Yi-6B-zhihu3 / trainer_state.json
yyx123's picture
Model save
5129bb9 verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.975609756097561,
"eval_steps": 500,
"global_step": 800,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"learning_rate": 2.4390243902439027e-06,
"loss": 2.6767,
"step": 1
},
{
"epoch": 0.01,
"learning_rate": 1.2195121951219513e-05,
"loss": 2.7823,
"step": 5
},
{
"epoch": 0.01,
"learning_rate": 2.4390243902439026e-05,
"loss": 2.4361,
"step": 10
},
{
"epoch": 0.02,
"learning_rate": 3.6585365853658535e-05,
"loss": 2.535,
"step": 15
},
{
"epoch": 0.02,
"learning_rate": 4.878048780487805e-05,
"loss": 2.5652,
"step": 20
},
{
"epoch": 0.03,
"learning_rate": 6.097560975609756e-05,
"loss": 2.6216,
"step": 25
},
{
"epoch": 0.04,
"learning_rate": 7.317073170731707e-05,
"loss": 2.1459,
"step": 30
},
{
"epoch": 0.04,
"learning_rate": 8.53658536585366e-05,
"loss": 2.5116,
"step": 35
},
{
"epoch": 0.05,
"learning_rate": 9.75609756097561e-05,
"loss": 2.5704,
"step": 40
},
{
"epoch": 0.05,
"learning_rate": 0.00010975609756097563,
"loss": 2.3636,
"step": 45
},
{
"epoch": 0.06,
"learning_rate": 0.00012195121951219512,
"loss": 2.6786,
"step": 50
},
{
"epoch": 0.07,
"learning_rate": 0.00013414634146341464,
"loss": 2.4003,
"step": 55
},
{
"epoch": 0.07,
"learning_rate": 0.00014634146341463414,
"loss": 2.4267,
"step": 60
},
{
"epoch": 0.08,
"learning_rate": 0.00015853658536585366,
"loss": 2.4149,
"step": 65
},
{
"epoch": 0.09,
"learning_rate": 0.0001707317073170732,
"loss": 2.1072,
"step": 70
},
{
"epoch": 0.09,
"learning_rate": 0.0001829268292682927,
"loss": 2.3557,
"step": 75
},
{
"epoch": 0.1,
"learning_rate": 0.0001951219512195122,
"loss": 2.3392,
"step": 80
},
{
"epoch": 0.1,
"learning_rate": 0.00019999184556954776,
"loss": 2.2043,
"step": 85
},
{
"epoch": 0.11,
"learning_rate": 0.0001999420177550043,
"loss": 2.2218,
"step": 90
},
{
"epoch": 0.12,
"learning_rate": 0.00019984691491033906,
"loss": 2.4057,
"step": 95
},
{
"epoch": 0.12,
"learning_rate": 0.00019970658011837404,
"loss": 2.3682,
"step": 100
},
{
"epoch": 0.13,
"learning_rate": 0.00019952107695258992,
"loss": 2.718,
"step": 105
},
{
"epoch": 0.13,
"learning_rate": 0.00019929048944832638,
"loss": 2.4517,
"step": 110
},
{
"epoch": 0.14,
"learning_rate": 0.00019901492206471325,
"loss": 2.5616,
"step": 115
},
{
"epoch": 0.15,
"learning_rate": 0.00019869449963734893,
"loss": 2.496,
"step": 120
},
{
"epoch": 0.15,
"learning_rate": 0.00019832936732174834,
"loss": 2.1452,
"step": 125
},
{
"epoch": 0.16,
"learning_rate": 0.00019791969052758562,
"loss": 2.2789,
"step": 130
},
{
"epoch": 0.16,
"learning_rate": 0.00019746565484376132,
"loss": 2.0612,
"step": 135
},
{
"epoch": 0.17,
"learning_rate": 0.00019696746595432828,
"loss": 2.2248,
"step": 140
},
{
"epoch": 0.18,
"learning_rate": 0.0001964253495453141,
"loss": 2.6121,
"step": 145
},
{
"epoch": 0.18,
"learning_rate": 0.00019583955120248237,
"loss": 2.4211,
"step": 150
},
{
"epoch": 0.19,
"learning_rate": 0.00019521033630007928,
"loss": 2.5135,
"step": 155
},
{
"epoch": 0.2,
"learning_rate": 0.00019453798988061535,
"loss": 2.343,
"step": 160
},
{
"epoch": 0.2,
"learning_rate": 0.00019382281652573785,
"loss": 2.5214,
"step": 165
},
{
"epoch": 0.21,
"learning_rate": 0.00019306514021825118,
"loss": 2.6032,
"step": 170
},
{
"epoch": 0.21,
"learning_rate": 0.00019226530419534833,
"loss": 2.7405,
"step": 175
},
{
"epoch": 0.22,
"learning_rate": 0.00019142367079312021,
"loss": 2.2154,
"step": 180
},
{
"epoch": 0.23,
"learning_rate": 0.00019054062128241264,
"loss": 2.5134,
"step": 185
},
{
"epoch": 0.23,
"learning_rate": 0.00018961655569610557,
"loss": 2.3687,
"step": 190
},
{
"epoch": 0.24,
"learning_rate": 0.0001886518926478932,
"loss": 2.4462,
"step": 195
},
{
"epoch": 0.24,
"learning_rate": 0.00018764706914264635,
"loss": 2.0328,
"step": 200
},
{
"epoch": 0.25,
"learning_rate": 0.00018660254037844388,
"loss": 2.5844,
"step": 205
},
{
"epoch": 0.26,
"learning_rate": 0.00018551877954036162,
"loss": 2.5333,
"step": 210
},
{
"epoch": 0.26,
"learning_rate": 0.00018439627758611385,
"loss": 2.3423,
"step": 215
},
{
"epoch": 0.27,
"learning_rate": 0.00018323554302364272,
"loss": 2.5999,
"step": 220
},
{
"epoch": 0.27,
"learning_rate": 0.00018203710168075788,
"loss": 2.421,
"step": 225
},
{
"epoch": 0.28,
"learning_rate": 0.0001808014964669293,
"loss": 2.5597,
"step": 230
},
{
"epoch": 0.29,
"learning_rate": 0.00017952928712734268,
"loss": 2.48,
"step": 235
},
{
"epoch": 0.29,
"learning_rate": 0.00017822104998932713,
"loss": 2.6025,
"step": 240
},
{
"epoch": 0.3,
"learning_rate": 0.00017687737770127185,
"loss": 2.3616,
"step": 245
},
{
"epoch": 0.3,
"learning_rate": 0.00017549887896414851,
"loss": 2.4311,
"step": 250
},
{
"epoch": 0.31,
"learning_rate": 0.0001740861782557618,
"loss": 2.5858,
"step": 255
},
{
"epoch": 0.32,
"learning_rate": 0.0001726399155478529,
"loss": 2.4726,
"step": 260
},
{
"epoch": 0.32,
"learning_rate": 0.00017116074601618417,
"loss": 2.4097,
"step": 265
},
{
"epoch": 0.33,
"learning_rate": 0.0001696493397437357,
"loss": 2.4452,
"step": 270
},
{
"epoch": 0.34,
"learning_rate": 0.00016810638141714934,
"loss": 2.341,
"step": 275
},
{
"epoch": 0.34,
"learning_rate": 0.00016653257001655652,
"loss": 2.2435,
"step": 280
},
{
"epoch": 0.35,
"learning_rate": 0.0001649286184989315,
"loss": 2.3826,
"step": 285
},
{
"epoch": 0.35,
"learning_rate": 0.0001632952534751122,
"loss": 2.3505,
"step": 290
},
{
"epoch": 0.36,
"learning_rate": 0.00016163321488063637,
"loss": 2.5566,
"step": 295
},
{
"epoch": 0.37,
"learning_rate": 0.00015994325564054122,
"loss": 2.3983,
"step": 300
},
{
"epoch": 0.37,
"learning_rate": 0.00015822614132827837,
"loss": 2.0597,
"step": 305
},
{
"epoch": 0.38,
"learning_rate": 0.00015648264981889934,
"loss": 2.3717,
"step": 310
},
{
"epoch": 0.38,
"learning_rate": 0.00015471357093666804,
"loss": 2.6116,
"step": 315
},
{
"epoch": 0.39,
"learning_rate": 0.00015291970609726007,
"loss": 2.3766,
"step": 320
},
{
"epoch": 0.4,
"learning_rate": 0.00015110186794471103,
"loss": 2.2623,
"step": 325
},
{
"epoch": 0.4,
"learning_rate": 0.00014926087998327837,
"loss": 2.3868,
"step": 330
},
{
"epoch": 0.41,
"learning_rate": 0.00014739757620438307,
"loss": 2.6517,
"step": 335
},
{
"epoch": 0.41,
"learning_rate": 0.0001455128007088009,
"loss": 2.355,
"step": 340
},
{
"epoch": 0.42,
"learning_rate": 0.00014360740732427367,
"loss": 2.3603,
"step": 345
},
{
"epoch": 0.43,
"learning_rate": 0.00014168225921871433,
"loss": 2.4304,
"step": 350
},
{
"epoch": 0.43,
"learning_rate": 0.00013973822850918055,
"loss": 2.3809,
"step": 355
},
{
"epoch": 0.44,
"learning_rate": 0.0001377761958667946,
"loss": 2.6495,
"step": 360
},
{
"epoch": 0.45,
"learning_rate": 0.00013579705011778766,
"loss": 2.4582,
"step": 365
},
{
"epoch": 0.45,
"learning_rate": 0.00013380168784085027,
"loss": 2.4106,
"step": 370
},
{
"epoch": 0.46,
"learning_rate": 0.00013179101296097035,
"loss": 2.1831,
"step": 375
},
{
"epoch": 0.46,
"learning_rate": 0.00012976593633994346,
"loss": 2.5222,
"step": 380
},
{
"epoch": 0.47,
"learning_rate": 0.0001277273753637408,
"loss": 2.4411,
"step": 385
},
{
"epoch": 0.48,
"learning_rate": 0.00012567625352692127,
"loss": 2.468,
"step": 390
},
{
"epoch": 0.48,
"learning_rate": 0.0001236135000142765,
"loss": 2.5124,
"step": 395
},
{
"epoch": 0.49,
"learning_rate": 0.00012154004927989815,
"loss": 2.2386,
"step": 400
},
{
"epoch": 0.49,
"learning_rate": 0.00011945684062385803,
"loss": 2.3717,
"step": 405
},
{
"epoch": 0.5,
"learning_rate": 0.00011736481776669306,
"loss": 2.4957,
"step": 410
},
{
"epoch": 0.51,
"learning_rate": 0.00011526492842188745,
"loss": 2.5043,
"step": 415
},
{
"epoch": 0.51,
"learning_rate": 0.0001131581238665465,
"loss": 2.4179,
"step": 420
},
{
"epoch": 0.52,
"learning_rate": 0.00011104535851045539,
"loss": 2.3551,
"step": 425
},
{
"epoch": 0.52,
"learning_rate": 0.00010892758946371944,
"loss": 2.3083,
"step": 430
},
{
"epoch": 0.53,
"learning_rate": 0.00010680577610318072,
"loss": 2.5975,
"step": 435
},
{
"epoch": 0.54,
"learning_rate": 0.00010468087963780789,
"loss": 2.2378,
"step": 440
},
{
"epoch": 0.54,
"learning_rate": 0.00010255386267325602,
"loss": 2.2045,
"step": 445
},
{
"epoch": 0.55,
"learning_rate": 0.00010042568877579388,
"loss": 2.0761,
"step": 450
},
{
"epoch": 0.55,
"learning_rate": 9.829732203579584e-05,
"loss": 2.3797,
"step": 455
},
{
"epoch": 0.56,
"learning_rate": 9.616972663099647e-05,
"loss": 2.457,
"step": 460
},
{
"epoch": 0.57,
"learning_rate": 9.404386638970542e-05,
"loss": 2.4869,
"step": 465
},
{
"epoch": 0.57,
"learning_rate": 9.192070435418079e-05,
"loss": 2.3134,
"step": 470
},
{
"epoch": 0.58,
"learning_rate": 8.980120234435849e-05,
"loss": 2.6018,
"step": 475
},
{
"epoch": 0.59,
"learning_rate": 8.768632052213531e-05,
"loss": 2.3912,
"step": 480
},
{
"epoch": 0.59,
"learning_rate": 8.557701695640321e-05,
"loss": 2.273,
"step": 485
},
{
"epoch": 0.6,
"learning_rate": 8.347424718903151e-05,
"loss": 2.4595,
"step": 490
},
{
"epoch": 0.6,
"learning_rate": 8.13789638019942e-05,
"loss": 2.4639,
"step": 495
},
{
"epoch": 0.61,
"learning_rate": 7.929211598583794e-05,
"loss": 2.3833,
"step": 500
},
{
"epoch": 0.62,
"learning_rate": 7.721464910968627e-05,
"loss": 2.1531,
"step": 505
},
{
"epoch": 0.62,
"learning_rate": 7.514750429297528e-05,
"loss": 1.9377,
"step": 510
},
{
"epoch": 0.63,
"learning_rate": 7.309161797911441e-05,
"loss": 2.3244,
"step": 515
},
{
"epoch": 0.63,
"learning_rate": 7.104792151126515e-05,
"loss": 2.452,
"step": 520
},
{
"epoch": 0.64,
"learning_rate": 6.901734071043071e-05,
"loss": 2.5114,
"step": 525
},
{
"epoch": 0.65,
"learning_rate": 6.700079545604708e-05,
"loss": 2.4173,
"step": 530
},
{
"epoch": 0.65,
"learning_rate": 6.499919926926566e-05,
"loss": 2.3595,
"step": 535
},
{
"epoch": 0.66,
"learning_rate": 6.301345889911637e-05,
"loss": 2.5965,
"step": 540
},
{
"epoch": 0.66,
"learning_rate": 6.104447391173858e-05,
"loss": 2.4791,
"step": 545
},
{
"epoch": 0.67,
"learning_rate": 5.909313628286601e-05,
"loss": 2.3998,
"step": 550
},
{
"epoch": 0.68,
"learning_rate": 5.716032999375006e-05,
"loss": 2.5947,
"step": 555
},
{
"epoch": 0.68,
"learning_rate": 5.524693063070492e-05,
"loss": 2.5223,
"step": 560
},
{
"epoch": 0.69,
"learning_rate": 5.335380498845559e-05,
"loss": 2.4139,
"step": 565
},
{
"epoch": 0.7,
"learning_rate": 5.148181067746862e-05,
"loss": 2.4851,
"step": 570
},
{
"epoch": 0.7,
"learning_rate": 4.963179573544357e-05,
"loss": 2.3187,
"step": 575
},
{
"epoch": 0.71,
"learning_rate": 4.7804598243140666e-05,
"loss": 2.5466,
"step": 580
},
{
"epoch": 0.71,
"learning_rate": 4.60010459447196e-05,
"loss": 2.5622,
"step": 585
},
{
"epoch": 0.72,
"learning_rate": 4.422195587276058e-05,
"loss": 2.3284,
"step": 590
},
{
"epoch": 0.73,
"learning_rate": 4.2468133978137945e-05,
"loss": 2.1308,
"step": 595
},
{
"epoch": 0.73,
"learning_rate": 4.0740374764914136e-05,
"loss": 2.5981,
"step": 600
},
{
"epoch": 0.74,
"learning_rate": 3.903946093041877e-05,
"loss": 2.1314,
"step": 605
},
{
"epoch": 0.74,
"learning_rate": 3.736616301067694e-05,
"loss": 2.2348,
"step": 610
},
{
"epoch": 0.75,
"learning_rate": 3.5721239031346066e-05,
"loss": 2.5269,
"step": 615
},
{
"epoch": 0.76,
"learning_rate": 3.410543416432069e-05,
"loss": 2.3657,
"step": 620
},
{
"epoch": 0.76,
"learning_rate": 3.2519480390159806e-05,
"loss": 2.4194,
"step": 625
},
{
"epoch": 0.77,
"learning_rate": 3.096409616649023e-05,
"loss": 2.4472,
"step": 630
},
{
"epoch": 0.77,
"learning_rate": 2.9439986102536043e-05,
"loss": 2.0381,
"step": 635
},
{
"epoch": 0.78,
"learning_rate": 2.794784063992131e-05,
"loss": 1.9038,
"step": 640
},
{
"epoch": 0.79,
"learning_rate": 2.6488335739891178e-05,
"loss": 2.5467,
"step": 645
},
{
"epoch": 0.79,
"learning_rate": 2.50621325770927e-05,
"loss": 2.3751,
"step": 650
},
{
"epoch": 0.8,
"learning_rate": 2.366987724005404e-05,
"loss": 2.2535,
"step": 655
},
{
"epoch": 0.8,
"learning_rate": 2.2312200438498043e-05,
"loss": 2.5885,
"step": 660
},
{
"epoch": 0.81,
"learning_rate": 2.0989717217622652e-05,
"loss": 2.4353,
"step": 665
},
{
"epoch": 0.82,
"learning_rate": 1.9703026679477256e-05,
"loss": 2.3635,
"step": 670
},
{
"epoch": 0.82,
"learning_rate": 1.8452711711561842e-05,
"loss": 2.4849,
"step": 675
},
{
"epoch": 0.83,
"learning_rate": 1.7239338722771327e-05,
"loss": 2.3917,
"step": 680
},
{
"epoch": 0.84,
"learning_rate": 1.6063457386805004e-05,
"loss": 2.7249,
"step": 685
},
{
"epoch": 0.84,
"learning_rate": 1.4925600393157324e-05,
"loss": 2.5142,
"step": 690
},
{
"epoch": 0.85,
"learning_rate": 1.3826283205802427e-05,
"loss": 2.5111,
"step": 695
},
{
"epoch": 0.85,
"learning_rate": 1.2766003829682505e-05,
"loss": 2.1423,
"step": 700
},
{
"epoch": 0.86,
"learning_rate": 1.1745242585104955e-05,
"loss": 2.2201,
"step": 705
},
{
"epoch": 0.87,
"learning_rate": 1.0764461890151112e-05,
"loss": 2.5609,
"step": 710
},
{
"epoch": 0.87,
"learning_rate": 9.824106051194859e-06,
"loss": 2.339,
"step": 715
},
{
"epoch": 0.88,
"learning_rate": 8.924601061626048e-06,
"loss": 2.5648,
"step": 720
},
{
"epoch": 0.88,
"learning_rate": 8.066354408870048e-06,
"loss": 2.3712,
"step": 725
},
{
"epoch": 0.89,
"learning_rate": 7.249754889790539e-06,
"loss": 2.4474,
"step": 730
},
{
"epoch": 0.9,
"learning_rate": 6.475172434559573e-06,
"loss": 2.4237,
"step": 735
},
{
"epoch": 0.9,
"learning_rate": 5.742957939074412e-06,
"loss": 2.401,
"step": 740
},
{
"epoch": 0.91,
"learning_rate": 5.0534431059970685e-06,
"loss": 2.6402,
"step": 745
},
{
"epoch": 0.91,
"learning_rate": 4.40694029448877e-06,
"loss": 2.2489,
"step": 750
},
{
"epoch": 0.92,
"learning_rate": 3.803742378707198e-06,
"loss": 2.1774,
"step": 755
},
{
"epoch": 0.93,
"learning_rate": 3.2441226151306404e-06,
"loss": 2.4729,
"step": 760
},
{
"epoch": 0.93,
"learning_rate": 2.7283345187693264e-06,
"loss": 2.1403,
"step": 765
},
{
"epoch": 0.94,
"learning_rate": 2.256611748319792e-06,
"loss": 2.818,
"step": 770
},
{
"epoch": 0.95,
"learning_rate": 1.8291680003145073e-06,
"loss": 2.8155,
"step": 775
},
{
"epoch": 0.95,
"learning_rate": 1.4461969123145457e-06,
"loss": 2.2691,
"step": 780
},
{
"epoch": 0.96,
"learning_rate": 1.107871975189234e-06,
"loss": 2.3321,
"step": 785
},
{
"epoch": 0.96,
"learning_rate": 8.143464545226298e-07,
"loss": 2.3751,
"step": 790
},
{
"epoch": 0.97,
"learning_rate": 5.657533211820942e-07,
"loss": 2.5208,
"step": 795
},
{
"epoch": 0.98,
"learning_rate": 3.622051910808666e-07,
"loss": 1.9324,
"step": 800
},
{
"epoch": 0.98,
"step": 800,
"total_flos": 5.73623349805056e+16,
"train_loss": 0.0,
"train_runtime": 12.6248,
"train_samples_per_second": 60.674,
"train_steps_per_second": 60.674
}
],
"logging_steps": 5,
"max_steps": 766,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 100,
"total_flos": 5.73623349805056e+16,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}