Go4miii's picture
commit from root
30fe73d
{
"best_metric": 0.41885045170783997,
"best_model_checkpoint": "output/Baichuan-13B-Chat_lora_wqs_nlp/checkpoint-2400",
"epoch": 1.9992665933259994,
"global_step": 2726,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"learning_rate": 4.999865525734509e-05,
"loss": 1.5452,
"step": 10
},
{
"epoch": 0.01,
"learning_rate": 4.9994621174046976e-05,
"loss": 1.3291,
"step": 20
},
{
"epoch": 0.02,
"learning_rate": 4.9986985231938546e-05,
"loss": 1.1048,
"step": 30
},
{
"epoch": 0.03,
"learning_rate": 4.99760306731191e-05,
"loss": 0.9155,
"step": 40
},
{
"epoch": 0.04,
"learning_rate": 4.9961758952505326e-05,
"loss": 0.8856,
"step": 50
},
{
"epoch": 0.04,
"learning_rate": 4.9944171965578836e-05,
"loss": 0.8052,
"step": 60
},
{
"epoch": 0.05,
"learning_rate": 4.992327204813435e-05,
"loss": 0.7332,
"step": 70
},
{
"epoch": 0.06,
"learning_rate": 4.989906197596955e-05,
"loss": 0.6982,
"step": 80
},
{
"epoch": 0.07,
"learning_rate": 4.987154496451635e-05,
"loss": 0.6811,
"step": 90
},
{
"epoch": 0.07,
"learning_rate": 4.984072466841389e-05,
"loss": 0.6323,
"step": 100
},
{
"epoch": 0.08,
"learning_rate": 4.981016546765289e-05,
"loss": 0.6234,
"step": 110
},
{
"epoch": 0.09,
"learning_rate": 4.977308057009269e-05,
"loss": 0.6125,
"step": 120
},
{
"epoch": 0.1,
"learning_rate": 4.97327054653146e-05,
"loss": 0.5977,
"step": 130
},
{
"epoch": 0.1,
"learning_rate": 4.968904551569013e-05,
"loss": 0.6179,
"step": 140
},
{
"epoch": 0.11,
"learning_rate": 4.9642106519863544e-05,
"loss": 0.5879,
"step": 150
},
{
"epoch": 0.12,
"learning_rate": 4.959189471198171e-05,
"loss": 0.5472,
"step": 160
},
{
"epoch": 0.12,
"learning_rate": 4.953841676086613e-05,
"loss": 0.602,
"step": 170
},
{
"epoch": 0.13,
"learning_rate": 4.9481679769127275e-05,
"loss": 0.5773,
"step": 180
},
{
"epoch": 0.14,
"learning_rate": 4.9421691272221167e-05,
"loss": 0.5615,
"step": 190
},
{
"epoch": 0.15,
"learning_rate": 4.935845923744865e-05,
"loss": 0.5704,
"step": 200
},
{
"epoch": 0.15,
"eval_loss": 0.5564362406730652,
"eval_runtime": 72.9469,
"eval_samples_per_second": 15.107,
"eval_steps_per_second": 1.522,
"step": 200
},
{
"epoch": 0.15,
"learning_rate": 4.9291992062897183e-05,
"loss": 0.5803,
"step": 210
},
{
"epoch": 0.16,
"learning_rate": 4.922229857632545e-05,
"loss": 0.5655,
"step": 220
},
{
"epoch": 0.17,
"learning_rate": 4.9149388033990966e-05,
"loss": 0.5769,
"step": 230
},
{
"epoch": 0.18,
"learning_rate": 4.9073270119420635e-05,
"loss": 0.5862,
"step": 240
},
{
"epoch": 0.18,
"learning_rate": 4.899395494212471e-05,
"loss": 0.5506,
"step": 250
},
{
"epoch": 0.19,
"learning_rate": 4.891145303625408e-05,
"loss": 0.5183,
"step": 260
},
{
"epoch": 0.2,
"learning_rate": 4.882577535920121e-05,
"loss": 0.575,
"step": 270
},
{
"epoch": 0.21,
"learning_rate": 4.8736933290144815e-05,
"loss": 0.5359,
"step": 280
},
{
"epoch": 0.21,
"learning_rate": 4.8644938628538606e-05,
"loss": 0.5302,
"step": 290
},
{
"epoch": 0.22,
"learning_rate": 4.8549803592544076e-05,
"loss": 0.5399,
"step": 300
},
{
"epoch": 0.23,
"learning_rate": 4.845154081740783e-05,
"loss": 0.5379,
"step": 310
},
{
"epoch": 0.23,
"learning_rate": 4.835016335378343e-05,
"loss": 0.5299,
"step": 320
},
{
"epoch": 0.24,
"learning_rate": 4.8245684665998073e-05,
"loss": 0.535,
"step": 330
},
{
"epoch": 0.25,
"learning_rate": 4.813811863026436e-05,
"loss": 0.5145,
"step": 340
},
{
"epoch": 0.26,
"learning_rate": 4.802747953283732e-05,
"loss": 0.5126,
"step": 350
},
{
"epoch": 0.26,
"learning_rate": 4.791378206811704e-05,
"loss": 0.5012,
"step": 360
},
{
"epoch": 0.27,
"learning_rate": 4.7797041336696995e-05,
"loss": 0.5373,
"step": 370
},
{
"epoch": 0.28,
"learning_rate": 4.767727284335852e-05,
"loss": 0.4778,
"step": 380
},
{
"epoch": 0.29,
"learning_rate": 4.755449249501155e-05,
"loss": 0.4894,
"step": 390
},
{
"epoch": 0.29,
"learning_rate": 4.7428716598581934e-05,
"loss": 0.4986,
"step": 400
},
{
"epoch": 0.29,
"eval_loss": 0.4967592656612396,
"eval_runtime": 70.0948,
"eval_samples_per_second": 15.722,
"eval_steps_per_second": 1.584,
"step": 400
},
{
"epoch": 0.3,
"learning_rate": 4.729996185884571e-05,
"loss": 0.4928,
"step": 410
},
{
"epoch": 0.31,
"learning_rate": 4.716824537621042e-05,
"loss": 0.4807,
"step": 420
},
{
"epoch": 0.32,
"learning_rate": 4.703358464444397e-05,
"loss": 0.4734,
"step": 430
},
{
"epoch": 0.32,
"learning_rate": 4.689599754835122e-05,
"loss": 0.4603,
"step": 440
},
{
"epoch": 0.33,
"learning_rate": 4.6755502361398616e-05,
"loss": 0.4661,
"step": 450
},
{
"epoch": 0.34,
"learning_rate": 4.6612117743287234e-05,
"loss": 0.5072,
"step": 460
},
{
"epoch": 0.34,
"learning_rate": 4.646586273747452e-05,
"loss": 0.5056,
"step": 470
},
{
"epoch": 0.35,
"learning_rate": 4.631675676864503e-05,
"loss": 0.4391,
"step": 480
},
{
"epoch": 0.36,
"learning_rate": 4.6164819640130595e-05,
"loss": 0.483,
"step": 490
},
{
"epoch": 0.37,
"learning_rate": 4.601007153128014e-05,
"loss": 0.4774,
"step": 500
},
{
"epoch": 0.37,
"learning_rate": 4.5852532994779606e-05,
"loss": 0.4848,
"step": 510
},
{
"epoch": 0.38,
"learning_rate": 4.5692224953922266e-05,
"loss": 0.5099,
"step": 520
},
{
"epoch": 0.39,
"learning_rate": 4.5529168699829805e-05,
"loss": 0.4851,
"step": 530
},
{
"epoch": 0.4,
"learning_rate": 4.536338588862459e-05,
"loss": 0.498,
"step": 540
},
{
"epoch": 0.4,
"learning_rate": 4.519489853855341e-05,
"loss": 0.5398,
"step": 550
},
{
"epoch": 0.41,
"learning_rate": 4.50237290270632e-05,
"loss": 0.4566,
"step": 560
},
{
"epoch": 0.42,
"learning_rate": 4.484990008782893e-05,
"loss": 0.4942,
"step": 570
},
{
"epoch": 0.43,
"learning_rate": 4.467343480773433e-05,
"loss": 0.4625,
"step": 580
},
{
"epoch": 0.43,
"learning_rate": 4.44943566238056e-05,
"loss": 0.4703,
"step": 590
},
{
"epoch": 0.44,
"learning_rate": 4.431268932009865e-05,
"loss": 0.464,
"step": 600
},
{
"epoch": 0.44,
"eval_loss": 0.46821752190589905,
"eval_runtime": 69.9505,
"eval_samples_per_second": 15.754,
"eval_steps_per_second": 1.587,
"step": 600
},
{
"epoch": 0.45,
"learning_rate": 4.412845702454024e-05,
"loss": 0.4739,
"step": 610
},
{
"epoch": 0.45,
"learning_rate": 4.394168420572349e-05,
"loss": 0.4529,
"step": 620
},
{
"epoch": 0.46,
"learning_rate": 4.3752395669658086e-05,
"loss": 0.4433,
"step": 630
},
{
"epoch": 0.47,
"learning_rate": 4.356061655647571e-05,
"loss": 0.459,
"step": 640
},
{
"epoch": 0.48,
"learning_rate": 4.336637233709107e-05,
"loss": 0.4877,
"step": 650
},
{
"epoch": 0.48,
"learning_rate": 4.316968880981903e-05,
"loss": 0.4676,
"step": 660
},
{
"epoch": 0.49,
"learning_rate": 4.2970592096948236e-05,
"loss": 0.4782,
"step": 670
},
{
"epoch": 0.5,
"learning_rate": 4.276910864127168e-05,
"loss": 0.4778,
"step": 680
},
{
"epoch": 0.51,
"learning_rate": 4.25652652025748e-05,
"loss": 0.454,
"step": 690
},
{
"epoch": 0.51,
"learning_rate": 4.235908885408133e-05,
"loss": 0.4399,
"step": 700
},
{
"epoch": 0.52,
"learning_rate": 4.215060697885767e-05,
"loss": 0.5106,
"step": 710
},
{
"epoch": 0.53,
"learning_rate": 4.1939847266176e-05,
"loss": 0.4683,
"step": 720
},
{
"epoch": 0.54,
"learning_rate": 4.172683770783677e-05,
"loss": 0.4896,
"step": 730
},
{
"epoch": 0.54,
"learning_rate": 4.1511606594451016e-05,
"loss": 0.4633,
"step": 740
},
{
"epoch": 0.55,
"learning_rate": 4.1294182511682946e-05,
"loss": 0.4486,
"step": 750
},
{
"epoch": 0.56,
"learning_rate": 4.1074594336453384e-05,
"loss": 0.4706,
"step": 760
},
{
"epoch": 0.56,
"learning_rate": 4.085287123310455e-05,
"loss": 0.4797,
"step": 770
},
{
"epoch": 0.57,
"learning_rate": 4.062904264952657e-05,
"loss": 0.4251,
"step": 780
},
{
"epoch": 0.58,
"learning_rate": 4.0403138313246435e-05,
"loss": 0.4687,
"step": 790
},
{
"epoch": 0.59,
"learning_rate": 4.017518822747976e-05,
"loss": 0.48,
"step": 800
},
{
"epoch": 0.59,
"eval_loss": 0.4554018974304199,
"eval_runtime": 70.2392,
"eval_samples_per_second": 15.689,
"eval_steps_per_second": 1.58,
"step": 800
},
{
"epoch": 0.59,
"learning_rate": 3.994522266714594e-05,
"loss": 0.4662,
"step": 810
},
{
"epoch": 0.6,
"learning_rate": 3.9713272174847246e-05,
"loss": 0.4837,
"step": 820
},
{
"epoch": 0.61,
"learning_rate": 3.947936755681229e-05,
"loss": 0.5014,
"step": 830
},
{
"epoch": 0.62,
"learning_rate": 3.92435398788046e-05,
"loss": 0.4764,
"step": 840
},
{
"epoch": 0.62,
"learning_rate": 3.9005820461996604e-05,
"loss": 0.4587,
"step": 850
},
{
"epoch": 0.63,
"learning_rate": 3.876624087880979e-05,
"loss": 0.484,
"step": 860
},
{
"epoch": 0.64,
"learning_rate": 3.85248329487214e-05,
"loss": 0.4285,
"step": 870
},
{
"epoch": 0.65,
"learning_rate": 3.828162873403843e-05,
"loss": 0.4742,
"step": 880
},
{
"epoch": 0.65,
"learning_rate": 3.803666053563926e-05,
"loss": 0.4533,
"step": 890
},
{
"epoch": 0.66,
"learning_rate": 3.778996088868365e-05,
"loss": 0.4395,
"step": 900
},
{
"epoch": 0.67,
"learning_rate": 3.754156255829167e-05,
"loss": 0.4519,
"step": 910
},
{
"epoch": 0.67,
"learning_rate": 3.7291498535191996e-05,
"loss": 0.4684,
"step": 920
},
{
"epoch": 0.68,
"learning_rate": 3.703980203134029e-05,
"loss": 0.4259,
"step": 930
},
{
"epoch": 0.69,
"learning_rate": 3.678650647550822e-05,
"loss": 0.4576,
"step": 940
},
{
"epoch": 0.7,
"learning_rate": 3.6531645508843636e-05,
"loss": 0.446,
"step": 950
},
{
"epoch": 0.7,
"learning_rate": 3.6275252980402544e-05,
"loss": 0.4439,
"step": 960
},
{
"epoch": 0.71,
"learning_rate": 3.601736294265354e-05,
"loss": 0.476,
"step": 970
},
{
"epoch": 0.72,
"learning_rate": 3.5758009646955115e-05,
"loss": 0.4153,
"step": 980
},
{
"epoch": 0.73,
"learning_rate": 3.5497227539006614e-05,
"loss": 0.4444,
"step": 990
},
{
"epoch": 0.73,
"learning_rate": 3.523505125427341e-05,
"loss": 0.4395,
"step": 1000
},
{
"epoch": 0.73,
"eval_loss": 0.44357746839523315,
"eval_runtime": 70.1513,
"eval_samples_per_second": 15.709,
"eval_steps_per_second": 1.582,
"step": 1000
},
{
"epoch": 0.74,
"learning_rate": 3.497151561338678e-05,
"loss": 0.4387,
"step": 1010
},
{
"epoch": 0.75,
"learning_rate": 3.470665561751928e-05,
"loss": 0.4559,
"step": 1020
},
{
"epoch": 0.76,
"learning_rate": 3.444050644373611e-05,
"loss": 0.4349,
"step": 1030
},
{
"epoch": 0.76,
"learning_rate": 3.417310344032309e-05,
"loss": 0.4661,
"step": 1040
},
{
"epoch": 0.77,
"learning_rate": 3.390448212209191e-05,
"loss": 0.4945,
"step": 1050
},
{
"epoch": 0.78,
"learning_rate": 3.3634678165663325e-05,
"loss": 0.4399,
"step": 1060
},
{
"epoch": 0.78,
"learning_rate": 3.336372740472877e-05,
"loss": 0.4399,
"step": 1070
},
{
"epoch": 0.79,
"learning_rate": 3.309166582529114e-05,
"loss": 0.4251,
"step": 1080
},
{
"epoch": 0.8,
"learning_rate": 3.281852956088537e-05,
"loss": 0.4385,
"step": 1090
},
{
"epoch": 0.81,
"learning_rate": 3.254435488777941e-05,
"loss": 0.4566,
"step": 1100
},
{
"epoch": 0.81,
"learning_rate": 3.226917822015623e-05,
"loss": 0.4352,
"step": 1110
},
{
"epoch": 0.82,
"learning_rate": 3.199303610527749e-05,
"loss": 0.4617,
"step": 1120
},
{
"epoch": 0.83,
"learning_rate": 3.1715965218629595e-05,
"loss": 0.476,
"step": 1130
},
{
"epoch": 0.84,
"learning_rate": 3.143800235905268e-05,
"loss": 0.4322,
"step": 1140
},
{
"epoch": 0.84,
"learning_rate": 3.115918444385315e-05,
"loss": 0.4736,
"step": 1150
},
{
"epoch": 0.85,
"learning_rate": 3.0879548503900665e-05,
"loss": 0.4322,
"step": 1160
},
{
"epoch": 0.86,
"learning_rate": 3.0599131678709836e-05,
"loss": 0.4152,
"step": 1170
},
{
"epoch": 0.87,
"learning_rate": 3.031797121150764e-05,
"loss": 0.4187,
"step": 1180
},
{
"epoch": 0.87,
"learning_rate": 3.0036104444286954e-05,
"loss": 0.4331,
"step": 1190
},
{
"epoch": 0.88,
"learning_rate": 2.9753568812847065e-05,
"loss": 0.45,
"step": 1200
},
{
"epoch": 0.88,
"eval_loss": 0.43782830238342285,
"eval_runtime": 70.0939,
"eval_samples_per_second": 15.722,
"eval_steps_per_second": 1.584,
"step": 1200
},
{
"epoch": 0.89,
"learning_rate": 2.9470401841821686e-05,
"loss": 0.4457,
"step": 1210
},
{
"epoch": 0.89,
"learning_rate": 2.9186641139695108e-05,
"loss": 0.4667,
"step": 1220
},
{
"epoch": 0.9,
"learning_rate": 2.8902324393807333e-05,
"loss": 0.4516,
"step": 1230
},
{
"epoch": 0.91,
"learning_rate": 2.861748936534867e-05,
"loss": 0.4695,
"step": 1240
},
{
"epoch": 0.92,
"learning_rate": 2.8332173884344477e-05,
"loss": 0.45,
"step": 1250
},
{
"epoch": 0.92,
"learning_rate": 2.8046415844630857e-05,
"loss": 0.4246,
"step": 1260
},
{
"epoch": 0.93,
"learning_rate": 2.7760253198821822e-05,
"loss": 0.4449,
"step": 1270
},
{
"epoch": 0.94,
"learning_rate": 2.7473723953268687e-05,
"loss": 0.4217,
"step": 1280
},
{
"epoch": 0.95,
"learning_rate": 2.7186866163012232e-05,
"loss": 0.4495,
"step": 1290
},
{
"epoch": 0.95,
"learning_rate": 2.6899717926728535e-05,
"loss": 0.4523,
"step": 1300
},
{
"epoch": 0.96,
"learning_rate": 2.6612317381668915e-05,
"loss": 0.4523,
"step": 1310
},
{
"epoch": 0.97,
"learning_rate": 2.632470269859478e-05,
"loss": 0.45,
"step": 1320
},
{
"epoch": 0.98,
"learning_rate": 2.603691207670803e-05,
"loss": 0.4371,
"step": 1330
},
{
"epoch": 0.98,
"learning_rate": 2.5748983738577653e-05,
"loss": 0.4221,
"step": 1340
},
{
"epoch": 0.99,
"learning_rate": 2.5460955925063268e-05,
"loss": 0.4565,
"step": 1350
},
{
"epoch": 1.0,
"learning_rate": 2.5172866890236203e-05,
"loss": 0.3892,
"step": 1360
},
{
"epoch": 1.0,
"learning_rate": 2.48847548962988e-05,
"loss": 0.3973,
"step": 1370
},
{
"epoch": 1.01,
"learning_rate": 2.4596658208502713e-05,
"loss": 0.4151,
"step": 1380
},
{
"epoch": 1.02,
"learning_rate": 2.4308615090066735e-05,
"loss": 0.3839,
"step": 1390
},
{
"epoch": 1.03,
"learning_rate": 2.4020663797094864e-05,
"loss": 0.4062,
"step": 1400
},
{
"epoch": 1.03,
"eval_loss": 0.43284282088279724,
"eval_runtime": 70.1658,
"eval_samples_per_second": 15.706,
"eval_steps_per_second": 1.582,
"step": 1400
},
{
"epoch": 1.03,
"learning_rate": 2.373284257349544e-05,
"loss": 0.4003,
"step": 1410
},
{
"epoch": 1.04,
"learning_rate": 2.3445189645901806e-05,
"loss": 0.4447,
"step": 1420
},
{
"epoch": 1.05,
"learning_rate": 2.3157743218595247e-05,
"loss": 0.4565,
"step": 1430
},
{
"epoch": 1.06,
"learning_rate": 2.287054146843097e-05,
"loss": 0.4265,
"step": 1440
},
{
"epoch": 1.06,
"learning_rate": 2.2583622539767668e-05,
"loss": 0.4056,
"step": 1450
},
{
"epoch": 1.07,
"learning_rate": 2.2297024539401463e-05,
"loss": 0.4074,
"step": 1460
},
{
"epoch": 1.08,
"learning_rate": 2.2010785531504716e-05,
"loss": 0.4281,
"step": 1470
},
{
"epoch": 1.09,
"learning_rate": 2.172494353257066e-05,
"loss": 0.4663,
"step": 1480
},
{
"epoch": 1.09,
"learning_rate": 2.1439536506364274e-05,
"loss": 0.4281,
"step": 1490
},
{
"epoch": 1.1,
"learning_rate": 2.1154602358880122e-05,
"loss": 0.4111,
"step": 1500
},
{
"epoch": 1.11,
"learning_rate": 2.0870178933307948e-05,
"loss": 0.4187,
"step": 1510
},
{
"epoch": 1.11,
"learning_rate": 2.0586304005006585e-05,
"loss": 0.3964,
"step": 1520
},
{
"epoch": 1.12,
"learning_rate": 2.030301527648684e-05,
"loss": 0.4331,
"step": 1530
},
{
"epoch": 1.13,
"learning_rate": 2.0020350372404102e-05,
"loss": 0.4432,
"step": 1540
},
{
"epoch": 1.14,
"learning_rate": 1.9738346834561254e-05,
"loss": 0.4056,
"step": 1550
},
{
"epoch": 1.14,
"learning_rate": 1.945704211692262e-05,
"loss": 0.4354,
"step": 1560
},
{
"epoch": 1.15,
"learning_rate": 1.9176473580639538e-05,
"loss": 0.4309,
"step": 1570
},
{
"epoch": 1.16,
"learning_rate": 1.8896678489088304e-05,
"loss": 0.4058,
"step": 1580
},
{
"epoch": 1.17,
"learning_rate": 1.8617694002921064e-05,
"loss": 0.4319,
"step": 1590
},
{
"epoch": 1.17,
"learning_rate": 1.8339557175130383e-05,
"loss": 0.4267,
"step": 1600
},
{
"epoch": 1.17,
"eval_loss": 0.42894992232322693,
"eval_runtime": 70.251,
"eval_samples_per_second": 15.687,
"eval_steps_per_second": 1.58,
"step": 1600
},
{
"epoch": 1.18,
"learning_rate": 1.8062304946128073e-05,
"loss": 0.3921,
"step": 1610
},
{
"epoch": 1.19,
"learning_rate": 1.7785974138839018e-05,
"loss": 0.4206,
"step": 1620
},
{
"epoch": 1.2,
"learning_rate": 1.7510601453810594e-05,
"loss": 0.4356,
"step": 1630
},
{
"epoch": 1.2,
"learning_rate": 1.723622346433828e-05,
"loss": 0.4167,
"step": 1640
},
{
"epoch": 1.21,
"learning_rate": 1.6962876611608262e-05,
"loss": 0.4233,
"step": 1650
},
{
"epoch": 1.22,
"learning_rate": 1.6690597199857523e-05,
"loss": 0.4176,
"step": 1660
},
{
"epoch": 1.22,
"learning_rate": 1.6419421391552142e-05,
"loss": 0.4241,
"step": 1670
},
{
"epoch": 1.23,
"learning_rate": 1.6149385202584423e-05,
"loss": 0.4524,
"step": 1680
},
{
"epoch": 1.24,
"learning_rate": 1.5880524497489474e-05,
"loss": 0.4647,
"step": 1690
},
{
"epoch": 1.25,
"learning_rate": 1.5612874984681923e-05,
"loss": 0.4036,
"step": 1700
},
{
"epoch": 1.25,
"learning_rate": 1.534647221171334e-05,
"loss": 0.4148,
"step": 1710
},
{
"epoch": 1.26,
"learning_rate": 1.5081351560551021e-05,
"loss": 0.408,
"step": 1720
},
{
"epoch": 1.27,
"learning_rate": 1.4817548242878759e-05,
"loss": 0.457,
"step": 1730
},
{
"epoch": 1.28,
"learning_rate": 1.45550972954203e-05,
"loss": 0.4193,
"step": 1740
},
{
"epoch": 1.28,
"learning_rate": 1.4294033575285914e-05,
"loss": 0.4149,
"step": 1750
},
{
"epoch": 1.29,
"learning_rate": 1.4034391755342972e-05,
"loss": 0.4032,
"step": 1760
},
{
"epoch": 1.3,
"learning_rate": 1.3776206319610823e-05,
"loss": 0.4078,
"step": 1770
},
{
"epoch": 1.31,
"learning_rate": 1.3519511558680892e-05,
"loss": 0.4123,
"step": 1780
},
{
"epoch": 1.31,
"learning_rate": 1.3264341565162422e-05,
"loss": 0.4089,
"step": 1790
},
{
"epoch": 1.32,
"learning_rate": 1.3010730229154445e-05,
"loss": 0.383,
"step": 1800
},
{
"epoch": 1.32,
"eval_loss": 0.4250344932079315,
"eval_runtime": 70.4814,
"eval_samples_per_second": 15.635,
"eval_steps_per_second": 1.575,
"step": 1800
},
{
"epoch": 1.33,
"learning_rate": 1.2758711233744783e-05,
"loss": 0.3936,
"step": 1810
},
{
"epoch": 1.33,
"learning_rate": 1.2508318050536421e-05,
"loss": 0.4127,
"step": 1820
},
{
"epoch": 1.34,
"learning_rate": 1.2259583935202062e-05,
"loss": 0.4015,
"step": 1830
},
{
"epoch": 1.35,
"learning_rate": 1.2012541923067244e-05,
"loss": 0.4203,
"step": 1840
},
{
"epoch": 1.36,
"learning_rate": 1.176722482472286e-05,
"loss": 0.4367,
"step": 1850
},
{
"epoch": 1.36,
"learning_rate": 1.1523665221667398e-05,
"loss": 0.4233,
"step": 1860
},
{
"epoch": 1.37,
"learning_rate": 1.1281895461979732e-05,
"loss": 0.4405,
"step": 1870
},
{
"epoch": 1.38,
"learning_rate": 1.104194765602281e-05,
"loss": 0.4379,
"step": 1880
},
{
"epoch": 1.39,
"learning_rate": 1.0803853672178946e-05,
"loss": 0.4146,
"step": 1890
},
{
"epoch": 1.39,
"learning_rate": 1.0567645132617316e-05,
"loss": 0.4438,
"step": 1900
},
{
"epoch": 1.4,
"learning_rate": 1.0333353409094015e-05,
"loss": 0.3915,
"step": 1910
},
{
"epoch": 1.41,
"learning_rate": 1.0101009618785528e-05,
"loss": 0.4063,
"step": 1920
},
{
"epoch": 1.42,
"learning_rate": 9.870644620155877e-06,
"loss": 0.3974,
"step": 1930
},
{
"epoch": 1.42,
"learning_rate": 9.642289008858244e-06,
"loss": 0.4015,
"step": 1940
},
{
"epoch": 1.43,
"learning_rate": 9.41597311367142e-06,
"loss": 0.4264,
"step": 1950
},
{
"epoch": 1.44,
"learning_rate": 9.191726992471725e-06,
"loss": 0.4334,
"step": 1960
},
{
"epoch": 1.44,
"learning_rate": 8.969580428240903e-06,
"loss": 0.4197,
"step": 1970
},
{
"epoch": 1.45,
"learning_rate": 8.74956292511056e-06,
"loss": 0.422,
"step": 1980
},
{
"epoch": 1.46,
"learning_rate": 8.531703704443575e-06,
"loss": 0.4077,
"step": 1990
},
{
"epoch": 1.47,
"learning_rate": 8.316031700953086e-06,
"loss": 0.4334,
"step": 2000
},
{
"epoch": 1.47,
"eval_loss": 0.4206145107746124,
"eval_runtime": 70.0533,
"eval_samples_per_second": 15.731,
"eval_steps_per_second": 1.585,
"step": 2000
},
{
"epoch": 1.47,
"learning_rate": 8.102575558859612e-06,
"loss": 0.4171,
"step": 2010
},
{
"epoch": 1.48,
"learning_rate": 7.891363628086671e-06,
"loss": 0.3957,
"step": 2020
},
{
"epoch": 1.49,
"learning_rate": 7.682423960495517e-06,
"loss": 0.4174,
"step": 2030
},
{
"epoch": 1.5,
"learning_rate": 7.475784306159478e-06,
"loss": 0.4187,
"step": 2040
},
{
"epoch": 1.5,
"learning_rate": 7.271472109678379e-06,
"loss": 0.3775,
"step": 2050
},
{
"epoch": 1.51,
"learning_rate": 7.0695145065334585e-06,
"loss": 0.4019,
"step": 2060
},
{
"epoch": 1.52,
"learning_rate": 6.869938319483471e-06,
"loss": 0.4143,
"step": 2070
},
{
"epoch": 1.53,
"learning_rate": 6.67277005500222e-06,
"loss": 0.3853,
"step": 2080
},
{
"epoch": 1.53,
"learning_rate": 6.478035899758139e-06,
"loss": 0.4403,
"step": 2090
},
{
"epoch": 1.54,
"learning_rate": 6.285761717136335e-06,
"loss": 0.3888,
"step": 2100
},
{
"epoch": 1.55,
"learning_rate": 6.095973043803577e-06,
"loss": 0.4123,
"step": 2110
},
{
"epoch": 1.55,
"learning_rate": 5.908695086316701e-06,
"loss": 0.4035,
"step": 2120
},
{
"epoch": 1.56,
"learning_rate": 5.723952717774763e-06,
"loss": 0.4219,
"step": 2130
},
{
"epoch": 1.57,
"learning_rate": 5.54177047451562e-06,
"loss": 0.4209,
"step": 2140
},
{
"epoch": 1.58,
"learning_rate": 5.362172552857128e-06,
"loss": 0.4286,
"step": 2150
},
{
"epoch": 1.58,
"learning_rate": 5.18518280588354e-06,
"loss": 0.4134,
"step": 2160
},
{
"epoch": 1.59,
"learning_rate": 5.010824740277501e-06,
"loss": 0.4005,
"step": 2170
},
{
"epoch": 1.6,
"learning_rate": 4.83912151319802e-06,
"loss": 0.3973,
"step": 2180
},
{
"epoch": 1.61,
"learning_rate": 4.6700959292048875e-06,
"loss": 0.3959,
"step": 2190
},
{
"epoch": 1.61,
"learning_rate": 4.503770437229918e-06,
"loss": 0.4017,
"step": 2200
},
{
"epoch": 1.61,
"eval_loss": 0.41970890760421753,
"eval_runtime": 69.9019,
"eval_samples_per_second": 15.765,
"eval_steps_per_second": 1.588,
"step": 2200
},
{
"epoch": 1.62,
"learning_rate": 4.340167127595407e-06,
"loss": 0.4013,
"step": 2210
},
{
"epoch": 1.63,
"learning_rate": 4.179307729080256e-06,
"loss": 0.4069,
"step": 2220
},
{
"epoch": 1.64,
"learning_rate": 4.021213606034064e-06,
"loss": 0.3857,
"step": 2230
},
{
"epoch": 1.64,
"learning_rate": 3.8659057555396645e-06,
"loss": 0.4262,
"step": 2240
},
{
"epoch": 1.65,
"learning_rate": 3.713404804624418e-06,
"loss": 0.3953,
"step": 2250
},
{
"epoch": 1.66,
"learning_rate": 3.5637310075206544e-06,
"loss": 0.4231,
"step": 2260
},
{
"epoch": 1.66,
"learning_rate": 3.416904242975635e-06,
"loss": 0.407,
"step": 2270
},
{
"epoch": 1.67,
"learning_rate": 3.2729440116113843e-06,
"loss": 0.4254,
"step": 2280
},
{
"epoch": 1.68,
"learning_rate": 3.131869433334725e-06,
"loss": 0.4054,
"step": 2290
},
{
"epoch": 1.69,
"learning_rate": 2.9936992447979068e-06,
"loss": 0.3801,
"step": 2300
},
{
"epoch": 1.69,
"learning_rate": 2.8584517969101054e-06,
"loss": 0.4335,
"step": 2310
},
{
"epoch": 1.7,
"learning_rate": 2.7261450524001807e-06,
"loss": 0.412,
"step": 2320
},
{
"epoch": 1.71,
"learning_rate": 2.596796583430969e-06,
"loss": 0.4456,
"step": 2330
},
{
"epoch": 1.72,
"learning_rate": 2.470423569265462e-06,
"loss": 0.3972,
"step": 2340
},
{
"epoch": 1.72,
"learning_rate": 2.34704279398516e-06,
"loss": 0.3935,
"step": 2350
},
{
"epoch": 1.73,
"learning_rate": 2.2266706442609226e-06,
"loss": 0.3974,
"step": 2360
},
{
"epoch": 1.74,
"learning_rate": 2.109323107176578e-06,
"loss": 0.4157,
"step": 2370
},
{
"epoch": 1.75,
"learning_rate": 1.9950157681056318e-06,
"loss": 0.4379,
"step": 2380
},
{
"epoch": 1.75,
"learning_rate": 1.8837638086413063e-06,
"loss": 0.4035,
"step": 2390
},
{
"epoch": 1.76,
"learning_rate": 1.7755820045802145e-06,
"loss": 0.4455,
"step": 2400
},
{
"epoch": 1.76,
"eval_loss": 0.41885045170783997,
"eval_runtime": 70.0957,
"eval_samples_per_second": 15.721,
"eval_steps_per_second": 1.584,
"step": 2400
},
{
"epoch": 1.77,
"learning_rate": 1.6704847239599364e-06,
"loss": 0.4532,
"step": 2410
},
{
"epoch": 1.77,
"learning_rate": 1.5684859251507394e-06,
"loss": 0.4213,
"step": 2420
},
{
"epoch": 1.78,
"learning_rate": 1.4695991550017164e-06,
"loss": 0.3873,
"step": 2430
},
{
"epoch": 1.79,
"learning_rate": 1.373837547041576e-06,
"loss": 0.4163,
"step": 2440
},
{
"epoch": 1.8,
"learning_rate": 1.2812138197343392e-06,
"loss": 0.3946,
"step": 2450
},
{
"epoch": 1.8,
"learning_rate": 1.1917402747901152e-06,
"loss": 0.41,
"step": 2460
},
{
"epoch": 1.81,
"learning_rate": 1.105428795531327e-06,
"loss": 0.4126,
"step": 2470
},
{
"epoch": 1.82,
"learning_rate": 1.0222908453143804e-06,
"loss": 0.4505,
"step": 2480
},
{
"epoch": 1.83,
"learning_rate": 9.423374660072065e-07,
"loss": 0.4303,
"step": 2490
},
{
"epoch": 1.83,
"learning_rate": 8.655792765227405e-07,
"loss": 0.4179,
"step": 2500
},
{
"epoch": 1.84,
"learning_rate": 7.920264714085828e-07,
"loss": 0.4376,
"step": 2510
},
{
"epoch": 1.85,
"learning_rate": 7.216888194930272e-07,
"loss": 0.4069,
"step": 2520
},
{
"epoch": 1.86,
"learning_rate": 6.545756625876031e-07,
"loss": 0.3887,
"step": 2530
},
{
"epoch": 1.86,
"learning_rate": 5.906959142463947e-07,
"loss": 0.3972,
"step": 2540
},
{
"epoch": 1.87,
"learning_rate": 5.300580585821696e-07,
"loss": 0.4078,
"step": 2550
},
{
"epoch": 1.88,
"learning_rate": 4.7267014913956463e-07,
"loss": 0.4467,
"step": 2560
},
{
"epoch": 1.88,
"learning_rate": 4.1853980782549097e-07,
"loss": 0.4567,
"step": 2570
},
{
"epoch": 1.89,
"learning_rate": 3.6767422389682173e-07,
"loss": 0.4494,
"step": 2580
},
{
"epoch": 1.9,
"learning_rate": 3.2008015300555306e-07,
"loss": 0.4344,
"step": 2590
},
{
"epoch": 1.91,
"learning_rate": 2.757639163015774e-07,
"loss": 0.4101,
"step": 2600
},
{
"epoch": 1.91,
"eval_loss": 0.41886425018310547,
"eval_runtime": 70.2179,
"eval_samples_per_second": 15.694,
"eval_steps_per_second": 1.581,
"step": 2600
},
{
"epoch": 1.91,
"learning_rate": 2.347313995931466e-07,
"loss": 0.4058,
"step": 2610
},
{
"epoch": 1.92,
"learning_rate": 1.9698805256513908e-07,
"loss": 0.414,
"step": 2620
},
{
"epoch": 1.93,
"learning_rate": 1.6253888805527474e-07,
"loss": 0.4254,
"step": 2630
},
{
"epoch": 1.94,
"learning_rate": 1.3138848138835313e-07,
"loss": 0.4017,
"step": 2640
},
{
"epoch": 1.94,
"learning_rate": 1.0354096976856186e-07,
"loss": 0.4258,
"step": 2650
},
{
"epoch": 1.95,
"learning_rate": 7.900005173002712e-08,
"loss": 0.4098,
"step": 2660
},
{
"epoch": 1.96,
"learning_rate": 5.776898664557051e-08,
"loss": 0.42,
"step": 2670
},
{
"epoch": 1.97,
"learning_rate": 3.985059429383875e-08,
"loss": 0.3886,
"step": 2680
},
{
"epoch": 1.97,
"learning_rate": 2.5247254484794813e-08,
"loss": 0.4276,
"step": 2690
},
{
"epoch": 1.98,
"learning_rate": 1.3960906743634706e-08,
"loss": 0.4006,
"step": 2700
},
{
"epoch": 1.99,
"learning_rate": 5.993050053204607e-09,
"loss": 0.4243,
"step": 2710
},
{
"epoch": 1.99,
"learning_rate": 1.3447426549129117e-09,
"loss": 0.4439,
"step": 2720
},
{
"epoch": 2.0,
"step": 2726,
"total_flos": 1.1289170586615415e+19,
"train_loss": 0.46561298847898175,
"train_runtime": 35477.4141,
"train_samples_per_second": 6.149,
"train_steps_per_second": 0.077
}
],
"max_steps": 2726,
"num_train_epochs": 2,
"total_flos": 1.1289170586615415e+19,
"trial_name": null,
"trial_params": null
}