Yi-6B-zhihu2 / trainer_state.json
yyx123's picture
Model save
d9e953d verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.0,
"eval_steps": 500,
"global_step": 793,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"learning_rate": 2.5e-06,
"loss": 2.731,
"step": 1
},
{
"epoch": 0.01,
"learning_rate": 1.25e-05,
"loss": 2.4673,
"step": 5
},
{
"epoch": 0.01,
"learning_rate": 2.5e-05,
"loss": 2.6706,
"step": 10
},
{
"epoch": 0.02,
"learning_rate": 3.7500000000000003e-05,
"loss": 2.6655,
"step": 15
},
{
"epoch": 0.03,
"learning_rate": 5e-05,
"loss": 2.6413,
"step": 20
},
{
"epoch": 0.03,
"learning_rate": 6.25e-05,
"loss": 2.5822,
"step": 25
},
{
"epoch": 0.04,
"learning_rate": 7.500000000000001e-05,
"loss": 2.5791,
"step": 30
},
{
"epoch": 0.04,
"learning_rate": 8.75e-05,
"loss": 2.4541,
"step": 35
},
{
"epoch": 0.05,
"learning_rate": 0.0001,
"loss": 2.2348,
"step": 40
},
{
"epoch": 0.06,
"learning_rate": 0.00011250000000000001,
"loss": 2.4262,
"step": 45
},
{
"epoch": 0.06,
"learning_rate": 0.000125,
"loss": 2.5609,
"step": 50
},
{
"epoch": 0.07,
"learning_rate": 0.0001375,
"loss": 2.3085,
"step": 55
},
{
"epoch": 0.08,
"learning_rate": 0.00015000000000000001,
"loss": 2.4856,
"step": 60
},
{
"epoch": 0.08,
"learning_rate": 0.00016250000000000002,
"loss": 2.5578,
"step": 65
},
{
"epoch": 0.09,
"learning_rate": 0.000175,
"loss": 2.5337,
"step": 70
},
{
"epoch": 0.09,
"learning_rate": 0.0001875,
"loss": 2.6725,
"step": 75
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 2.0052,
"step": 80
},
{
"epoch": 0.11,
"learning_rate": 0.0001999757331652354,
"loss": 2.8976,
"step": 85
},
{
"epoch": 0.11,
"learning_rate": 0.00019990294443852685,
"loss": 2.7538,
"step": 90
},
{
"epoch": 0.12,
"learning_rate": 0.00019978166914691453,
"loss": 2.11,
"step": 95
},
{
"epoch": 0.13,
"learning_rate": 0.00019961196614974767,
"loss": 2.4711,
"step": 100
},
{
"epoch": 0.13,
"learning_rate": 0.00019939391781011807,
"loss": 2.1637,
"step": 105
},
{
"epoch": 0.14,
"learning_rate": 0.00019912762995488633,
"loss": 2.6475,
"step": 110
},
{
"epoch": 0.15,
"learning_rate": 0.00019881323182332006,
"loss": 2.3961,
"step": 115
},
{
"epoch": 0.15,
"learning_rate": 0.00019845087600436947,
"loss": 2.6875,
"step": 120
},
{
"epoch": 0.16,
"learning_rate": 0.00019804073836261025,
"loss": 2.6841,
"step": 125
},
{
"epoch": 0.16,
"learning_rate": 0.0001975830179528901,
"loss": 2.6747,
"step": 130
},
{
"epoch": 0.17,
"learning_rate": 0.00019707793692372,
"loss": 2.6586,
"step": 135
},
{
"epoch": 0.18,
"learning_rate": 0.00019652574040945745,
"loss": 2.4275,
"step": 140
},
{
"epoch": 0.18,
"learning_rate": 0.00019592669641133395,
"loss": 2.4158,
"step": 145
},
{
"epoch": 0.19,
"learning_rate": 0.00019528109566738382,
"loss": 2.2207,
"step": 150
},
{
"epoch": 0.2,
"learning_rate": 0.0001945892515113386,
"loss": 2.3564,
"step": 155
},
{
"epoch": 0.2,
"learning_rate": 0.00019385149972055466,
"loss": 2.9308,
"step": 160
},
{
"epoch": 0.21,
"learning_rate": 0.0001930681983530481,
"loss": 2.7843,
"step": 165
},
{
"epoch": 0.21,
"learning_rate": 0.000192239727573716,
"loss": 2.5293,
"step": 170
},
{
"epoch": 0.22,
"learning_rate": 0.0001913664894698286,
"loss": 2.374,
"step": 175
},
{
"epoch": 0.23,
"learning_rate": 0.0001904489078558814,
"loss": 2.6003,
"step": 180
},
{
"epoch": 0.23,
"learning_rate": 0.0001894874280679026,
"loss": 2.3425,
"step": 185
},
{
"epoch": 0.24,
"learning_rate": 0.00018848251674731507,
"loss": 2.5159,
"step": 190
},
{
"epoch": 0.25,
"learning_rate": 0.00018743466161445823,
"loss": 2.6329,
"step": 195
},
{
"epoch": 0.25,
"learning_rate": 0.00018634437123187937,
"loss": 2.3351,
"step": 200
},
{
"epoch": 0.26,
"learning_rate": 0.00018521217475750973,
"loss": 2.5704,
"step": 205
},
{
"epoch": 0.26,
"learning_rate": 0.00018403862168784457,
"loss": 2.4092,
"step": 210
},
{
"epoch": 0.27,
"learning_rate": 0.00018282428159125248,
"loss": 2.4303,
"step": 215
},
{
"epoch": 0.28,
"learning_rate": 0.0001815697438315429,
"loss": 2.4577,
"step": 220
},
{
"epoch": 0.28,
"learning_rate": 0.00018027561728192626,
"loss": 2.46,
"step": 225
},
{
"epoch": 0.29,
"learning_rate": 0.00017894253002950542,
"loss": 2.5202,
"step": 230
},
{
"epoch": 0.3,
"learning_rate": 0.000177571129070442,
"loss": 2.6827,
"step": 235
},
{
"epoch": 0.3,
"learning_rate": 0.0001761620799959454,
"loss": 2.989,
"step": 240
},
{
"epoch": 0.31,
"learning_rate": 0.000174716066669237,
"loss": 2.7209,
"step": 245
},
{
"epoch": 0.32,
"learning_rate": 0.00017323379089364602,
"loss": 2.6086,
"step": 250
},
{
"epoch": 0.32,
"learning_rate": 0.000171715972071999,
"loss": 2.6182,
"step": 255
},
{
"epoch": 0.33,
"learning_rate": 0.00017016334685746685,
"loss": 2.5666,
"step": 260
},
{
"epoch": 0.33,
"learning_rate": 0.0001685766687960402,
"loss": 2.4663,
"step": 265
},
{
"epoch": 0.34,
"learning_rate": 0.00016695670796080592,
"loss": 2.3282,
"step": 270
},
{
"epoch": 0.35,
"learning_rate": 0.00016530425057820227,
"loss": 2.4567,
"step": 275
},
{
"epoch": 0.35,
"learning_rate": 0.00016362009864643443,
"loss": 2.2274,
"step": 280
},
{
"epoch": 0.36,
"learning_rate": 0.0001619050695462353,
"loss": 2.2665,
"step": 285
},
{
"epoch": 0.37,
"learning_rate": 0.00016015999564416078,
"loss": 2.046,
"step": 290
},
{
"epoch": 0.37,
"learning_rate": 0.00015838572388861147,
"loss": 2.2182,
"step": 295
},
{
"epoch": 0.38,
"learning_rate": 0.0001565831153987778,
"loss": 2.4857,
"step": 300
},
{
"epoch": 0.38,
"learning_rate": 0.0001547530450467071,
"loss": 2.3942,
"step": 305
},
{
"epoch": 0.39,
"learning_rate": 0.00015289640103269625,
"loss": 2.635,
"step": 310
},
{
"epoch": 0.4,
"learning_rate": 0.0001510140844542153,
"loss": 2.4103,
"step": 315
},
{
"epoch": 0.4,
"learning_rate": 0.00014910700886857197,
"loss": 2.5356,
"step": 320
},
{
"epoch": 0.41,
"learning_rate": 0.00014717609984952866,
"loss": 2.6382,
"step": 325
},
{
"epoch": 0.42,
"learning_rate": 0.00014522229453808758,
"loss": 2.1514,
"step": 330
},
{
"epoch": 0.42,
"learning_rate": 0.0001432465411876618,
"loss": 2.5063,
"step": 335
},
{
"epoch": 0.43,
"learning_rate": 0.0001412497987038532,
"loss": 2.4789,
"step": 340
},
{
"epoch": 0.44,
"learning_rate": 0.00013923303617906015,
"loss": 2.3494,
"step": 345
},
{
"epoch": 0.44,
"learning_rate": 0.00013719723242214168,
"loss": 2.443,
"step": 350
},
{
"epoch": 0.45,
"learning_rate": 0.0001351433754833654,
"loss": 2.382,
"step": 355
},
{
"epoch": 0.45,
"learning_rate": 0.0001330724621748706,
"loss": 2.5961,
"step": 360
},
{
"epoch": 0.46,
"learning_rate": 0.00013098549758687865,
"loss": 2.8336,
"step": 365
},
{
"epoch": 0.47,
"learning_rate": 0.0001288834945998859,
"loss": 2.5518,
"step": 370
},
{
"epoch": 0.47,
"learning_rate": 0.00012676747339307554,
"loss": 2.1127,
"step": 375
},
{
"epoch": 0.48,
"learning_rate": 0.00012463846094918723,
"loss": 2.6908,
"step": 380
},
{
"epoch": 0.49,
"learning_rate": 0.00012249749055608474,
"loss": 2.6419,
"step": 385
},
{
"epoch": 0.49,
"learning_rate": 0.0001203456013052634,
"loss": 2.4063,
"step": 390
},
{
"epoch": 0.5,
"learning_rate": 0.00011818383758754082,
"loss": 2.2766,
"step": 395
},
{
"epoch": 0.5,
"learning_rate": 0.00011601324858617572,
"loss": 2.5315,
"step": 400
},
{
"epoch": 0.51,
"learning_rate": 0.00011383488776766093,
"loss": 2.8151,
"step": 405
},
{
"epoch": 0.52,
"learning_rate": 0.00011164981237043728,
"loss": 2.6057,
"step": 410
},
{
"epoch": 0.52,
"learning_rate": 0.00010945908289177694,
"loss": 2.4863,
"step": 415
},
{
"epoch": 0.53,
"learning_rate": 0.00010726376257308544,
"loss": 2.2287,
"step": 420
},
{
"epoch": 0.54,
"learning_rate": 0.00010506491688387127,
"loss": 2.3064,
"step": 425
},
{
"epoch": 0.54,
"learning_rate": 0.00010286361300463481,
"loss": 2.4449,
"step": 430
},
{
"epoch": 0.55,
"learning_rate": 0.00010066091930892605,
"loss": 2.7777,
"step": 435
},
{
"epoch": 0.55,
"learning_rate": 9.845790484482404e-05,
"loss": 2.5837,
"step": 440
},
{
"epoch": 0.56,
"learning_rate": 9.625563881608847e-05,
"loss": 2.4259,
"step": 445
},
{
"epoch": 0.57,
"learning_rate": 9.405519006323588e-05,
"loss": 2.7834,
"step": 450
},
{
"epoch": 0.57,
"learning_rate": 9.185762654479219e-05,
"loss": 2.6675,
"step": 455
},
{
"epoch": 0.58,
"learning_rate": 8.966401481897304e-05,
"loss": 2.5957,
"step": 460
},
{
"epoch": 0.59,
"learning_rate": 8.747541952604423e-05,
"loss": 2.3405,
"step": 465
},
{
"epoch": 0.59,
"learning_rate": 8.52929028716126e-05,
"loss": 2.4473,
"step": 470
},
{
"epoch": 0.6,
"learning_rate": 8.311752411109863e-05,
"loss": 2.2817,
"step": 475
},
{
"epoch": 0.61,
"learning_rate": 8.095033903564091e-05,
"loss": 2.7025,
"step": 480
},
{
"epoch": 0.61,
"learning_rate": 7.879239945968209e-05,
"loss": 2.3289,
"step": 485
},
{
"epoch": 0.62,
"learning_rate": 7.664475271048458e-05,
"loss": 2.6471,
"step": 490
},
{
"epoch": 0.62,
"learning_rate": 7.450844111982436e-05,
"loss": 2.5394,
"step": 495
},
{
"epoch": 0.63,
"learning_rate": 7.238450151810885e-05,
"loss": 2.422,
"step": 500
},
{
"epoch": 0.64,
"learning_rate": 7.02739647311654e-05,
"loss": 2.4525,
"step": 505
},
{
"epoch": 0.64,
"learning_rate": 6.817785507994349e-05,
"loss": 2.5048,
"step": 510
},
{
"epoch": 0.65,
"learning_rate": 6.609718988337423e-05,
"loss": 2.5573,
"step": 515
},
{
"epoch": 0.66,
"learning_rate": 6.403297896462807e-05,
"loss": 2.6636,
"step": 520
},
{
"epoch": 0.66,
"learning_rate": 6.198622416101077e-05,
"loss": 2.4896,
"step": 525
},
{
"epoch": 0.67,
"learning_rate": 5.995791883773474e-05,
"loss": 2.5308,
"step": 530
},
{
"epoch": 0.67,
"learning_rate": 5.794904740580269e-05,
"loss": 2.4272,
"step": 535
},
{
"epoch": 0.68,
"learning_rate": 5.596058484423656e-05,
"loss": 2.4119,
"step": 540
},
{
"epoch": 0.69,
"learning_rate": 5.399349622688479e-05,
"loss": 2.7247,
"step": 545
},
{
"epoch": 0.69,
"learning_rate": 5.20487362540362e-05,
"loss": 2.4899,
"step": 550
},
{
"epoch": 0.7,
"learning_rate": 5.012724878906926e-05,
"loss": 2.4428,
"step": 555
},
{
"epoch": 0.71,
"learning_rate": 4.82299664003601e-05,
"loss": 2.613,
"step": 560
},
{
"epoch": 0.71,
"learning_rate": 4.6357809908673435e-05,
"loss": 2.5947,
"step": 565
},
{
"epoch": 0.72,
"learning_rate": 4.451168794025391e-05,
"loss": 2.3137,
"step": 570
},
{
"epoch": 0.73,
"learning_rate": 4.269249648583692e-05,
"loss": 2.3568,
"step": 575
},
{
"epoch": 0.73,
"learning_rate": 4.0901118465790875e-05,
"loss": 1.9487,
"step": 580
},
{
"epoch": 0.74,
"learning_rate": 3.9138423301604154e-05,
"loss": 2.3949,
"step": 585
},
{
"epoch": 0.74,
"learning_rate": 3.740526649392256e-05,
"loss": 2.6208,
"step": 590
},
{
"epoch": 0.75,
"learning_rate": 3.570248920734354e-05,
"loss": 2.3749,
"step": 595
},
{
"epoch": 0.76,
"learning_rate": 3.4030917862168165e-05,
"loss": 2.5309,
"step": 600
},
{
"epoch": 0.76,
"learning_rate": 3.239136373330903e-05,
"loss": 2.6263,
"step": 605
},
{
"epoch": 0.77,
"learning_rate": 3.0784622556548856e-05,
"loss": 2.5203,
"step": 610
},
{
"epoch": 0.78,
"learning_rate": 2.921147414234051e-05,
"loss": 2.3685,
"step": 615
},
{
"epoch": 0.78,
"learning_rate": 2.767268199733657e-05,
"loss": 2.7285,
"step": 620
},
{
"epoch": 0.79,
"learning_rate": 2.6168992953831396e-05,
"loss": 2.3638,
"step": 625
},
{
"epoch": 0.79,
"learning_rate": 2.4701136807296177e-05,
"loss": 2.1867,
"step": 630
},
{
"epoch": 0.8,
"learning_rate": 2.326982596218219e-05,
"loss": 2.6267,
"step": 635
},
{
"epoch": 0.81,
"learning_rate": 2.187575508616493e-05,
"loss": 2.6749,
"step": 640
},
{
"epoch": 0.81,
"learning_rate": 2.0519600772996405e-05,
"loss": 2.4208,
"step": 645
},
{
"epoch": 0.82,
"learning_rate": 1.920202121412924e-05,
"loss": 2.6287,
"step": 650
},
{
"epoch": 0.83,
"learning_rate": 1.7923655879272393e-05,
"loss": 2.6315,
"step": 655
},
{
"epoch": 0.83,
"learning_rate": 1.668512520603276e-05,
"loss": 2.5708,
"step": 660
},
{
"epoch": 0.84,
"learning_rate": 1.548703029879437e-05,
"loss": 2.3316,
"step": 665
},
{
"epoch": 0.84,
"learning_rate": 1.4329952636980115e-05,
"loss": 2.1933,
"step": 670
},
{
"epoch": 0.85,
"learning_rate": 1.321445379283861e-05,
"loss": 2.1816,
"step": 675
},
{
"epoch": 0.86,
"learning_rate": 1.214107515889248e-05,
"loss": 2.474,
"step": 680
},
{
"epoch": 0.86,
"learning_rate": 1.111033768518066e-05,
"loss": 2.5638,
"step": 685
},
{
"epoch": 0.87,
"learning_rate": 1.0122741626422438e-05,
"loss": 2.5808,
"step": 690
},
{
"epoch": 0.88,
"learning_rate": 9.178766299225205e-06,
"loss": 2.1113,
"step": 695
},
{
"epoch": 0.88,
"learning_rate": 8.278869849454718e-06,
"loss": 2.6343,
"step": 700
},
{
"epoch": 0.89,
"learning_rate": 7.423489029879982e-06,
"loss": 2.3117,
"step": 705
},
{
"epoch": 0.9,
"learning_rate": 6.613038988201237e-06,
"loss": 2.7684,
"step": 710
},
{
"epoch": 0.9,
"learning_rate": 5.847913065563382e-06,
"loss": 2.5271,
"step": 715
},
{
"epoch": 0.91,
"learning_rate": 5.128482605653173e-06,
"loss": 2.2342,
"step": 720
},
{
"epoch": 0.91,
"learning_rate": 4.455096774472567e-06,
"loss": 2.2851,
"step": 725
},
{
"epoch": 0.92,
"learning_rate": 3.828082390875487e-06,
"loss": 2.4365,
"step": 730
},
{
"epoch": 0.93,
"learning_rate": 3.247743767950795e-06,
"loss": 2.6543,
"step": 735
},
{
"epoch": 0.93,
"learning_rate": 2.714362565327866e-06,
"loss": 2.6314,
"step": 740
},
{
"epoch": 0.94,
"learning_rate": 2.2281976524769266e-06,
"loss": 2.1431,
"step": 745
},
{
"epoch": 0.95,
"learning_rate": 1.7894849830701443e-06,
"loss": 2.4283,
"step": 750
},
{
"epoch": 0.95,
"learning_rate": 1.398437480464676e-06,
"loss": 2.5455,
"step": 755
},
{
"epoch": 0.96,
"learning_rate": 1.055244934363131e-06,
"loss": 2.4203,
"step": 760
},
{
"epoch": 0.96,
"learning_rate": 7.600739087016862e-07,
"loss": 2.5425,
"step": 765
},
{
"epoch": 0.97,
"learning_rate": 5.130676608104845e-07,
"loss": 2.623,
"step": 770
},
{
"epoch": 0.98,
"learning_rate": 3.143460718855962e-07,
"loss": 2.5706,
"step": 775
},
{
"epoch": 0.98,
"learning_rate": 1.6400558880627082e-07,
"loss": 2.4692,
"step": 780
},
{
"epoch": 0.99,
"learning_rate": 6.21191773257368e-08,
"loss": 2.684,
"step": 785
},
{
"epoch": 1.0,
"learning_rate": 8.736286658228565e-09,
"loss": 2.3449,
"step": 790
},
{
"epoch": 1.0,
"eval_loss": 2.400331974029541,
"eval_runtime": 240.8241,
"eval_samples_per_second": 3.293,
"eval_steps_per_second": 3.293,
"step": 793
},
{
"epoch": 1.0,
"step": 793,
"total_flos": 5.686041454942618e+16,
"train_loss": 0.28998795253181214,
"train_runtime": 345.0365,
"train_samples_per_second": 2.298,
"train_steps_per_second": 2.298
}
],
"logging_steps": 5,
"max_steps": 793,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 100,
"total_flos": 5.686041454942618e+16,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}