grammar-synthesis-base / trainer_state.json
pszemraj's picture
upd chkpt +3 epochs additional finetuning
f070b50
raw
history blame
21.3 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.9971509971509972,
"global_step": 350,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"learning_rate": 2e-05,
"loss": 0.097,
"step": 2
},
{
"epoch": 0.01,
"learning_rate": 4e-05,
"loss": 0.1096,
"step": 4
},
{
"epoch": 0.02,
"learning_rate": 3.999672139632675e-05,
"loss": 0.098,
"step": 6
},
{
"epoch": 0.02,
"learning_rate": 3.9986886660231184e-05,
"loss": 0.0934,
"step": 8
},
{
"epoch": 0.03,
"learning_rate": 3.997049901613351e-05,
"loss": 0.1051,
"step": 10
},
{
"epoch": 0.03,
"learning_rate": 3.9947563836892725e-05,
"loss": 0.1093,
"step": 12
},
{
"epoch": 0.04,
"learning_rate": 3.9918088642045126e-05,
"loss": 0.1113,
"step": 14
},
{
"epoch": 0.05,
"learning_rate": 3.9882083095338934e-05,
"loss": 0.1097,
"step": 16
},
{
"epoch": 0.05,
"learning_rate": 3.98395590015659e-05,
"loss": 0.0988,
"step": 18
},
{
"epoch": 0.06,
"learning_rate": 3.979053030269103e-05,
"loss": 0.1046,
"step": 20
},
{
"epoch": 0.06,
"learning_rate": 3.9735013073281564e-05,
"loss": 0.102,
"step": 22
},
{
"epoch": 0.07,
"learning_rate": 3.967302551523671e-05,
"loss": 0.1029,
"step": 24
},
{
"epoch": 0.07,
"learning_rate": 3.960458795182003e-05,
"loss": 0.1102,
"step": 26
},
{
"epoch": 0.08,
"learning_rate": 3.95297228209962e-05,
"loss": 0.1021,
"step": 28
},
{
"epoch": 0.09,
"learning_rate": 3.944845466807451e-05,
"loss": 0.1015,
"step": 30
},
{
"epoch": 0.09,
"learning_rate": 3.936081013766143e-05,
"loss": 0.109,
"step": 32
},
{
"epoch": 0.1,
"learning_rate": 3.9266817964924905e-05,
"loss": 0.1131,
"step": 34
},
{
"epoch": 0.1,
"learning_rate": 3.91665089661732e-05,
"loss": 0.1019,
"step": 36
},
{
"epoch": 0.11,
"learning_rate": 3.9059916028751496e-05,
"loss": 0.1032,
"step": 38
},
{
"epoch": 0.11,
"learning_rate": 3.894707410025941e-05,
"loss": 0.104,
"step": 40
},
{
"epoch": 0.12,
"learning_rate": 3.882802017709307e-05,
"loss": 0.1015,
"step": 42
},
{
"epoch": 0.13,
"learning_rate": 3.870279329231546e-05,
"loss": 0.106,
"step": 44
},
{
"epoch": 0.13,
"learning_rate": 3.857143450285901e-05,
"loss": 0.0968,
"step": 46
},
{
"epoch": 0.14,
"learning_rate": 3.84339868760647e-05,
"loss": 0.0982,
"step": 48
},
{
"epoch": 0.14,
"learning_rate": 3.829049547556193e-05,
"loss": 0.1069,
"step": 50
},
{
"epoch": 0.15,
"learning_rate": 3.8141007346493964e-05,
"loss": 0.1048,
"step": 52
},
{
"epoch": 0.15,
"learning_rate": 3.798557150009373e-05,
"loss": 0.1054,
"step": 54
},
{
"epoch": 0.16,
"learning_rate": 3.782423889761492e-05,
"loss": 0.1146,
"step": 56
},
{
"epoch": 0.17,
"learning_rate": 3.7657062433623825e-05,
"loss": 0.1015,
"step": 58
},
{
"epoch": 0.17,
"learning_rate": 3.748409691865737e-05,
"loss": 0.0976,
"step": 60
},
{
"epoch": 0.18,
"learning_rate": 3.7305399061252795e-05,
"loss": 0.1108,
"step": 62
},
{
"epoch": 0.18,
"learning_rate": 3.712102744935529e-05,
"loss": 0.1041,
"step": 64
},
{
"epoch": 0.19,
"learning_rate": 3.6931042531109246e-05,
"loss": 0.1061,
"step": 66
},
{
"epoch": 0.19,
"learning_rate": 3.673550659503975e-05,
"loss": 0.0952,
"step": 68
},
{
"epoch": 0.2,
"learning_rate": 3.6534483749630624e-05,
"loss": 0.1023,
"step": 70
},
{
"epoch": 0.21,
"learning_rate": 3.6328039902305806e-05,
"loss": 0.0961,
"step": 72
},
{
"epoch": 0.21,
"learning_rate": 3.611624273782092e-05,
"loss": 0.0971,
"step": 74
},
{
"epoch": 0.22,
"learning_rate": 3.589916169607209e-05,
"loss": 0.1019,
"step": 76
},
{
"epoch": 0.22,
"learning_rate": 3.567686794932943e-05,
"loss": 0.1016,
"step": 78
},
{
"epoch": 0.23,
"learning_rate": 3.544943437890238e-05,
"loss": 0.1057,
"step": 80
},
{
"epoch": 0.23,
"learning_rate": 3.5216935551244896e-05,
"loss": 0.104,
"step": 82
},
{
"epoch": 0.24,
"learning_rate": 3.4979447693508e-05,
"loss": 0.103,
"step": 84
},
{
"epoch": 0.25,
"learning_rate": 3.4737048668547995e-05,
"loss": 0.1039,
"step": 86
},
{
"epoch": 0.25,
"learning_rate": 3.4489817949398224e-05,
"loss": 0.0955,
"step": 88
},
{
"epoch": 0.26,
"learning_rate": 3.423783659321307e-05,
"loss": 0.1059,
"step": 90
},
{
"epoch": 0.26,
"learning_rate": 3.398118721469255e-05,
"loss": 0.1028,
"step": 92
},
{
"epoch": 0.27,
"learning_rate": 3.371995395899618e-05,
"loss": 0.1056,
"step": 94
},
{
"epoch": 0.27,
"learning_rate": 3.345422247415512e-05,
"loss": 0.0985,
"step": 96
},
{
"epoch": 0.28,
"learning_rate": 3.3184079882991606e-05,
"loss": 0.106,
"step": 98
},
{
"epoch": 0.28,
"learning_rate": 3.29096147545548e-05,
"loss": 0.0983,
"step": 100
},
{
"epoch": 0.29,
"learning_rate": 3.2630917075082545e-05,
"loss": 0.0979,
"step": 102
},
{
"epoch": 0.3,
"learning_rate": 3.234807821849838e-05,
"loss": 0.0987,
"step": 104
},
{
"epoch": 0.3,
"learning_rate": 3.2061190916453745e-05,
"loss": 0.1096,
"step": 106
},
{
"epoch": 0.31,
"learning_rate": 3.1770349227924854e-05,
"loss": 0.11,
"step": 108
},
{
"epoch": 0.31,
"learning_rate": 3.147564850837455e-05,
"loss": 0.1004,
"step": 110
},
{
"epoch": 0.32,
"learning_rate": 3.1177185378488984e-05,
"loss": 0.0939,
"step": 112
},
{
"epoch": 0.32,
"learning_rate": 3.0875057692499566e-05,
"loss": 0.0944,
"step": 114
},
{
"epoch": 0.33,
"learning_rate": 3.05693645061004e-05,
"loss": 0.1117,
"step": 116
},
{
"epoch": 0.34,
"learning_rate": 3.0260206043971857e-05,
"loss": 0.0962,
"step": 118
},
{
"epoch": 0.34,
"learning_rate": 2.9947683666920913e-05,
"loss": 0.0993,
"step": 120
},
{
"epoch": 0.35,
"learning_rate": 2.9631899838648887e-05,
"loss": 0.0946,
"step": 122
},
{
"epoch": 0.35,
"learning_rate": 2.9312958092157724e-05,
"loss": 0.1003,
"step": 124
},
{
"epoch": 0.36,
"learning_rate": 2.8990962995805577e-05,
"loss": 0.1009,
"step": 126
},
{
"epoch": 0.36,
"learning_rate": 2.866602011902301e-05,
"loss": 0.0913,
"step": 128
},
{
"epoch": 0.37,
"learning_rate": 2.833823599770098e-05,
"loss": 0.0961,
"step": 130
},
{
"epoch": 0.38,
"learning_rate": 2.8007718099261886e-05,
"loss": 0.1042,
"step": 132
},
{
"epoch": 0.38,
"learning_rate": 2.767457478742533e-05,
"loss": 0.1049,
"step": 134
},
{
"epoch": 0.39,
"learning_rate": 2.733891528667991e-05,
"loss": 0.1063,
"step": 136
},
{
"epoch": 0.39,
"learning_rate": 2.7000849646472826e-05,
"loss": 0.1028,
"step": 138
},
{
"epoch": 0.4,
"learning_rate": 2.6660488705129054e-05,
"loss": 0.0973,
"step": 140
},
{
"epoch": 0.4,
"learning_rate": 2.6317944053511853e-05,
"loss": 0.1022,
"step": 142
},
{
"epoch": 0.41,
"learning_rate": 2.5973327998436527e-05,
"loss": 0.1044,
"step": 144
},
{
"epoch": 0.42,
"learning_rate": 2.562675352584947e-05,
"loss": 0.104,
"step": 146
},
{
"epoch": 0.42,
"learning_rate": 2.5278334263784587e-05,
"loss": 0.1015,
"step": 148
},
{
"epoch": 0.43,
"learning_rate": 2.4928184445109108e-05,
"loss": 0.1026,
"step": 150
},
{
"epoch": 0.43,
"learning_rate": 2.457641887007121e-05,
"loss": 0.1108,
"step": 152
},
{
"epoch": 0.44,
"learning_rate": 2.4223152868661535e-05,
"loss": 0.104,
"step": 154
},
{
"epoch": 0.44,
"learning_rate": 2.3868502262801065e-05,
"loss": 0.1013,
"step": 156
},
{
"epoch": 0.45,
"learning_rate": 2.3512583328367717e-05,
"loss": 0.1004,
"step": 158
},
{
"epoch": 0.46,
"learning_rate": 2.3155512757074065e-05,
"loss": 0.0986,
"step": 160
},
{
"epoch": 0.46,
"learning_rate": 2.2797407618208784e-05,
"loss": 0.0986,
"step": 162
},
{
"epoch": 0.47,
"learning_rate": 2.2438385320254234e-05,
"loss": 0.1106,
"step": 164
},
{
"epoch": 0.47,
"learning_rate": 2.2078563572392907e-05,
"loss": 0.097,
"step": 166
},
{
"epoch": 0.48,
"learning_rate": 2.171806034591522e-05,
"loss": 0.0981,
"step": 168
},
{
"epoch": 0.48,
"learning_rate": 2.135699383554144e-05,
"loss": 0.1088,
"step": 170
},
{
"epoch": 0.49,
"learning_rate": 2.099548242067028e-05,
"loss": 0.0911,
"step": 172
},
{
"epoch": 0.5,
"learning_rate": 2.0633644626567007e-05,
"loss": 0.0978,
"step": 174
},
{
"epoch": 0.5,
"learning_rate": 2.0271599085503722e-05,
"loss": 0.0912,
"step": 176
},
{
"epoch": 0.51,
"learning_rate": 1.9909464497864487e-05,
"loss": 0.107,
"step": 178
},
{
"epoch": 0.51,
"learning_rate": 1.954735959322825e-05,
"loss": 0.1027,
"step": 180
},
{
"epoch": 0.52,
"learning_rate": 1.9185403091442044e-05,
"loss": 0.1048,
"step": 182
},
{
"epoch": 0.52,
"learning_rate": 1.882371366369749e-05,
"loss": 0.0915,
"step": 184
},
{
"epoch": 0.53,
"learning_rate": 1.846240989362325e-05,
"loss": 0.1041,
"step": 186
},
{
"epoch": 0.54,
"learning_rate": 1.810161023840607e-05,
"loss": 0.1001,
"step": 188
},
{
"epoch": 0.54,
"learning_rate": 1.774143298995346e-05,
"loss": 0.1043,
"step": 190
},
{
"epoch": 0.55,
"learning_rate": 1.7381996236110386e-05,
"loss": 0.1067,
"step": 192
},
{
"epoch": 0.55,
"learning_rate": 1.702341782194301e-05,
"loss": 0.1095,
"step": 194
},
{
"epoch": 0.56,
"learning_rate": 1.6665815311101896e-05,
"loss": 0.1016,
"step": 196
},
{
"epoch": 0.56,
"learning_rate": 1.630930594727762e-05,
"loss": 0.0963,
"step": 198
},
{
"epoch": 0.57,
"learning_rate": 1.5954006615761158e-05,
"loss": 0.1036,
"step": 200
},
{
"epoch": 0.58,
"learning_rate": 1.560003380512185e-05,
"loss": 0.1136,
"step": 202
},
{
"epoch": 0.58,
"learning_rate": 1.5247503569015413e-05,
"loss": 0.0947,
"step": 204
},
{
"epoch": 0.59,
"learning_rate": 1.489653148813455e-05,
"loss": 0.1105,
"step": 206
},
{
"epoch": 0.59,
"learning_rate": 1.4547232632314624e-05,
"loss": 0.1033,
"step": 208
},
{
"epoch": 0.6,
"learning_rate": 1.4199721522806807e-05,
"loss": 0.102,
"step": 210
},
{
"epoch": 0.6,
"learning_rate": 1.3854112094731116e-05,
"loss": 0.1037,
"step": 212
},
{
"epoch": 0.61,
"learning_rate": 1.3510517659721583e-05,
"loss": 0.1005,
"step": 214
},
{
"epoch": 0.62,
"learning_rate": 1.316905086877589e-05,
"loss": 0.0979,
"step": 216
},
{
"epoch": 0.62,
"learning_rate": 1.2829823675321535e-05,
"loss": 0.1007,
"step": 218
},
{
"epoch": 0.63,
"learning_rate": 1.2492947298510783e-05,
"loss": 0.1002,
"step": 220
},
{
"epoch": 0.63,
"learning_rate": 1.2158532186756275e-05,
"loss": 0.1037,
"step": 222
},
{
"epoch": 0.64,
"learning_rate": 1.182668798151939e-05,
"loss": 0.1018,
"step": 224
},
{
"epoch": 0.64,
"learning_rate": 1.1497523481363146e-05,
"loss": 0.1002,
"step": 226
},
{
"epoch": 0.65,
"learning_rate": 1.1171146606281482e-05,
"loss": 0.0982,
"step": 228
},
{
"epoch": 0.66,
"learning_rate": 1.0847664362316549e-05,
"loss": 0.102,
"step": 230
},
{
"epoch": 0.66,
"learning_rate": 1.0527182806475662e-05,
"loss": 0.0928,
"step": 232
},
{
"epoch": 0.67,
"learning_rate": 1.020980701195946e-05,
"loss": 0.0996,
"step": 234
},
{
"epoch": 0.67,
"learning_rate": 9.895641033712507e-06,
"loss": 0.1014,
"step": 236
},
{
"epoch": 0.68,
"learning_rate": 9.584787874307828e-06,
"loss": 0.0994,
"step": 238
},
{
"epoch": 0.68,
"learning_rate": 9.277349450176445e-06,
"loss": 0.1092,
"step": 240
},
{
"epoch": 0.69,
"learning_rate": 8.97342655819303e-06,
"loss": 0.0992,
"step": 242
},
{
"epoch": 0.7,
"learning_rate": 8.673118842628595e-06,
"loss": 0.0892,
"step": 244
},
{
"epoch": 0.7,
"learning_rate": 8.376524762481069e-06,
"loss": 0.0975,
"step": 246
},
{
"epoch": 0.71,
"learning_rate": 8.083741559194515e-06,
"loss": 0.0982,
"step": 248
},
{
"epoch": 0.71,
"learning_rate": 7.794865224777504e-06,
"loss": 0.1026,
"step": 250
},
{
"epoch": 0.72,
"learning_rate": 7.509990470331159e-06,
"loss": 0.0973,
"step": 252
},
{
"epoch": 0.72,
"learning_rate": 7.229210694997113e-06,
"loss": 0.0985,
"step": 254
},
{
"epoch": 0.73,
"learning_rate": 6.952617955335641e-06,
"loss": 0.1005,
"step": 256
},
{
"epoch": 0.74,
"learning_rate": 6.680302935143963e-06,
"loss": 0.0968,
"step": 258
},
{
"epoch": 0.74,
"learning_rate": 6.412354915724642e-06,
"loss": 0.1079,
"step": 260
},
{
"epoch": 0.75,
"learning_rate": 6.14886174661373e-06,
"loss": 0.0994,
"step": 262
},
{
"epoch": 0.75,
"learning_rate": 5.889909816778458e-06,
"loss": 0.0991,
"step": 264
},
{
"epoch": 0.76,
"learning_rate": 5.635584026293655e-06,
"loss": 0.098,
"step": 266
},
{
"epoch": 0.76,
"learning_rate": 5.385967758506407e-06,
"loss": 0.1035,
"step": 268
},
{
"epoch": 0.77,
"learning_rate": 5.141142852697956e-06,
"loss": 0.1022,
"step": 270
},
{
"epoch": 0.77,
"learning_rate": 4.901189577251864e-06,
"loss": 0.0938,
"step": 272
},
{
"epoch": 0.78,
"learning_rate": 4.6661866033371506e-06,
"loss": 0.1026,
"step": 274
},
{
"epoch": 0.79,
"learning_rate": 4.4362109791151695e-06,
"loss": 0.0983,
"step": 276
},
{
"epoch": 0.79,
"learning_rate": 4.211338104478548e-06,
"loss": 0.1036,
"step": 278
},
{
"epoch": 0.8,
"learning_rate": 3.991641706330575e-06,
"loss": 0.092,
"step": 280
},
{
"epoch": 0.8,
"learning_rate": 3.777193814413045e-06,
"loss": 0.1038,
"step": 282
},
{
"epoch": 0.81,
"learning_rate": 3.5680647376905666e-06,
"loss": 0.1,
"step": 284
},
{
"epoch": 0.81,
"learning_rate": 3.3643230412990625e-06,
"loss": 0.1053,
"step": 286
},
{
"epoch": 0.82,
"learning_rate": 3.1660355240659423e-06,
"loss": 0.1029,
"step": 288
},
{
"epoch": 0.83,
"learning_rate": 2.973267196609453e-06,
"loss": 0.0974,
"step": 290
},
{
"epoch": 0.83,
"learning_rate": 2.786081260024236e-06,
"loss": 0.0966,
"step": 292
},
{
"epoch": 0.84,
"learning_rate": 2.604539085160218e-06,
"loss": 0.1029,
"step": 294
},
{
"epoch": 0.84,
"learning_rate": 2.428700192501534e-06,
"loss": 0.1031,
"step": 296
},
{
"epoch": 0.85,
"learning_rate": 2.2586222326521277e-06,
"loss": 0.1017,
"step": 298
},
{
"epoch": 0.85,
"learning_rate": 2.0943609674343833e-06,
"loss": 0.1021,
"step": 300
},
{
"epoch": 0.86,
"learning_rate": 1.9359702516070553e-06,
"loss": 0.103,
"step": 302
},
{
"epoch": 0.87,
"learning_rate": 1.7835020152084116e-06,
"loss": 0.1023,
"step": 304
},
{
"epoch": 0.87,
"learning_rate": 1.6370062465304503e-06,
"loss": 0.1034,
"step": 306
},
{
"epoch": 0.88,
"learning_rate": 1.496530975729693e-06,
"loss": 0.0926,
"step": 308
},
{
"epoch": 0.88,
"learning_rate": 1.3621222590800342e-06,
"loss": 0.099,
"step": 310
},
{
"epoch": 0.89,
"learning_rate": 1.2338241638726811e-06,
"loss": 0.0985,
"step": 312
},
{
"epoch": 0.89,
"learning_rate": 1.1116787539682571e-06,
"loss": 0.1044,
"step": 314
},
{
"epoch": 0.9,
"learning_rate": 9.957260760057164e-07,
"loss": 0.0929,
"step": 316
},
{
"epoch": 0.91,
"learning_rate": 8.860041462726543e-07,
"loss": 0.0988,
"step": 318
},
{
"epoch": 0.91,
"learning_rate": 7.825489382412521e-07,
"loss": 0.1043,
"step": 320
},
{
"epoch": 0.92,
"learning_rate": 6.853943707740218e-07,
"loss": 0.0975,
"step": 322
},
{
"epoch": 0.92,
"learning_rate": 5.945722970031332e-07,
"loss": 0.1065,
"step": 324
},
{
"epoch": 0.93,
"learning_rate": 5.101124938870605e-07,
"loss": 0.0978,
"step": 326
},
{
"epoch": 0.93,
"learning_rate": 4.320426524478749e-07,
"loss": 0.1,
"step": 328
},
{
"epoch": 0.94,
"learning_rate": 3.603883686924681e-07,
"loss": 0.0967,
"step": 330
},
{
"epoch": 0.95,
"learning_rate": 2.951731352206322e-07,
"loss": 0.0954,
"step": 332
},
{
"epoch": 0.95,
"learning_rate": 2.3641833352276768e-07,
"loss": 0.1016,
"step": 334
},
{
"epoch": 0.96,
"learning_rate": 1.841432269697463e-07,
"loss": 0.0993,
"step": 336
},
{
"epoch": 0.96,
"learning_rate": 1.3836495449719878e-07,
"loss": 0.1002,
"step": 338
},
{
"epoch": 0.97,
"learning_rate": 9.90985249863563e-08,
"loss": 0.1043,
"step": 340
},
{
"epoch": 0.97,
"learning_rate": 6.635681234321789e-08,
"loss": 0.0984,
"step": 342
},
{
"epoch": 0.98,
"learning_rate": 4.0150551277724494e-08,
"loss": 0.1046,
"step": 344
},
{
"epoch": 0.99,
"learning_rate": 2.0488333784249858e-08,
"loss": 0.0948,
"step": 346
},
{
"epoch": 0.99,
"learning_rate": 7.376606324644986e-09,
"loss": 0.0897,
"step": 348
},
{
"epoch": 1.0,
"learning_rate": 8.196677146932175e-10,
"loss": 0.0971,
"step": 350
}
],
"max_steps": 351,
"num_train_epochs": 1,
"total_flos": 2.454170351173632e+17,
"trial_name": null,
"trial_params": null
}