stealth-rag-v1 / trainer_state.json
jan-hq's picture
Model save
02376c3 verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.913937547600914,
"eval_steps": 500,
"global_step": 900,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"learning_rate": 6.756756756756757e-08,
"loss": 2.0057,
"step": 1
},
{
"epoch": 0.01,
"learning_rate": 3.378378378378379e-07,
"loss": 2.0564,
"step": 5
},
{
"epoch": 0.01,
"learning_rate": 6.756756756756758e-07,
"loss": 2.0403,
"step": 10
},
{
"epoch": 0.02,
"learning_rate": 1.0135135135135136e-06,
"loss": 2.0396,
"step": 15
},
{
"epoch": 0.02,
"learning_rate": 1.3513513513513515e-06,
"loss": 1.9947,
"step": 20
},
{
"epoch": 0.03,
"learning_rate": 1.6891891891891894e-06,
"loss": 2.0448,
"step": 25
},
{
"epoch": 0.03,
"learning_rate": 2.0270270270270273e-06,
"loss": 2.0019,
"step": 30
},
{
"epoch": 0.04,
"learning_rate": 2.364864864864865e-06,
"loss": 1.9959,
"step": 35
},
{
"epoch": 0.04,
"learning_rate": 2.702702702702703e-06,
"loss": 2.0093,
"step": 40
},
{
"epoch": 0.05,
"learning_rate": 3.040540540540541e-06,
"loss": 2.0242,
"step": 45
},
{
"epoch": 0.05,
"learning_rate": 3.3783783783783788e-06,
"loss": 1.9715,
"step": 50
},
{
"epoch": 0.06,
"learning_rate": 3.7162162162162162e-06,
"loss": 1.9773,
"step": 55
},
{
"epoch": 0.06,
"learning_rate": 4.0540540540540545e-06,
"loss": 1.9811,
"step": 60
},
{
"epoch": 0.07,
"learning_rate": 4.391891891891892e-06,
"loss": 1.9947,
"step": 65
},
{
"epoch": 0.07,
"learning_rate": 4.72972972972973e-06,
"loss": 1.988,
"step": 70
},
{
"epoch": 0.08,
"learning_rate": 5.067567567567568e-06,
"loss": 1.9718,
"step": 75
},
{
"epoch": 0.08,
"learning_rate": 5.405405405405406e-06,
"loss": 1.9505,
"step": 80
},
{
"epoch": 0.09,
"learning_rate": 5.743243243243244e-06,
"loss": 1.9706,
"step": 85
},
{
"epoch": 0.09,
"learning_rate": 6.081081081081082e-06,
"loss": 1.9506,
"step": 90
},
{
"epoch": 0.1,
"learning_rate": 6.41891891891892e-06,
"loss": 1.9243,
"step": 95
},
{
"epoch": 0.1,
"learning_rate": 6.7567567567567575e-06,
"loss": 1.9785,
"step": 100
},
{
"epoch": 0.11,
"learning_rate": 7.0945945945945946e-06,
"loss": 1.8531,
"step": 105
},
{
"epoch": 0.11,
"learning_rate": 7.4324324324324324e-06,
"loss": 1.9446,
"step": 110
},
{
"epoch": 0.12,
"learning_rate": 7.77027027027027e-06,
"loss": 1.9,
"step": 115
},
{
"epoch": 0.12,
"learning_rate": 8.108108108108109e-06,
"loss": 1.8533,
"step": 120
},
{
"epoch": 0.13,
"learning_rate": 8.445945945945948e-06,
"loss": 1.92,
"step": 125
},
{
"epoch": 0.13,
"learning_rate": 8.783783783783785e-06,
"loss": 1.8792,
"step": 130
},
{
"epoch": 0.14,
"learning_rate": 9.121621621621622e-06,
"loss": 1.9132,
"step": 135
},
{
"epoch": 0.14,
"learning_rate": 9.45945945945946e-06,
"loss": 1.8309,
"step": 140
},
{
"epoch": 0.15,
"learning_rate": 9.797297297297298e-06,
"loss": 1.8194,
"step": 145
},
{
"epoch": 0.15,
"learning_rate": 1.0135135135135136e-05,
"loss": 1.7605,
"step": 150
},
{
"epoch": 0.16,
"learning_rate": 1.0472972972972975e-05,
"loss": 1.806,
"step": 155
},
{
"epoch": 0.16,
"learning_rate": 1.0810810810810812e-05,
"loss": 1.879,
"step": 160
},
{
"epoch": 0.17,
"learning_rate": 1.114864864864865e-05,
"loss": 1.7906,
"step": 165
},
{
"epoch": 0.17,
"learning_rate": 1.1486486486486488e-05,
"loss": 1.7865,
"step": 170
},
{
"epoch": 0.18,
"learning_rate": 1.1824324324324325e-05,
"loss": 1.855,
"step": 175
},
{
"epoch": 0.18,
"learning_rate": 1.2162162162162164e-05,
"loss": 1.7428,
"step": 180
},
{
"epoch": 0.19,
"learning_rate": 1.25e-05,
"loss": 1.8219,
"step": 185
},
{
"epoch": 0.19,
"learning_rate": 1.283783783783784e-05,
"loss": 1.7607,
"step": 190
},
{
"epoch": 0.2,
"learning_rate": 1.3175675675675676e-05,
"loss": 1.8289,
"step": 195
},
{
"epoch": 0.2,
"learning_rate": 1.3513513513513515e-05,
"loss": 1.8311,
"step": 200
},
{
"epoch": 0.21,
"learning_rate": 1.3851351351351352e-05,
"loss": 1.7481,
"step": 205
},
{
"epoch": 0.21,
"learning_rate": 1.4189189189189189e-05,
"loss": 1.795,
"step": 210
},
{
"epoch": 0.22,
"learning_rate": 1.4527027027027028e-05,
"loss": 1.7709,
"step": 215
},
{
"epoch": 0.22,
"learning_rate": 1.4864864864864865e-05,
"loss": 1.7569,
"step": 220
},
{
"epoch": 0.23,
"learning_rate": 1.5202702702702704e-05,
"loss": 1.7625,
"step": 225
},
{
"epoch": 0.23,
"learning_rate": 1.554054054054054e-05,
"loss": 1.7773,
"step": 230
},
{
"epoch": 0.24,
"learning_rate": 1.587837837837838e-05,
"loss": 1.7449,
"step": 235
},
{
"epoch": 0.24,
"learning_rate": 1.6216216216216218e-05,
"loss": 1.7536,
"step": 240
},
{
"epoch": 0.25,
"learning_rate": 1.6554054054054057e-05,
"loss": 1.7815,
"step": 245
},
{
"epoch": 0.25,
"learning_rate": 1.6891891891891896e-05,
"loss": 1.81,
"step": 250
},
{
"epoch": 0.26,
"learning_rate": 1.722972972972973e-05,
"loss": 1.7928,
"step": 255
},
{
"epoch": 0.26,
"learning_rate": 1.756756756756757e-05,
"loss": 1.7427,
"step": 260
},
{
"epoch": 0.27,
"learning_rate": 1.790540540540541e-05,
"loss": 1.6618,
"step": 265
},
{
"epoch": 0.27,
"learning_rate": 1.8243243243243244e-05,
"loss": 1.7179,
"step": 270
},
{
"epoch": 0.28,
"learning_rate": 1.8581081081081082e-05,
"loss": 1.7514,
"step": 275
},
{
"epoch": 0.28,
"learning_rate": 1.891891891891892e-05,
"loss": 1.7457,
"step": 280
},
{
"epoch": 0.29,
"learning_rate": 1.925675675675676e-05,
"loss": 1.8168,
"step": 285
},
{
"epoch": 0.29,
"learning_rate": 1.9594594594594595e-05,
"loss": 1.7466,
"step": 290
},
{
"epoch": 0.3,
"learning_rate": 1.9932432432432434e-05,
"loss": 1.7226,
"step": 295
},
{
"epoch": 0.3,
"learning_rate": 1.9999888073536733e-05,
"loss": 1.7554,
"step": 300
},
{
"epoch": 0.31,
"learning_rate": 1.9999433376573803e-05,
"loss": 1.8218,
"step": 305
},
{
"epoch": 0.31,
"learning_rate": 1.9998628929598956e-05,
"loss": 1.7816,
"step": 310
},
{
"epoch": 0.32,
"learning_rate": 1.9997474760749314e-05,
"loss": 1.7414,
"step": 315
},
{
"epoch": 0.32,
"learning_rate": 1.9995970910394228e-05,
"loss": 1.778,
"step": 320
},
{
"epoch": 0.33,
"learning_rate": 1.9994117431133843e-05,
"loss": 1.7267,
"step": 325
},
{
"epoch": 0.34,
"learning_rate": 1.9991914387797266e-05,
"loss": 1.6588,
"step": 330
},
{
"epoch": 0.34,
"learning_rate": 1.9989361857440318e-05,
"loss": 1.7615,
"step": 335
},
{
"epoch": 0.35,
"learning_rate": 1.99864599293428e-05,
"loss": 1.7038,
"step": 340
},
{
"epoch": 0.35,
"learning_rate": 1.998320870500541e-05,
"loss": 1.751,
"step": 345
},
{
"epoch": 0.36,
"learning_rate": 1.997960829814616e-05,
"loss": 1.695,
"step": 350
},
{
"epoch": 0.36,
"learning_rate": 1.997565883469643e-05,
"loss": 1.6773,
"step": 355
},
{
"epoch": 0.37,
"learning_rate": 1.9971360452796523e-05,
"loss": 1.7159,
"step": 360
},
{
"epoch": 0.37,
"learning_rate": 1.9966713302790875e-05,
"loss": 1.7268,
"step": 365
},
{
"epoch": 0.38,
"learning_rate": 1.9961717547222775e-05,
"loss": 1.7641,
"step": 370
},
{
"epoch": 0.38,
"learning_rate": 1.995637336082868e-05,
"loss": 1.7376,
"step": 375
},
{
"epoch": 0.39,
"learning_rate": 1.9950680930532107e-05,
"loss": 1.7687,
"step": 380
},
{
"epoch": 0.39,
"learning_rate": 1.994464045543709e-05,
"loss": 1.7248,
"step": 385
},
{
"epoch": 0.4,
"learning_rate": 1.9938252146821236e-05,
"loss": 1.7281,
"step": 390
},
{
"epoch": 0.4,
"learning_rate": 1.9931516228128295e-05,
"loss": 1.7265,
"step": 395
},
{
"epoch": 0.41,
"learning_rate": 1.9924432934960384e-05,
"loss": 1.7459,
"step": 400
},
{
"epoch": 0.41,
"learning_rate": 1.9917002515069732e-05,
"loss": 1.719,
"step": 405
},
{
"epoch": 0.42,
"learning_rate": 1.9909225228350007e-05,
"loss": 1.6819,
"step": 410
},
{
"epoch": 0.42,
"learning_rate": 1.9901101346827233e-05,
"loss": 1.6849,
"step": 415
},
{
"epoch": 0.43,
"learning_rate": 1.989263115465028e-05,
"loss": 1.7905,
"step": 420
},
{
"epoch": 0.43,
"learning_rate": 1.9883814948080918e-05,
"loss": 1.7414,
"step": 425
},
{
"epoch": 0.44,
"learning_rate": 1.987465303548345e-05,
"loss": 1.7082,
"step": 430
},
{
"epoch": 0.44,
"learning_rate": 1.9865145737313937e-05,
"loss": 1.743,
"step": 435
},
{
"epoch": 0.45,
"learning_rate": 1.9855293386108995e-05,
"loss": 1.7089,
"step": 440
},
{
"epoch": 0.45,
"learning_rate": 1.9845096326474135e-05,
"loss": 1.7491,
"step": 445
},
{
"epoch": 0.46,
"learning_rate": 1.9834554915071745e-05,
"loss": 1.6897,
"step": 450
},
{
"epoch": 0.46,
"learning_rate": 1.982366952060859e-05,
"loss": 1.7759,
"step": 455
},
{
"epoch": 0.47,
"learning_rate": 1.981244052382293e-05,
"loss": 1.7774,
"step": 460
},
{
"epoch": 0.47,
"learning_rate": 1.9800868317471196e-05,
"loss": 1.6844,
"step": 465
},
{
"epoch": 0.48,
"learning_rate": 1.978895330631425e-05,
"loss": 1.7419,
"step": 470
},
{
"epoch": 0.48,
"learning_rate": 1.977669590710324e-05,
"loss": 1.7966,
"step": 475
},
{
"epoch": 0.49,
"learning_rate": 1.976409654856501e-05,
"loss": 1.6607,
"step": 480
},
{
"epoch": 0.49,
"learning_rate": 1.975115567138711e-05,
"loss": 1.8121,
"step": 485
},
{
"epoch": 0.5,
"learning_rate": 1.9737873728202376e-05,
"loss": 1.7282,
"step": 490
},
{
"epoch": 0.5,
"learning_rate": 1.972425118357312e-05,
"loss": 1.6864,
"step": 495
},
{
"epoch": 0.51,
"learning_rate": 1.9710288513974846e-05,
"loss": 1.7665,
"step": 500
},
{
"epoch": 0.51,
"learning_rate": 1.969598620777962e-05,
"loss": 1.6843,
"step": 505
},
{
"epoch": 0.52,
"learning_rate": 1.9681344765238958e-05,
"loss": 1.6675,
"step": 510
},
{
"epoch": 0.52,
"learning_rate": 1.966636469846635e-05,
"loss": 1.7891,
"step": 515
},
{
"epoch": 0.53,
"learning_rate": 1.9651046531419335e-05,
"loss": 1.6728,
"step": 520
},
{
"epoch": 0.53,
"learning_rate": 1.9635390799881186e-05,
"loss": 1.7289,
"step": 525
},
{
"epoch": 0.54,
"learning_rate": 1.961939805144216e-05,
"loss": 1.7241,
"step": 530
},
{
"epoch": 0.54,
"learning_rate": 1.9603068845480347e-05,
"loss": 1.7378,
"step": 535
},
{
"epoch": 0.55,
"learning_rate": 1.9586403753142104e-05,
"loss": 1.7717,
"step": 540
},
{
"epoch": 0.55,
"learning_rate": 1.956940335732209e-05,
"loss": 1.7401,
"step": 545
},
{
"epoch": 0.56,
"learning_rate": 1.9552068252642858e-05,
"loss": 1.6758,
"step": 550
},
{
"epoch": 0.56,
"learning_rate": 1.9534399045434073e-05,
"loss": 1.7241,
"step": 555
},
{
"epoch": 0.57,
"learning_rate": 1.9516396353711297e-05,
"loss": 1.7026,
"step": 560
},
{
"epoch": 0.57,
"learning_rate": 1.9498060807154368e-05,
"loss": 1.7662,
"step": 565
},
{
"epoch": 0.58,
"learning_rate": 1.9479393047085392e-05,
"loss": 1.6935,
"step": 570
},
{
"epoch": 0.58,
"learning_rate": 1.94603937264463e-05,
"loss": 1.7303,
"step": 575
},
{
"epoch": 0.59,
"learning_rate": 1.9441063509776003e-05,
"loss": 1.7758,
"step": 580
},
{
"epoch": 0.59,
"learning_rate": 1.9421403073187162e-05,
"loss": 1.7017,
"step": 585
},
{
"epoch": 0.6,
"learning_rate": 1.9401413104342535e-05,
"loss": 1.7726,
"step": 590
},
{
"epoch": 0.6,
"learning_rate": 1.938109430243093e-05,
"loss": 1.6689,
"step": 595
},
{
"epoch": 0.61,
"learning_rate": 1.936044737814273e-05,
"loss": 1.6734,
"step": 600
},
{
"epoch": 0.61,
"learning_rate": 1.9339473053645056e-05,
"loss": 1.656,
"step": 605
},
{
"epoch": 0.62,
"learning_rate": 1.931817206255651e-05,
"loss": 1.7196,
"step": 610
},
{
"epoch": 0.62,
"learning_rate": 1.9296545149921488e-05,
"loss": 1.686,
"step": 615
},
{
"epoch": 0.63,
"learning_rate": 1.9274593072184152e-05,
"loss": 1.7709,
"step": 620
},
{
"epoch": 0.63,
"learning_rate": 1.9252316597161947e-05,
"loss": 1.6834,
"step": 625
},
{
"epoch": 0.64,
"learning_rate": 1.922971650401877e-05,
"loss": 1.6787,
"step": 630
},
{
"epoch": 0.64,
"learning_rate": 1.920679358323769e-05,
"loss": 1.6922,
"step": 635
},
{
"epoch": 0.65,
"learning_rate": 1.9183548636593322e-05,
"loss": 1.7244,
"step": 640
},
{
"epoch": 0.65,
"learning_rate": 1.9159982477123776e-05,
"loss": 1.7187,
"step": 645
},
{
"epoch": 0.66,
"learning_rate": 1.9136095929102204e-05,
"loss": 1.6827,
"step": 650
},
{
"epoch": 0.67,
"learning_rate": 1.9111889828007997e-05,
"loss": 1.7466,
"step": 655
},
{
"epoch": 0.67,
"learning_rate": 1.908736502049754e-05,
"loss": 1.7892,
"step": 660
},
{
"epoch": 0.68,
"learning_rate": 1.9062522364374617e-05,
"loss": 1.744,
"step": 665
},
{
"epoch": 0.68,
"learning_rate": 1.903736272856038e-05,
"loss": 1.6438,
"step": 670
},
{
"epoch": 0.69,
"learning_rate": 1.9011886993062994e-05,
"loss": 1.7538,
"step": 675
},
{
"epoch": 0.69,
"learning_rate": 1.8986096048946826e-05,
"loss": 1.6889,
"step": 680
},
{
"epoch": 0.7,
"learning_rate": 1.8959990798301286e-05,
"loss": 1.7199,
"step": 685
},
{
"epoch": 0.7,
"learning_rate": 1.893357215420929e-05,
"loss": 1.7801,
"step": 690
},
{
"epoch": 0.71,
"learning_rate": 1.8906841040715304e-05,
"loss": 1.7559,
"step": 695
},
{
"epoch": 0.71,
"learning_rate": 1.8879798392793033e-05,
"loss": 1.6639,
"step": 700
},
{
"epoch": 0.72,
"learning_rate": 1.8852445156312713e-05,
"loss": 1.687,
"step": 705
},
{
"epoch": 0.72,
"learning_rate": 1.8824782288008038e-05,
"loss": 1.7619,
"step": 710
},
{
"epoch": 0.73,
"learning_rate": 1.8796810755442675e-05,
"loss": 1.7529,
"step": 715
},
{
"epoch": 0.73,
"learning_rate": 1.8768531536976452e-05,
"loss": 1.7513,
"step": 720
},
{
"epoch": 0.74,
"learning_rate": 1.8739945621731114e-05,
"loss": 1.7322,
"step": 725
},
{
"epoch": 0.74,
"learning_rate": 1.8711054009555736e-05,
"loss": 1.7762,
"step": 730
},
{
"epoch": 0.75,
"learning_rate": 1.8681857710991745e-05,
"loss": 1.6949,
"step": 735
},
{
"epoch": 0.75,
"learning_rate": 1.865235774723758e-05,
"loss": 1.6526,
"step": 740
},
{
"epoch": 0.76,
"learning_rate": 1.862255515011298e-05,
"loss": 1.7155,
"step": 745
},
{
"epoch": 0.76,
"learning_rate": 1.8592450962022872e-05,
"loss": 1.8251,
"step": 750
},
{
"epoch": 0.77,
"learning_rate": 1.8562046235920938e-05,
"loss": 1.7077,
"step": 755
},
{
"epoch": 0.77,
"learning_rate": 1.8531342035272768e-05,
"loss": 1.6627,
"step": 760
},
{
"epoch": 0.78,
"learning_rate": 1.850033943401867e-05,
"loss": 1.7163,
"step": 765
},
{
"epoch": 0.78,
"learning_rate": 1.84690395165361e-05,
"loss": 1.713,
"step": 770
},
{
"epoch": 0.79,
"learning_rate": 1.8437443377601736e-05,
"loss": 1.7024,
"step": 775
},
{
"epoch": 0.79,
"learning_rate": 1.8405552122353212e-05,
"loss": 1.7253,
"step": 780
},
{
"epoch": 0.8,
"learning_rate": 1.8373366866250407e-05,
"loss": 1.7109,
"step": 785
},
{
"epoch": 0.8,
"learning_rate": 1.8340888735036485e-05,
"loss": 1.7563,
"step": 790
},
{
"epoch": 0.81,
"learning_rate": 1.830811886469849e-05,
"loss": 1.7479,
"step": 795
},
{
"epoch": 0.81,
"learning_rate": 1.8275058401427622e-05,
"loss": 1.7315,
"step": 800
},
{
"epoch": 0.82,
"learning_rate": 1.8241708501579146e-05,
"loss": 1.7126,
"step": 805
},
{
"epoch": 0.82,
"learning_rate": 1.8208070331631943e-05,
"loss": 1.6866,
"step": 810
},
{
"epoch": 0.83,
"learning_rate": 1.817414506814772e-05,
"loss": 1.6708,
"step": 815
},
{
"epoch": 0.83,
"learning_rate": 1.8139933897729833e-05,
"loss": 1.6868,
"step": 820
},
{
"epoch": 0.84,
"learning_rate": 1.8105438016981816e-05,
"loss": 1.7493,
"step": 825
},
{
"epoch": 0.84,
"learning_rate": 1.807065863246551e-05,
"loss": 1.7707,
"step": 830
},
{
"epoch": 0.85,
"learning_rate": 1.8035596960658856e-05,
"loss": 1.7692,
"step": 835
},
{
"epoch": 0.85,
"learning_rate": 1.8000254227913346e-05,
"loss": 1.7761,
"step": 840
},
{
"epoch": 0.86,
"learning_rate": 1.7964631670411154e-05,
"loss": 1.6765,
"step": 845
},
{
"epoch": 0.86,
"learning_rate": 1.7928730534121872e-05,
"loss": 1.6437,
"step": 850
},
{
"epoch": 0.87,
"learning_rate": 1.7892552074758932e-05,
"loss": 1.6804,
"step": 855
},
{
"epoch": 0.87,
"learning_rate": 1.7856097557735697e-05,
"loss": 1.7373,
"step": 860
},
{
"epoch": 0.88,
"learning_rate": 1.7819368258121188e-05,
"loss": 1.7316,
"step": 865
},
{
"epoch": 0.88,
"learning_rate": 1.77823654605955e-05,
"loss": 1.6764,
"step": 870
},
{
"epoch": 0.89,
"learning_rate": 1.7745090459404853e-05,
"loss": 1.7456,
"step": 875
},
{
"epoch": 0.89,
"learning_rate": 1.7707544558316332e-05,
"loss": 1.6518,
"step": 880
},
{
"epoch": 0.9,
"learning_rate": 1.766972907057229e-05,
"loss": 1.7705,
"step": 885
},
{
"epoch": 0.9,
"learning_rate": 1.763164531884439e-05,
"loss": 1.7431,
"step": 890
},
{
"epoch": 0.91,
"learning_rate": 1.7593294635187385e-05,
"loss": 1.6928,
"step": 895
},
{
"epoch": 0.91,
"learning_rate": 1.7554678360992475e-05,
"loss": 1.7483,
"step": 900
},
{
"epoch": 0.91,
"step": 900,
"total_flos": 1.2656331456763658e+18,
"train_loss": 0.0,
"train_runtime": 3.9268,
"train_samples_per_second": 12034.857,
"train_steps_per_second": 187.937
}
],
"logging_steps": 5,
"max_steps": 738,
"num_input_tokens_seen": 0,
"num_train_epochs": 3,
"save_steps": 100,
"total_flos": 1.2656331456763658e+18,
"train_batch_size": 2,
"trial_name": null,
"trial_params": null
}