798260807991028,
|
|
"num_tokens": 5933787.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 6.533902311325074,
|
|
"epoch": 3.016907011437096,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044620542293297974,
|
|
"loss": 5.8488,
|
|
"mean_token_accuracy": 0.16344929188489915,
|
|
"num_tokens": 5944186.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 6.450361442565918,
|
|
"epoch": 3.021879661859771,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004459517685898312,
|
|
"loss": 5.861,
|
|
"mean_token_accuracy": 0.1669216573238373,
|
|
"num_tokens": 5954465.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 6.597113370895386,
|
|
"epoch": 3.0268523122824464,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00044569759923618113,
|
|
"loss": 5.8856,
|
|
"mean_token_accuracy": 0.16505503058433532,
|
|
"num_tokens": 5965047.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 6.400167417526245,
|
|
"epoch": 3.0318249627051217,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004454429156377419,
|
|
"loss": 5.7082,
|
|
"mean_token_accuracy": 0.18034494966268538,
|
|
"num_tokens": 5974855.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 6.318440580368042,
|
|
"epoch": 3.036797613127797,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004451877185617751,
|
|
"loss": 5.7092,
|
|
"mean_token_accuracy": 0.17241442650556565,
|
|
"num_tokens": 5985965.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 6.519586038589478,
|
|
"epoch": 3.041770263550472,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004449320087770894,
|
|
"loss": 5.9224,
|
|
"mean_token_accuracy": 0.16718820482492447,
|
|
"num_tokens": 5996646.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 6.430587863922119,
|
|
"epoch": 3.046742913973148,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004446757870540376,
|
|
"loss": 5.7209,
|
|
"mean_token_accuracy": 0.18517163544893264,
|
|
"num_tokens": 6005825.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 6.301305437088013,
|
|
"epoch": 3.051715564395823,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004444190541645153,
|
|
"loss": 5.654,
|
|
"mean_token_accuracy": 0.18023774027824402,
|
|
"num_tokens": 6016909.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 6.412121677398682,
|
|
"epoch": 3.0566882148184984,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00044416181088195776,
|
|
"loss": 5.8561,
|
|
"mean_token_accuracy": 0.16778892725706102,
|
|
"num_tokens": 6025511.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 6.455734491348267,
|
|
"epoch": 3.0616608652411736,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004439040579813379,
|
|
"loss": 5.7884,
|
|
"mean_token_accuracy": 0.1773159459233284,
|
|
"num_tokens": 6035114.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 6.4119494438171385,
|
|
"epoch": 3.066633515663849,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00044364579623916385,
|
|
"loss": 5.8111,
|
|
"mean_token_accuracy": 0.1709066703915596,
|
|
"num_tokens": 6044963.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 6.320486927032471,
|
|
"epoch": 3.071606166086524,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004433870264334767,
|
|
"loss": 5.6496,
|
|
"mean_token_accuracy": 0.19023120850324632,
|
|
"num_tokens": 6054660.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 6.408462810516357,
|
|
"epoch": 3.0765788165091994,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004431277493438484,
|
|
"loss": 5.7656,
|
|
"mean_token_accuracy": 0.1724860906600952,
|
|
"num_tokens": 6064288.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 6.379149484634399,
|
|
"epoch": 3.0815514669318746,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004428679657513789,
|
|
"loss": 5.7784,
|
|
"mean_token_accuracy": 0.16862039268016815,
|
|
"num_tokens": 6074133.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 6.35357437133789,
|
|
"epoch": 3.08652411735455,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004426076764386939,
|
|
"loss": 5.8141,
|
|
"mean_token_accuracy": 0.17215913981199266,
|
|
"num_tokens": 6084314.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 6.342261219024659,
|
|
"epoch": 3.091496767777225,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044234688218994293,
|
|
"loss": 5.7642,
|
|
"mean_token_accuracy": 0.17807788103818895,
|
|
"num_tokens": 6094616.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 6.401951313018799,
|
|
"epoch": 3.0964694181999004,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004420855837907967,
|
|
"loss": 5.803,
|
|
"mean_token_accuracy": 0.16995078772306443,
|
|
"num_tokens": 6104167.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 6.245429515838623,
|
|
"epoch": 3.1014420686225757,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00044182378202844427,
|
|
"loss": 5.5667,
|
|
"mean_token_accuracy": 0.19368275851011277,
|
|
"num_tokens": 6113970.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 6.413508367538452,
|
|
"epoch": 3.106414719045251,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00044156147769159183,
|
|
"loss": 5.7748,
|
|
"mean_token_accuracy": 0.170443195104599,
|
|
"num_tokens": 6123482.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 6.342976856231689,
|
|
"epoch": 3.1113873694679266,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004412986715704591,
|
|
"loss": 5.7363,
|
|
"mean_token_accuracy": 0.17443411201238632,
|
|
"num_tokens": 6133682.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 6.343202543258667,
|
|
"epoch": 3.116360019890602,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004410353644567777,
|
|
"loss": 5.7714,
|
|
"mean_token_accuracy": 0.17632579505443574,
|
|
"num_tokens": 6143876.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 6.401024723052979,
|
|
"epoch": 3.121332670313277,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044077155714378836,
|
|
"loss": 5.8809,
|
|
"mean_token_accuracy": 0.17110844552516938,
|
|
"num_tokens": 6153615.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 6.372533798217773,
|
|
"epoch": 3.1263053207359524,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004405072504262391,
|
|
"loss": 5.7717,
|
|
"mean_token_accuracy": 0.1809968501329422,
|
|
"num_tokens": 6163572.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 6.338580369949341,
|
|
"epoch": 3.1312779711586276,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000440242445100382,
|
|
"loss": 5.7526,
|
|
"mean_token_accuracy": 0.17514811754226683,
|
|
"num_tokens": 6173085.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 6.313119173049927,
|
|
"epoch": 3.136250621581303,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00043997714196397143,
|
|
"loss": 5.8182,
|
|
"mean_token_accuracy": 0.1669571615755558,
|
|
"num_tokens": 6183124.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 6.477424287796021,
|
|
"epoch": 3.141223272003978,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00043971134181626173,
|
|
"loss": 5.8487,
|
|
"mean_token_accuracy": 0.17093345373868943,
|
|
"num_tokens": 6192817.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 6.369207191467285,
|
|
"epoch": 3.1461959224266534,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00043944504545800396,
|
|
"loss": 5.8123,
|
|
"mean_token_accuracy": 0.17716975510120392,
|
|
"num_tokens": 6203643.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 6.388614988327026,
|
|
"epoch": 3.1511685728493286,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004391782536914447,
|
|
"loss": 5.8151,
|
|
"mean_token_accuracy": 0.17258669584989547,
|
|
"num_tokens": 6213712.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 6.3534238815307615,
|
|
"epoch": 3.156141223272004,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00043891096732032257,
|
|
"loss": 5.8964,
|
|
"mean_token_accuracy": 0.168116694688797,
|
|
"num_tokens": 6223962.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 6.374617624282837,
|
|
"epoch": 3.161113873694679,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00043864318714986625,
|
|
"loss": 5.8928,
|
|
"mean_token_accuracy": 0.16508829444646836,
|
|
"num_tokens": 6235273.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 6.344814777374268,
|
|
"epoch": 3.1660865241173544,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00043837491398679234,
|
|
"loss": 5.6851,
|
|
"mean_token_accuracy": 0.17990447580814362,
|
|
"num_tokens": 6245161.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 6.348833274841309,
|
|
"epoch": 3.17105917454003,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004381061486393025,
|
|
"loss": 5.765,
|
|
"mean_token_accuracy": 0.17466068267822266,
|
|
"num_tokens": 6253300.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 6.300552940368652,
|
|
"epoch": 3.1760318249627053,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004378368919170809,
|
|
"loss": 5.8003,
|
|
"mean_token_accuracy": 0.17742593139410018,
|
|
"num_tokens": 6263525.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 6.31963038444519,
|
|
"epoch": 3.1810044753853806,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00043756714463129244,
|
|
"loss": 5.7793,
|
|
"mean_token_accuracy": 0.17233418375253678,
|
|
"num_tokens": 6274010.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 6.415994453430176,
|
|
"epoch": 3.185977125808056,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004372969075945796,
|
|
"loss": 5.8407,
|
|
"mean_token_accuracy": 0.17350470274686813,
|
|
"num_tokens": 6283442.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 6.321871423721314,
|
|
"epoch": 3.190949776230731,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004370261816210604,
|
|
"loss": 5.7812,
|
|
"mean_token_accuracy": 0.1779524564743042,
|
|
"num_tokens": 6293498.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 6.263674354553222,
|
|
"epoch": 3.1959224266534063,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004367549675263258,
|
|
"loss": 5.6227,
|
|
"mean_token_accuracy": 0.192646786570549,
|
|
"num_tokens": 6303557.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 6.441506242752075,
|
|
"epoch": 3.2008950770760816,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004364832661274373,
|
|
"loss": 5.9608,
|
|
"mean_token_accuracy": 0.16531638875603677,
|
|
"num_tokens": 6314304.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 6.403697538375854,
|
|
"epoch": 3.205867727498757,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00043621107824292444,
|
|
"loss": 5.8202,
|
|
"mean_token_accuracy": 0.16749281734228133,
|
|
"num_tokens": 6324959.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 6.218290615081787,
|
|
"epoch": 3.210840377921432,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004359384046927824,
|
|
"loss": 5.6145,
|
|
"mean_token_accuracy": 0.19874607473611833,
|
|
"num_tokens": 6335909.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 6.376282358169556,
|
|
"epoch": 3.2158130283441073,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004356652462984693,
|
|
"loss": 5.8622,
|
|
"mean_token_accuracy": 0.16914740651845933,
|
|
"num_tokens": 6346659.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 6.313331174850464,
|
|
"epoch": 3.2207856787667826,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00043539160388290416,
|
|
"loss": 5.668,
|
|
"mean_token_accuracy": 0.19056421816349028,
|
|
"num_tokens": 6357264.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 6.374173021316528,
|
|
"epoch": 3.225758329189458,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004351174782704641,
|
|
"loss": 5.8946,
|
|
"mean_token_accuracy": 0.1673770561814308,
|
|
"num_tokens": 6366107.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 6.260113477706909,
|
|
"epoch": 3.230730979612133,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00043484287028698176,
|
|
"loss": 5.6841,
|
|
"mean_token_accuracy": 0.18578468412160873,
|
|
"num_tokens": 6375505.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 6.3877862930297855,
|
|
"epoch": 3.2357036300348083,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00043456778075974306,
|
|
"loss": 5.9168,
|
|
"mean_token_accuracy": 0.1649317756295204,
|
|
"num_tokens": 6384701.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 6.3717625617980955,
|
|
"epoch": 3.240676280457484,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00043429221051748487,
|
|
"loss": 5.7359,
|
|
"mean_token_accuracy": 0.1791525200009346,
|
|
"num_tokens": 6393707.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 6.3085297584533695,
|
|
"epoch": 3.2456489308801593,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00043401616039039174,
|
|
"loss": 5.8397,
|
|
"mean_token_accuracy": 0.17122748792171477,
|
|
"num_tokens": 6404277.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 6.367399024963379,
|
|
"epoch": 3.2506215813028345,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004337396312100944,
|
|
"loss": 5.8923,
|
|
"mean_token_accuracy": 0.16508898735046387,
|
|
"num_tokens": 6415275.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 6.381771469116211,
|
|
"epoch": 3.2555942317255098,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004334626238096665,
|
|
"loss": 5.7959,
|
|
"mean_token_accuracy": 0.17386487424373626,
|
|
"num_tokens": 6425598.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 6.365439796447754,
|
|
"epoch": 3.260566882148185,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004331851390236227,
|
|
"loss": 5.7627,
|
|
"mean_token_accuracy": 0.17297363728284837,
|
|
"num_tokens": 6435685.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 6.293175268173218,
|
|
"epoch": 3.2655395325708603,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00043290717768791533,
|
|
"loss": 5.7446,
|
|
"mean_token_accuracy": 0.17963172793388366,
|
|
"num_tokens": 6444950.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 6.280172300338745,
|
|
"epoch": 3.2705121829935355,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00043262874063993285,
|
|
"loss": 5.7602,
|
|
"mean_token_accuracy": 0.1741028070449829,
|
|
"num_tokens": 6454066.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 6.365146112442017,
|
|
"epoch": 3.2754848334162108,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004323498287184967,
|
|
"loss": 5.892,
|
|
"mean_token_accuracy": 0.16845283210277556,
|
|
"num_tokens": 6464912.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 6.392491197586059,
|
|
"epoch": 3.280457483838886,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00043207044276385884,
|
|
"loss": 5.9174,
|
|
"mean_token_accuracy": 0.16207081079483032,
|
|
"num_tokens": 6474178.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 6.223371267318726,
|
|
"epoch": 3.2854301342615613,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00043179058361769924,
|
|
"loss": 5.6226,
|
|
"mean_token_accuracy": 0.18952620178461074,
|
|
"num_tokens": 6485703.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 6.358218193054199,
|
|
"epoch": 3.2904027846842365,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004315102521231237,
|
|
"loss": 5.7437,
|
|
"mean_token_accuracy": 0.17418602854013443,
|
|
"num_tokens": 6495309.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 6.390391540527344,
|
|
"epoch": 3.295375435106912,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004312294491246609,
|
|
"loss": 5.7104,
|
|
"mean_token_accuracy": 0.17614705562591554,
|
|
"num_tokens": 6505498.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 6.290208196640014,
|
|
"epoch": 3.3003480855295875,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004309481754682597,
|
|
"loss": 5.7438,
|
|
"mean_token_accuracy": 0.17657722383737565,
|
|
"num_tokens": 6514656.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 6.314484691619873,
|
|
"epoch": 3.3053207359522627,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00043066643200128724,
|
|
"loss": 5.7987,
|
|
"mean_token_accuracy": 0.17480697482824326,
|
|
"num_tokens": 6524549.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 6.294785118103027,
|
|
"epoch": 3.310293386374938,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004303842195725259,
|
|
"loss": 5.7117,
|
|
"mean_token_accuracy": 0.18283183127641678,
|
|
"num_tokens": 6533861.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 6.392320060729981,
|
|
"epoch": 3.3152660367976132,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00043010153903217074,
|
|
"loss": 5.8971,
|
|
"mean_token_accuracy": 0.1664445698261261,
|
|
"num_tokens": 6543676.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 6.380225276947021,
|
|
"epoch": 3.3202386872202885,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004298183912318273,
|
|
"loss": 5.799,
|
|
"mean_token_accuracy": 0.17061721831560134,
|
|
"num_tokens": 6552263.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 6.277283096313477,
|
|
"epoch": 3.3252113376429637,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00042953477702450845,
|
|
"loss": 5.7652,
|
|
"mean_token_accuracy": 0.17937974035739898,
|
|
"num_tokens": 6560960.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 6.237183904647827,
|
|
"epoch": 3.330183988065639,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004292506972646324,
|
|
"loss": 5.6554,
|
|
"mean_token_accuracy": 0.19072091430425644,
|
|
"num_tokens": 6572345.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 6.33965368270874,
|
|
"epoch": 3.3351566384883142,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00042896615280802006,
|
|
"loss": 5.8389,
|
|
"mean_token_accuracy": 0.17175715118646623,
|
|
"num_tokens": 6582274.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 6.288807916641235,
|
|
"epoch": 3.3401292889109895,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00042868114451189164,
|
|
"loss": 5.7665,
|
|
"mean_token_accuracy": 0.17306731641292572,
|
|
"num_tokens": 6592486.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 6.266016578674316,
|
|
"epoch": 3.3451019393336647,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004283956732348655,
|
|
"loss": 5.689,
|
|
"mean_token_accuracy": 0.19121331870555877,
|
|
"num_tokens": 6600876.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 6.2881519317626955,
|
|
"epoch": 3.35007458975634,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00042810973983695426,
|
|
"loss": 5.7044,
|
|
"mean_token_accuracy": 0.18148678690195083,
|
|
"num_tokens": 6610679.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 6.333586311340332,
|
|
"epoch": 3.3550472401790152,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004278233451795629,
|
|
"loss": 5.8321,
|
|
"mean_token_accuracy": 0.16926302909851074,
|
|
"num_tokens": 6620779.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 6.37353744506836,
|
|
"epoch": 3.3600198906016905,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000427536490125486,
|
|
"loss": 5.78,
|
|
"mean_token_accuracy": 0.17491379976272584,
|
|
"num_tokens": 6630807.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 6.2095420360565186,
|
|
"epoch": 3.364992541024366,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00042724917553890507,
|
|
"loss": 5.686,
|
|
"mean_token_accuracy": 0.18185881078243255,
|
|
"num_tokens": 6640530.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 6.267149972915649,
|
|
"epoch": 3.3699651914470414,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00042696140228538616,
|
|
"loss": 5.7399,
|
|
"mean_token_accuracy": 0.1769838273525238,
|
|
"num_tokens": 6650845.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 6.341059350967408,
|
|
"epoch": 3.3749378418697167,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00042667317123187696,
|
|
"loss": 5.7859,
|
|
"mean_token_accuracy": 0.1721637010574341,
|
|
"num_tokens": 6660455.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 6.278128623962402,
|
|
"epoch": 3.379910492292392,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00042638448324670424,
|
|
"loss": 5.7455,
|
|
"mean_token_accuracy": 0.17768496423959732,
|
|
"num_tokens": 6670493.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 6.324783849716186,
|
|
"epoch": 3.384883142715067,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004260953391995716,
|
|
"loss": 5.8658,
|
|
"mean_token_accuracy": 0.16838005781173707,
|
|
"num_tokens": 6680598.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 6.159350728988647,
|
|
"epoch": 3.3898557931377424,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00042580573996155646,
|
|
"loss": 5.6165,
|
|
"mean_token_accuracy": 0.19578235149383544,
|
|
"num_tokens": 6691569.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 6.37211012840271,
|
|
"epoch": 3.3948284435604177,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004255156864051074,
|
|
"loss": 5.8948,
|
|
"mean_token_accuracy": 0.16866730600595475,
|
|
"num_tokens": 6702570.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 6.179108190536499,
|
|
"epoch": 3.399801093983093,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00042522517940404186,
|
|
"loss": 5.5951,
|
|
"mean_token_accuracy": 0.19658319503068925,
|
|
"num_tokens": 6711628.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 6.346974563598633,
|
|
"epoch": 3.404773744405768,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004249342198335433,
|
|
"loss": 5.7863,
|
|
"mean_token_accuracy": 0.16636384278535843,
|
|
"num_tokens": 6721828.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 6.178861045837403,
|
|
"epoch": 3.4097463948284434,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00042464280857015856,
|
|
"loss": 5.6316,
|
|
"mean_token_accuracy": 0.1895842969417572,
|
|
"num_tokens": 6731903.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 6.228585338592529,
|
|
"epoch": 3.4147190452511187,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00042435094649179525,
|
|
"loss": 5.7922,
|
|
"mean_token_accuracy": 0.1843317061662674,
|
|
"num_tokens": 6743236.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 6.1815673351287845,
|
|
"epoch": 3.419691695673794,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004240586344777191,
|
|
"loss": 5.6228,
|
|
"mean_token_accuracy": 0.18854227513074875,
|
|
"num_tokens": 6753291.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 6.3971258163452145,
|
|
"epoch": 3.4246643460964696,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004237658734085514,
|
|
"loss": 5.8132,
|
|
"mean_token_accuracy": 0.17136881053447722,
|
|
"num_tokens": 6762635.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 6.33477029800415,
|
|
"epoch": 3.429636996519145,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00042347266416626624,
|
|
"loss": 5.774,
|
|
"mean_token_accuracy": 0.17653224617242813,
|
|
"num_tokens": 6772101.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 6.262850046157837,
|
|
"epoch": 3.43460964694182,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004231790076341877,
|
|
"loss": 5.7936,
|
|
"mean_token_accuracy": 0.17215746492147446,
|
|
"num_tokens": 6781563.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 6.283638191223145,
|
|
"epoch": 3.4395822973644954,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004228849046969876,
|
|
"loss": 5.7582,
|
|
"mean_token_accuracy": 0.17933691143989564,
|
|
"num_tokens": 6790589.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 6.355904245376587,
|
|
"epoch": 3.4445549477871706,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004225903562406826,
|
|
"loss": 5.7866,
|
|
"mean_token_accuracy": 0.17500382363796235,
|
|
"num_tokens": 6799842.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 6.313158845901489,
|
|
"epoch": 3.449527598209846,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004222953631526313,
|
|
"loss": 5.833,
|
|
"mean_token_accuracy": 0.174507075548172,
|
|
"num_tokens": 6809166.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 6.3001923084259035,
|
|
"epoch": 3.454500248632521,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000421999926321532,
|
|
"loss": 5.8405,
|
|
"mean_token_accuracy": 0.1693264812231064,
|
|
"num_tokens": 6818255.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 6.32542085647583,
|
|
"epoch": 3.4594728990551964,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004217040466374199,
|
|
"loss": 5.748,
|
|
"mean_token_accuracy": 0.17985807806253434,
|
|
"num_tokens": 6828628.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 6.3387291431427,
|
|
"epoch": 3.4644455494778716,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00042140772499166403,
|
|
"loss": 5.867,
|
|
"mean_token_accuracy": 0.17349916845560073,
|
|
"num_tokens": 6839053.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 6.344242334365845,
|
|
"epoch": 3.469418199900547,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004211109622769652,
|
|
"loss": 5.8169,
|
|
"mean_token_accuracy": 0.17196138203144073,
|
|
"num_tokens": 6849226.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 6.276549530029297,
|
|
"epoch": 3.474390850323222,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00042081375938735283,
|
|
"loss": 5.771,
|
|
"mean_token_accuracy": 0.18062664419412613,
|
|
"num_tokens": 6858867.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 6.315723562240601,
|
|
"epoch": 3.4793635007458974,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00042051611721818233,
|
|
"loss": 5.806,
|
|
"mean_token_accuracy": 0.17377726137638091,
|
|
"num_tokens": 6869171.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 6.281809997558594,
|
|
"epoch": 3.4843361511685726,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00042021803666613273,
|
|
"loss": 5.7653,
|
|
"mean_token_accuracy": 0.17706142365932465,
|
|
"num_tokens": 6878582.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 6.252849864959717,
|
|
"epoch": 3.489308801591248,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00041991951862920365,
|
|
"loss": 5.8257,
|
|
"mean_token_accuracy": 0.1746981516480446,
|
|
"num_tokens": 6887522.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 6.287425756454468,
|
|
"epoch": 3.4942814520139236,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004196205640067124,
|
|
"loss": 5.7268,
|
|
"mean_token_accuracy": 0.18144605904817582,
|
|
"num_tokens": 6897556.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 6.30746054649353,
|
|
"epoch": 3.499254102436599,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000419321173699292,
|
|
"loss": 5.7822,
|
|
"mean_token_accuracy": 0.18155322074890137,
|
|
"num_tokens": 6908219.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 6.350363159179688,
|
|
"epoch": 3.504226752859274,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004190213486088877,
|
|
"loss": 5.8127,
|
|
"mean_token_accuracy": 0.16932816952466964,
|
|
"num_tokens": 6918175.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 6.341069507598877,
|
|
"epoch": 3.5091994032819493,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004187210896387546,
|
|
"loss": 5.9188,
|
|
"mean_token_accuracy": 0.17309466004371643,
|
|
"num_tokens": 6927076.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 6.342219400405884,
|
|
"epoch": 3.5141720537046246,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004184203976934552,
|
|
"loss": 5.8236,
|
|
"mean_token_accuracy": 0.17986441254615784,
|
|
"num_tokens": 6936568.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 6.246902418136597,
|
|
"epoch": 3.5191447041273,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00041811927367885605,
|
|
"loss": 5.7792,
|
|
"mean_token_accuracy": 0.1782641366124153,
|
|
"num_tokens": 6945494.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 6.284408283233643,
|
|
"epoch": 3.524117354549975,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004178177185021254,
|
|
"loss": 5.7449,
|
|
"mean_token_accuracy": 0.18381952196359636,
|
|
"num_tokens": 6955397.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 6.331853818893433,
|
|
"epoch": 3.5290900049726504,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004175157330717305,
|
|
"loss": 5.8726,
|
|
"mean_token_accuracy": 0.16879275739192962,
|
|
"num_tokens": 6965431.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 6.254835796356201,
|
|
"epoch": 3.5340626553953256,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000417213318297435,
|
|
"loss": 5.734,
|
|
"mean_token_accuracy": 0.18405893594026565,
|
|
"num_tokens": 6975457.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 6.238683795928955,
|
|
"epoch": 3.539035305818001,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00041691047509029555,
|
|
"loss": 5.6356,
|
|
"mean_token_accuracy": 0.19078688472509384,
|
|
"num_tokens": 6984691.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 6.194393873214722,
|
|
"epoch": 3.5440079562406765,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00041660720436265987,
|
|
"loss": 5.6732,
|
|
"mean_token_accuracy": 0.1908874660730362,
|
|
"num_tokens": 6994052.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 6.280835151672363,
|
|
"epoch": 3.548980606663352,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004163035070281634,
|
|
"loss": 5.8025,
|
|
"mean_token_accuracy": 0.1759354829788208,
|
|
"num_tokens": 7003625.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 6.263123559951782,
|
|
"epoch": 3.553953257086027,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00041599938400172686,
|
|
"loss": 5.8452,
|
|
"mean_token_accuracy": 0.17291782945394515,
|
|
"num_tokens": 7011967.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 6.335304021835327,
|
|
"epoch": 3.5589259075087023,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00041569483619955354,
|
|
"loss": 5.8741,
|
|
"mean_token_accuracy": 0.17249398678541183,
|
|
"num_tokens": 7021722.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 6.414807987213135,
|
|
"epoch": 3.5638985579313776,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00041538986453912616,
|
|
"loss": 5.8848,
|
|
"mean_token_accuracy": 0.16595202386379243,
|
|
"num_tokens": 7031720.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 6.275649070739746,
|
|
"epoch": 3.568871208354053,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004150844699392045,
|
|
"loss": 5.8624,
|
|
"mean_token_accuracy": 0.17008156031370164,
|
|
"num_tokens": 7040575.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 6.42878041267395,
|
|
"epoch": 3.573843858776728,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00041477865331982264,
|
|
"loss": 5.8294,
|
|
"mean_token_accuracy": 0.17087699770927428,
|
|
"num_tokens": 7050035.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 6.188215017318726,
|
|
"epoch": 3.5788165091994033,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004144724156022857,
|
|
"loss": 5.6305,
|
|
"mean_token_accuracy": 0.19610500782728196,
|
|
"num_tokens": 7060950.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 6.227851533889771,
|
|
"epoch": 3.5837891596220786,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00041416575770916763,
|
|
"loss": 5.7944,
|
|
"mean_token_accuracy": 0.17367122918367386,
|
|
"num_tokens": 7070384.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 6.392584705352784,
|
|
"epoch": 3.588761810044754,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00041385868056430827,
|
|
"loss": 5.9226,
|
|
"mean_token_accuracy": 0.16312913447618485,
|
|
"num_tokens": 7081118.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 6.331424045562744,
|
|
"epoch": 3.593734460467429,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004135511850928102,
|
|
"loss": 5.8598,
|
|
"mean_token_accuracy": 0.17069692462682723,
|
|
"num_tokens": 7091157.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 6.265208959579468,
|
|
"epoch": 3.5987071108901043,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00041324327222103655,
|
|
"loss": 5.8433,
|
|
"mean_token_accuracy": 0.17199520915746688,
|
|
"num_tokens": 7101005.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 6.253416061401367,
|
|
"epoch": 3.6036797613127796,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00041293494287660785,
|
|
"loss": 5.7831,
|
|
"mean_token_accuracy": 0.17246333211660386,
|
|
"num_tokens": 7109944.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 6.402906084060669,
|
|
"epoch": 3.608652411735455,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00041262619798839914,
|
|
"loss": 5.8677,
|
|
"mean_token_accuracy": 0.1633292630314827,
|
|
"num_tokens": 7119073.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 6.335119819641113,
|
|
"epoch": 3.61362506215813,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00041231703848653746,
|
|
"loss": 5.8724,
|
|
"mean_token_accuracy": 0.16824883073568345,
|
|
"num_tokens": 7128920.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 6.328790330886841,
|
|
"epoch": 3.6185977125808053,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00041200746530239903,
|
|
"loss": 5.844,
|
|
"mean_token_accuracy": 0.1672176331281662,
|
|
"num_tokens": 7138691.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 6.313627433776856,
|
|
"epoch": 3.623570363003481,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004116974793686061,
|
|
"loss": 5.7691,
|
|
"mean_token_accuracy": 0.18123773485422134,
|
|
"num_tokens": 7148362.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 6.271098709106445,
|
|
"epoch": 3.6285430134261563,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004113870816190244,
|
|
"loss": 5.777,
|
|
"mean_token_accuracy": 0.17695932537317277,
|
|
"num_tokens": 7157138.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 6.3142557621002195,
|
|
"epoch": 3.6335156638488315,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004110762729887605,
|
|
"loss": 5.913,
|
|
"mean_token_accuracy": 0.16751261055469513,
|
|
"num_tokens": 7167425.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 6.35818510055542,
|
|
"epoch": 3.6384883142715068,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00041076505441415873,
|
|
"loss": 5.8203,
|
|
"mean_token_accuracy": 0.17349623292684554,
|
|
"num_tokens": 7177059.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 6.3564306735992435,
|
|
"epoch": 3.643460964694182,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004104534268327982,
|
|
"loss": 5.8577,
|
|
"mean_token_accuracy": 0.17099332064390182,
|
|
"num_tokens": 7186435.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 6.258958625793457,
|
|
"epoch": 3.6484336151168573,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004101413911834904,
|
|
"loss": 5.8502,
|
|
"mean_token_accuracy": 0.17133126854896547,
|
|
"num_tokens": 7197248.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 6.272427892684936,
|
|
"epoch": 3.6534062655395325,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00040982894840627614,
|
|
"loss": 5.8106,
|
|
"mean_token_accuracy": 0.17008461952209472,
|
|
"num_tokens": 7206042.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 6.3389380931854244,
|
|
"epoch": 3.6583789159622078,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004095160994424228,
|
|
"loss": 5.8545,
|
|
"mean_token_accuracy": 0.17546366304159164,
|
|
"num_tokens": 7215822.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 6.33992714881897,
|
|
"epoch": 3.663351566384883,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00040920284523442125,
|
|
"loss": 5.8585,
|
|
"mean_token_accuracy": 0.17041762918233871,
|
|
"num_tokens": 7226131.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 6.2338059425354,
|
|
"epoch": 3.6683242168075583,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00040888918672598337,
|
|
"loss": 5.7669,
|
|
"mean_token_accuracy": 0.17887679636478424,
|
|
"num_tokens": 7234095.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 6.234047937393188,
|
|
"epoch": 3.673296867230234,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000408575124862039,
|
|
"loss": 5.7044,
|
|
"mean_token_accuracy": 0.1929001197218895,
|
|
"num_tokens": 7243184.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 6.334602546691895,
|
|
"epoch": 3.678269517652909,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004082606605887331,
|
|
"loss": 5.8225,
|
|
"mean_token_accuracy": 0.16810275614261627,
|
|
"num_tokens": 7253654.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 6.226493406295776,
|
|
"epoch": 3.6832421680755845,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00040794579485342304,
|
|
"loss": 5.6829,
|
|
"mean_token_accuracy": 0.17336571961641312,
|
|
"num_tokens": 7262063.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 6.21433391571045,
|
|
"epoch": 3.6882148184982597,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00040763052860467547,
|
|
"loss": 5.6907,
|
|
"mean_token_accuracy": 0.19129979312419892,
|
|
"num_tokens": 7272490.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 6.303772115707398,
|
|
"epoch": 3.693187468920935,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004073148627922637,
|
|
"loss": 5.9151,
|
|
"mean_token_accuracy": 0.1665804371237755,
|
|
"num_tokens": 7282938.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 6.319575929641724,
|
|
"epoch": 3.69816011934361,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004069987983671648,
|
|
"loss": 5.7644,
|
|
"mean_token_accuracy": 0.17417771071195604,
|
|
"num_tokens": 7293486.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 6.2420995235443115,
|
|
"epoch": 3.7031327697662855,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00040668233628155664,
|
|
"loss": 5.7574,
|
|
"mean_token_accuracy": 0.1717290535569191,
|
|
"num_tokens": 7302806.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 6.228424978256226,
|
|
"epoch": 3.7081054201889607,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004063654774888153,
|
|
"loss": 5.8522,
|
|
"mean_token_accuracy": 0.17007196992635726,
|
|
"num_tokens": 7312250.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 6.3695556640625,
|
|
"epoch": 3.713078070611636,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004060482229435116,
|
|
"loss": 5.7519,
|
|
"mean_token_accuracy": 0.18264814466238022,
|
|
"num_tokens": 7321512.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 6.2825846672058105,
|
|
"epoch": 3.7180507210343112,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004057305736014089,
|
|
"loss": 5.8203,
|
|
"mean_token_accuracy": 0.18326488882303238,
|
|
"num_tokens": 7331467.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 6.222785568237304,
|
|
"epoch": 3.7230233714569865,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004054125304194598,
|
|
"loss": 5.7654,
|
|
"mean_token_accuracy": 0.1726515606045723,
|
|
"num_tokens": 7340064.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 6.250120067596436,
|
|
"epoch": 3.7279960218796617,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00040509409435580355,
|
|
"loss": 5.7606,
|
|
"mean_token_accuracy": 0.17830842286348342,
|
|
"num_tokens": 7350138.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 6.308509349822998,
|
|
"epoch": 3.732968672302337,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004047752663697627,
|
|
"loss": 5.8486,
|
|
"mean_token_accuracy": 0.1745805710554123,
|
|
"num_tokens": 7359613.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 6.264008808135986,
|
|
"epoch": 3.7379413227250122,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004044560474218409,
|
|
"loss": 5.774,
|
|
"mean_token_accuracy": 0.18045201301574706,
|
|
"num_tokens": 7368754.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 6.322387933731079,
|
|
"epoch": 3.7429139731476875,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00040413643847371923,
|
|
"loss": 5.7531,
|
|
"mean_token_accuracy": 0.18121736347675324,
|
|
"num_tokens": 7378930.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 6.3046191215515135,
|
|
"epoch": 3.7478866235703627,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004038164404882539,
|
|
"loss": 5.8356,
|
|
"mean_token_accuracy": 0.1690583661198616,
|
|
"num_tokens": 7387281.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 6.2881440162658695,
|
|
"epoch": 3.7528592739930384,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004034960544294731,
|
|
"loss": 5.7047,
|
|
"mean_token_accuracy": 0.17966454327106476,
|
|
"num_tokens": 7397039.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 6.201868534088135,
|
|
"epoch": 3.7578319244157137,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00040317528126257413,
|
|
"loss": 5.802,
|
|
"mean_token_accuracy": 0.17610154002904893,
|
|
"num_tokens": 7406771.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 6.249643325805664,
|
|
"epoch": 3.762804574838389,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004028541219539205,
|
|
"loss": 5.7576,
|
|
"mean_token_accuracy": 0.18376846015453338,
|
|
"num_tokens": 7415743.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 6.065288257598877,
|
|
"epoch": 3.767777225261064,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004025325774710391,
|
|
"loss": 5.5524,
|
|
"mean_token_accuracy": 0.18897473812103271,
|
|
"num_tokens": 7424243.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 6.223342084884644,
|
|
"epoch": 3.7727498756837394,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00040221064878261684,
|
|
"loss": 5.8498,
|
|
"mean_token_accuracy": 0.16722080260515212,
|
|
"num_tokens": 7434273.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 6.2889751434326175,
|
|
"epoch": 3.7777225261064147,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004018883368584985,
|
|
"loss": 5.7868,
|
|
"mean_token_accuracy": 0.17143706381320953,
|
|
"num_tokens": 7444640.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 6.214770412445068,
|
|
"epoch": 3.78269517652909,
|
|
"grad_norm": 1.90625,
|
|
"learning_rate": 0.00040156564266968336,
|
|
"loss": 5.634,
|
|
"mean_token_accuracy": 0.18543930649757384,
|
|
"num_tokens": 7453774.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 6.308246803283692,
|
|
"epoch": 3.787667826951765,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00040124256718832214,
|
|
"loss": 5.7815,
|
|
"mean_token_accuracy": 0.17472663819789885,
|
|
"num_tokens": 7464288.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 6.310828971862793,
|
|
"epoch": 3.7926404773744404,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004009191113877141,
|
|
"loss": 5.8292,
|
|
"mean_token_accuracy": 0.16994530111551284,
|
|
"num_tokens": 7474626.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 6.305044078826905,
|
|
"epoch": 3.797613127797116,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004005952762423048,
|
|
"loss": 5.8238,
|
|
"mean_token_accuracy": 0.18264661580324174,
|
|
"num_tokens": 7483393.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 6.333598709106445,
|
|
"epoch": 3.8025857782197914,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004002710627276821,
|
|
"loss": 5.8508,
|
|
"mean_token_accuracy": 0.17023134231567383,
|
|
"num_tokens": 7492197.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 6.270931243896484,
|
|
"epoch": 3.8075584286424666,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003999464718205739,
|
|
"loss": 5.8076,
|
|
"mean_token_accuracy": 0.1808437779545784,
|
|
"num_tokens": 7501822.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 6.202649068832398,
|
|
"epoch": 3.812531079065142,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003996215044988451,
|
|
"loss": 5.7133,
|
|
"mean_token_accuracy": 0.18330524265766143,
|
|
"num_tokens": 7512686.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 6.299923658370972,
|
|
"epoch": 3.817503729487817,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00039929616174149454,
|
|
"loss": 5.804,
|
|
"mean_token_accuracy": 0.17051225155591965,
|
|
"num_tokens": 7522593.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 6.2756274223327635,
|
|
"epoch": 3.8224763799104924,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00039897044452865207,
|
|
"loss": 5.8739,
|
|
"mean_token_accuracy": 0.16918620243668556,
|
|
"num_tokens": 7532448.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 6.214023590087891,
|
|
"epoch": 3.8274490303331676,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003986443538415756,
|
|
"loss": 5.5884,
|
|
"mean_token_accuracy": 0.2026273339986801,
|
|
"num_tokens": 7543008.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 6.196967935562133,
|
|
"epoch": 3.832421680755843,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003983178906626484,
|
|
"loss": 5.7169,
|
|
"mean_token_accuracy": 0.1819205328822136,
|
|
"num_tokens": 7553236.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 6.278046464920044,
|
|
"epoch": 3.837394331178518,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00039799105597537564,
|
|
"loss": 5.8411,
|
|
"mean_token_accuracy": 0.1808039054274559,
|
|
"num_tokens": 7561583.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 6.256569576263428,
|
|
"epoch": 3.8423669816011934,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003976638507643818,
|
|
"loss": 5.8255,
|
|
"mean_token_accuracy": 0.16512110233306884,
|
|
"num_tokens": 7571290.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 6.274621915817261,
|
|
"epoch": 3.8473396320238686,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003973362760154076,
|
|
"loss": 5.8424,
|
|
"mean_token_accuracy": 0.1667559936642647,
|
|
"num_tokens": 7579976.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 6.2478879451751705,
|
|
"epoch": 3.852312282446544,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003970083327153071,
|
|
"loss": 5.7957,
|
|
"mean_token_accuracy": 0.1678186073899269,
|
|
"num_tokens": 7589796.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 6.26448540687561,
|
|
"epoch": 3.857284932869219,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003966800218520445,
|
|
"loss": 5.7173,
|
|
"mean_token_accuracy": 0.1817557007074356,
|
|
"num_tokens": 7598976.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 6.263399839401245,
|
|
"epoch": 3.8622575832918944,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003963513444146914,
|
|
"loss": 5.8773,
|
|
"mean_token_accuracy": 0.16959742456674576,
|
|
"num_tokens": 7607955.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 6.230166006088257,
|
|
"epoch": 3.8672302337145696,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003960223013934239,
|
|
"loss": 5.7692,
|
|
"mean_token_accuracy": 0.17484400421380997,
|
|
"num_tokens": 7619822.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 6.192731952667236,
|
|
"epoch": 3.872202884137245,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003956928937795192,
|
|
"loss": 5.6739,
|
|
"mean_token_accuracy": 0.18913860917091369,
|
|
"num_tokens": 7630481.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 6.278459310531616,
|
|
"epoch": 3.8771755345599206,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000395363122565353,
|
|
"loss": 5.7804,
|
|
"mean_token_accuracy": 0.17349368929862977,
|
|
"num_tokens": 7640312.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 6.403591871261597,
|
|
"epoch": 3.882148184982596,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00039503298874439655,
|
|
"loss": 6.0579,
|
|
"mean_token_accuracy": 0.15335207358002662,
|
|
"num_tokens": 7649792.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 6.2781657695770265,
|
|
"epoch": 3.887120835405271,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003947024933112131,
|
|
"loss": 5.776,
|
|
"mean_token_accuracy": 0.17664752751588822,
|
|
"num_tokens": 7659458.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 6.2505881786346436,
|
|
"epoch": 3.8920934858279463,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003943716372614556,
|
|
"loss": 5.7807,
|
|
"mean_token_accuracy": 0.18036242574453354,
|
|
"num_tokens": 7668770.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 6.1901017189025875,
|
|
"epoch": 3.8970661362506216,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003940404215918634,
|
|
"loss": 5.7094,
|
|
"mean_token_accuracy": 0.18567833602428435,
|
|
"num_tokens": 7677067.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 6.201345634460449,
|
|
"epoch": 3.902038786673297,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00039370884730025897,
|
|
"loss": 5.7724,
|
|
"mean_token_accuracy": 0.17716639637947082,
|
|
"num_tokens": 7687723.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 6.328340768814087,
|
|
"epoch": 3.907011437095972,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003933769153855455,
|
|
"loss": 5.9196,
|
|
"mean_token_accuracy": 0.1697556868195534,
|
|
"num_tokens": 7696706.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 6.318368530273437,
|
|
"epoch": 3.9119840875186473,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003930446268477034,
|
|
"loss": 5.8101,
|
|
"mean_token_accuracy": 0.17165198028087617,
|
|
"num_tokens": 7706356.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 6.261753511428833,
|
|
"epoch": 3.9169567379413226,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003927119826877875,
|
|
"loss": 5.8525,
|
|
"mean_token_accuracy": 0.17633430808782577,
|
|
"num_tokens": 7717255.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 6.246714735031128,
|
|
"epoch": 3.921929388363998,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003923789839079238,
|
|
"loss": 5.8357,
|
|
"mean_token_accuracy": 0.17457175105810166,
|
|
"num_tokens": 7728493.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 6.325566530227661,
|
|
"epoch": 3.9269020387866735,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00039204563151130683,
|
|
"loss": 5.7962,
|
|
"mean_token_accuracy": 0.1695742502808571,
|
|
"num_tokens": 7738582.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 6.282317733764648,
|
|
"epoch": 3.931874689209349,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003917119265021965,
|
|
"loss": 5.8696,
|
|
"mean_token_accuracy": 0.1741955652832985,
|
|
"num_tokens": 7748486.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 6.142000389099121,
|
|
"epoch": 3.936847339632024,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003913778698859146,
|
|
"loss": 5.609,
|
|
"mean_token_accuracy": 0.19151835590600969,
|
|
"num_tokens": 7758033.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 6.269939136505127,
|
|
"epoch": 3.9418199900546993,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003910434626688427,
|
|
"loss": 5.6832,
|
|
"mean_token_accuracy": 0.17772685885429382,
|
|
"num_tokens": 7768577.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 6.2766197681427,
|
|
"epoch": 3.9467926404773745,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003907087058584183,
|
|
"loss": 5.8067,
|
|
"mean_token_accuracy": 0.173308926820755,
|
|
"num_tokens": 7778546.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 6.226923084259033,
|
|
"epoch": 3.95176529090005,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00039037360046313186,
|
|
"loss": 5.8146,
|
|
"mean_token_accuracy": 0.17733479887247086,
|
|
"num_tokens": 7787523.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 6.341509914398193,
|
|
"epoch": 3.956737941322725,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 0.00039003814749252454,
|
|
"loss": 5.7826,
|
|
"mean_token_accuracy": 0.17707646638154984,
|
|
"num_tokens": 7798198.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 6.383081293106079,
|
|
"epoch": 3.9617105917454003,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00038970234795718424,
|
|
"loss": 5.9214,
|
|
"mean_token_accuracy": 0.1642968773841858,
|
|
"num_tokens": 7808137.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 6.2498194694519045,
|
|
"epoch": 3.9666832421680756,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000389366202868743,
|
|
"loss": 5.7844,
|
|
"mean_token_accuracy": 0.18175089359283447,
|
|
"num_tokens": 7817367.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 6.093843603134156,
|
|
"epoch": 3.971655892590751,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003890297132398739,
|
|
"loss": 5.6148,
|
|
"mean_token_accuracy": 0.19028337746858598,
|
|
"num_tokens": 7826419.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 6.225203609466552,
|
|
"epoch": 3.976628543013426,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003886928800842879,
|
|
"loss": 5.8011,
|
|
"mean_token_accuracy": 0.17368798553943635,
|
|
"num_tokens": 7834984.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 3.976628543013426,
|
|
"eval_entropy": 6.111848868532665,
|
|
"eval_loss": 6.457040309906006,
|
|
"eval_mean_token_accuracy": 0.14850503286080702,
|
|
"eval_num_tokens": 7834984.0,
|
|
"eval_runtime": 2.1072,
|
|
"eval_samples_per_second": 1644.341,
|
|
"eval_steps_per_second": 205.958,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"entropy": 6.287959051132202,
|
|
"epoch": 3.9816011934361013,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00038835570441673095,
|
|
"loss": 5.7881,
|
|
"mean_token_accuracy": 0.1751801162958145,
|
|
"num_tokens": 7844781.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 6.274995803833008,
|
|
"epoch": 3.9865738438587766,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003880181872529809,
|
|
"loss": 5.8438,
|
|
"mean_token_accuracy": 0.17689413130283355,
|
|
"num_tokens": 7854553.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 6.2566625595092775,
|
|
"epoch": 3.991546494281452,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003876803296098443,
|
|
"loss": 5.7969,
|
|
"mean_token_accuracy": 0.17074501216411592,
|
|
"num_tokens": 7863900.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 6.312726736068726,
|
|
"epoch": 3.996519144704127,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003873421325051535,
|
|
"loss": 5.7608,
|
|
"mean_token_accuracy": 0.1786616250872612,
|
|
"num_tokens": 7873008.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 6.253713024987115,
|
|
"epoch": 4.000994530084535,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00038700359695776346,
|
|
"loss": 5.7526,
|
|
"mean_token_accuracy": 0.17578770882553524,
|
|
"num_tokens": 7882489.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 6.223328876495361,
|
|
"epoch": 4.00596718050721,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003866647239875487,
|
|
"loss": 5.5557,
|
|
"mean_token_accuracy": 0.1826157808303833,
|
|
"num_tokens": 7893370.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 6.27371039390564,
|
|
"epoch": 4.010939830929885,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00038632551461540023,
|
|
"loss": 5.4422,
|
|
"mean_token_accuracy": 0.19857455790042877,
|
|
"num_tokens": 7903815.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 6.327260875701905,
|
|
"epoch": 4.015912481352561,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003859859698632228,
|
|
"loss": 5.5687,
|
|
"mean_token_accuracy": 0.1848178595304489,
|
|
"num_tokens": 7913365.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 6.265515995025635,
|
|
"epoch": 4.020885131775236,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003856460907539312,
|
|
"loss": 5.5218,
|
|
"mean_token_accuracy": 0.19614047706127166,
|
|
"num_tokens": 7922973.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 6.34101128578186,
|
|
"epoch": 4.025857782197911,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003853058783114478,
|
|
"loss": 5.5011,
|
|
"mean_token_accuracy": 0.19272979497909545,
|
|
"num_tokens": 7932860.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 6.26291298866272,
|
|
"epoch": 4.030830432620586,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003849653335606987,
|
|
"loss": 5.5034,
|
|
"mean_token_accuracy": 0.1955903023481369,
|
|
"num_tokens": 7942524.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 6.3063267230987545,
|
|
"epoch": 4.035803083043262,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003846244575276117,
|
|
"loss": 5.5819,
|
|
"mean_token_accuracy": 0.1855929285287857,
|
|
"num_tokens": 7952289.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 6.162996053695679,
|
|
"epoch": 4.040775733465938,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00038428325123911205,
|
|
"loss": 5.4379,
|
|
"mean_token_accuracy": 0.19461295306682586,
|
|
"num_tokens": 7962176.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 6.241329956054687,
|
|
"epoch": 4.045748383888613,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003839417157231205,
|
|
"loss": 5.535,
|
|
"mean_token_accuracy": 0.19296381920576094,
|
|
"num_tokens": 7971300.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 6.148099517822265,
|
|
"epoch": 4.050721034311288,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003835998520085492,
|
|
"loss": 5.4365,
|
|
"mean_token_accuracy": 0.197830231487751,
|
|
"num_tokens": 7981135.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 6.226283121109009,
|
|
"epoch": 4.0556936847339635,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003832576611252993,
|
|
"loss": 5.4716,
|
|
"mean_token_accuracy": 0.1944435253739357,
|
|
"num_tokens": 7991327.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 6.274404668807984,
|
|
"epoch": 4.060666335156639,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003829151441042572,
|
|
"loss": 5.5296,
|
|
"mean_token_accuracy": 0.18904608488082886,
|
|
"num_tokens": 8001318.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 6.160905122756958,
|
|
"epoch": 4.065638985579314,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00038257230197729237,
|
|
"loss": 5.3484,
|
|
"mean_token_accuracy": 0.20476668030023576,
|
|
"num_tokens": 8010964.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 6.127015733718872,
|
|
"epoch": 4.070611636001989,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00038222913577725325,
|
|
"loss": 5.4886,
|
|
"mean_token_accuracy": 0.1971145510673523,
|
|
"num_tokens": 8020894.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 6.165887928009033,
|
|
"epoch": 4.0755842864246645,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00038188564653796487,
|
|
"loss": 5.4702,
|
|
"mean_token_accuracy": 0.19622017592191696,
|
|
"num_tokens": 8030635.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 6.219968461990357,
|
|
"epoch": 4.08055693684734,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003815418352942252,
|
|
"loss": 5.4062,
|
|
"mean_token_accuracy": 0.19922858327627183,
|
|
"num_tokens": 8040626.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 5.989646482467651,
|
|
"epoch": 4.085529587270015,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003811977030818026,
|
|
"loss": 5.1539,
|
|
"mean_token_accuracy": 0.22911933660507203,
|
|
"num_tokens": 8051835.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 6.177275514602661,
|
|
"epoch": 4.09050223769269,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000380853250937432,
|
|
"loss": 5.4825,
|
|
"mean_token_accuracy": 0.19567041844129562,
|
|
"num_tokens": 8061818.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 6.216147613525391,
|
|
"epoch": 4.0954748881153655,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00038050847989881236,
|
|
"loss": 5.5625,
|
|
"mean_token_accuracy": 0.1903337776660919,
|
|
"num_tokens": 8071693.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 6.114920616149902,
|
|
"epoch": 4.100447538538041,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003801633910046035,
|
|
"loss": 5.3475,
|
|
"mean_token_accuracy": 0.21196105033159257,
|
|
"num_tokens": 8082304.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 6.237759733200074,
|
|
"epoch": 4.105420188960716,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003798179852944225,
|
|
"loss": 5.4578,
|
|
"mean_token_accuracy": 0.19175122082233428,
|
|
"num_tokens": 8092319.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 6.238774919509888,
|
|
"epoch": 4.110392839383391,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00037947226380884115,
|
|
"loss": 5.4756,
|
|
"mean_token_accuracy": 0.19483212977647782,
|
|
"num_tokens": 8100894.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 6.1681348323822025,
|
|
"epoch": 4.1153654898060665,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00037912622758938243,
|
|
"loss": 5.5746,
|
|
"mean_token_accuracy": 0.18238226920366288,
|
|
"num_tokens": 8110029.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 6.220950698852539,
|
|
"epoch": 4.120338140228742,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003787798776785174,
|
|
"loss": 5.4544,
|
|
"mean_token_accuracy": 0.19198264330625534,
|
|
"num_tokens": 8118715.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 6.147511672973633,
|
|
"epoch": 4.125310790651417,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003784332151196622,
|
|
"loss": 5.4541,
|
|
"mean_token_accuracy": 0.19794805496931075,
|
|
"num_tokens": 8128778.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 6.167963600158691,
|
|
"epoch": 4.130283441074092,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00037808624095717503,
|
|
"loss": 5.4797,
|
|
"mean_token_accuracy": 0.1910448744893074,
|
|
"num_tokens": 8138126.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 6.163439559936523,
|
|
"epoch": 4.1352560914967675,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003777389562363526,
|
|
"loss": 5.4568,
|
|
"mean_token_accuracy": 0.20317636728286742,
|
|
"num_tokens": 8147971.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 6.261989545822144,
|
|
"epoch": 4.140228741919443,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00037739136200342743,
|
|
"loss": 5.5351,
|
|
"mean_token_accuracy": 0.18970374315977095,
|
|
"num_tokens": 8157853.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 6.125674200057984,
|
|
"epoch": 4.145201392342118,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00037704345930556426,
|
|
"loss": 5.4398,
|
|
"mean_token_accuracy": 0.19906748682260514,
|
|
"num_tokens": 8167187.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 6.1246006965637205,
|
|
"epoch": 4.150174042764793,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00037669524919085716,
|
|
"loss": 5.4056,
|
|
"mean_token_accuracy": 0.19908996820449829,
|
|
"num_tokens": 8175492.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 6.149858570098877,
|
|
"epoch": 4.1551466931874685,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003763467327083264,
|
|
"loss": 5.5026,
|
|
"mean_token_accuracy": 0.19565404802560807,
|
|
"num_tokens": 8185356.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 6.0935914516448975,
|
|
"epoch": 4.160119343610144,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00037599791090791527,
|
|
"loss": 5.326,
|
|
"mean_token_accuracy": 0.20395482033491136,
|
|
"num_tokens": 8194833.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 6.283611631393432,
|
|
"epoch": 4.165091994032819,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003756487848404868,
|
|
"loss": 5.5712,
|
|
"mean_token_accuracy": 0.18659697473049164,
|
|
"num_tokens": 8205040.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 6.227370882034302,
|
|
"epoch": 4.170064644455495,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003752993555578205,
|
|
"loss": 5.5442,
|
|
"mean_token_accuracy": 0.18751176148653032,
|
|
"num_tokens": 8214769.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 6.050363206863404,
|
|
"epoch": 4.17503729487817,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003749496241126095,
|
|
"loss": 5.3785,
|
|
"mean_token_accuracy": 0.21155274510383607,
|
|
"num_tokens": 8225007.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 6.243951654434204,
|
|
"epoch": 4.180009945300846,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00037459959155845726,
|
|
"loss": 5.5092,
|
|
"mean_token_accuracy": 0.18836152106523513,
|
|
"num_tokens": 8234543.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 6.215252113342285,
|
|
"epoch": 4.184982595723521,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003742492589498744,
|
|
"loss": 5.4113,
|
|
"mean_token_accuracy": 0.19762518703937532,
|
|
"num_tokens": 8243790.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 6.1484856605529785,
|
|
"epoch": 4.189955246146196,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00037389862734227525,
|
|
"loss": 5.4952,
|
|
"mean_token_accuracy": 0.1932399034500122,
|
|
"num_tokens": 8253144.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 6.1374287605285645,
|
|
"epoch": 4.194927896568871,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00037354769779197514,
|
|
"loss": 5.4776,
|
|
"mean_token_accuracy": 0.19037245959043503,
|
|
"num_tokens": 8262782.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 6.186536407470703,
|
|
"epoch": 4.199900546991547,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00037319647135618695,
|
|
"loss": 5.4898,
|
|
"mean_token_accuracy": 0.1883343502879143,
|
|
"num_tokens": 8271544.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 6.206919765472412,
|
|
"epoch": 4.204873197414222,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00037284494909301793,
|
|
"loss": 5.4699,
|
|
"mean_token_accuracy": 0.19875955134630202,
|
|
"num_tokens": 8280574.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 6.167245197296142,
|
|
"epoch": 4.209845847836897,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003724931320614663,
|
|
"loss": 5.4442,
|
|
"mean_token_accuracy": 0.20010717064142228,
|
|
"num_tokens": 8290318.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 6.125017404556274,
|
|
"epoch": 4.214818498259572,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00037214102132141874,
|
|
"loss": 5.4769,
|
|
"mean_token_accuracy": 0.19579645544290541,
|
|
"num_tokens": 8299619.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 6.10298752784729,
|
|
"epoch": 4.219791148682248,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003717886179336465,
|
|
"loss": 5.4293,
|
|
"mean_token_accuracy": 0.20446307808160782,
|
|
"num_tokens": 8310085.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 6.142875385284424,
|
|
"epoch": 4.224763799104923,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00037143592295980243,
|
|
"loss": 5.4911,
|
|
"mean_token_accuracy": 0.19921409487724304,
|
|
"num_tokens": 8320673.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 6.159069061279297,
|
|
"epoch": 4.229736449527598,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003710829374624181,
|
|
"loss": 5.463,
|
|
"mean_token_accuracy": 0.19641628563404084,
|
|
"num_tokens": 8331073.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 6.187798881530762,
|
|
"epoch": 4.2347090999502734,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003707296625048999,
|
|
"loss": 5.5419,
|
|
"mean_token_accuracy": 0.18706087172031402,
|
|
"num_tokens": 8341313.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 6.177285528182983,
|
|
"epoch": 4.239681750372949,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003703760991515264,
|
|
"loss": 5.5564,
|
|
"mean_token_accuracy": 0.18458047956228257,
|
|
"num_tokens": 8351004.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 6.048095178604126,
|
|
"epoch": 4.244654400795624,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00037002224846744526,
|
|
"loss": 5.3015,
|
|
"mean_token_accuracy": 0.2101175755262375,
|
|
"num_tokens": 8361083.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 6.188832950592041,
|
|
"epoch": 4.249627051218299,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00036966811151866943,
|
|
"loss": 5.5709,
|
|
"mean_token_accuracy": 0.1868265077471733,
|
|
"num_tokens": 8372162.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 6.230817747116089,
|
|
"epoch": 4.2545997016409745,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00036931368937207436,
|
|
"loss": 5.574,
|
|
"mean_token_accuracy": 0.18524915128946304,
|
|
"num_tokens": 8382259.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 6.23739914894104,
|
|
"epoch": 4.25957235206365,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003689589830953948,
|
|
"loss": 5.5033,
|
|
"mean_token_accuracy": 0.1943778559565544,
|
|
"num_tokens": 8392275.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 6.153167295455932,
|
|
"epoch": 4.264545002486325,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003686039937572213,
|
|
"loss": 5.4909,
|
|
"mean_token_accuracy": 0.19662642776966094,
|
|
"num_tokens": 8402481.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 6.106589221954346,
|
|
"epoch": 4.269517652909,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003682487224269972,
|
|
"loss": 5.4891,
|
|
"mean_token_accuracy": 0.1939071610569954,
|
|
"num_tokens": 8412609.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 6.175233459472656,
|
|
"epoch": 4.2744903033316755,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003678931701750155,
|
|
"loss": 5.5332,
|
|
"mean_token_accuracy": 0.18761128634214402,
|
|
"num_tokens": 8421396.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 6.1679497241973875,
|
|
"epoch": 4.279462953754351,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00036753733807241536,
|
|
"loss": 5.4829,
|
|
"mean_token_accuracy": 0.19499483704566956,
|
|
"num_tokens": 8430076.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 6.094186782836914,
|
|
"epoch": 4.284435604177026,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00036718122719117907,
|
|
"loss": 5.3046,
|
|
"mean_token_accuracy": 0.2123893141746521,
|
|
"num_tokens": 8440199.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 6.150802040100098,
|
|
"epoch": 4.289408254599701,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003668248386041288,
|
|
"loss": 5.47,
|
|
"mean_token_accuracy": 0.19511553794145584,
|
|
"num_tokens": 8449463.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 6.151372623443604,
|
|
"epoch": 4.294380905022377,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00036646817338492334,
|
|
"loss": 5.3647,
|
|
"mean_token_accuracy": 0.19646665304899216,
|
|
"num_tokens": 8459824.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 6.206710958480835,
|
|
"epoch": 4.299353555445053,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003661112326080548,
|
|
"loss": 5.5021,
|
|
"mean_token_accuracy": 0.19226923882961272,
|
|
"num_tokens": 8468778.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 6.155476474761963,
|
|
"epoch": 4.304326205867728,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003657540173488455,
|
|
"loss": 5.4371,
|
|
"mean_token_accuracy": 0.19976142346858977,
|
|
"num_tokens": 8478542.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 6.078841829299927,
|
|
"epoch": 4.309298856290403,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003653965286834446,
|
|
"loss": 5.3646,
|
|
"mean_token_accuracy": 0.20851506292819977,
|
|
"num_tokens": 8489652.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 6.215034437179566,
|
|
"epoch": 4.314271506713078,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003650387676888251,
|
|
"loss": 5.544,
|
|
"mean_token_accuracy": 0.18692104518413544,
|
|
"num_tokens": 8499951.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 6.2712866306304935,
|
|
"epoch": 4.319244157135754,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003646807354427802,
|
|
"loss": 5.6774,
|
|
"mean_token_accuracy": 0.1810026213526726,
|
|
"num_tokens": 8510664.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 6.256493663787841,
|
|
"epoch": 4.324216807558429,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00036432243302392025,
|
|
"loss": 5.5207,
|
|
"mean_token_accuracy": 0.19319245219230652,
|
|
"num_tokens": 8519768.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 6.232345676422119,
|
|
"epoch": 4.329189457981104,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00036396386151166975,
|
|
"loss": 5.6041,
|
|
"mean_token_accuracy": 0.18509972840547562,
|
|
"num_tokens": 8530909.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 6.209866094589233,
|
|
"epoch": 4.334162108403779,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003636050219862638,
|
|
"loss": 5.6105,
|
|
"mean_token_accuracy": 0.18592436909675597,
|
|
"num_tokens": 8541357.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 6.188391590118409,
|
|
"epoch": 4.339134758826455,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00036324591552874493,
|
|
"loss": 5.5106,
|
|
"mean_token_accuracy": 0.19171480238437652,
|
|
"num_tokens": 8550819.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 6.256089687347412,
|
|
"epoch": 4.34410740924913,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00036288654322095973,
|
|
"loss": 5.5751,
|
|
"mean_token_accuracy": 0.18829000890254974,
|
|
"num_tokens": 8561577.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 6.260975456237793,
|
|
"epoch": 4.349080059671805,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00036252690614555557,
|
|
"loss": 5.5818,
|
|
"mean_token_accuracy": 0.18135609477758408,
|
|
"num_tokens": 8572590.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 6.190154266357422,
|
|
"epoch": 4.35405271009448,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003621670053859778,
|
|
"loss": 5.5371,
|
|
"mean_token_accuracy": 0.1896975517272949,
|
|
"num_tokens": 8583559.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 6.183533191680908,
|
|
"epoch": 4.359025360517156,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000361806842026466,
|
|
"loss": 5.5762,
|
|
"mean_token_accuracy": 0.18470533341169357,
|
|
"num_tokens": 8592845.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 6.0286328315734865,
|
|
"epoch": 4.363998010939831,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00036144641715205077,
|
|
"loss": 5.3088,
|
|
"mean_token_accuracy": 0.20722802579402924,
|
|
"num_tokens": 8602862.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 6.012164354324341,
|
|
"epoch": 4.368970661362506,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0003610857318485507,
|
|
"loss": 5.4124,
|
|
"mean_token_accuracy": 0.2122264638543129,
|
|
"num_tokens": 8613793.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 6.021222686767578,
|
|
"epoch": 4.373943311785181,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00036072478720256885,
|
|
"loss": 5.288,
|
|
"mean_token_accuracy": 0.21395353227853775,
|
|
"num_tokens": 8623086.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 6.0325323104858395,
|
|
"epoch": 4.378915962207857,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0003603635843014895,
|
|
"loss": 5.3659,
|
|
"mean_token_accuracy": 0.20556246489286423,
|
|
"num_tokens": 8632392.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 6.202382659912109,
|
|
"epoch": 4.383888612630532,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00036000212423347523,
|
|
"loss": 5.6186,
|
|
"mean_token_accuracy": 0.18220187723636627,
|
|
"num_tokens": 8642452.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 6.174440145492554,
|
|
"epoch": 4.388861263053207,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000359640408087463,
|
|
"loss": 5.5639,
|
|
"mean_token_accuracy": 0.18803748041391372,
|
|
"num_tokens": 8651431.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 6.238477087020874,
|
|
"epoch": 4.393833913475882,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003592784369531615,
|
|
"loss": 5.6507,
|
|
"mean_token_accuracy": 0.18258023709058763,
|
|
"num_tokens": 8661348.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 6.209085750579834,
|
|
"epoch": 4.398806563898558,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00035891621192104764,
|
|
"loss": 5.5627,
|
|
"mean_token_accuracy": 0.18565418869256972,
|
|
"num_tokens": 8671038.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 6.1717499732971195,
|
|
"epoch": 4.403779214321233,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00035855373408236294,
|
|
"loss": 5.5199,
|
|
"mean_token_accuracy": 0.1885978251695633,
|
|
"num_tokens": 8680115.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 6.180452013015747,
|
|
"epoch": 4.408751864743908,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003581910045291107,
|
|
"loss": 5.5764,
|
|
"mean_token_accuracy": 0.1837707966566086,
|
|
"num_tokens": 8689665.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 6.185402774810791,
|
|
"epoch": 4.413724515166583,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003578280243540526,
|
|
"loss": 5.5443,
|
|
"mean_token_accuracy": 0.18745121210813523,
|
|
"num_tokens": 8700114.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 6.028589248657227,
|
|
"epoch": 4.4186971655892595,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00035746479465070515,
|
|
"loss": 5.3018,
|
|
"mean_token_accuracy": 0.21425554305315017,
|
|
"num_tokens": 8710154.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 6.187015962600708,
|
|
"epoch": 4.423669816011935,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00035710131651333683,
|
|
"loss": 5.6444,
|
|
"mean_token_accuracy": 0.18069725781679152,
|
|
"num_tokens": 8720787.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 6.144265460968017,
|
|
"epoch": 4.42864246643461,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00035673759103696447,
|
|
"loss": 5.4893,
|
|
"mean_token_accuracy": 0.19696639031171798,
|
|
"num_tokens": 8731069.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 6.165715980529785,
|
|
"epoch": 4.433615116857285,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003563736193173499,
|
|
"loss": 5.478,
|
|
"mean_token_accuracy": 0.19825471192598343,
|
|
"num_tokens": 8740149.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 6.245183610916138,
|
|
"epoch": 4.4385877672799605,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003560094024509969,
|
|
"loss": 5.5719,
|
|
"mean_token_accuracy": 0.18873225152492523,
|
|
"num_tokens": 8749622.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 6.090638875961304,
|
|
"epoch": 4.443560417702636,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00035564494153514777,
|
|
"loss": 5.4815,
|
|
"mean_token_accuracy": 0.19764526635408403,
|
|
"num_tokens": 8759521.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 6.128550958633423,
|
|
"epoch": 4.448533068125311,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00035528023766778,
|
|
"loss": 5.5604,
|
|
"mean_token_accuracy": 0.19401091784238816,
|
|
"num_tokens": 8769387.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 6.208818340301514,
|
|
"epoch": 4.453505718547986,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0003549152919476031,
|
|
"loss": 5.5132,
|
|
"mean_token_accuracy": 0.19581702202558518,
|
|
"num_tokens": 8777871.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 6.168417453765869,
|
|
"epoch": 4.4584783689706615,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00035455010547405513,
|
|
"loss": 5.5699,
|
|
"mean_token_accuracy": 0.1929297089576721,
|
|
"num_tokens": 8787613.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 6.181653642654419,
|
|
"epoch": 4.463451019393337,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00035418467934729934,
|
|
"loss": 5.5296,
|
|
"mean_token_accuracy": 0.17932227849960328,
|
|
"num_tokens": 8797075.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 6.1851804733276365,
|
|
"epoch": 4.468423669816012,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00035381901466822113,
|
|
"loss": 5.6121,
|
|
"mean_token_accuracy": 0.1888619914650917,
|
|
"num_tokens": 8806087.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 6.243667411804199,
|
|
"epoch": 4.473396320238687,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003534531125384244,
|
|
"loss": 5.5558,
|
|
"mean_token_accuracy": 0.18323901444673538,
|
|
"num_tokens": 8815358.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"entropy": 6.238092947006225,
|
|
"epoch": 4.4783689706613625,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00035308697406022865,
|
|
"loss": 5.5786,
|
|
"mean_token_accuracy": 0.19499890357255936,
|
|
"num_tokens": 8825559.0,
|
|
"step": 4505
|
|
},
|
|
{
|
|
"entropy": 6.133929824829101,
|
|
"epoch": 4.483341621084038,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00035272060033666503,
|
|
"loss": 5.4717,
|
|
"mean_token_accuracy": 0.19509387612342835,
|
|
"num_tokens": 8835136.0,
|
|
"step": 4510
|
|
},
|
|
{
|
|
"entropy": 6.113640356063843,
|
|
"epoch": 4.488314271506713,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003523539924714738,
|
|
"loss": 5.5068,
|
|
"mean_token_accuracy": 0.1966530665755272,
|
|
"num_tokens": 8843936.0,
|
|
"step": 4515
|
|
},
|
|
{
|
|
"entropy": 6.205671882629394,
|
|
"epoch": 4.493286921929388,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00035198715156910016,
|
|
"loss": 5.6372,
|
|
"mean_token_accuracy": 0.17724415361881257,
|
|
"num_tokens": 8853990.0,
|
|
"step": 4520
|
|
},
|
|
{
|
|
"entropy": 6.14102029800415,
|
|
"epoch": 4.4982595723520635,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00035162007873469175,
|
|
"loss": 5.5845,
|
|
"mean_token_accuracy": 0.18677018731832504,
|
|
"num_tokens": 8863856.0,
|
|
"step": 4525
|
|
},
|
|
{
|
|
"entropy": 6.263545846939087,
|
|
"epoch": 4.503232222774739,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00035125277507409483,
|
|
"loss": 5.5458,
|
|
"mean_token_accuracy": 0.1901472330093384,
|
|
"num_tokens": 8872846.0,
|
|
"step": 4530
|
|
},
|
|
{
|
|
"entropy": 6.02675576210022,
|
|
"epoch": 4.508204873197414,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00035088524169385075,
|
|
"loss": 5.3822,
|
|
"mean_token_accuracy": 0.20844630151987076,
|
|
"num_tokens": 8883323.0,
|
|
"step": 4535
|
|
},
|
|
{
|
|
"entropy": 6.168177556991577,
|
|
"epoch": 4.513177523620089,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00035051747970119354,
|
|
"loss": 5.6506,
|
|
"mean_token_accuracy": 0.18423720747232436,
|
|
"num_tokens": 8893346.0,
|
|
"step": 4540
|
|
},
|
|
{
|
|
"entropy": 6.263223505020141,
|
|
"epoch": 4.5181501740427645,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00035014949020404523,
|
|
"loss": 5.5612,
|
|
"mean_token_accuracy": 0.1866194412112236,
|
|
"num_tokens": 8902480.0,
|
|
"step": 4545
|
|
},
|
|
{
|
|
"entropy": 6.1644909381866455,
|
|
"epoch": 4.52312282446544,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00034978127431101365,
|
|
"loss": 5.4457,
|
|
"mean_token_accuracy": 0.18779982924461364,
|
|
"num_tokens": 8913496.0,
|
|
"step": 4550
|
|
},
|
|
{
|
|
"entropy": 6.17883186340332,
|
|
"epoch": 4.528095474888115,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00034941283313138865,
|
|
"loss": 5.5747,
|
|
"mean_token_accuracy": 0.18634498566389085,
|
|
"num_tokens": 8924344.0,
|
|
"step": 4555
|
|
},
|
|
{
|
|
"entropy": 6.1152716159820555,
|
|
"epoch": 4.53306812531079,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003490441677751387,
|
|
"loss": 5.5128,
|
|
"mean_token_accuracy": 0.1943649798631668,
|
|
"num_tokens": 8933602.0,
|
|
"step": 4560
|
|
},
|
|
{
|
|
"entropy": 6.211251211166382,
|
|
"epoch": 4.5380407757334655,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00034867527935290773,
|
|
"loss": 5.5425,
|
|
"mean_token_accuracy": 0.18967919498682023,
|
|
"num_tokens": 8943722.0,
|
|
"step": 4565
|
|
},
|
|
{
|
|
"entropy": 6.188485622406006,
|
|
"epoch": 4.543013426156142,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00034830616897601146,
|
|
"loss": 5.4524,
|
|
"mean_token_accuracy": 0.19909740686416627,
|
|
"num_tokens": 8953535.0,
|
|
"step": 4570
|
|
},
|
|
{
|
|
"entropy": 6.028005361557007,
|
|
"epoch": 4.547986076578816,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003479368377564344,
|
|
"loss": 5.3295,
|
|
"mean_token_accuracy": 0.20190407931804658,
|
|
"num_tokens": 8963775.0,
|
|
"step": 4575
|
|
},
|
|
{
|
|
"entropy": 6.043946027755737,
|
|
"epoch": 4.552958727001492,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003475672868068265,
|
|
"loss": 5.4119,
|
|
"mean_token_accuracy": 0.19683539718389512,
|
|
"num_tokens": 8972822.0,
|
|
"step": 4580
|
|
},
|
|
{
|
|
"entropy": 6.094542932510376,
|
|
"epoch": 4.557931377424167,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00034719751724049955,
|
|
"loss": 5.572,
|
|
"mean_token_accuracy": 0.18594588935375214,
|
|
"num_tokens": 8983058.0,
|
|
"step": 4585
|
|
},
|
|
{
|
|
"entropy": 6.172282361984253,
|
|
"epoch": 4.562904027846843,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00034682753017142387,
|
|
"loss": 5.593,
|
|
"mean_token_accuracy": 0.18477794528007507,
|
|
"num_tokens": 8993069.0,
|
|
"step": 4590
|
|
},
|
|
{
|
|
"entropy": 6.215897083282471,
|
|
"epoch": 4.567876678269518,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00034645732671422517,
|
|
"loss": 5.6037,
|
|
"mean_token_accuracy": 0.1842757597565651,
|
|
"num_tokens": 9001471.0,
|
|
"step": 4595
|
|
},
|
|
{
|
|
"entropy": 6.189833164215088,
|
|
"epoch": 4.572849328692193,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003460869079841809,
|
|
"loss": 5.5515,
|
|
"mean_token_accuracy": 0.18947257101535797,
|
|
"num_tokens": 9012411.0,
|
|
"step": 4600
|
|
},
|
|
{
|
|
"entropy": 6.152330207824707,
|
|
"epoch": 4.577821979114868,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003457162750972172,
|
|
"loss": 5.4499,
|
|
"mean_token_accuracy": 0.20111989974975586,
|
|
"num_tokens": 9024126.0,
|
|
"step": 4605
|
|
},
|
|
{
|
|
"entropy": 6.100563716888428,
|
|
"epoch": 4.582794629537544,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00034534542916990534,
|
|
"loss": 5.4616,
|
|
"mean_token_accuracy": 0.19486365020275115,
|
|
"num_tokens": 9034355.0,
|
|
"step": 4610
|
|
},
|
|
{
|
|
"entropy": 6.154762935638428,
|
|
"epoch": 4.587767279960219,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00034497437131945837,
|
|
"loss": 5.4923,
|
|
"mean_token_accuracy": 0.19576872140169144,
|
|
"num_tokens": 9043772.0,
|
|
"step": 4615
|
|
},
|
|
{
|
|
"entropy": 6.1851600170135494,
|
|
"epoch": 4.592739930382894,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003446031026637277,
|
|
"loss": 5.492,
|
|
"mean_token_accuracy": 0.20086588859558105,
|
|
"num_tokens": 9053397.0,
|
|
"step": 4620
|
|
},
|
|
{
|
|
"entropy": 6.101235866546631,
|
|
"epoch": 4.597712580805569,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00034423162432119995,
|
|
"loss": 5.4478,
|
|
"mean_token_accuracy": 0.18851661384105683,
|
|
"num_tokens": 9063904.0,
|
|
"step": 4625
|
|
},
|
|
{
|
|
"entropy": 6.150432920455932,
|
|
"epoch": 4.602685231228245,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003438599374109934,
|
|
"loss": 5.4808,
|
|
"mean_token_accuracy": 0.19160433262586593,
|
|
"num_tokens": 9073146.0,
|
|
"step": 4630
|
|
},
|
|
{
|
|
"entropy": 6.016639852523804,
|
|
"epoch": 4.60765788165092,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00034348804305285466,
|
|
"loss": 5.34,
|
|
"mean_token_accuracy": 0.21535953283309936,
|
|
"num_tokens": 9083617.0,
|
|
"step": 4635
|
|
},
|
|
{
|
|
"entropy": 6.134166049957275,
|
|
"epoch": 4.612630532073595,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003431159423671552,
|
|
"loss": 5.5369,
|
|
"mean_token_accuracy": 0.18740968853235246,
|
|
"num_tokens": 9092639.0,
|
|
"step": 4640
|
|
},
|
|
{
|
|
"entropy": 6.151765012741089,
|
|
"epoch": 4.61760318249627,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00034274363647488826,
|
|
"loss": 5.6032,
|
|
"mean_token_accuracy": 0.18610710203647612,
|
|
"num_tokens": 9103377.0,
|
|
"step": 4645
|
|
},
|
|
{
|
|
"entropy": 6.10326623916626,
|
|
"epoch": 4.622575832918946,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003423711264976652,
|
|
"loss": 5.4292,
|
|
"mean_token_accuracy": 0.20029692947864533,
|
|
"num_tokens": 9112681.0,
|
|
"step": 4650
|
|
},
|
|
{
|
|
"entropy": 6.122379398345947,
|
|
"epoch": 4.627548483341621,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00034199841355771226,
|
|
"loss": 5.4801,
|
|
"mean_token_accuracy": 0.19609153866767884,
|
|
"num_tokens": 9121079.0,
|
|
"step": 4655
|
|
},
|
|
{
|
|
"entropy": 6.155416345596313,
|
|
"epoch": 4.632521133764296,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00034162549877786685,
|
|
"loss": 5.5562,
|
|
"mean_token_accuracy": 0.19373055398464203,
|
|
"num_tokens": 9131121.0,
|
|
"step": 4660
|
|
},
|
|
{
|
|
"entropy": 6.210113191604615,
|
|
"epoch": 4.6374937841869714,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003412523832815748,
|
|
"loss": 5.5932,
|
|
"mean_token_accuracy": 0.1828755885362625,
|
|
"num_tokens": 9140352.0,
|
|
"step": 4665
|
|
},
|
|
{
|
|
"entropy": 6.1812022686004635,
|
|
"epoch": 4.642466434609647,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003408790681928865,
|
|
"loss": 5.5603,
|
|
"mean_token_accuracy": 0.18068954944610596,
|
|
"num_tokens": 9150892.0,
|
|
"step": 4670
|
|
},
|
|
{
|
|
"entropy": 6.126920127868653,
|
|
"epoch": 4.647439085032322,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003405055546364535,
|
|
"loss": 5.5979,
|
|
"mean_token_accuracy": 0.18845024704933167,
|
|
"num_tokens": 9160943.0,
|
|
"step": 4675
|
|
},
|
|
{
|
|
"entropy": 6.2431694030761715,
|
|
"epoch": 4.652411735454997,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00034013184373752557,
|
|
"loss": 5.6381,
|
|
"mean_token_accuracy": 0.1869217947125435,
|
|
"num_tokens": 9171461.0,
|
|
"step": 4680
|
|
},
|
|
{
|
|
"entropy": 6.16233115196228,
|
|
"epoch": 4.6573843858776725,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003397579366219466,
|
|
"loss": 5.5536,
|
|
"mean_token_accuracy": 0.1911210760474205,
|
|
"num_tokens": 9181991.0,
|
|
"step": 4685
|
|
},
|
|
{
|
|
"entropy": 6.198425340652466,
|
|
"epoch": 4.662357036300348,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 0.0003393838344161519,
|
|
"loss": 5.3917,
|
|
"mean_token_accuracy": 0.20256743133068084,
|
|
"num_tokens": 9191715.0,
|
|
"step": 4690
|
|
},
|
|
{
|
|
"entropy": 6.151905107498169,
|
|
"epoch": 4.667329686723024,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00033900953824716443,
|
|
"loss": 5.4497,
|
|
"mean_token_accuracy": 0.19835282862186432,
|
|
"num_tokens": 9201338.0,
|
|
"step": 4695
|
|
},
|
|
{
|
|
"entropy": 6.198702096939087,
|
|
"epoch": 4.672302337145698,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003386350492425913,
|
|
"loss": 5.5929,
|
|
"mean_token_accuracy": 0.18748149126768113,
|
|
"num_tokens": 9210970.0,
|
|
"step": 4700
|
|
},
|
|
{
|
|
"entropy": 6.1795793056488035,
|
|
"epoch": 4.677274987568374,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00033826036853062095,
|
|
"loss": 5.5642,
|
|
"mean_token_accuracy": 0.187570258975029,
|
|
"num_tokens": 9221304.0,
|
|
"step": 4705
|
|
},
|
|
{
|
|
"entropy": 6.181730794906616,
|
|
"epoch": 4.68224763799105,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00033788549724001897,
|
|
"loss": 5.5114,
|
|
"mean_token_accuracy": 0.19059522598981857,
|
|
"num_tokens": 9231386.0,
|
|
"step": 4710
|
|
},
|
|
{
|
|
"entropy": 6.211932420730591,
|
|
"epoch": 4.687220288413725,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00033751043650012534,
|
|
"loss": 5.5796,
|
|
"mean_token_accuracy": 0.18814879655838013,
|
|
"num_tokens": 9240203.0,
|
|
"step": 4715
|
|
},
|
|
{
|
|
"entropy": 6.175481939315796,
|
|
"epoch": 4.6921929388364,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003371351874408506,
|
|
"loss": 5.5279,
|
|
"mean_token_accuracy": 0.18962276428937913,
|
|
"num_tokens": 9249623.0,
|
|
"step": 4720
|
|
},
|
|
{
|
|
"entropy": 6.207551670074463,
|
|
"epoch": 4.697165589259075,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0003367597511926728,
|
|
"loss": 5.6098,
|
|
"mean_token_accuracy": 0.18015818148851395,
|
|
"num_tokens": 9259359.0,
|
|
"step": 4725
|
|
},
|
|
{
|
|
"entropy": 6.155758666992187,
|
|
"epoch": 4.702138239681751,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00033638412888663374,
|
|
"loss": 5.4498,
|
|
"mean_token_accuracy": 0.1938175678253174,
|
|
"num_tokens": 9267897.0,
|
|
"step": 4730
|
|
},
|
|
{
|
|
"entropy": 6.1013831615448,
|
|
"epoch": 4.707110890104426,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003360083216543358,
|
|
"loss": 5.4252,
|
|
"mean_token_accuracy": 0.19910849630832672,
|
|
"num_tokens": 9277520.0,
|
|
"step": 4735
|
|
},
|
|
{
|
|
"entropy": 6.270988416671753,
|
|
"epoch": 4.712083540527101,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003356323306279387,
|
|
"loss": 5.7335,
|
|
"mean_token_accuracy": 0.17054524272680283,
|
|
"num_tokens": 9288157.0,
|
|
"step": 4740
|
|
},
|
|
{
|
|
"entropy": 6.205646514892578,
|
|
"epoch": 4.717056190949776,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00033525615694015545,
|
|
"loss": 5.4748,
|
|
"mean_token_accuracy": 0.19892797172069548,
|
|
"num_tokens": 9296698.0,
|
|
"step": 4745
|
|
},
|
|
{
|
|
"entropy": 6.204740333557129,
|
|
"epoch": 4.722028841372452,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00033487980172424955,
|
|
"loss": 5.5891,
|
|
"mean_token_accuracy": 0.18285587579011917,
|
|
"num_tokens": 9305880.0,
|
|
"step": 4750
|
|
},
|
|
{
|
|
"entropy": 6.216586780548096,
|
|
"epoch": 4.727001491795127,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00033450326611403136,
|
|
"loss": 5.619,
|
|
"mean_token_accuracy": 0.18418166935443878,
|
|
"num_tokens": 9316123.0,
|
|
"step": 4755
|
|
},
|
|
{
|
|
"entropy": 6.219891548156738,
|
|
"epoch": 4.731974142217802,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00033412655124385494,
|
|
"loss": 5.6111,
|
|
"mean_token_accuracy": 0.1840257540345192,
|
|
"num_tokens": 9325914.0,
|
|
"step": 4760
|
|
},
|
|
{
|
|
"entropy": 6.035458135604858,
|
|
"epoch": 4.736946792640477,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003337496582486139,
|
|
"loss": 5.4949,
|
|
"mean_token_accuracy": 0.1982277199625969,
|
|
"num_tokens": 9336330.0,
|
|
"step": 4765
|
|
},
|
|
{
|
|
"entropy": 6.1835180759429935,
|
|
"epoch": 4.741919443063153,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00033337258826373896,
|
|
"loss": 5.5561,
|
|
"mean_token_accuracy": 0.1863594353199005,
|
|
"num_tokens": 9345358.0,
|
|
"step": 4770
|
|
},
|
|
{
|
|
"entropy": 6.174221038818359,
|
|
"epoch": 4.746892093485828,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00033299534242519355,
|
|
"loss": 5.4826,
|
|
"mean_token_accuracy": 0.19162244945764542,
|
|
"num_tokens": 9355610.0,
|
|
"step": 4775
|
|
},
|
|
{
|
|
"entropy": 6.132575750350952,
|
|
"epoch": 4.751864743908503,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00033261792186947133,
|
|
"loss": 5.5631,
|
|
"mean_token_accuracy": 0.19499763250350952,
|
|
"num_tokens": 9364450.0,
|
|
"step": 4780
|
|
},
|
|
{
|
|
"entropy": 6.142422151565552,
|
|
"epoch": 4.756837394331178,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00033224032773359203,
|
|
"loss": 5.5959,
|
|
"mean_token_accuracy": 0.18504454642534257,
|
|
"num_tokens": 9374426.0,
|
|
"step": 4785
|
|
},
|
|
{
|
|
"entropy": 6.0517254829406735,
|
|
"epoch": 4.761810044753854,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00033186256115509845,
|
|
"loss": 5.3753,
|
|
"mean_token_accuracy": 0.21169984638690947,
|
|
"num_tokens": 9384067.0,
|
|
"step": 4790
|
|
},
|
|
{
|
|
"entropy": 6.0825094223022464,
|
|
"epoch": 4.766782695176529,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003314846232720528,
|
|
"loss": 5.3013,
|
|
"mean_token_accuracy": 0.21154463440179824,
|
|
"num_tokens": 9394269.0,
|
|
"step": 4795
|
|
},
|
|
{
|
|
"entropy": 6.167383146286011,
|
|
"epoch": 4.771755345599204,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00033110651522303346,
|
|
"loss": 5.475,
|
|
"mean_token_accuracy": 0.1990559607744217,
|
|
"num_tokens": 9403930.0,
|
|
"step": 4800
|
|
},
|
|
{
|
|
"entropy": 6.144851016998291,
|
|
"epoch": 4.776727996021879,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003307282381471314,
|
|
"loss": 5.5407,
|
|
"mean_token_accuracy": 0.17982854396104814,
|
|
"num_tokens": 9413520.0,
|
|
"step": 4805
|
|
},
|
|
{
|
|
"entropy": 6.094368743896484,
|
|
"epoch": 4.781700646444555,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00033034979318394676,
|
|
"loss": 5.5085,
|
|
"mean_token_accuracy": 0.19185639619827272,
|
|
"num_tokens": 9423945.0,
|
|
"step": 4810
|
|
},
|
|
{
|
|
"entropy": 6.144041872024536,
|
|
"epoch": 4.78667329686723,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00032997118147358554,
|
|
"loss": 5.5742,
|
|
"mean_token_accuracy": 0.19159136861562728,
|
|
"num_tokens": 9434139.0,
|
|
"step": 4815
|
|
},
|
|
{
|
|
"entropy": 6.182383394241333,
|
|
"epoch": 4.791645947289906,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000329592404156656,
|
|
"loss": 5.5682,
|
|
"mean_token_accuracy": 0.1973691999912262,
|
|
"num_tokens": 9443807.0,
|
|
"step": 4820
|
|
},
|
|
{
|
|
"entropy": 6.179702949523926,
|
|
"epoch": 4.79661859771258,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00032921346237426536,
|
|
"loss": 5.5541,
|
|
"mean_token_accuracy": 0.18842471092939378,
|
|
"num_tokens": 9452608.0,
|
|
"step": 4825
|
|
},
|
|
{
|
|
"entropy": 6.142550945281982,
|
|
"epoch": 4.8015912481352565,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00032883435726801635,
|
|
"loss": 5.5891,
|
|
"mean_token_accuracy": 0.1811933770775795,
|
|
"num_tokens": 9462233.0,
|
|
"step": 4830
|
|
},
|
|
{
|
|
"entropy": 6.1972095489501955,
|
|
"epoch": 4.806563898557932,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003284550899800036,
|
|
"loss": 5.5971,
|
|
"mean_token_accuracy": 0.18287865221500396,
|
|
"num_tokens": 9472926.0,
|
|
"step": 4835
|
|
},
|
|
{
|
|
"entropy": 6.1560694694519045,
|
|
"epoch": 4.811536548980607,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00032807566165281055,
|
|
"loss": 5.5041,
|
|
"mean_token_accuracy": 0.19433143734931946,
|
|
"num_tokens": 9481711.0,
|
|
"step": 4840
|
|
},
|
|
{
|
|
"entropy": 6.149320030212403,
|
|
"epoch": 4.816509199403282,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003276960734295054,
|
|
"loss": 5.4646,
|
|
"mean_token_accuracy": 0.1864618644118309,
|
|
"num_tokens": 9492655.0,
|
|
"step": 4845
|
|
},
|
|
{
|
|
"entropy": 6.123106288909912,
|
|
"epoch": 4.8214818498259575,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00032731632645363856,
|
|
"loss": 5.4488,
|
|
"mean_token_accuracy": 0.1973360151052475,
|
|
"num_tokens": 9502852.0,
|
|
"step": 4850
|
|
},
|
|
{
|
|
"entropy": 6.08588342666626,
|
|
"epoch": 4.826454500248633,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003269364218692384,
|
|
"loss": 5.4209,
|
|
"mean_token_accuracy": 0.20411384850740433,
|
|
"num_tokens": 9511781.0,
|
|
"step": 4855
|
|
},
|
|
{
|
|
"entropy": 6.115477657318115,
|
|
"epoch": 4.831427150671308,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003265563608208081,
|
|
"loss": 5.5201,
|
|
"mean_token_accuracy": 0.19253501147031785,
|
|
"num_tokens": 9521436.0,
|
|
"step": 4860
|
|
},
|
|
{
|
|
"entropy": 6.144715213775635,
|
|
"epoch": 4.836399801093983,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003261761444533221,
|
|
"loss": 5.4692,
|
|
"mean_token_accuracy": 0.1997506707906723,
|
|
"num_tokens": 9531432.0,
|
|
"step": 4865
|
|
},
|
|
{
|
|
"entropy": 6.195819091796875,
|
|
"epoch": 4.8413724515166585,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003257957739122231,
|
|
"loss": 5.6241,
|
|
"mean_token_accuracy": 0.1885920986533165,
|
|
"num_tokens": 9540976.0,
|
|
"step": 4870
|
|
},
|
|
{
|
|
"entropy": 6.127127695083618,
|
|
"epoch": 4.846345101939334,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00032541525034341805,
|
|
"loss": 5.3947,
|
|
"mean_token_accuracy": 0.19630798995494841,
|
|
"num_tokens": 9549566.0,
|
|
"step": 4875
|
|
},
|
|
{
|
|
"entropy": 6.072380542755127,
|
|
"epoch": 4.851317752362009,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003250345748932749,
|
|
"loss": 5.5005,
|
|
"mean_token_accuracy": 0.19089152663946152,
|
|
"num_tokens": 9558937.0,
|
|
"step": 4880
|
|
},
|
|
{
|
|
"entropy": 6.044972324371338,
|
|
"epoch": 4.856290402784684,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00032465374870861913,
|
|
"loss": 5.4857,
|
|
"mean_token_accuracy": 0.19674136787652968,
|
|
"num_tokens": 9568374.0,
|
|
"step": 4885
|
|
},
|
|
{
|
|
"entropy": 6.0751487731933596,
|
|
"epoch": 4.8612630532073595,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00032427277293673053,
|
|
"loss": 5.3763,
|
|
"mean_token_accuracy": 0.20399460196495056,
|
|
"num_tokens": 9577835.0,
|
|
"step": 4890
|
|
},
|
|
{
|
|
"entropy": 6.188834095001221,
|
|
"epoch": 4.866235703630035,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003238916487253393,
|
|
"loss": 5.5422,
|
|
"mean_token_accuracy": 0.18308869302272796,
|
|
"num_tokens": 9586787.0,
|
|
"step": 4895
|
|
},
|
|
{
|
|
"entropy": 6.130209732055664,
|
|
"epoch": 4.87120835405271,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003235103772226229,
|
|
"loss": 5.5671,
|
|
"mean_token_accuracy": 0.18771662265062333,
|
|
"num_tokens": 9597669.0,
|
|
"step": 4900
|
|
},
|
|
{
|
|
"entropy": 6.129355573654175,
|
|
"epoch": 4.876181004475385,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003231289595772026,
|
|
"loss": 5.4574,
|
|
"mean_token_accuracy": 0.1916378542780876,
|
|
"num_tokens": 9607535.0,
|
|
"step": 4905
|
|
},
|
|
{
|
|
"entropy": 6.1369863986969,
|
|
"epoch": 4.8811536548980605,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0003227473969381398,
|
|
"loss": 5.5249,
|
|
"mean_token_accuracy": 0.18959954380989075,
|
|
"num_tokens": 9616745.0,
|
|
"step": 4910
|
|
},
|
|
{
|
|
"entropy": 6.173167657852173,
|
|
"epoch": 4.886126305320736,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00032236569045493283,
|
|
"loss": 5.6081,
|
|
"mean_token_accuracy": 0.18494287878274918,
|
|
"num_tokens": 9626445.0,
|
|
"step": 4915
|
|
},
|
|
{
|
|
"entropy": 6.130174779891968,
|
|
"epoch": 4.891098955743411,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00032198384127751344,
|
|
"loss": 5.4481,
|
|
"mean_token_accuracy": 0.20307512134313582,
|
|
"num_tokens": 9635083.0,
|
|
"step": 4920
|
|
},
|
|
{
|
|
"entropy": 6.12288498878479,
|
|
"epoch": 4.896071606166086,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000321601850556243,
|
|
"loss": 5.6143,
|
|
"mean_token_accuracy": 0.18231757879257202,
|
|
"num_tokens": 9645410.0,
|
|
"step": 4925
|
|
},
|
|
{
|
|
"entropy": 6.165859842300415,
|
|
"epoch": 4.9010442565887615,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003212197194419095,
|
|
"loss": 5.54,
|
|
"mean_token_accuracy": 0.18427771627902984,
|
|
"num_tokens": 9654826.0,
|
|
"step": 4930
|
|
},
|
|
{
|
|
"entropy": 6.15842924118042,
|
|
"epoch": 4.906016907011437,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003208374490857238,
|
|
"loss": 5.5637,
|
|
"mean_token_accuracy": 0.17900614589452743,
|
|
"num_tokens": 9664609.0,
|
|
"step": 4935
|
|
},
|
|
{
|
|
"entropy": 6.011914491653442,
|
|
"epoch": 4.910989557434112,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003204550406393164,
|
|
"loss": 5.3664,
|
|
"mean_token_accuracy": 0.21100896149873732,
|
|
"num_tokens": 9672963.0,
|
|
"step": 4940
|
|
},
|
|
{
|
|
"entropy": 6.148446178436279,
|
|
"epoch": 4.915962207856787,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003200724952547335,
|
|
"loss": 5.4365,
|
|
"mean_token_accuracy": 0.20017173439264296,
|
|
"num_tokens": 9681662.0,
|
|
"step": 4945
|
|
},
|
|
{
|
|
"entropy": 6.063104677200317,
|
|
"epoch": 4.9209348582794625,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003196898140844341,
|
|
"loss": 5.362,
|
|
"mean_token_accuracy": 0.20787673443555832,
|
|
"num_tokens": 9693204.0,
|
|
"step": 4950
|
|
},
|
|
{
|
|
"entropy": 6.103886318206787,
|
|
"epoch": 4.925907508702139,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00031930699828128614,
|
|
"loss": 5.4675,
|
|
"mean_token_accuracy": 0.19004283249378204,
|
|
"num_tokens": 9702964.0,
|
|
"step": 4955
|
|
},
|
|
{
|
|
"entropy": 6.164912462234497,
|
|
"epoch": 4.930880159124813,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003189240489985633,
|
|
"loss": 5.4926,
|
|
"mean_token_accuracy": 0.2028622269630432,
|
|
"num_tokens": 9712407.0,
|
|
"step": 4960
|
|
},
|
|
{
|
|
"entropy": 6.154390859603882,
|
|
"epoch": 4.935852809547489,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003185409673899412,
|
|
"loss": 5.5423,
|
|
"mean_token_accuracy": 0.18527751564979553,
|
|
"num_tokens": 9722123.0,
|
|
"step": 4965
|
|
},
|
|
{
|
|
"entropy": 6.181591939926148,
|
|
"epoch": 4.940825459970164,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00031815775460949426,
|
|
"loss": 5.5018,
|
|
"mean_token_accuracy": 0.19198473989963533,
|
|
"num_tokens": 9732641.0,
|
|
"step": 4970
|
|
},
|
|
{
|
|
"entropy": 6.2073655128479,
|
|
"epoch": 4.94579811039284,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003177744118116919,
|
|
"loss": 5.5615,
|
|
"mean_token_accuracy": 0.1889641135931015,
|
|
"num_tokens": 9741498.0,
|
|
"step": 4975
|
|
},
|
|
{
|
|
"entropy": 6.161973905563355,
|
|
"epoch": 4.950770760815515,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0003173909401513954,
|
|
"loss": 5.5511,
|
|
"mean_token_accuracy": 0.18777817487716675,
|
|
"num_tokens": 9750811.0,
|
|
"step": 4980
|
|
},
|
|
{
|
|
"entropy": 5.9492865085601805,
|
|
"epoch": 4.95574341123819,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00031700734078385416,
|
|
"loss": 5.3846,
|
|
"mean_token_accuracy": 0.20711942464113237,
|
|
"num_tokens": 9761233.0,
|
|
"step": 4985
|
|
},
|
|
{
|
|
"entropy": 6.111984348297119,
|
|
"epoch": 4.960716061660865,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00031662361486470243,
|
|
"loss": 5.4301,
|
|
"mean_token_accuracy": 0.19681327193975448,
|
|
"num_tokens": 9771258.0,
|
|
"step": 4990
|
|
},
|
|
{
|
|
"entropy": 6.066908836364746,
|
|
"epoch": 4.965688712083541,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003162397635499555,
|
|
"loss": 5.4547,
|
|
"mean_token_accuracy": 0.19821509718894958,
|
|
"num_tokens": 9781501.0,
|
|
"step": 4995
|
|
},
|
|
{
|
|
"entropy": 6.144715309143066,
|
|
"epoch": 4.970661362506216,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00031585578799600653,
|
|
"loss": 5.7171,
|
|
"mean_token_accuracy": 0.1770050749182701,
|
|
"num_tokens": 9792488.0,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"epoch": 4.970661362506216,
|
|
"eval_entropy": 6.001577816800587,
|
|
"eval_loss": 6.4145588874816895,
|
|
"eval_mean_token_accuracy": 0.15077919325005898,
|
|
"eval_num_tokens": 9792488.0,
|
|
"eval_runtime": 1.92,
|
|
"eval_samples_per_second": 1804.67,
|
|
"eval_steps_per_second": 226.039,
|
|
"step": 5000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 10050,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 1000,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 2381576968488960.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|