zephyr-7b-sft-qlora-50p / trainer_state.json
terry69's picture
Model save
d581a03 verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.0,
"eval_steps": 500,
"global_step": 812,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"grad_norm": 0.22422564560903913,
"learning_rate": 2.4390243902439027e-06,
"loss": 0.8165,
"step": 1
},
{
"epoch": 0.01,
"grad_norm": 0.24393612584966326,
"learning_rate": 1.2195121951219513e-05,
"loss": 0.8171,
"step": 5
},
{
"epoch": 0.01,
"grad_norm": 0.32976460652590056,
"learning_rate": 2.4390243902439026e-05,
"loss": 0.8001,
"step": 10
},
{
"epoch": 0.02,
"grad_norm": 0.22143488102760825,
"learning_rate": 3.6585365853658535e-05,
"loss": 0.7749,
"step": 15
},
{
"epoch": 0.02,
"grad_norm": 0.27241208120124494,
"learning_rate": 4.878048780487805e-05,
"loss": 0.7399,
"step": 20
},
{
"epoch": 0.03,
"grad_norm": 0.22476497302964527,
"learning_rate": 6.097560975609756e-05,
"loss": 0.765,
"step": 25
},
{
"epoch": 0.04,
"grad_norm": 0.23267051655279075,
"learning_rate": 7.317073170731707e-05,
"loss": 0.7434,
"step": 30
},
{
"epoch": 0.04,
"grad_norm": 0.22352973519663066,
"learning_rate": 8.53658536585366e-05,
"loss": 0.7884,
"step": 35
},
{
"epoch": 0.05,
"grad_norm": 0.23366158183718752,
"learning_rate": 9.75609756097561e-05,
"loss": 0.767,
"step": 40
},
{
"epoch": 0.06,
"grad_norm": 0.21853843372631854,
"learning_rate": 0.00010975609756097563,
"loss": 0.76,
"step": 45
},
{
"epoch": 0.06,
"grad_norm": 0.1997879106921521,
"learning_rate": 0.00012195121951219512,
"loss": 0.7516,
"step": 50
},
{
"epoch": 0.07,
"grad_norm": 0.2251793056206077,
"learning_rate": 0.00013414634146341464,
"loss": 0.7173,
"step": 55
},
{
"epoch": 0.07,
"grad_norm": 0.21081648300798228,
"learning_rate": 0.00014634146341463414,
"loss": 0.7528,
"step": 60
},
{
"epoch": 0.08,
"grad_norm": 0.23666214760062876,
"learning_rate": 0.00015853658536585366,
"loss": 0.7404,
"step": 65
},
{
"epoch": 0.09,
"grad_norm": 0.23722379969745064,
"learning_rate": 0.0001707317073170732,
"loss": 0.7612,
"step": 70
},
{
"epoch": 0.09,
"grad_norm": 0.24333660134577043,
"learning_rate": 0.0001829268292682927,
"loss": 0.717,
"step": 75
},
{
"epoch": 0.1,
"grad_norm": 0.24793258937714172,
"learning_rate": 0.0001951219512195122,
"loss": 0.7233,
"step": 80
},
{
"epoch": 0.1,
"grad_norm": 0.21823934898390882,
"learning_rate": 0.0001999916658654738,
"loss": 0.7168,
"step": 85
},
{
"epoch": 0.11,
"grad_norm": 0.2468304527466075,
"learning_rate": 0.0001999407400739705,
"loss": 0.7519,
"step": 90
},
{
"epoch": 0.12,
"grad_norm": 0.2481744524594569,
"learning_rate": 0.00019984354211555644,
"loss": 0.7212,
"step": 95
},
{
"epoch": 0.12,
"grad_norm": 0.22553056449126135,
"learning_rate": 0.00019970011699250152,
"loss": 0.7351,
"step": 100
},
{
"epoch": 0.13,
"grad_norm": 0.20832972377638914,
"learning_rate": 0.00019951053111006976,
"loss": 0.7611,
"step": 105
},
{
"epoch": 0.14,
"grad_norm": 0.2276942421072934,
"learning_rate": 0.00019927487224577402,
"loss": 0.7373,
"step": 110
},
{
"epoch": 0.14,
"grad_norm": 0.22785202882473857,
"learning_rate": 0.0001989932495087353,
"loss": 0.7777,
"step": 115
},
{
"epoch": 0.15,
"grad_norm": 0.23403114621707224,
"learning_rate": 0.0001986657932891657,
"loss": 0.7814,
"step": 120
},
{
"epoch": 0.15,
"grad_norm": 0.21032447129887216,
"learning_rate": 0.0001982926551979982,
"loss": 0.7601,
"step": 125
},
{
"epoch": 0.16,
"grad_norm": 0.21778287055015214,
"learning_rate": 0.00019787400799669154,
"loss": 0.7465,
"step": 130
},
{
"epoch": 0.17,
"grad_norm": 0.23068074124934998,
"learning_rate": 0.00019741004551724207,
"loss": 0.7671,
"step": 135
},
{
"epoch": 0.17,
"grad_norm": 0.21498665804807518,
"learning_rate": 0.00019690098257244064,
"loss": 0.7448,
"step": 140
},
{
"epoch": 0.18,
"grad_norm": 0.2217460126233852,
"learning_rate": 0.00019634705485641488,
"loss": 0.7508,
"step": 145
},
{
"epoch": 0.18,
"grad_norm": 0.22491198578150204,
"learning_rate": 0.00019574851883550395,
"loss": 0.7528,
"step": 150
},
{
"epoch": 0.19,
"grad_norm": 0.21656015924096556,
"learning_rate": 0.00019510565162951537,
"loss": 0.7528,
"step": 155
},
{
"epoch": 0.2,
"grad_norm": 0.21231516089906458,
"learning_rate": 0.00019441875088341997,
"loss": 0.7334,
"step": 160
},
{
"epoch": 0.2,
"grad_norm": 0.20043526486232421,
"learning_rate": 0.00019368813462954316,
"loss": 0.7092,
"step": 165
},
{
"epoch": 0.21,
"grad_norm": 0.2066905562271494,
"learning_rate": 0.00019291414114031743,
"loss": 0.69,
"step": 170
},
{
"epoch": 0.22,
"grad_norm": 0.2409977578699428,
"learning_rate": 0.00019209712877166349,
"loss": 0.7274,
"step": 175
},
{
"epoch": 0.22,
"grad_norm": 0.21478894589801012,
"learning_rate": 0.00019123747579707275,
"loss": 0.702,
"step": 180
},
{
"epoch": 0.23,
"grad_norm": 0.21830847034248066,
"learning_rate": 0.00019033558023246844,
"loss": 0.7474,
"step": 185
},
{
"epoch": 0.23,
"grad_norm": 0.22565415608082542,
"learning_rate": 0.0001893918596519257,
"loss": 0.7263,
"step": 190
},
{
"epoch": 0.24,
"grad_norm": 0.22451701734198667,
"learning_rate": 0.00018840675099433636,
"loss": 0.7376,
"step": 195
},
{
"epoch": 0.25,
"grad_norm": 0.22016338896296778,
"learning_rate": 0.00018738071036110808,
"loss": 0.6939,
"step": 200
},
{
"epoch": 0.25,
"grad_norm": 0.20607069335042727,
"learning_rate": 0.00018631421280499116,
"loss": 0.7351,
"step": 205
},
{
"epoch": 0.26,
"grad_norm": 0.20601552052171884,
"learning_rate": 0.00018520775211013093,
"loss": 0.7082,
"step": 210
},
{
"epoch": 0.26,
"grad_norm": 0.21774646892016394,
"learning_rate": 0.00018406184056344782,
"loss": 0.745,
"step": 215
},
{
"epoch": 0.27,
"grad_norm": 0.23099746146509867,
"learning_rate": 0.00018287700871745036,
"loss": 0.7707,
"step": 220
},
{
"epoch": 0.28,
"grad_norm": 0.2225513878139953,
"learning_rate": 0.0001816538051445916,
"loss": 0.72,
"step": 225
},
{
"epoch": 0.28,
"grad_norm": 0.2810648109589189,
"learning_rate": 0.00018039279618328212,
"loss": 0.7159,
"step": 230
},
{
"epoch": 0.29,
"grad_norm": 0.23360820743854752,
"learning_rate": 0.00017909456567567772,
"loss": 0.7916,
"step": 235
},
{
"epoch": 0.3,
"grad_norm": 0.2346981575638156,
"learning_rate": 0.0001777597146973627,
"loss": 0.7543,
"step": 240
},
{
"epoch": 0.3,
"grad_norm": 0.22472791649678256,
"learning_rate": 0.00017638886127905427,
"loss": 0.7145,
"step": 245
},
{
"epoch": 0.31,
"grad_norm": 0.2330382456080435,
"learning_rate": 0.00017498264012045687,
"loss": 0.6951,
"step": 250
},
{
"epoch": 0.31,
"grad_norm": 0.28462945710570964,
"learning_rate": 0.00017354170229639856,
"loss": 0.7218,
"step": 255
},
{
"epoch": 0.32,
"grad_norm": 0.23120564986808903,
"learning_rate": 0.00017206671495538612,
"loss": 0.6954,
"step": 260
},
{
"epoch": 0.33,
"grad_norm": 0.21787726424323484,
"learning_rate": 0.0001705583610107178,
"loss": 0.6993,
"step": 265
},
{
"epoch": 0.33,
"grad_norm": 0.22337741579847503,
"learning_rate": 0.0001690173388242972,
"loss": 0.7742,
"step": 270
},
{
"epoch": 0.34,
"grad_norm": 0.2418352997312942,
"learning_rate": 0.00016744436188329456,
"loss": 0.7385,
"step": 275
},
{
"epoch": 0.34,
"grad_norm": 0.25104847122600993,
"learning_rate": 0.0001658401584698049,
"loss": 0.7142,
"step": 280
},
{
"epoch": 0.35,
"grad_norm": 0.21545698908596508,
"learning_rate": 0.00016420547132365635,
"loss": 0.7043,
"step": 285
},
{
"epoch": 0.36,
"grad_norm": 0.24534521084917407,
"learning_rate": 0.00016254105729852464,
"loss": 0.6988,
"step": 290
},
{
"epoch": 0.36,
"grad_norm": 0.24621959592804224,
"learning_rate": 0.00016084768701151263,
"loss": 0.701,
"step": 295
},
{
"epoch": 0.37,
"grad_norm": 0.2441445089915858,
"learning_rate": 0.00015912614448635782,
"loss": 0.713,
"step": 300
},
{
"epoch": 0.38,
"grad_norm": 0.23144503509205255,
"learning_rate": 0.00015737722679043248,
"loss": 0.7185,
"step": 305
},
{
"epoch": 0.38,
"grad_norm": 0.23779196053246493,
"learning_rate": 0.00015560174366570446,
"loss": 0.6585,
"step": 310
},
{
"epoch": 0.39,
"grad_norm": 0.23922306500283833,
"learning_rate": 0.00015380051715382996,
"loss": 0.7025,
"step": 315
},
{
"epoch": 0.39,
"grad_norm": 0.25732276520746106,
"learning_rate": 0.0001519743812155516,
"loss": 0.6496,
"step": 320
},
{
"epoch": 0.4,
"grad_norm": 0.25676931335616193,
"learning_rate": 0.00015012418134457755,
"loss": 0.6946,
"step": 325
},
{
"epoch": 0.41,
"grad_norm": 0.25925282419918827,
"learning_rate": 0.00014825077417612186,
"loss": 0.7309,
"step": 330
},
{
"epoch": 0.41,
"grad_norm": 0.25723625109413206,
"learning_rate": 0.0001463550270902851,
"loss": 0.6929,
"step": 335
},
{
"epoch": 0.42,
"grad_norm": 0.2758091340120669,
"learning_rate": 0.00014443781781046136,
"loss": 0.7016,
"step": 340
},
{
"epoch": 0.42,
"grad_norm": 0.2368505670464136,
"learning_rate": 0.0001425000339969554,
"loss": 0.703,
"step": 345
},
{
"epoch": 0.43,
"grad_norm": 0.23084399243795903,
"learning_rate": 0.00014054257283599973,
"loss": 0.7102,
"step": 350
},
{
"epoch": 0.44,
"grad_norm": 0.2503463601524904,
"learning_rate": 0.0001385663406243607,
"loss": 0.7278,
"step": 355
},
{
"epoch": 0.44,
"grad_norm": 0.22226083699821775,
"learning_rate": 0.00013657225234972695,
"loss": 0.6753,
"step": 360
},
{
"epoch": 0.45,
"grad_norm": 0.26300812718557964,
"learning_rate": 0.00013456123126707334,
"loss": 0.6871,
"step": 365
},
{
"epoch": 0.46,
"grad_norm": 0.257794988184064,
"learning_rate": 0.00013253420847119803,
"loss": 0.737,
"step": 370
},
{
"epoch": 0.46,
"grad_norm": 0.23283253484592037,
"learning_rate": 0.0001304921224656289,
"loss": 0.6948,
"step": 375
},
{
"epoch": 0.47,
"grad_norm": 0.2489222484332262,
"learning_rate": 0.0001284359187281004,
"loss": 0.6807,
"step": 380
},
{
"epoch": 0.47,
"grad_norm": 0.24773496287366326,
"learning_rate": 0.00012636654927280073,
"loss": 0.7322,
"step": 385
},
{
"epoch": 0.48,
"grad_norm": 0.25284782064647693,
"learning_rate": 0.0001242849722095936,
"loss": 0.7038,
"step": 390
},
{
"epoch": 0.49,
"grad_norm": 0.2640705980904302,
"learning_rate": 0.00012219215130041656,
"loss": 0.6818,
"step": 395
},
{
"epoch": 0.49,
"grad_norm": 0.244353569188838,
"learning_rate": 0.00012008905551306356,
"loss": 0.6888,
"step": 400
},
{
"epoch": 0.5,
"grad_norm": 0.2582385397817055,
"learning_rate": 0.00011797665857255621,
"loss": 0.7347,
"step": 405
},
{
"epoch": 0.5,
"grad_norm": 0.2657939654189472,
"learning_rate": 0.00011585593851031347,
"loss": 0.7292,
"step": 410
},
{
"epoch": 0.51,
"grad_norm": 0.2412402754418769,
"learning_rate": 0.00011372787721132648,
"loss": 0.6793,
"step": 415
},
{
"epoch": 0.52,
"grad_norm": 0.25269309847633625,
"learning_rate": 0.00011159345995955006,
"loss": 0.6811,
"step": 420
},
{
"epoch": 0.52,
"grad_norm": 0.3555199335436584,
"learning_rate": 0.00010945367498171993,
"loss": 0.6632,
"step": 425
},
{
"epoch": 0.53,
"grad_norm": 0.2807299993278487,
"learning_rate": 0.00010730951298980776,
"loss": 0.7121,
"step": 430
},
{
"epoch": 0.54,
"grad_norm": 0.2676168527706757,
"learning_rate": 0.00010516196672232539,
"loss": 0.6897,
"step": 435
},
{
"epoch": 0.54,
"grad_norm": 0.2696033985636684,
"learning_rate": 0.00010301203048469083,
"loss": 0.7149,
"step": 440
},
{
"epoch": 0.55,
"grad_norm": 0.264343040293582,
"learning_rate": 0.00010086069968886885,
"loss": 0.697,
"step": 445
},
{
"epoch": 0.55,
"grad_norm": 0.24941452883154777,
"learning_rate": 9.870897039249911e-05,
"loss": 0.6739,
"step": 450
},
{
"epoch": 0.56,
"grad_norm": 0.27184754304212144,
"learning_rate": 9.655783883772545e-05,
"loss": 0.6771,
"step": 455
},
{
"epoch": 0.57,
"grad_norm": 0.25069232194770324,
"learning_rate": 9.440830098993969e-05,
"loss": 0.6628,
"step": 460
},
{
"epoch": 0.57,
"grad_norm": 0.26945962491120906,
"learning_rate": 9.22613520766537e-05,
"loss": 0.6501,
"step": 465
},
{
"epoch": 0.58,
"grad_norm": 0.27595385280343615,
"learning_rate": 9.011798612671286e-05,
"loss": 0.6923,
"step": 470
},
{
"epoch": 0.58,
"grad_norm": 0.23254427170051437,
"learning_rate": 8.797919551006475e-05,
"loss": 0.6687,
"step": 475
},
{
"epoch": 0.59,
"grad_norm": 0.2700843553748168,
"learning_rate": 8.58459704782957e-05,
"loss": 0.6922,
"step": 480
},
{
"epoch": 0.6,
"grad_norm": 0.29075126290562925,
"learning_rate": 8.371929870614833e-05,
"loss": 0.6838,
"step": 485
},
{
"epoch": 0.6,
"grad_norm": 0.2582533489019304,
"learning_rate": 8.160016483423199e-05,
"loss": 0.709,
"step": 490
},
{
"epoch": 0.61,
"grad_norm": 0.24306629955384693,
"learning_rate": 7.948955001313811e-05,
"loss": 0.6644,
"step": 495
},
{
"epoch": 0.62,
"grad_norm": 0.2792860639380007,
"learning_rate": 7.738843144917119e-05,
"loss": 0.7122,
"step": 500
},
{
"epoch": 0.62,
"grad_norm": 0.2717162268363558,
"learning_rate": 7.529778195190645e-05,
"loss": 0.7235,
"step": 505
},
{
"epoch": 0.63,
"grad_norm": 0.284678632001059,
"learning_rate": 7.321856948378259e-05,
"loss": 0.6916,
"step": 510
},
{
"epoch": 0.63,
"grad_norm": 0.26262649770739277,
"learning_rate": 7.115175671193913e-05,
"loss": 0.6741,
"step": 515
},
{
"epoch": 0.64,
"grad_norm": 0.2529748057506211,
"learning_rate": 6.909830056250527e-05,
"loss": 0.7025,
"step": 520
},
{
"epoch": 0.65,
"grad_norm": 0.28817490304932813,
"learning_rate": 6.7059151777547e-05,
"loss": 0.7071,
"step": 525
},
{
"epoch": 0.65,
"grad_norm": 0.28586738996038136,
"learning_rate": 6.503525447487715e-05,
"loss": 0.7223,
"step": 530
},
{
"epoch": 0.66,
"grad_norm": 0.27355383837426006,
"learning_rate": 6.30275457109327e-05,
"loss": 0.6878,
"step": 535
},
{
"epoch": 0.67,
"grad_norm": 0.2696146366032459,
"learning_rate": 6.103695504692122e-05,
"loss": 0.666,
"step": 540
},
{
"epoch": 0.67,
"grad_norm": 0.293084107951244,
"learning_rate": 5.906440411843787e-05,
"loss": 0.7165,
"step": 545
},
{
"epoch": 0.68,
"grad_norm": 0.2744477897912315,
"learning_rate": 5.7110806208751655e-05,
"loss": 0.6791,
"step": 550
},
{
"epoch": 0.68,
"grad_norm": 0.2719074571288193,
"learning_rate": 5.5177065825958966e-05,
"loss": 0.7183,
"step": 555
},
{
"epoch": 0.69,
"grad_norm": 0.29658031172179306,
"learning_rate": 5.326407828419979e-05,
"loss": 0.6741,
"step": 560
},
{
"epoch": 0.7,
"grad_norm": 0.25304272831172997,
"learning_rate": 5.137272928913097e-05,
"loss": 0.68,
"step": 565
},
{
"epoch": 0.7,
"grad_norm": 0.2676461633502972,
"learning_rate": 4.9503894527847964e-05,
"loss": 0.6593,
"step": 570
},
{
"epoch": 0.71,
"grad_norm": 0.31247663625681804,
"learning_rate": 4.765843926344501e-05,
"loss": 0.6863,
"step": 575
},
{
"epoch": 0.71,
"grad_norm": 0.2955325664002294,
"learning_rate": 4.583721793440188e-05,
"loss": 0.6728,
"step": 580
},
{
"epoch": 0.72,
"grad_norm": 0.2807174542636257,
"learning_rate": 4.4041073758982335e-05,
"loss": 0.6913,
"step": 585
},
{
"epoch": 0.73,
"grad_norm": 0.2989395763460932,
"learning_rate": 4.227083834482728e-05,
"loss": 0.6085,
"step": 590
},
{
"epoch": 0.73,
"grad_norm": 0.2636409419874345,
"learning_rate": 4.052733130392367e-05,
"loss": 0.676,
"step": 595
},
{
"epoch": 0.74,
"grad_norm": 0.2735419958012749,
"learning_rate": 3.881135987312757e-05,
"loss": 0.697,
"step": 600
},
{
"epoch": 0.75,
"grad_norm": 0.29812585759610105,
"learning_rate": 3.712371854041654e-05,
"loss": 0.6788,
"step": 605
},
{
"epoch": 0.75,
"grad_norm": 0.2858805204008935,
"learning_rate": 3.546518867704499e-05,
"loss": 0.6891,
"step": 610
},
{
"epoch": 0.76,
"grad_norm": 0.2711930626215174,
"learning_rate": 3.383653817577216e-05,
"loss": 0.6626,
"step": 615
},
{
"epoch": 0.76,
"grad_norm": 0.30583390146359735,
"learning_rate": 3.223852109533112e-05,
"loss": 0.6977,
"step": 620
},
{
"epoch": 0.77,
"grad_norm": 0.31918222012772746,
"learning_rate": 3.0671877311302244e-05,
"loss": 0.6546,
"step": 625
},
{
"epoch": 0.78,
"grad_norm": 0.27113125300550006,
"learning_rate": 2.9137332173554043e-05,
"loss": 0.6718,
"step": 630
},
{
"epoch": 0.78,
"grad_norm": 0.2933116316225334,
"learning_rate": 2.763559617040876e-05,
"loss": 0.677,
"step": 635
},
{
"epoch": 0.79,
"grad_norm": 0.3236593994578141,
"learning_rate": 2.616736459968936e-05,
"loss": 0.6685,
"step": 640
},
{
"epoch": 0.79,
"grad_norm": 0.30406592976690633,
"learning_rate": 2.473331724679917e-05,
"loss": 0.6746,
"step": 645
},
{
"epoch": 0.8,
"grad_norm": 0.27899362724771787,
"learning_rate": 2.33341180699841e-05,
"loss": 0.6667,
"step": 650
},
{
"epoch": 0.81,
"grad_norm": 0.2954598561484786,
"learning_rate": 2.1970414892922442e-05,
"loss": 0.7175,
"step": 655
},
{
"epoch": 0.81,
"grad_norm": 0.3100581301740447,
"learning_rate": 2.0642839104785272e-05,
"loss": 0.6979,
"step": 660
},
{
"epoch": 0.82,
"grad_norm": 0.2759099798186259,
"learning_rate": 1.9352005367905536e-05,
"loss": 0.7098,
"step": 665
},
{
"epoch": 0.83,
"grad_norm": 0.30931195338097267,
"learning_rate": 1.8098511333192024e-05,
"loss": 0.6593,
"step": 670
},
{
"epoch": 0.83,
"grad_norm": 0.32368942926694433,
"learning_rate": 1.6882937363419203e-05,
"loss": 0.6869,
"step": 675
},
{
"epoch": 0.84,
"grad_norm": 0.2776307386865438,
"learning_rate": 1.570584626452173e-05,
"loss": 0.6717,
"step": 680
},
{
"epoch": 0.84,
"grad_norm": 0.30301564173702233,
"learning_rate": 1.4567783025017301e-05,
"loss": 0.6681,
"step": 685
},
{
"epoch": 0.85,
"grad_norm": 0.31904417802084467,
"learning_rate": 1.3469274563679402e-05,
"loss": 0.6851,
"step": 690
},
{
"epoch": 0.86,
"grad_norm": 0.2774036633854793,
"learning_rate": 1.2410829485575704e-05,
"loss": 0.7023,
"step": 695
},
{
"epoch": 0.86,
"grad_norm": 0.2965220463492153,
"learning_rate": 1.1392937846586215e-05,
"loss": 0.6565,
"step": 700
},
{
"epoch": 0.87,
"grad_norm": 0.32675908474718157,
"learning_rate": 1.0416070926509113e-05,
"loss": 0.6487,
"step": 705
},
{
"epoch": 0.87,
"grad_norm": 0.3090900072007822,
"learning_rate": 9.48068101086026e-06,
"loss": 0.6976,
"step": 710
},
{
"epoch": 0.88,
"grad_norm": 0.309037705156712,
"learning_rate": 8.58720118146662e-06,
"loss": 0.66,
"step": 715
},
{
"epoch": 0.89,
"grad_norm": 0.3087738006371124,
"learning_rate": 7.736045115951251e-06,
"loss": 0.7067,
"step": 720
},
{
"epoch": 0.89,
"grad_norm": 0.3443047214751086,
"learning_rate": 6.927606896202066e-06,
"loss": 0.6968,
"step": 725
},
{
"epoch": 0.9,
"grad_norm": 0.30866169616965344,
"learning_rate": 6.16226082591359e-06,
"loss": 0.6595,
"step": 730
},
{
"epoch": 0.91,
"grad_norm": 0.26453835920017715,
"learning_rate": 5.440361257285742e-06,
"loss": 0.6873,
"step": 735
},
{
"epoch": 0.91,
"grad_norm": 0.2870240594644306,
"learning_rate": 4.762242426960262e-06,
"loss": 0.6897,
"step": 740
},
{
"epoch": 0.92,
"grad_norm": 0.31907803929807727,
"learning_rate": 4.128218301270359e-06,
"loss": 0.7062,
"step": 745
},
{
"epoch": 0.92,
"grad_norm": 0.3461230801807586,
"learning_rate": 3.5385824308756587e-06,
"loss": 0.6506,
"step": 750
},
{
"epoch": 0.93,
"grad_norm": 0.2982663104307397,
"learning_rate": 2.9936078148492973e-06,
"loss": 0.6722,
"step": 755
},
{
"epoch": 0.94,
"grad_norm": 0.27990503389995286,
"learning_rate": 2.493546774280531e-06,
"loss": 0.7049,
"step": 760
},
{
"epoch": 0.94,
"grad_norm": 0.3112912132178525,
"learning_rate": 2.0386308354509942e-06,
"loss": 0.7125,
"step": 765
},
{
"epoch": 0.95,
"grad_norm": 0.31758074243604545,
"learning_rate": 1.6290706226390285e-06,
"loss": 0.6857,
"step": 770
},
{
"epoch": 0.95,
"grad_norm": 0.3210813944651733,
"learning_rate": 1.2650557606013635e-06,
"loss": 0.6829,
"step": 775
},
{
"epoch": 0.96,
"grad_norm": 0.2880839335185284,
"learning_rate": 9.46754786777726e-07,
"loss": 0.7146,
"step": 780
},
{
"epoch": 0.97,
"grad_norm": 0.29564960884643837,
"learning_rate": 6.74315073258569e-07,
"loss": 0.7101,
"step": 785
},
{
"epoch": 0.97,
"grad_norm": 0.29256820612374973,
"learning_rate": 4.4786275855247527e-07,
"loss": 0.6478,
"step": 790
},
{
"epoch": 0.98,
"grad_norm": 0.33071981852297627,
"learning_rate": 2.675026891844512e-07,
"loss": 0.672,
"step": 795
},
{
"epoch": 0.99,
"grad_norm": 0.329286091376579,
"learning_rate": 1.333183711524133e-07,
"loss": 0.6438,
"step": 800
},
{
"epoch": 0.99,
"grad_norm": 0.3104373424505981,
"learning_rate": 4.5371931264270864e-08,
"loss": 0.6555,
"step": 805
},
{
"epoch": 1.0,
"grad_norm": 0.2971195933385636,
"learning_rate": 3.7040883734462683e-09,
"loss": 0.6774,
"step": 810
},
{
"epoch": 1.0,
"eval_loss": 0.926139771938324,
"eval_runtime": 459.5724,
"eval_samples_per_second": 5.026,
"eval_steps_per_second": 0.316,
"step": 812
},
{
"epoch": 1.0,
"step": 812,
"total_flos": 9607955794100224.0,
"train_loss": 0.7061199183240899,
"train_runtime": 29649.4111,
"train_samples_per_second": 1.753,
"train_steps_per_second": 0.027
}
],
"logging_steps": 5,
"max_steps": 812,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 100,
"total_flos": 9607955794100224.0,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}