cvt-13-finetuned-ibird / trainer_state.json
hemlatachanne's picture
End of training
cbea244 verified
{
"best_metric": 0.9866785079928952,
"best_model_checkpoint": "cvt-13-finetuned-ibird/checkpoint-1696",
"epoch": 4.995579133510168,
"eval_steps": 500,
"global_step": 2825,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.02,
"grad_norm": 3.22133731842041,
"learning_rate": 1.76678445229682e-06,
"loss": 3.2244,
"step": 10
},
{
"epoch": 0.04,
"grad_norm": 2.790149688720703,
"learning_rate": 3.53356890459364e-06,
"loss": 3.2169,
"step": 20
},
{
"epoch": 0.05,
"grad_norm": 2.872779369354248,
"learning_rate": 5.30035335689046e-06,
"loss": 3.2236,
"step": 30
},
{
"epoch": 0.07,
"grad_norm": 2.464066982269287,
"learning_rate": 7.06713780918728e-06,
"loss": 3.2237,
"step": 40
},
{
"epoch": 0.09,
"grad_norm": 3.8840532302856445,
"learning_rate": 8.8339222614841e-06,
"loss": 3.2091,
"step": 50
},
{
"epoch": 0.11,
"grad_norm": 2.897418975830078,
"learning_rate": 1.060070671378092e-05,
"loss": 3.2041,
"step": 60
},
{
"epoch": 0.12,
"grad_norm": 2.91304874420166,
"learning_rate": 1.236749116607774e-05,
"loss": 3.1712,
"step": 70
},
{
"epoch": 0.14,
"grad_norm": 3.0000202655792236,
"learning_rate": 1.413427561837456e-05,
"loss": 3.1854,
"step": 80
},
{
"epoch": 0.16,
"grad_norm": 2.865732192993164,
"learning_rate": 1.5901060070671377e-05,
"loss": 3.1636,
"step": 90
},
{
"epoch": 0.18,
"grad_norm": 3.4034979343414307,
"learning_rate": 1.76678445229682e-05,
"loss": 3.128,
"step": 100
},
{
"epoch": 0.19,
"grad_norm": 3.288928747177124,
"learning_rate": 1.9434628975265016e-05,
"loss": 3.1292,
"step": 110
},
{
"epoch": 0.21,
"grad_norm": 5.1810078620910645,
"learning_rate": 2.120141342756184e-05,
"loss": 3.0734,
"step": 120
},
{
"epoch": 0.23,
"grad_norm": 2.9931440353393555,
"learning_rate": 2.296819787985866e-05,
"loss": 3.0704,
"step": 130
},
{
"epoch": 0.25,
"grad_norm": 3.1638436317443848,
"learning_rate": 2.473498233215548e-05,
"loss": 3.0107,
"step": 140
},
{
"epoch": 0.27,
"grad_norm": 2.532331705093384,
"learning_rate": 2.6501766784452298e-05,
"loss": 2.9914,
"step": 150
},
{
"epoch": 0.28,
"grad_norm": 3.235107421875,
"learning_rate": 2.826855123674912e-05,
"loss": 2.9416,
"step": 160
},
{
"epoch": 0.3,
"grad_norm": 3.254007339477539,
"learning_rate": 3.003533568904594e-05,
"loss": 2.9103,
"step": 170
},
{
"epoch": 0.32,
"grad_norm": 3.413905143737793,
"learning_rate": 3.1802120141342755e-05,
"loss": 2.8261,
"step": 180
},
{
"epoch": 0.34,
"grad_norm": 3.3824758529663086,
"learning_rate": 3.356890459363958e-05,
"loss": 2.7711,
"step": 190
},
{
"epoch": 0.35,
"grad_norm": 3.0212182998657227,
"learning_rate": 3.53356890459364e-05,
"loss": 2.7134,
"step": 200
},
{
"epoch": 0.37,
"grad_norm": 3.3175628185272217,
"learning_rate": 3.710247349823322e-05,
"loss": 2.5992,
"step": 210
},
{
"epoch": 0.39,
"grad_norm": 3.2799651622772217,
"learning_rate": 3.886925795053003e-05,
"loss": 2.565,
"step": 220
},
{
"epoch": 0.41,
"grad_norm": 5.323554039001465,
"learning_rate": 4.063604240282686e-05,
"loss": 2.4975,
"step": 230
},
{
"epoch": 0.42,
"grad_norm": 3.1259026527404785,
"learning_rate": 4.240282685512368e-05,
"loss": 2.3784,
"step": 240
},
{
"epoch": 0.44,
"grad_norm": 3.178640842437744,
"learning_rate": 4.416961130742049e-05,
"loss": 2.3291,
"step": 250
},
{
"epoch": 0.46,
"grad_norm": 3.33782696723938,
"learning_rate": 4.593639575971732e-05,
"loss": 2.2431,
"step": 260
},
{
"epoch": 0.48,
"grad_norm": 3.880281686782837,
"learning_rate": 4.7703180212014135e-05,
"loss": 2.135,
"step": 270
},
{
"epoch": 0.5,
"grad_norm": 3.968682050704956,
"learning_rate": 4.946996466431096e-05,
"loss": 2.0128,
"step": 280
},
{
"epoch": 0.51,
"grad_norm": 4.611078262329102,
"learning_rate": 4.9862313139260423e-05,
"loss": 2.0385,
"step": 290
},
{
"epoch": 0.53,
"grad_norm": 3.0588765144348145,
"learning_rate": 4.9665617623918175e-05,
"loss": 1.8919,
"step": 300
},
{
"epoch": 0.55,
"grad_norm": 3.524935722351074,
"learning_rate": 4.9468922108575926e-05,
"loss": 1.8712,
"step": 310
},
{
"epoch": 0.57,
"grad_norm": 3.9682562351226807,
"learning_rate": 4.927222659323368e-05,
"loss": 1.7009,
"step": 320
},
{
"epoch": 0.58,
"grad_norm": 3.5031826496124268,
"learning_rate": 4.907553107789143e-05,
"loss": 1.5617,
"step": 330
},
{
"epoch": 0.6,
"grad_norm": 3.6093549728393555,
"learning_rate": 4.887883556254917e-05,
"loss": 1.6847,
"step": 340
},
{
"epoch": 0.62,
"grad_norm": 4.248983383178711,
"learning_rate": 4.8682140047206924e-05,
"loss": 1.5254,
"step": 350
},
{
"epoch": 0.64,
"grad_norm": 3.7051937580108643,
"learning_rate": 4.8485444531864675e-05,
"loss": 1.4551,
"step": 360
},
{
"epoch": 0.65,
"grad_norm": 3.0025603771209717,
"learning_rate": 4.8288749016522426e-05,
"loss": 1.5533,
"step": 370
},
{
"epoch": 0.67,
"grad_norm": 3.4976930618286133,
"learning_rate": 4.809205350118017e-05,
"loss": 1.3265,
"step": 380
},
{
"epoch": 0.69,
"grad_norm": 3.950221061706543,
"learning_rate": 4.789535798583792e-05,
"loss": 1.2725,
"step": 390
},
{
"epoch": 0.71,
"grad_norm": 4.968952178955078,
"learning_rate": 4.769866247049567e-05,
"loss": 1.2879,
"step": 400
},
{
"epoch": 0.73,
"grad_norm": 3.3156518936157227,
"learning_rate": 4.7501966955153424e-05,
"loss": 1.1826,
"step": 410
},
{
"epoch": 0.74,
"grad_norm": 3.190901756286621,
"learning_rate": 4.7305271439811175e-05,
"loss": 1.1947,
"step": 420
},
{
"epoch": 0.76,
"grad_norm": 2.0758602619171143,
"learning_rate": 4.7108575924468926e-05,
"loss": 1.2037,
"step": 430
},
{
"epoch": 0.78,
"grad_norm": 2.5758004188537598,
"learning_rate": 4.691188040912668e-05,
"loss": 1.1067,
"step": 440
},
{
"epoch": 0.8,
"grad_norm": 3.4717776775360107,
"learning_rate": 4.671518489378442e-05,
"loss": 1.2233,
"step": 450
},
{
"epoch": 0.81,
"grad_norm": 2.4771013259887695,
"learning_rate": 4.651848937844217e-05,
"loss": 1.0819,
"step": 460
},
{
"epoch": 0.83,
"grad_norm": 2.409405469894409,
"learning_rate": 4.6321793863099924e-05,
"loss": 1.2001,
"step": 470
},
{
"epoch": 0.85,
"grad_norm": 2.6338610649108887,
"learning_rate": 4.6125098347757675e-05,
"loss": 1.0693,
"step": 480
},
{
"epoch": 0.87,
"grad_norm": 3.0745251178741455,
"learning_rate": 4.5928402832415426e-05,
"loss": 1.1109,
"step": 490
},
{
"epoch": 0.88,
"grad_norm": 2.047109842300415,
"learning_rate": 4.573170731707318e-05,
"loss": 1.0295,
"step": 500
},
{
"epoch": 0.9,
"grad_norm": 4.1728410720825195,
"learning_rate": 4.553501180173092e-05,
"loss": 1.0533,
"step": 510
},
{
"epoch": 0.92,
"grad_norm": 3.705878973007202,
"learning_rate": 4.533831628638867e-05,
"loss": 1.0001,
"step": 520
},
{
"epoch": 0.94,
"grad_norm": 2.719705820083618,
"learning_rate": 4.5141620771046424e-05,
"loss": 1.0657,
"step": 530
},
{
"epoch": 0.95,
"grad_norm": 3.654970169067383,
"learning_rate": 4.4944925255704175e-05,
"loss": 0.9326,
"step": 540
},
{
"epoch": 0.97,
"grad_norm": 5.275374889373779,
"learning_rate": 4.4748229740361926e-05,
"loss": 1.0006,
"step": 550
},
{
"epoch": 0.99,
"grad_norm": 3.071272373199463,
"learning_rate": 4.455153422501967e-05,
"loss": 1.0089,
"step": 560
},
{
"epoch": 1.0,
"eval_accuracy": 0.9649200710479574,
"eval_loss": 0.2877684533596039,
"eval_runtime": 86.5125,
"eval_samples_per_second": 26.031,
"eval_steps_per_second": 3.26,
"step": 565
},
{
"epoch": 1.01,
"grad_norm": 3.3852458000183105,
"learning_rate": 4.435483870967742e-05,
"loss": 0.865,
"step": 570
},
{
"epoch": 1.03,
"grad_norm": 2.0667455196380615,
"learning_rate": 4.415814319433517e-05,
"loss": 0.9117,
"step": 580
},
{
"epoch": 1.04,
"grad_norm": 1.9478442668914795,
"learning_rate": 4.3961447678992924e-05,
"loss": 0.8391,
"step": 590
},
{
"epoch": 1.06,
"grad_norm": 0.9961579442024231,
"learning_rate": 4.376475216365067e-05,
"loss": 0.8528,
"step": 600
},
{
"epoch": 1.08,
"grad_norm": 3.6760354042053223,
"learning_rate": 4.356805664830842e-05,
"loss": 0.8387,
"step": 610
},
{
"epoch": 1.1,
"grad_norm": 3.119180917739868,
"learning_rate": 4.337136113296617e-05,
"loss": 0.7549,
"step": 620
},
{
"epoch": 1.11,
"grad_norm": 3.2463724613189697,
"learning_rate": 4.317466561762392e-05,
"loss": 0.9329,
"step": 630
},
{
"epoch": 1.13,
"grad_norm": 1.7437245845794678,
"learning_rate": 4.297797010228167e-05,
"loss": 0.824,
"step": 640
},
{
"epoch": 1.15,
"grad_norm": 2.5226683616638184,
"learning_rate": 4.278127458693942e-05,
"loss": 0.8805,
"step": 650
},
{
"epoch": 1.17,
"grad_norm": 2.30314302444458,
"learning_rate": 4.258457907159717e-05,
"loss": 0.9097,
"step": 660
},
{
"epoch": 1.18,
"grad_norm": 2.0539016723632812,
"learning_rate": 4.238788355625492e-05,
"loss": 0.8936,
"step": 670
},
{
"epoch": 1.2,
"grad_norm": 3.0018861293792725,
"learning_rate": 4.219118804091267e-05,
"loss": 0.9214,
"step": 680
},
{
"epoch": 1.22,
"grad_norm": 2.0286073684692383,
"learning_rate": 4.1994492525570416e-05,
"loss": 0.7991,
"step": 690
},
{
"epoch": 1.24,
"grad_norm": 0.8428072929382324,
"learning_rate": 4.179779701022817e-05,
"loss": 0.756,
"step": 700
},
{
"epoch": 1.26,
"grad_norm": 1.7304598093032837,
"learning_rate": 4.160110149488592e-05,
"loss": 0.7738,
"step": 710
},
{
"epoch": 1.27,
"grad_norm": 3.970334768295288,
"learning_rate": 4.140440597954367e-05,
"loss": 0.7426,
"step": 720
},
{
"epoch": 1.29,
"grad_norm": 2.3020598888397217,
"learning_rate": 4.1207710464201413e-05,
"loss": 0.7884,
"step": 730
},
{
"epoch": 1.31,
"grad_norm": 1.7566787004470825,
"learning_rate": 4.1011014948859165e-05,
"loss": 0.7569,
"step": 740
},
{
"epoch": 1.33,
"grad_norm": 3.269785165786743,
"learning_rate": 4.0814319433516916e-05,
"loss": 0.8191,
"step": 750
},
{
"epoch": 1.34,
"grad_norm": 2.191699504852295,
"learning_rate": 4.061762391817467e-05,
"loss": 0.8077,
"step": 760
},
{
"epoch": 1.36,
"grad_norm": 1.7478408813476562,
"learning_rate": 4.042092840283242e-05,
"loss": 0.7621,
"step": 770
},
{
"epoch": 1.38,
"grad_norm": 2.7008702754974365,
"learning_rate": 4.022423288749016e-05,
"loss": 0.81,
"step": 780
},
{
"epoch": 1.4,
"grad_norm": 1.9396083354949951,
"learning_rate": 4.0027537372147914e-05,
"loss": 0.7493,
"step": 790
},
{
"epoch": 1.41,
"grad_norm": 2.55366587638855,
"learning_rate": 3.9830841856805665e-05,
"loss": 0.7471,
"step": 800
},
{
"epoch": 1.43,
"grad_norm": 1.6567083597183228,
"learning_rate": 3.9634146341463416e-05,
"loss": 0.9089,
"step": 810
},
{
"epoch": 1.45,
"grad_norm": 4.915744781494141,
"learning_rate": 3.943745082612117e-05,
"loss": 0.8211,
"step": 820
},
{
"epoch": 1.47,
"grad_norm": 1.2864570617675781,
"learning_rate": 3.924075531077892e-05,
"loss": 0.7662,
"step": 830
},
{
"epoch": 1.49,
"grad_norm": 1.4654157161712646,
"learning_rate": 3.904405979543666e-05,
"loss": 0.7942,
"step": 840
},
{
"epoch": 1.5,
"grad_norm": 5.769187927246094,
"learning_rate": 3.8847364280094414e-05,
"loss": 0.73,
"step": 850
},
{
"epoch": 1.52,
"grad_norm": 0.6875308156013489,
"learning_rate": 3.8650668764752165e-05,
"loss": 0.7394,
"step": 860
},
{
"epoch": 1.54,
"grad_norm": 1.2160513401031494,
"learning_rate": 3.8453973249409916e-05,
"loss": 0.846,
"step": 870
},
{
"epoch": 1.56,
"grad_norm": 1.4410216808319092,
"learning_rate": 3.825727773406767e-05,
"loss": 0.7366,
"step": 880
},
{
"epoch": 1.57,
"grad_norm": 4.212517261505127,
"learning_rate": 3.806058221872542e-05,
"loss": 0.8275,
"step": 890
},
{
"epoch": 1.59,
"grad_norm": 0.9933316111564636,
"learning_rate": 3.786388670338317e-05,
"loss": 0.7688,
"step": 900
},
{
"epoch": 1.61,
"grad_norm": 2.2545719146728516,
"learning_rate": 3.7667191188040914e-05,
"loss": 0.734,
"step": 910
},
{
"epoch": 1.63,
"grad_norm": 1.0783274173736572,
"learning_rate": 3.7470495672698665e-05,
"loss": 0.8099,
"step": 920
},
{
"epoch": 1.64,
"grad_norm": 1.6741780042648315,
"learning_rate": 3.7273800157356416e-05,
"loss": 0.9695,
"step": 930
},
{
"epoch": 1.66,
"grad_norm": 3.448066473007202,
"learning_rate": 3.707710464201417e-05,
"loss": 0.7506,
"step": 940
},
{
"epoch": 1.68,
"grad_norm": 9.674800872802734,
"learning_rate": 3.688040912667191e-05,
"loss": 0.6566,
"step": 950
},
{
"epoch": 1.7,
"grad_norm": 6.721800327301025,
"learning_rate": 3.668371361132966e-05,
"loss": 0.7075,
"step": 960
},
{
"epoch": 1.72,
"grad_norm": 5.588386058807373,
"learning_rate": 3.6487018095987414e-05,
"loss": 0.8246,
"step": 970
},
{
"epoch": 1.73,
"grad_norm": 1.4187864065170288,
"learning_rate": 3.6290322580645165e-05,
"loss": 0.7808,
"step": 980
},
{
"epoch": 1.75,
"grad_norm": 0.7722711563110352,
"learning_rate": 3.6093627065302916e-05,
"loss": 0.8957,
"step": 990
},
{
"epoch": 1.77,
"grad_norm": 1.5007630586624146,
"learning_rate": 3.589693154996066e-05,
"loss": 0.7536,
"step": 1000
},
{
"epoch": 1.79,
"grad_norm": 0.7198464870452881,
"learning_rate": 3.570023603461841e-05,
"loss": 0.6727,
"step": 1010
},
{
"epoch": 1.8,
"grad_norm": 6.015268802642822,
"learning_rate": 3.550354051927616e-05,
"loss": 0.737,
"step": 1020
},
{
"epoch": 1.82,
"grad_norm": 0.5802467465400696,
"learning_rate": 3.5306845003933914e-05,
"loss": 0.6929,
"step": 1030
},
{
"epoch": 1.84,
"grad_norm": 6.546850681304932,
"learning_rate": 3.511014948859166e-05,
"loss": 0.7529,
"step": 1040
},
{
"epoch": 1.86,
"grad_norm": 7.255277156829834,
"learning_rate": 3.491345397324941e-05,
"loss": 0.8639,
"step": 1050
},
{
"epoch": 1.87,
"grad_norm": 0.564946711063385,
"learning_rate": 3.471675845790716e-05,
"loss": 0.7234,
"step": 1060
},
{
"epoch": 1.89,
"grad_norm": 1.2637860774993896,
"learning_rate": 3.452006294256491e-05,
"loss": 0.9388,
"step": 1070
},
{
"epoch": 1.91,
"grad_norm": 1.095790147781372,
"learning_rate": 3.432336742722266e-05,
"loss": 0.7637,
"step": 1080
},
{
"epoch": 1.93,
"grad_norm": 4.213192462921143,
"learning_rate": 3.412667191188041e-05,
"loss": 0.7857,
"step": 1090
},
{
"epoch": 1.95,
"grad_norm": 9.215189933776855,
"learning_rate": 3.392997639653816e-05,
"loss": 0.6239,
"step": 1100
},
{
"epoch": 1.96,
"grad_norm": 2.098184585571289,
"learning_rate": 3.373328088119591e-05,
"loss": 0.5597,
"step": 1110
},
{
"epoch": 1.98,
"grad_norm": 1.9442644119262695,
"learning_rate": 3.353658536585366e-05,
"loss": 0.7885,
"step": 1120
},
{
"epoch": 2.0,
"grad_norm": 6.179882526397705,
"learning_rate": 3.3339889850511406e-05,
"loss": 0.7855,
"step": 1130
},
{
"epoch": 2.0,
"eval_accuracy": 0.9791296625222025,
"eval_loss": 0.10786169767379761,
"eval_runtime": 86.2088,
"eval_samples_per_second": 26.123,
"eval_steps_per_second": 3.271,
"step": 1131
},
{
"epoch": 2.02,
"grad_norm": 0.36537787318229675,
"learning_rate": 3.314319433516916e-05,
"loss": 0.8073,
"step": 1140
},
{
"epoch": 2.03,
"grad_norm": 0.9752954840660095,
"learning_rate": 3.294649881982691e-05,
"loss": 0.586,
"step": 1150
},
{
"epoch": 2.05,
"grad_norm": 0.6317528486251831,
"learning_rate": 3.274980330448466e-05,
"loss": 0.7136,
"step": 1160
},
{
"epoch": 2.07,
"grad_norm": 0.4661918580532074,
"learning_rate": 3.255310778914241e-05,
"loss": 0.7598,
"step": 1170
},
{
"epoch": 2.09,
"grad_norm": 0.6691815853118896,
"learning_rate": 3.2356412273800155e-05,
"loss": 0.9078,
"step": 1180
},
{
"epoch": 2.1,
"grad_norm": 2.7598626613616943,
"learning_rate": 3.2159716758457906e-05,
"loss": 0.5849,
"step": 1190
},
{
"epoch": 2.12,
"grad_norm": 0.4353080093860626,
"learning_rate": 3.196302124311566e-05,
"loss": 0.8519,
"step": 1200
},
{
"epoch": 2.14,
"grad_norm": 0.5575789213180542,
"learning_rate": 3.176632572777341e-05,
"loss": 0.7124,
"step": 1210
},
{
"epoch": 2.16,
"grad_norm": 2.028132677078247,
"learning_rate": 3.156963021243116e-05,
"loss": 0.8845,
"step": 1220
},
{
"epoch": 2.18,
"grad_norm": 2.471712350845337,
"learning_rate": 3.137293469708891e-05,
"loss": 0.7683,
"step": 1230
},
{
"epoch": 2.19,
"grad_norm": 3.2386505603790283,
"learning_rate": 3.1176239181746655e-05,
"loss": 0.6573,
"step": 1240
},
{
"epoch": 2.21,
"grad_norm": 0.6279830932617188,
"learning_rate": 3.0979543666404406e-05,
"loss": 0.6261,
"step": 1250
},
{
"epoch": 2.23,
"grad_norm": 1.3760743141174316,
"learning_rate": 3.078284815106216e-05,
"loss": 0.8395,
"step": 1260
},
{
"epoch": 2.25,
"grad_norm": 0.6067408919334412,
"learning_rate": 3.058615263571991e-05,
"loss": 0.6631,
"step": 1270
},
{
"epoch": 2.26,
"grad_norm": 1.8222625255584717,
"learning_rate": 3.0389457120377656e-05,
"loss": 0.7618,
"step": 1280
},
{
"epoch": 2.28,
"grad_norm": 0.4705863296985626,
"learning_rate": 3.0192761605035407e-05,
"loss": 0.657,
"step": 1290
},
{
"epoch": 2.3,
"grad_norm": 0.885718047618866,
"learning_rate": 2.999606608969316e-05,
"loss": 0.6618,
"step": 1300
},
{
"epoch": 2.32,
"grad_norm": 0.5552624464035034,
"learning_rate": 2.9799370574350903e-05,
"loss": 0.7135,
"step": 1310
},
{
"epoch": 2.33,
"grad_norm": 0.8557870388031006,
"learning_rate": 2.9602675059008654e-05,
"loss": 0.6492,
"step": 1320
},
{
"epoch": 2.35,
"grad_norm": 1.226236343383789,
"learning_rate": 2.9405979543666405e-05,
"loss": 0.5645,
"step": 1330
},
{
"epoch": 2.37,
"grad_norm": 0.4508123993873596,
"learning_rate": 2.9209284028324156e-05,
"loss": 0.653,
"step": 1340
},
{
"epoch": 2.39,
"grad_norm": 5.647830963134766,
"learning_rate": 2.9012588512981904e-05,
"loss": 0.6795,
"step": 1350
},
{
"epoch": 2.4,
"grad_norm": 0.6091474294662476,
"learning_rate": 2.8815892997639655e-05,
"loss": 0.7905,
"step": 1360
},
{
"epoch": 2.42,
"grad_norm": 0.5606883764266968,
"learning_rate": 2.8619197482297406e-05,
"loss": 0.6878,
"step": 1370
},
{
"epoch": 2.44,
"grad_norm": 2.7051727771759033,
"learning_rate": 2.8422501966955157e-05,
"loss": 0.6996,
"step": 1380
},
{
"epoch": 2.46,
"grad_norm": 0.7471837401390076,
"learning_rate": 2.822580645161291e-05,
"loss": 0.706,
"step": 1390
},
{
"epoch": 2.48,
"grad_norm": 2.322110176086426,
"learning_rate": 2.8029110936270653e-05,
"loss": 0.9161,
"step": 1400
},
{
"epoch": 2.49,
"grad_norm": 0.7188151478767395,
"learning_rate": 2.7832415420928404e-05,
"loss": 0.6905,
"step": 1410
},
{
"epoch": 2.51,
"grad_norm": 10.193795204162598,
"learning_rate": 2.7635719905586155e-05,
"loss": 0.7407,
"step": 1420
},
{
"epoch": 2.53,
"grad_norm": 0.9392734169960022,
"learning_rate": 2.7439024390243906e-05,
"loss": 0.6731,
"step": 1430
},
{
"epoch": 2.55,
"grad_norm": 0.5180323719978333,
"learning_rate": 2.724232887490165e-05,
"loss": 0.5429,
"step": 1440
},
{
"epoch": 2.56,
"grad_norm": 0.4639306664466858,
"learning_rate": 2.7045633359559402e-05,
"loss": 0.7713,
"step": 1450
},
{
"epoch": 2.58,
"grad_norm": 3.5896573066711426,
"learning_rate": 2.6848937844217153e-05,
"loss": 0.7773,
"step": 1460
},
{
"epoch": 2.6,
"grad_norm": 0.7149112820625305,
"learning_rate": 2.6652242328874904e-05,
"loss": 0.6672,
"step": 1470
},
{
"epoch": 2.62,
"grad_norm": 0.8536508679389954,
"learning_rate": 2.645554681353265e-05,
"loss": 0.7086,
"step": 1480
},
{
"epoch": 2.63,
"grad_norm": 1.054091215133667,
"learning_rate": 2.62588512981904e-05,
"loss": 0.6519,
"step": 1490
},
{
"epoch": 2.65,
"grad_norm": 0.39519792795181274,
"learning_rate": 2.606215578284815e-05,
"loss": 0.6434,
"step": 1500
},
{
"epoch": 2.67,
"grad_norm": 0.4575929343700409,
"learning_rate": 2.5865460267505902e-05,
"loss": 0.6553,
"step": 1510
},
{
"epoch": 2.69,
"grad_norm": 0.534916877746582,
"learning_rate": 2.5668764752163653e-05,
"loss": 0.7915,
"step": 1520
},
{
"epoch": 2.71,
"grad_norm": 0.7122427821159363,
"learning_rate": 2.54720692368214e-05,
"loss": 0.6913,
"step": 1530
},
{
"epoch": 2.72,
"grad_norm": 0.49569156765937805,
"learning_rate": 2.5275373721479152e-05,
"loss": 0.6085,
"step": 1540
},
{
"epoch": 2.74,
"grad_norm": 0.466847687959671,
"learning_rate": 2.5078678206136903e-05,
"loss": 0.6222,
"step": 1550
},
{
"epoch": 2.76,
"grad_norm": 0.34097394347190857,
"learning_rate": 2.488198269079465e-05,
"loss": 0.6883,
"step": 1560
},
{
"epoch": 2.78,
"grad_norm": 1.601562261581421,
"learning_rate": 2.4685287175452402e-05,
"loss": 0.6904,
"step": 1570
},
{
"epoch": 2.79,
"grad_norm": 0.5125943422317505,
"learning_rate": 2.448859166011015e-05,
"loss": 0.6484,
"step": 1580
},
{
"epoch": 2.81,
"grad_norm": 0.4327985644340515,
"learning_rate": 2.42918961447679e-05,
"loss": 0.7579,
"step": 1590
},
{
"epoch": 2.83,
"grad_norm": 0.4168800115585327,
"learning_rate": 2.4095200629425652e-05,
"loss": 0.7012,
"step": 1600
},
{
"epoch": 2.85,
"grad_norm": 2.620213270187378,
"learning_rate": 2.38985051140834e-05,
"loss": 0.6107,
"step": 1610
},
{
"epoch": 2.86,
"grad_norm": 0.47252315282821655,
"learning_rate": 2.370180959874115e-05,
"loss": 0.63,
"step": 1620
},
{
"epoch": 2.88,
"grad_norm": 3.600597620010376,
"learning_rate": 2.35051140833989e-05,
"loss": 0.7014,
"step": 1630
},
{
"epoch": 2.9,
"grad_norm": 0.9168177843093872,
"learning_rate": 2.330841856805665e-05,
"loss": 0.6946,
"step": 1640
},
{
"epoch": 2.92,
"grad_norm": 2.111616611480713,
"learning_rate": 2.3111723052714398e-05,
"loss": 0.5557,
"step": 1650
},
{
"epoch": 2.94,
"grad_norm": 0.4891892671585083,
"learning_rate": 2.291502753737215e-05,
"loss": 0.6718,
"step": 1660
},
{
"epoch": 2.95,
"grad_norm": 1.4157081842422485,
"learning_rate": 2.2718332022029897e-05,
"loss": 0.7549,
"step": 1670
},
{
"epoch": 2.97,
"grad_norm": 0.3661864399909973,
"learning_rate": 2.2521636506687648e-05,
"loss": 0.8072,
"step": 1680
},
{
"epoch": 2.99,
"grad_norm": 2.6403448581695557,
"learning_rate": 2.2324940991345396e-05,
"loss": 0.6657,
"step": 1690
},
{
"epoch": 3.0,
"eval_accuracy": 0.9866785079928952,
"eval_loss": 0.07058515399694443,
"eval_runtime": 87.367,
"eval_samples_per_second": 25.776,
"eval_steps_per_second": 3.228,
"step": 1696
},
{
"epoch": 3.01,
"grad_norm": 0.5366094708442688,
"learning_rate": 2.2128245476003147e-05,
"loss": 0.7239,
"step": 1700
},
{
"epoch": 3.02,
"grad_norm": 0.5593940019607544,
"learning_rate": 2.1931549960660898e-05,
"loss": 0.8491,
"step": 1710
},
{
"epoch": 3.04,
"grad_norm": 0.2935149371623993,
"learning_rate": 2.1734854445318646e-05,
"loss": 0.6358,
"step": 1720
},
{
"epoch": 3.06,
"grad_norm": 0.3379581570625305,
"learning_rate": 2.1538158929976397e-05,
"loss": 0.6124,
"step": 1730
},
{
"epoch": 3.08,
"grad_norm": 0.6903249621391296,
"learning_rate": 2.134146341463415e-05,
"loss": 0.7021,
"step": 1740
},
{
"epoch": 3.09,
"grad_norm": 0.5648875832557678,
"learning_rate": 2.11447678992919e-05,
"loss": 0.7315,
"step": 1750
},
{
"epoch": 3.11,
"grad_norm": 3.177462100982666,
"learning_rate": 2.0948072383949647e-05,
"loss": 0.7836,
"step": 1760
},
{
"epoch": 3.13,
"grad_norm": 2.3751437664031982,
"learning_rate": 2.07513768686074e-05,
"loss": 0.5984,
"step": 1770
},
{
"epoch": 3.15,
"grad_norm": 0.39180535078048706,
"learning_rate": 2.0554681353265146e-05,
"loss": 0.6462,
"step": 1780
},
{
"epoch": 3.17,
"grad_norm": 0.7616192102432251,
"learning_rate": 2.0357985837922897e-05,
"loss": 0.648,
"step": 1790
},
{
"epoch": 3.18,
"grad_norm": 0.7866030931472778,
"learning_rate": 2.0161290322580645e-05,
"loss": 0.6992,
"step": 1800
},
{
"epoch": 3.2,
"grad_norm": 2.5894863605499268,
"learning_rate": 1.9964594807238396e-05,
"loss": 0.7294,
"step": 1810
},
{
"epoch": 3.22,
"grad_norm": 0.33011573553085327,
"learning_rate": 1.9767899291896147e-05,
"loss": 0.6715,
"step": 1820
},
{
"epoch": 3.24,
"grad_norm": 1.21015465259552,
"learning_rate": 1.9571203776553895e-05,
"loss": 0.6617,
"step": 1830
},
{
"epoch": 3.25,
"grad_norm": 0.3521706461906433,
"learning_rate": 1.9374508261211646e-05,
"loss": 0.7361,
"step": 1840
},
{
"epoch": 3.27,
"grad_norm": 1.986220359802246,
"learning_rate": 1.9177812745869394e-05,
"loss": 0.782,
"step": 1850
},
{
"epoch": 3.29,
"grad_norm": 0.6578150391578674,
"learning_rate": 1.8981117230527145e-05,
"loss": 0.6532,
"step": 1860
},
{
"epoch": 3.31,
"grad_norm": 0.5501819252967834,
"learning_rate": 1.8784421715184893e-05,
"loss": 0.5756,
"step": 1870
},
{
"epoch": 3.32,
"grad_norm": 1.0883512496948242,
"learning_rate": 1.8587726199842644e-05,
"loss": 0.6532,
"step": 1880
},
{
"epoch": 3.34,
"grad_norm": 0.49157479405403137,
"learning_rate": 1.8391030684500392e-05,
"loss": 0.6951,
"step": 1890
},
{
"epoch": 3.36,
"grad_norm": 0.6763016581535339,
"learning_rate": 1.8194335169158143e-05,
"loss": 0.5232,
"step": 1900
},
{
"epoch": 3.38,
"grad_norm": 1.0173105001449585,
"learning_rate": 1.799763965381589e-05,
"loss": 0.6729,
"step": 1910
},
{
"epoch": 3.4,
"grad_norm": 0.49198317527770996,
"learning_rate": 1.7800944138473642e-05,
"loss": 0.6914,
"step": 1920
},
{
"epoch": 3.41,
"grad_norm": 0.7755669355392456,
"learning_rate": 1.7604248623131393e-05,
"loss": 0.6728,
"step": 1930
},
{
"epoch": 3.43,
"grad_norm": 0.4817189574241638,
"learning_rate": 1.7407553107789144e-05,
"loss": 0.7147,
"step": 1940
},
{
"epoch": 3.45,
"grad_norm": 1.0511811971664429,
"learning_rate": 1.7210857592446896e-05,
"loss": 0.7447,
"step": 1950
},
{
"epoch": 3.47,
"grad_norm": 0.4621807038784027,
"learning_rate": 1.7014162077104643e-05,
"loss": 0.7905,
"step": 1960
},
{
"epoch": 3.48,
"grad_norm": 0.42619049549102783,
"learning_rate": 1.6817466561762395e-05,
"loss": 0.6953,
"step": 1970
},
{
"epoch": 3.5,
"grad_norm": 0.4883289933204651,
"learning_rate": 1.6620771046420142e-05,
"loss": 0.7246,
"step": 1980
},
{
"epoch": 3.52,
"grad_norm": 5.140800476074219,
"learning_rate": 1.6424075531077893e-05,
"loss": 0.677,
"step": 1990
},
{
"epoch": 3.54,
"grad_norm": 0.7420614361763,
"learning_rate": 1.622738001573564e-05,
"loss": 0.6223,
"step": 2000
},
{
"epoch": 3.55,
"grad_norm": 0.5618042945861816,
"learning_rate": 1.6030684500393392e-05,
"loss": 0.7706,
"step": 2010
},
{
"epoch": 3.57,
"grad_norm": 8.696101188659668,
"learning_rate": 1.583398898505114e-05,
"loss": 0.5982,
"step": 2020
},
{
"epoch": 3.59,
"grad_norm": 0.613085925579071,
"learning_rate": 1.563729346970889e-05,
"loss": 0.5607,
"step": 2030
},
{
"epoch": 3.61,
"grad_norm": 0.37136101722717285,
"learning_rate": 1.5440597954366642e-05,
"loss": 0.7066,
"step": 2040
},
{
"epoch": 3.63,
"grad_norm": 1.0370784997940063,
"learning_rate": 1.524390243902439e-05,
"loss": 0.6117,
"step": 2050
},
{
"epoch": 3.64,
"grad_norm": 3.805783748626709,
"learning_rate": 1.5047206923682141e-05,
"loss": 0.7247,
"step": 2060
},
{
"epoch": 3.66,
"grad_norm": 0.3479630947113037,
"learning_rate": 1.485051140833989e-05,
"loss": 0.7094,
"step": 2070
},
{
"epoch": 3.68,
"grad_norm": 0.27747422456741333,
"learning_rate": 1.465381589299764e-05,
"loss": 0.6548,
"step": 2080
},
{
"epoch": 3.7,
"grad_norm": 0.4813224971294403,
"learning_rate": 1.445712037765539e-05,
"loss": 0.5747,
"step": 2090
},
{
"epoch": 3.71,
"grad_norm": 1.6526601314544678,
"learning_rate": 1.4260424862313141e-05,
"loss": 0.8374,
"step": 2100
},
{
"epoch": 3.73,
"grad_norm": 0.6736083626747131,
"learning_rate": 1.4063729346970889e-05,
"loss": 0.7677,
"step": 2110
},
{
"epoch": 3.75,
"grad_norm": 1.8762106895446777,
"learning_rate": 1.386703383162864e-05,
"loss": 0.5897,
"step": 2120
},
{
"epoch": 3.77,
"grad_norm": 0.622199296951294,
"learning_rate": 1.3670338316286388e-05,
"loss": 0.7407,
"step": 2130
},
{
"epoch": 3.78,
"grad_norm": 0.3692738115787506,
"learning_rate": 1.3473642800944139e-05,
"loss": 0.8474,
"step": 2140
},
{
"epoch": 3.8,
"grad_norm": 0.5641698241233826,
"learning_rate": 1.327694728560189e-05,
"loss": 0.6674,
"step": 2150
},
{
"epoch": 3.82,
"grad_norm": 0.7227357029914856,
"learning_rate": 1.3080251770259638e-05,
"loss": 0.6852,
"step": 2160
},
{
"epoch": 3.84,
"grad_norm": 0.80736243724823,
"learning_rate": 1.2883556254917389e-05,
"loss": 0.7686,
"step": 2170
},
{
"epoch": 3.85,
"grad_norm": 0.7187584042549133,
"learning_rate": 1.2686860739575138e-05,
"loss": 0.7859,
"step": 2180
},
{
"epoch": 3.87,
"grad_norm": 0.6288447976112366,
"learning_rate": 1.249016522423289e-05,
"loss": 0.683,
"step": 2190
},
{
"epoch": 3.89,
"grad_norm": 0.6657484769821167,
"learning_rate": 1.2293469708890639e-05,
"loss": 0.6807,
"step": 2200
},
{
"epoch": 3.91,
"grad_norm": 0.3658302128314972,
"learning_rate": 1.2096774193548388e-05,
"loss": 0.6274,
"step": 2210
},
{
"epoch": 3.93,
"grad_norm": 0.7069531083106995,
"learning_rate": 1.1900078678206138e-05,
"loss": 0.7722,
"step": 2220
},
{
"epoch": 3.94,
"grad_norm": 0.2362523227930069,
"learning_rate": 1.1703383162863887e-05,
"loss": 0.7107,
"step": 2230
},
{
"epoch": 3.96,
"grad_norm": 0.49751946330070496,
"learning_rate": 1.1506687647521637e-05,
"loss": 0.6972,
"step": 2240
},
{
"epoch": 3.98,
"grad_norm": 0.4281612038612366,
"learning_rate": 1.1309992132179386e-05,
"loss": 0.6935,
"step": 2250
},
{
"epoch": 4.0,
"grad_norm": 0.47943803668022156,
"learning_rate": 1.1113296616837136e-05,
"loss": 0.7475,
"step": 2260
},
{
"epoch": 4.0,
"eval_accuracy": 0.9866785079928952,
"eval_loss": 0.06677553802728653,
"eval_runtime": 87.1331,
"eval_samples_per_second": 25.846,
"eval_steps_per_second": 3.236,
"step": 2262
},
{
"epoch": 4.01,
"grad_norm": 1.7420241832733154,
"learning_rate": 1.0916601101494885e-05,
"loss": 0.7495,
"step": 2270
},
{
"epoch": 4.03,
"grad_norm": 0.43095558881759644,
"learning_rate": 1.0719905586152636e-05,
"loss": 0.6875,
"step": 2280
},
{
"epoch": 4.05,
"grad_norm": 4.584779739379883,
"learning_rate": 1.0523210070810386e-05,
"loss": 0.8217,
"step": 2290
},
{
"epoch": 4.07,
"grad_norm": 0.3836701810359955,
"learning_rate": 1.0326514555468137e-05,
"loss": 0.5724,
"step": 2300
},
{
"epoch": 4.08,
"grad_norm": 0.8410959243774414,
"learning_rate": 1.0129819040125886e-05,
"loss": 0.6068,
"step": 2310
},
{
"epoch": 4.1,
"grad_norm": 0.2401944100856781,
"learning_rate": 9.933123524783636e-06,
"loss": 0.6172,
"step": 2320
},
{
"epoch": 4.12,
"grad_norm": 1.9553325176239014,
"learning_rate": 9.736428009441385e-06,
"loss": 0.5384,
"step": 2330
},
{
"epoch": 4.14,
"grad_norm": 0.7399584650993347,
"learning_rate": 9.539732494099135e-06,
"loss": 0.6562,
"step": 2340
},
{
"epoch": 4.16,
"grad_norm": 0.6674773097038269,
"learning_rate": 9.343036978756884e-06,
"loss": 0.736,
"step": 2350
},
{
"epoch": 4.17,
"grad_norm": 0.35315585136413574,
"learning_rate": 9.146341463414634e-06,
"loss": 0.6334,
"step": 2360
},
{
"epoch": 4.19,
"grad_norm": 0.8621203303337097,
"learning_rate": 8.949645948072383e-06,
"loss": 0.7214,
"step": 2370
},
{
"epoch": 4.21,
"grad_norm": 6.545629501342773,
"learning_rate": 8.752950432730134e-06,
"loss": 0.5391,
"step": 2380
},
{
"epoch": 4.23,
"grad_norm": 0.5393751859664917,
"learning_rate": 8.556254917387884e-06,
"loss": 0.6851,
"step": 2390
},
{
"epoch": 4.24,
"grad_norm": 0.8293077945709229,
"learning_rate": 8.359559402045635e-06,
"loss": 0.791,
"step": 2400
},
{
"epoch": 4.26,
"grad_norm": 0.44837647676467896,
"learning_rate": 8.162863886703385e-06,
"loss": 0.5622,
"step": 2410
},
{
"epoch": 4.28,
"grad_norm": 0.5060232281684875,
"learning_rate": 7.966168371361134e-06,
"loss": 0.7254,
"step": 2420
},
{
"epoch": 4.3,
"grad_norm": 0.647249698638916,
"learning_rate": 7.769472856018883e-06,
"loss": 0.5765,
"step": 2430
},
{
"epoch": 4.31,
"grad_norm": 0.7065703868865967,
"learning_rate": 7.572777340676633e-06,
"loss": 0.7337,
"step": 2440
},
{
"epoch": 4.33,
"grad_norm": 0.6291216015815735,
"learning_rate": 7.376081825334382e-06,
"loss": 0.649,
"step": 2450
},
{
"epoch": 4.35,
"grad_norm": 0.6098889708518982,
"learning_rate": 7.179386309992133e-06,
"loss": 0.6712,
"step": 2460
},
{
"epoch": 4.37,
"grad_norm": 0.49595367908477783,
"learning_rate": 6.982690794649882e-06,
"loss": 0.5932,
"step": 2470
},
{
"epoch": 4.39,
"grad_norm": 0.34834733605384827,
"learning_rate": 6.785995279307632e-06,
"loss": 0.5702,
"step": 2480
},
{
"epoch": 4.4,
"grad_norm": 0.30979278683662415,
"learning_rate": 6.589299763965381e-06,
"loss": 0.6423,
"step": 2490
},
{
"epoch": 4.42,
"grad_norm": 1.0691258907318115,
"learning_rate": 6.392604248623131e-06,
"loss": 0.6254,
"step": 2500
},
{
"epoch": 4.44,
"grad_norm": 0.6674053072929382,
"learning_rate": 6.195908733280882e-06,
"loss": 0.6869,
"step": 2510
},
{
"epoch": 4.46,
"grad_norm": 0.4127647280693054,
"learning_rate": 5.999213217938631e-06,
"loss": 0.6614,
"step": 2520
},
{
"epoch": 4.47,
"grad_norm": 1.5234615802764893,
"learning_rate": 5.802517702596381e-06,
"loss": 0.716,
"step": 2530
},
{
"epoch": 4.49,
"grad_norm": 0.3525974750518799,
"learning_rate": 5.605822187254131e-06,
"loss": 0.8085,
"step": 2540
},
{
"epoch": 4.51,
"grad_norm": 0.9571001529693604,
"learning_rate": 5.40912667191188e-06,
"loss": 0.6391,
"step": 2550
},
{
"epoch": 4.53,
"grad_norm": 0.5705153942108154,
"learning_rate": 5.212431156569631e-06,
"loss": 0.5403,
"step": 2560
},
{
"epoch": 4.54,
"grad_norm": 0.6829048991203308,
"learning_rate": 5.01573564122738e-06,
"loss": 0.6538,
"step": 2570
},
{
"epoch": 4.56,
"grad_norm": 1.4802197217941284,
"learning_rate": 4.81904012588513e-06,
"loss": 0.7467,
"step": 2580
},
{
"epoch": 4.58,
"grad_norm": 1.3378098011016846,
"learning_rate": 4.62234461054288e-06,
"loss": 0.7882,
"step": 2590
},
{
"epoch": 4.6,
"grad_norm": 0.3483523428440094,
"learning_rate": 4.425649095200629e-06,
"loss": 0.6253,
"step": 2600
},
{
"epoch": 4.62,
"grad_norm": 0.7070222496986389,
"learning_rate": 4.22895357985838e-06,
"loss": 0.8082,
"step": 2610
},
{
"epoch": 4.63,
"grad_norm": 0.2973116934299469,
"learning_rate": 4.032258064516129e-06,
"loss": 0.7245,
"step": 2620
},
{
"epoch": 4.65,
"grad_norm": 0.6432918310165405,
"learning_rate": 3.835562549173879e-06,
"loss": 0.5764,
"step": 2630
},
{
"epoch": 4.67,
"grad_norm": 0.8339099287986755,
"learning_rate": 3.638867033831629e-06,
"loss": 0.607,
"step": 2640
},
{
"epoch": 4.69,
"grad_norm": 0.5163792371749878,
"learning_rate": 3.442171518489379e-06,
"loss": 0.6602,
"step": 2650
},
{
"epoch": 4.7,
"grad_norm": 1.1758856773376465,
"learning_rate": 3.2454760031471283e-06,
"loss": 0.8349,
"step": 2660
},
{
"epoch": 4.72,
"grad_norm": 0.5800232887268066,
"learning_rate": 3.0487804878048782e-06,
"loss": 0.6944,
"step": 2670
},
{
"epoch": 4.74,
"grad_norm": 0.342181921005249,
"learning_rate": 2.852084972462628e-06,
"loss": 0.6855,
"step": 2680
},
{
"epoch": 4.76,
"grad_norm": 0.5758516788482666,
"learning_rate": 2.655389457120378e-06,
"loss": 0.5884,
"step": 2690
},
{
"epoch": 4.77,
"grad_norm": 0.42382463812828064,
"learning_rate": 2.4586939417781275e-06,
"loss": 0.7499,
"step": 2700
},
{
"epoch": 4.79,
"grad_norm": 0.5452685952186584,
"learning_rate": 2.2619984264358773e-06,
"loss": 0.6471,
"step": 2710
},
{
"epoch": 4.81,
"grad_norm": 1.8580130338668823,
"learning_rate": 2.0653029110936272e-06,
"loss": 0.6492,
"step": 2720
},
{
"epoch": 4.83,
"grad_norm": 1.0084635019302368,
"learning_rate": 1.868607395751377e-06,
"loss": 0.5958,
"step": 2730
},
{
"epoch": 4.85,
"grad_norm": 0.44602665305137634,
"learning_rate": 1.6719118804091268e-06,
"loss": 0.6962,
"step": 2740
},
{
"epoch": 4.86,
"grad_norm": 0.45221227407455444,
"learning_rate": 1.4752163650668765e-06,
"loss": 0.6997,
"step": 2750
},
{
"epoch": 4.88,
"grad_norm": 1.639831304550171,
"learning_rate": 1.2785208497246264e-06,
"loss": 0.6848,
"step": 2760
},
{
"epoch": 4.9,
"grad_norm": 0.6076949238777161,
"learning_rate": 1.0818253343823763e-06,
"loss": 0.6033,
"step": 2770
},
{
"epoch": 4.92,
"grad_norm": 0.46627792716026306,
"learning_rate": 8.85129819040126e-07,
"loss": 0.6447,
"step": 2780
},
{
"epoch": 4.93,
"grad_norm": 1.5518214702606201,
"learning_rate": 6.884343036978757e-07,
"loss": 0.7177,
"step": 2790
},
{
"epoch": 4.95,
"grad_norm": 0.4077186584472656,
"learning_rate": 4.917387883556255e-07,
"loss": 0.6174,
"step": 2800
},
{
"epoch": 4.97,
"grad_norm": 0.6033698916435242,
"learning_rate": 2.9504327301337533e-07,
"loss": 0.5902,
"step": 2810
},
{
"epoch": 4.99,
"grad_norm": 0.20630072057247162,
"learning_rate": 9.834775767112511e-08,
"loss": 0.6021,
"step": 2820
},
{
"epoch": 5.0,
"eval_accuracy": 0.9853463587921847,
"eval_loss": 0.0637223869562149,
"eval_runtime": 86.6358,
"eval_samples_per_second": 25.994,
"eval_steps_per_second": 3.255,
"step": 2825
},
{
"epoch": 5.0,
"step": 2825,
"total_flos": 1.6016486531331686e+18,
"train_loss": 0.9856665741236864,
"train_runtime": 4486.694,
"train_samples_per_second": 20.16,
"train_steps_per_second": 0.63
}
],
"logging_steps": 10,
"max_steps": 2825,
"num_input_tokens_seen": 0,
"num_train_epochs": 5,
"save_steps": 500,
"total_flos": 1.6016486531331686e+18,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}