eminem / trainer_state.json
AlekseyKorshuk's picture
huggingartists
742dd5b
raw
history blame
11.7 kB
{
"best_metric": 1.8596649169921875,
"best_model_checkpoint": "output/eminem/checkpoint-457",
"epoch": 1.0,
"global_step": 457,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"learning_rate": 0.00013401701259429288,
"loss": 2.1682,
"step": 5
},
{
"epoch": 0.02,
"learning_rate": 0.0001332598992582699,
"loss": 2.1572,
"step": 10
},
{
"epoch": 0.03,
"learning_rate": 0.00013242470333428216,
"loss": 2.2158,
"step": 15
},
{
"epoch": 0.04,
"learning_rate": 0.0001315124333958035,
"loss": 2.1302,
"step": 20
},
{
"epoch": 0.06,
"learning_rate": 0.0001305241910900455,
"loss": 2.2621,
"step": 25
},
{
"epoch": 0.07,
"learning_rate": 0.0001294611698076205,
"loss": 2.1793,
"step": 30
},
{
"epoch": 0.08,
"learning_rate": 0.00012832465324141612,
"loss": 2.2551,
"step": 35
},
{
"epoch": 0.09,
"learning_rate": 0.0001271160138364218,
"loss": 2.1327,
"step": 40
},
{
"epoch": 0.1,
"learning_rate": 0.00012583671113237919,
"loss": 2.189,
"step": 45
},
{
"epoch": 0.11,
"learning_rate": 0.00012448829000125786,
"loss": 2.3515,
"step": 50
},
{
"epoch": 0.12,
"learning_rate": 0.00012307237878168522,
"loss": 2.2902,
"step": 55
},
{
"epoch": 0.13,
"learning_rate": 0.00012159068731258093,
"loss": 2.1917,
"step": 60
},
{
"epoch": 0.14,
"learning_rate": 0.00012004500486837715,
"loss": 2.0402,
"step": 65
},
{
"epoch": 0.15,
"learning_rate": 0.00011843719799830738,
"loss": 2.3137,
"step": 70
},
{
"epoch": 0.17,
"learning_rate": 0.00011676920827238334,
"loss": 2.1433,
"step": 75
},
{
"epoch": 0.18,
"learning_rate": 0.00011504304993677638,
"loss": 2.1509,
"step": 80
},
{
"epoch": 0.19,
"learning_rate": 0.00011326080748143405,
"loss": 2.1987,
"step": 85
},
{
"epoch": 0.2,
"learning_rate": 0.00011142463312287667,
"loss": 2.1876,
"step": 90
},
{
"epoch": 0.21,
"learning_rate": 0.00010953674420519918,
"loss": 2.3293,
"step": 95
},
{
"epoch": 0.22,
"learning_rate": 0.00010759942052243419,
"loss": 2.307,
"step": 100
},
{
"epoch": 0.23,
"learning_rate": 0.00010561500156549789,
"loss": 2.3729,
"step": 105
},
{
"epoch": 0.24,
"learning_rate": 0.00010358588369704255,
"loss": 2.3181,
"step": 110
},
{
"epoch": 0.25,
"learning_rate": 0.00010151451725763938,
"loss": 2.2169,
"step": 115
},
{
"epoch": 0.27,
"learning_rate": 9.940340360677185e-05,
"loss": 2.1579,
"step": 120
},
{
"epoch": 0.28,
"learning_rate": 9.725509210222182e-05,
"loss": 2.0936,
"step": 125
},
{
"epoch": 0.29,
"learning_rate": 9.507217702149286e-05,
"loss": 2.2298,
"step": 130
},
{
"epoch": 0.3,
"learning_rate": 9.28572944289891e-05,
"loss": 2.1296,
"step": 135
},
{
"epoch": 0.31,
"learning_rate": 9.061311899273271e-05,
"loss": 2.146,
"step": 140
},
{
"epoch": 0.32,
"learning_rate": 8.83423607544639e-05,
"loss": 2.2279,
"step": 145
},
{
"epoch": 0.33,
"learning_rate": 8.604776185702469e-05,
"loss": 2.1469,
"step": 150
},
{
"epoch": 0.34,
"learning_rate": 8.373209323297477e-05,
"loss": 2.2793,
"step": 155
},
{
"epoch": 0.35,
"learning_rate": 8.139815125844835e-05,
"loss": 1.9848,
"step": 160
},
{
"epoch": 0.37,
"learning_rate": 7.904875437627664e-05,
"loss": 2.6294,
"step": 165
},
{
"epoch": 0.38,
"learning_rate": 7.6686739692467e-05,
"loss": 2.2562,
"step": 170
},
{
"epoch": 0.39,
"learning_rate": 7.431495955014962e-05,
"loss": 2.337,
"step": 175
},
{
"epoch": 0.4,
"learning_rate": 7.193627808511231e-05,
"loss": 2.3747,
"step": 180
},
{
"epoch": 0.41,
"learning_rate": 6.955356776710623e-05,
"loss": 2.2054,
"step": 185
},
{
"epoch": 0.42,
"learning_rate": 6.716970593108287e-05,
"loss": 2.021,
"step": 190
},
{
"epoch": 0.43,
"learning_rate": 6.478757130255133e-05,
"loss": 2.3831,
"step": 195
},
{
"epoch": 0.44,
"learning_rate": 6.24100405212658e-05,
"loss": 2.3325,
"step": 200
},
{
"epoch": 0.45,
"learning_rate": 6.003998466742432e-05,
"loss": 2.0344,
"step": 205
},
{
"epoch": 0.46,
"learning_rate": 5.768026579458423e-05,
"loss": 2.1013,
"step": 210
},
{
"epoch": 0.48,
"learning_rate": 5.5333733473477525e-05,
"loss": 2.1656,
"step": 215
},
{
"epoch": 0.49,
"learning_rate": 5.300322135090038e-05,
"loss": 2.2654,
"step": 220
},
{
"epoch": 0.5,
"learning_rate": 5.069154372783225e-05,
"loss": 2.0446,
"step": 225
},
{
"epoch": 0.51,
"learning_rate": 4.840149216091673e-05,
"loss": 2.347,
"step": 230
},
{
"epoch": 0.52,
"learning_rate": 4.613583209140827e-05,
"loss": 2.0384,
"step": 235
},
{
"epoch": 0.53,
"learning_rate": 4.3897299505655516e-05,
"loss": 2.4872,
"step": 240
},
{
"epoch": 0.54,
"learning_rate": 4.1688597631154076e-05,
"loss": 2.2689,
"step": 245
},
{
"epoch": 0.55,
"learning_rate": 3.9512393672158556e-05,
"loss": 2.2838,
"step": 250
},
{
"epoch": 0.56,
"learning_rate": 3.7371315588795804e-05,
"loss": 2.0626,
"step": 255
},
{
"epoch": 0.58,
"learning_rate": 3.5267948923569414e-05,
"loss": 2.1429,
"step": 260
},
{
"epoch": 0.59,
"learning_rate": 3.320483367908473e-05,
"loss": 2.3835,
"step": 265
},
{
"epoch": 0.6,
"learning_rate": 3.118446125077409e-05,
"loss": 2.2844,
"step": 270
},
{
"epoch": 0.61,
"learning_rate": 2.9209271418309033e-05,
"loss": 2.1332,
"step": 275
},
{
"epoch": 0.62,
"learning_rate": 2.7281649399353233e-05,
"loss": 2.0147,
"step": 280
},
{
"epoch": 0.63,
"learning_rate": 2.5403922969199553e-05,
"loss": 2.0417,
"step": 285
},
{
"epoch": 0.64,
"learning_rate": 2.357835964977058e-05,
"loss": 2.2009,
"step": 290
},
{
"epoch": 0.65,
"learning_rate": 2.180716397138782e-05,
"loss": 2.1734,
"step": 295
},
{
"epoch": 0.66,
"learning_rate": 2.0092474810602975e-05,
"loss": 2.2568,
"step": 300
},
{
"epoch": 0.67,
"learning_rate": 1.843636280731442e-05,
"loss": 2.1657,
"step": 305
},
{
"epoch": 0.69,
"learning_rate": 1.6840827864285067e-05,
"loss": 2.1509,
"step": 310
},
{
"epoch": 0.7,
"learning_rate": 1.5307796732081633e-05,
"loss": 2.2271,
"step": 315
},
{
"epoch": 0.71,
"learning_rate": 1.3839120682351706e-05,
"loss": 2.2315,
"step": 320
},
{
"epoch": 0.72,
"learning_rate": 1.2436573272248327e-05,
"loss": 2.2438,
"step": 325
},
{
"epoch": 0.73,
"learning_rate": 1.1101848202701753e-05,
"loss": 2.2333,
"step": 330
},
{
"epoch": 0.74,
"learning_rate": 9.83655727312471e-06,
"loss": 2.3438,
"step": 335
},
{
"epoch": 0.75,
"learning_rate": 8.642228435021026e-06,
"loss": 1.9256,
"step": 340
},
{
"epoch": 0.76,
"learning_rate": 7.520303946848332e-06,
"loss": 2.1889,
"step": 345
},
{
"epoch": 0.77,
"learning_rate": 6.472138632361373e-06,
"loss": 2.1674,
"step": 350
},
{
"epoch": 0.79,
"learning_rate": 5.49899824454404e-06,
"loss": 2.099,
"step": 355
},
{
"epoch": 0.8,
"learning_rate": 4.602057937096952e-06,
"loss": 2.1071,
"step": 360
},
{
"epoch": 0.81,
"learning_rate": 3.7824008453367434e-06,
"loss": 2.2894,
"step": 365
},
{
"epoch": 0.82,
"learning_rate": 3.041016778213552e-06,
"loss": 2.3941,
"step": 370
},
{
"epoch": 0.83,
"learning_rate": 2.3788010230270542e-06,
"loss": 2.2867,
"step": 375
},
{
"epoch": 0.84,
"learning_rate": 1.7965532642889826e-06,
"loss": 1.972,
"step": 380
},
{
"epoch": 0.85,
"learning_rate": 1.2949766180319986e-06,
"loss": 2.2289,
"step": 385
},
{
"epoch": 0.86,
"learning_rate": 8.746767827345464e-07,
"loss": 2.212,
"step": 390
},
{
"epoch": 0.87,
"learning_rate": 5.36161307886377e-07,
"loss": 2.1037,
"step": 395
},
{
"epoch": 0.88,
"learning_rate": 2.798389810765762e-07,
"loss": 2.0082,
"step": 400
},
{
"epoch": 0.9,
"learning_rate": 1.0601933434685087e-07,
"loss": 2.1227,
"step": 405
},
{
"epoch": 0.91,
"learning_rate": 1.4912270403714612e-08,
"loss": 2.279,
"step": 410
},
{
"epoch": 0.92,
"learning_rate": 6.627809142258534e-09,
"loss": 2.1128,
"step": 415
},
{
"epoch": 0.93,
"learning_rate": 8.11759547874179e-08,
"loss": 2.1338,
"step": 420
},
{
"epoch": 0.94,
"learning_rate": 2.384666838129088e-07,
"loss": 2.1859,
"step": 425
},
{
"epoch": 0.95,
"learning_rate": 4.783100536526866e-07,
"loss": 2.1442,
"step": 430
},
{
"epoch": 0.96,
"learning_rate": 8.004164320735097e-07,
"loss": 2.3821,
"step": 435
},
{
"epoch": 0.97,
"learning_rate": 1.2043968469315794e-06,
"loss": 2.1965,
"step": 440
},
{
"epoch": 0.98,
"learning_rate": 1.6897634558914746e-06,
"loss": 2.2908,
"step": 445
},
{
"epoch": 1.0,
"learning_rate": 2.2559301355383585e-06,
"loss": 2.1072,
"step": 450
},
{
"epoch": 1.0,
"eval_loss": 2.026026487350464,
"eval_runtime": 32.0027,
"eval_samples_per_second": 20.779,
"eval_steps_per_second": 2.625,
"step": 452
},
{
"epoch": 1.0,
"learning_rate": 1.757624254440622e-06,
"loss": 2.2171,
"step": 455
},
{
"epoch": 1.0,
"eval_loss": 1.8596649169921875,
"eval_runtime": 29.8564,
"eval_samples_per_second": 20.9,
"eval_steps_per_second": 2.613,
"step": 457
}
],
"max_steps": 914,
"num_train_epochs": 2,
"total_flos": 476857958400000.0,
"trial_name": null,
"trial_params": null
}