eminem / trainer_state.json
AlekseyKorshuk's picture
huggingartists
fc38921
raw
history blame
11.5 kB
{
"best_metric": 0.22213919460773468,
"best_model_checkpoint": "output/eminem/checkpoint-455",
"epoch": 1.0,
"global_step": 455,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"learning_rate": 0.00013711989076069754,
"loss": 0.4887,
"step": 5
},
{
"epoch": 0.02,
"learning_rate": 0.00013719345930102362,
"loss": 0.4564,
"step": 10
},
{
"epoch": 0.03,
"learning_rate": 0.00013718528371963108,
"loss": 0.4132,
"step": 15
},
{
"epoch": 0.04,
"learning_rate": 0.00013709537375951485,
"loss": 0.4758,
"step": 20
},
{
"epoch": 0.05,
"learning_rate": 0.00013692383656807334,
"loss": 0.4929,
"step": 25
},
{
"epoch": 0.07,
"learning_rate": 0.00013667087656941957,
"loss": 0.4404,
"step": 30
},
{
"epoch": 0.08,
"learning_rate": 0.0001363367952207642,
"loss": 0.4726,
"step": 35
},
{
"epoch": 0.09,
"learning_rate": 0.00013592199065316386,
"loss": 0.4448,
"step": 40
},
{
"epoch": 0.1,
"learning_rate": 0.00013542695719706024,
"loss": 0.4852,
"step": 45
},
{
"epoch": 0.11,
"learning_rate": 0.00013485228479317936,
"loss": 0.4865,
"step": 50
},
{
"epoch": 0.12,
"learning_rate": 0.00013419865828948315,
"loss": 0.5125,
"step": 55
},
{
"epoch": 0.13,
"learning_rate": 0.00013346685662502967,
"loss": 0.5314,
"step": 60
},
{
"epoch": 0.14,
"learning_rate": 0.00013265775190169146,
"loss": 0.4813,
"step": 65
},
{
"epoch": 0.15,
"learning_rate": 0.00013177230834485767,
"loss": 0.4348,
"step": 70
},
{
"epoch": 0.16,
"learning_rate": 0.00013081158115434713,
"loss": 0.4715,
"step": 75
},
{
"epoch": 0.18,
"learning_rate": 0.00012977671524691005,
"loss": 0.4864,
"step": 80
},
{
"epoch": 0.19,
"learning_rate": 0.0001286689438918043,
"loss": 0.4764,
"step": 85
},
{
"epoch": 0.2,
"learning_rate": 0.0001274895872410883,
"loss": 0.5284,
"step": 90
},
{
"epoch": 0.21,
"learning_rate": 0.0001262400507563689,
"loss": 0.5136,
"step": 95
},
{
"epoch": 0.22,
"learning_rate": 0.00012492182353389172,
"loss": 0.5015,
"step": 100
},
{
"epoch": 0.23,
"learning_rate": 0.00012353647652994513,
"loss": 0.5157,
"step": 105
},
{
"epoch": 0.24,
"learning_rate": 0.00012208566068873714,
"loss": 0.5087,
"step": 110
},
{
"epoch": 0.25,
"learning_rate": 0.00012057110497492192,
"loss": 0.5977,
"step": 115
},
{
"epoch": 0.26,
"learning_rate": 0.00011899461431316551,
"loss": 0.4527,
"step": 120
},
{
"epoch": 0.27,
"learning_rate": 0.00011735806743718042,
"loss": 0.4758,
"step": 125
},
{
"epoch": 0.29,
"learning_rate": 0.00011566341465080868,
"loss": 0.4566,
"step": 130
},
{
"epoch": 0.3,
"learning_rate": 0.00011391267550379983,
"loss": 0.4963,
"step": 135
},
{
"epoch": 0.31,
"learning_rate": 0.00011210793638508056,
"loss": 0.4994,
"step": 140
},
{
"epoch": 0.32,
"learning_rate": 0.00011025134803636436,
"loss": 0.5175,
"step": 145
},
{
"epoch": 0.33,
"learning_rate": 0.00010834512298907191,
"loss": 0.5139,
"step": 150
},
{
"epoch": 0.34,
"learning_rate": 0.00010639153292761406,
"loss": 0.515,
"step": 155
},
{
"epoch": 0.35,
"learning_rate": 0.0001043929059821889,
"loss": 0.4672,
"step": 160
},
{
"epoch": 0.36,
"learning_rate": 0.00010235162395429551,
"loss": 0.5449,
"step": 165
},
{
"epoch": 0.37,
"learning_rate": 0.00010027011947830235,
"loss": 0.5187,
"step": 170
},
{
"epoch": 0.38,
"learning_rate": 9.81508731224277e-05,
"loss": 0.5283,
"step": 175
},
{
"epoch": 0.4,
"learning_rate": 9.599641043260833e-05,
"loss": 0.4869,
"step": 180
},
{
"epoch": 0.41,
"learning_rate": 9.380929892273937e-05,
"loss": 0.4807,
"step": 185
},
{
"epoch": 0.42,
"learning_rate": 9.15921450149442e-05,
"loss": 0.5009,
"step": 190
},
{
"epoch": 0.43,
"learning_rate": 8.93475909334324e-05,
"loss": 0.5661,
"step": 195
},
{
"epoch": 0.44,
"learning_rate": 8.70783115557202e-05,
"loss": 0.5181,
"step": 200
},
{
"epoch": 0.45,
"learning_rate": 8.478701122492464e-05,
"loss": 0.5132,
"step": 205
},
{
"epoch": 0.46,
"learning_rate": 8.247642052695435e-05,
"loss": 0.4738,
"step": 210
},
{
"epoch": 0.47,
"learning_rate": 8.014929303640746e-05,
"loss": 0.4718,
"step": 215
},
{
"epoch": 0.48,
"learning_rate": 7.780840203509217e-05,
"loss": 0.5202,
"step": 220
},
{
"epoch": 0.49,
"learning_rate": 7.545653720705064e-05,
"loss": 0.4643,
"step": 225
},
{
"epoch": 0.51,
"learning_rate": 7.309650131404948e-05,
"loss": 0.4846,
"step": 230
},
{
"epoch": 0.52,
"learning_rate": 7.073110685545424e-05,
"loss": 0.4528,
"step": 235
},
{
"epoch": 0.53,
"learning_rate": 6.836317271654793e-05,
"loss": 0.4695,
"step": 240
},
{
"epoch": 0.54,
"learning_rate": 6.599552080919084e-05,
"loss": 0.4723,
"step": 245
},
{
"epoch": 0.55,
"learning_rate": 6.363097270890452e-05,
"loss": 0.4357,
"step": 250
},
{
"epoch": 0.56,
"learning_rate": 6.127234629234303e-05,
"loss": 0.5521,
"step": 255
},
{
"epoch": 0.57,
"learning_rate": 5.892245237918387e-05,
"loss": 0.484,
"step": 260
},
{
"epoch": 0.58,
"learning_rate": 5.6584091382409376e-05,
"loss": 0.5221,
"step": 265
},
{
"epoch": 0.59,
"learning_rate": 5.426004997100939e-05,
"loss": 0.4865,
"step": 270
},
{
"epoch": 0.6,
"learning_rate": 5.195309774905142e-05,
"loss": 0.4395,
"step": 275
},
{
"epoch": 0.62,
"learning_rate": 4.966598395510049e-05,
"loss": 0.4693,
"step": 280
},
{
"epoch": 0.63,
"learning_rate": 4.7401434185878876e-05,
"loss": 0.5105,
"step": 285
},
{
"epoch": 0.64,
"learning_rate": 4.516214714814733e-05,
"loss": 0.406,
"step": 290
},
{
"epoch": 0.65,
"learning_rate": 4.2950791442584565e-05,
"loss": 0.4573,
"step": 295
},
{
"epoch": 0.66,
"learning_rate": 4.0770002383574794e-05,
"loss": 0.4153,
"step": 300
},
{
"epoch": 0.67,
"learning_rate": 3.862237885865012e-05,
"loss": 0.4713,
"step": 305
},
{
"epoch": 0.68,
"learning_rate": 3.651048023135423e-05,
"loss": 0.4615,
"step": 310
},
{
"epoch": 0.69,
"learning_rate": 3.4436823291189834e-05,
"loss": 0.4731,
"step": 315
},
{
"epoch": 0.7,
"learning_rate": 3.2403879254320225e-05,
"loss": 0.4743,
"step": 320
},
{
"epoch": 0.71,
"learning_rate": 3.0414070818570786e-05,
"loss": 0.4617,
"step": 325
},
{
"epoch": 0.73,
"learning_rate": 2.84697692762624e-05,
"loss": 0.4779,
"step": 330
},
{
"epoch": 0.74,
"learning_rate": 2.657329168827975e-05,
"loss": 0.4796,
"step": 335
},
{
"epoch": 0.75,
"learning_rate": 2.4726898122808552e-05,
"loss": 0.451,
"step": 340
},
{
"epoch": 0.76,
"learning_rate": 2.2932788961952088e-05,
"loss": 0.5316,
"step": 345
},
{
"epoch": 0.77,
"learning_rate": 2.1193102279502168e-05,
"loss": 0.4846,
"step": 350
},
{
"epoch": 0.78,
"learning_rate": 1.9509911292953254e-05,
"loss": 0.4435,
"step": 355
},
{
"epoch": 0.79,
"learning_rate": 1.7885221892815502e-05,
"loss": 0.4859,
"step": 360
},
{
"epoch": 0.8,
"learning_rate": 1.6320970252148786e-05,
"loss": 0.5439,
"step": 365
},
{
"epoch": 0.81,
"learning_rate": 1.48190205191943e-05,
"loss": 0.4491,
"step": 370
},
{
"epoch": 0.82,
"learning_rate": 1.3381162595831284e-05,
"loss": 0.4599,
"step": 375
},
{
"epoch": 0.84,
"learning_rate": 1.2009110004522916e-05,
"loss": 0.4688,
"step": 380
},
{
"epoch": 0.85,
"learning_rate": 1.0704497846266602e-05,
"loss": 0.445,
"step": 385
},
{
"epoch": 0.86,
"learning_rate": 9.468880852029007e-06,
"loss": 0.4632,
"step": 390
},
{
"epoch": 0.87,
"learning_rate": 8.303731529931584e-06,
"loss": 0.4699,
"step": 395
},
{
"epoch": 0.88,
"learning_rate": 7.210438410440295e-06,
"loss": 0.4994,
"step": 400
},
{
"epoch": 0.89,
"learning_rate": 6.190304391625742e-06,
"loss": 0.4727,
"step": 405
},
{
"epoch": 0.9,
"learning_rate": 5.24454518647817e-06,
"loss": 0.4766,
"step": 410
},
{
"epoch": 0.91,
"learning_rate": 4.374287874114302e-06,
"loss": 0.5309,
"step": 415
},
{
"epoch": 0.92,
"learning_rate": 3.5805695566192774e-06,
"loss": 0.4346,
"step": 420
},
{
"epoch": 0.93,
"learning_rate": 2.8643361231110826e-06,
"loss": 0.4458,
"step": 425
},
{
"epoch": 0.95,
"learning_rate": 2.2264411225093805e-06,
"loss": 0.416,
"step": 430
},
{
"epoch": 0.96,
"learning_rate": 1.6676447463391393e-06,
"loss": 0.524,
"step": 435
},
{
"epoch": 0.97,
"learning_rate": 1.1886129228033631e-06,
"loss": 0.4421,
"step": 440
},
{
"epoch": 0.98,
"learning_rate": 7.89916523178351e-07,
"loss": 0.4325,
"step": 445
},
{
"epoch": 0.99,
"learning_rate": 4.7203068149803844e-07,
"loss": 0.476,
"step": 450
},
{
"epoch": 1.0,
"learning_rate": 2.3533422832716514e-07,
"loss": 0.4506,
"step": 455
},
{
"epoch": 1.0,
"eval_loss": 0.22213919460773468,
"eval_runtime": 29.39,
"eval_samples_per_second": 21.946,
"eval_steps_per_second": 2.756,
"step": 455
}
],
"max_steps": 1365,
"num_train_epochs": 3,
"total_flos": 474636976128000.0,
"trial_name": null,
"trial_params": null
}