Files
Qwen3-4B-Fncall-SFT/trainer_state.json
ModelHub XC 9d1c73eda9 初始化项目,由ModelHub XC社区提供模型
Model: TabibitoQZP/Qwen3-4B-Fncall-SFT
Source: Original Platform
2026-09-09 08:20:12 +08:00

2837 lines
68 KiB
JSON

{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 2.0,
"eval_steps": 500,
"global_step": 3996,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0050062578222778474,
"grad_norm": 41.714508056640625,
"learning_rate": 2.2500000000000002e-07,
"loss": 1.2685,
"step": 10
},
{
"epoch": 0.010012515644555695,
"grad_norm": 35.297401428222656,
"learning_rate": 4.7500000000000006e-07,
"loss": 1.2116,
"step": 20
},
{
"epoch": 0.015018773466833541,
"grad_norm": 12.612136840820312,
"learning_rate": 7.25e-07,
"loss": 0.926,
"step": 30
},
{
"epoch": 0.02002503128911139,
"grad_norm": 2.90101957321167,
"learning_rate": 9.750000000000002e-07,
"loss": 0.7177,
"step": 40
},
{
"epoch": 0.025031289111389236,
"grad_norm": 1.4915510416030884,
"learning_rate": 1.2250000000000001e-06,
"loss": 0.6422,
"step": 50
},
{
"epoch": 0.030037546933667083,
"grad_norm": 1.2885878086090088,
"learning_rate": 1.475e-06,
"loss": 0.6043,
"step": 60
},
{
"epoch": 0.03504380475594493,
"grad_norm": 1.2088379859924316,
"learning_rate": 1.725e-06,
"loss": 0.5831,
"step": 70
},
{
"epoch": 0.04005006257822278,
"grad_norm": 1.2009735107421875,
"learning_rate": 1.975e-06,
"loss": 0.5818,
"step": 80
},
{
"epoch": 0.04505632040050062,
"grad_norm": 1.103482961654663,
"learning_rate": 2.2250000000000003e-06,
"loss": 0.5894,
"step": 90
},
{
"epoch": 0.05006257822277847,
"grad_norm": 1.148666501045227,
"learning_rate": 2.475e-06,
"loss": 0.5869,
"step": 100
},
{
"epoch": 0.05506883604505632,
"grad_norm": 1.1669937372207642,
"learning_rate": 2.7250000000000006e-06,
"loss": 0.5632,
"step": 110
},
{
"epoch": 0.060075093867334166,
"grad_norm": 0.9447499513626099,
"learning_rate": 2.9750000000000003e-06,
"loss": 0.5901,
"step": 120
},
{
"epoch": 0.06508135168961202,
"grad_norm": 1.0674545764923096,
"learning_rate": 3.2250000000000005e-06,
"loss": 0.5598,
"step": 130
},
{
"epoch": 0.07008760951188986,
"grad_norm": 1.0488513708114624,
"learning_rate": 3.475e-06,
"loss": 0.5755,
"step": 140
},
{
"epoch": 0.07509386733416772,
"grad_norm": 1.1044185161590576,
"learning_rate": 3.7250000000000003e-06,
"loss": 0.5539,
"step": 150
},
{
"epoch": 0.08010012515644556,
"grad_norm": 1.0041446685791016,
"learning_rate": 3.975000000000001e-06,
"loss": 0.56,
"step": 160
},
{
"epoch": 0.0851063829787234,
"grad_norm": 1.1575604677200317,
"learning_rate": 4.225e-06,
"loss": 0.5418,
"step": 170
},
{
"epoch": 0.09011264080100125,
"grad_norm": 1.0825908184051514,
"learning_rate": 4.475e-06,
"loss": 0.5589,
"step": 180
},
{
"epoch": 0.0951188986232791,
"grad_norm": 1.1007875204086304,
"learning_rate": 4.7250000000000005e-06,
"loss": 0.5485,
"step": 190
},
{
"epoch": 0.10012515644555695,
"grad_norm": 1.0647069215774536,
"learning_rate": 4.975000000000001e-06,
"loss": 0.5575,
"step": 200
},
{
"epoch": 0.10513141426783479,
"grad_norm": 1.1160515546798706,
"learning_rate": 5.225e-06,
"loss": 0.5567,
"step": 210
},
{
"epoch": 0.11013767209011265,
"grad_norm": 1.1660758256912231,
"learning_rate": 5.475e-06,
"loss": 0.5591,
"step": 220
},
{
"epoch": 0.11514392991239049,
"grad_norm": 1.0049713850021362,
"learning_rate": 5.725e-06,
"loss": 0.5712,
"step": 230
},
{
"epoch": 0.12015018773466833,
"grad_norm": 0.9587867856025696,
"learning_rate": 5.975e-06,
"loss": 0.5322,
"step": 240
},
{
"epoch": 0.1251564455569462,
"grad_norm": 1.1788851022720337,
"learning_rate": 6.225000000000001e-06,
"loss": 0.5503,
"step": 250
},
{
"epoch": 0.13016270337922403,
"grad_norm": 1.0932830572128296,
"learning_rate": 6.475e-06,
"loss": 0.564,
"step": 260
},
{
"epoch": 0.13516896120150187,
"grad_norm": 0.9803727865219116,
"learning_rate": 6.725000000000001e-06,
"loss": 0.5521,
"step": 270
},
{
"epoch": 0.14017521902377972,
"grad_norm": 1.0000665187835693,
"learning_rate": 6.975000000000001e-06,
"loss": 0.5599,
"step": 280
},
{
"epoch": 0.14518147684605756,
"grad_norm": 1.163681983947754,
"learning_rate": 7.225000000000001e-06,
"loss": 0.55,
"step": 290
},
{
"epoch": 0.15018773466833543,
"grad_norm": 1.100650429725647,
"learning_rate": 7.475000000000001e-06,
"loss": 0.5571,
"step": 300
},
{
"epoch": 0.15519399249061328,
"grad_norm": 1.2197952270507812,
"learning_rate": 7.725e-06,
"loss": 0.5378,
"step": 310
},
{
"epoch": 0.16020025031289112,
"grad_norm": 1.0715725421905518,
"learning_rate": 7.975e-06,
"loss": 0.5537,
"step": 320
},
{
"epoch": 0.16520650813516896,
"grad_norm": 1.1516462564468384,
"learning_rate": 8.225e-06,
"loss": 0.5557,
"step": 330
},
{
"epoch": 0.1702127659574468,
"grad_norm": 0.9862958788871765,
"learning_rate": 8.475000000000001e-06,
"loss": 0.5519,
"step": 340
},
{
"epoch": 0.17521902377972465,
"grad_norm": 1.076583981513977,
"learning_rate": 8.725000000000002e-06,
"loss": 0.5627,
"step": 350
},
{
"epoch": 0.1802252816020025,
"grad_norm": 1.110701560974121,
"learning_rate": 8.975e-06,
"loss": 0.5476,
"step": 360
},
{
"epoch": 0.18523153942428036,
"grad_norm": 1.0044567584991455,
"learning_rate": 9.225e-06,
"loss": 0.5351,
"step": 370
},
{
"epoch": 0.1902377972465582,
"grad_norm": 1.0845321416854858,
"learning_rate": 9.475000000000002e-06,
"loss": 0.554,
"step": 380
},
{
"epoch": 0.19524405506883605,
"grad_norm": 0.9674211144447327,
"learning_rate": 9.725000000000001e-06,
"loss": 0.5422,
"step": 390
},
{
"epoch": 0.2002503128911139,
"grad_norm": 1.0883331298828125,
"learning_rate": 9.975000000000002e-06,
"loss": 0.5652,
"step": 400
},
{
"epoch": 0.20525657071339173,
"grad_norm": 1.0060268640518188,
"learning_rate": 9.999845444960881e-06,
"loss": 0.5393,
"step": 410
},
{
"epoch": 0.21026282853566958,
"grad_norm": 1.0195198059082031,
"learning_rate": 9.99931119289516e-06,
"loss": 0.5451,
"step": 420
},
{
"epoch": 0.21526908635794745,
"grad_norm": 1.19355309009552,
"learning_rate": 9.998395376482152e-06,
"loss": 0.5638,
"step": 430
},
{
"epoch": 0.2202753441802253,
"grad_norm": 1.0359495878219604,
"learning_rate": 9.997098065620068e-06,
"loss": 0.542,
"step": 440
},
{
"epoch": 0.22528160200250313,
"grad_norm": 0.9536132216453552,
"learning_rate": 9.995419359324061e-06,
"loss": 0.5349,
"step": 450
},
{
"epoch": 0.23028785982478098,
"grad_norm": 1.055509090423584,
"learning_rate": 9.993359385718675e-06,
"loss": 0.535,
"step": 460
},
{
"epoch": 0.23529411764705882,
"grad_norm": 1.0224448442459106,
"learning_rate": 9.990918302028068e-06,
"loss": 0.5551,
"step": 470
},
{
"epoch": 0.24030037546933666,
"grad_norm": 1.0108073949813843,
"learning_rate": 9.988096294564006e-06,
"loss": 0.5354,
"step": 480
},
{
"epoch": 0.2453066332916145,
"grad_norm": 1.0135139226913452,
"learning_rate": 9.984893578711647e-06,
"loss": 0.5411,
"step": 490
},
{
"epoch": 0.2503128911138924,
"grad_norm": 1.0536352396011353,
"learning_rate": 9.981310398913107e-06,
"loss": 0.5423,
"step": 500
},
{
"epoch": 0.2553191489361702,
"grad_norm": 0.9891515970230103,
"learning_rate": 9.97734702864879e-06,
"loss": 0.5441,
"step": 510
},
{
"epoch": 0.26032540675844806,
"grad_norm": 0.9594472050666809,
"learning_rate": 9.973003770416524e-06,
"loss": 0.5486,
"step": 520
},
{
"epoch": 0.26533166458072593,
"grad_norm": 0.9631192088127136,
"learning_rate": 9.968280955708478e-06,
"loss": 0.5386,
"step": 530
},
{
"epoch": 0.27033792240300375,
"grad_norm": 1.041158676147461,
"learning_rate": 9.963178944985851e-06,
"loss": 0.5382,
"step": 540
},
{
"epoch": 0.2753441802252816,
"grad_norm": 0.9818730354309082,
"learning_rate": 9.957698127651366e-06,
"loss": 0.5525,
"step": 550
},
{
"epoch": 0.28035043804755944,
"grad_norm": 0.9888095259666443,
"learning_rate": 9.951838922019548e-06,
"loss": 0.5437,
"step": 560
},
{
"epoch": 0.2853566958698373,
"grad_norm": 1.0410245656967163,
"learning_rate": 9.945601775284797e-06,
"loss": 0.5396,
"step": 570
},
{
"epoch": 0.2903629536921151,
"grad_norm": 1.0097676515579224,
"learning_rate": 9.938987163487258e-06,
"loss": 0.5399,
"step": 580
},
{
"epoch": 0.295369211514393,
"grad_norm": 1.0537450313568115,
"learning_rate": 9.931995591476487e-06,
"loss": 0.5429,
"step": 590
},
{
"epoch": 0.30037546933667086,
"grad_norm": 0.9741904139518738,
"learning_rate": 9.924627592872917e-06,
"loss": 0.5443,
"step": 600
},
{
"epoch": 0.3053817271589487,
"grad_norm": 0.9675753116607666,
"learning_rate": 9.916883730027136e-06,
"loss": 0.5415,
"step": 610
},
{
"epoch": 0.31038798498122655,
"grad_norm": 0.9772769808769226,
"learning_rate": 9.908764593976959e-06,
"loss": 0.5439,
"step": 620
},
{
"epoch": 0.31539424280350437,
"grad_norm": 0.9558619260787964,
"learning_rate": 9.900270804402325e-06,
"loss": 0.5365,
"step": 630
},
{
"epoch": 0.32040050062578224,
"grad_norm": 0.9302605390548706,
"learning_rate": 9.891403009577994e-06,
"loss": 0.5474,
"step": 640
},
{
"epoch": 0.32540675844806005,
"grad_norm": 0.9413759708404541,
"learning_rate": 9.882161886324077e-06,
"loss": 0.5418,
"step": 650
},
{
"epoch": 0.3304130162703379,
"grad_norm": 0.8883467316627502,
"learning_rate": 9.872548139954368e-06,
"loss": 0.5475,
"step": 660
},
{
"epoch": 0.3354192740926158,
"grad_norm": 0.9297903180122375,
"learning_rate": 9.862562504222525e-06,
"loss": 0.551,
"step": 670
},
{
"epoch": 0.3404255319148936,
"grad_norm": 0.9027453660964966,
"learning_rate": 9.852205741266058e-06,
"loss": 0.5484,
"step": 680
},
{
"epoch": 0.3454317897371715,
"grad_norm": 0.8880664706230164,
"learning_rate": 9.841478641548159e-06,
"loss": 0.5351,
"step": 690
},
{
"epoch": 0.3504380475594493,
"grad_norm": 1.1402093172073364,
"learning_rate": 9.830382023797378e-06,
"loss": 0.5281,
"step": 700
},
{
"epoch": 0.35544430538172717,
"grad_norm": 1.055342674255371,
"learning_rate": 9.818916734945132e-06,
"loss": 0.5296,
"step": 710
},
{
"epoch": 0.360450563204005,
"grad_norm": 0.8715190291404724,
"learning_rate": 9.807083650061063e-06,
"loss": 0.5321,
"step": 720
},
{
"epoch": 0.36545682102628285,
"grad_norm": 1.0059534311294556,
"learning_rate": 9.794883672286252e-06,
"loss": 0.5342,
"step": 730
},
{
"epoch": 0.3704630788485607,
"grad_norm": 1.0429677963256836,
"learning_rate": 9.782317732764285e-06,
"loss": 0.5374,
"step": 740
},
{
"epoch": 0.37546933667083854,
"grad_norm": 0.9086937308311462,
"learning_rate": 9.769386790570188e-06,
"loss": 0.5448,
"step": 750
},
{
"epoch": 0.3804755944931164,
"grad_norm": 0.9410043954849243,
"learning_rate": 9.756091832637226e-06,
"loss": 0.5311,
"step": 760
},
{
"epoch": 0.3854818523153942,
"grad_norm": 0.9807677865028381,
"learning_rate": 9.742433873681577e-06,
"loss": 0.5395,
"step": 770
},
{
"epoch": 0.3904881101376721,
"grad_norm": 1.0230454206466675,
"learning_rate": 9.728413956124884e-06,
"loss": 0.5242,
"step": 780
},
{
"epoch": 0.3954943679599499,
"grad_norm": 0.8439666032791138,
"learning_rate": 9.7140331500147e-06,
"loss": 0.5438,
"step": 790
},
{
"epoch": 0.4005006257822278,
"grad_norm": 0.9394951462745667,
"learning_rate": 9.699292552942807e-06,
"loss": 0.5365,
"step": 800
},
{
"epoch": 0.40550688360450565,
"grad_norm": 0.8979880213737488,
"learning_rate": 9.684193289961455e-06,
"loss": 0.539,
"step": 810
},
{
"epoch": 0.41051314142678347,
"grad_norm": 0.9597287774085999,
"learning_rate": 9.66873651349749e-06,
"loss": 0.5188,
"step": 820
},
{
"epoch": 0.41551939924906134,
"grad_norm": 0.9950388669967651,
"learning_rate": 9.652923403264392e-06,
"loss": 0.5468,
"step": 830
},
{
"epoch": 0.42052565707133915,
"grad_norm": 0.8931434750556946,
"learning_rate": 9.636755166172248e-06,
"loss": 0.5397,
"step": 840
},
{
"epoch": 0.425531914893617,
"grad_norm": 0.9110593795776367,
"learning_rate": 9.62023303623562e-06,
"loss": 0.5291,
"step": 850
},
{
"epoch": 0.4305381727158949,
"grad_norm": 1.1425279378890991,
"learning_rate": 9.603358274479377e-06,
"loss": 0.5097,
"step": 860
},
{
"epoch": 0.4355444305381727,
"grad_norm": 0.9236230850219727,
"learning_rate": 9.586132168842435e-06,
"loss": 0.5404,
"step": 870
},
{
"epoch": 0.4405506883604506,
"grad_norm": 0.9602265954017639,
"learning_rate": 9.568556034079471e-06,
"loss": 0.5328,
"step": 880
},
{
"epoch": 0.4455569461827284,
"grad_norm": 0.9402238726615906,
"learning_rate": 9.550631211660562e-06,
"loss": 0.5349,
"step": 890
},
{
"epoch": 0.45056320400500627,
"grad_norm": 0.9396120309829712,
"learning_rate": 9.532359069668812e-06,
"loss": 0.5423,
"step": 900
},
{
"epoch": 0.4555694618272841,
"grad_norm": 0.9568923115730286,
"learning_rate": 9.51374100269593e-06,
"loss": 0.5331,
"step": 910
},
{
"epoch": 0.46057571964956195,
"grad_norm": 0.9286158680915833,
"learning_rate": 9.494778431735786e-06,
"loss": 0.5231,
"step": 920
},
{
"epoch": 0.4655819774718398,
"grad_norm": 0.9904299974441528,
"learning_rate": 9.475472804075964e-06,
"loss": 0.516,
"step": 930
},
{
"epoch": 0.47058823529411764,
"grad_norm": 0.8927105069160461,
"learning_rate": 9.455825593187295e-06,
"loss": 0.5219,
"step": 940
},
{
"epoch": 0.4755944931163955,
"grad_norm": 0.9750750660896301,
"learning_rate": 9.435838298611395e-06,
"loss": 0.5376,
"step": 950
},
{
"epoch": 0.4806007509386733,
"grad_norm": 0.9918380975723267,
"learning_rate": 9.415512445846225e-06,
"loss": 0.5358,
"step": 960
},
{
"epoch": 0.4856070087609512,
"grad_norm": 1.0143691301345825,
"learning_rate": 9.394849586229645e-06,
"loss": 0.5402,
"step": 970
},
{
"epoch": 0.490613266583229,
"grad_norm": 0.947730541229248,
"learning_rate": 9.373851296821022e-06,
"loss": 0.5316,
"step": 980
},
{
"epoch": 0.4956195244055069,
"grad_norm": 0.9850479364395142,
"learning_rate": 9.352519180280862e-06,
"loss": 0.5304,
"step": 990
},
{
"epoch": 0.5006257822277848,
"grad_norm": 0.8721357583999634,
"learning_rate": 9.33085486474848e-06,
"loss": 0.5456,
"step": 1000
},
{
"epoch": 0.5056320400500626,
"grad_norm": 0.9587690830230713,
"learning_rate": 9.308860003717748e-06,
"loss": 0.5148,
"step": 1010
},
{
"epoch": 0.5106382978723404,
"grad_norm": 0.8999404907226562,
"learning_rate": 9.286536275910891e-06,
"loss": 0.5302,
"step": 1020
},
{
"epoch": 0.5156445556946183,
"grad_norm": 0.9121967554092407,
"learning_rate": 9.263885385150354e-06,
"loss": 0.5289,
"step": 1030
},
{
"epoch": 0.5206508135168961,
"grad_norm": 1.0111724138259888,
"learning_rate": 9.240909060228765e-06,
"loss": 0.5211,
"step": 1040
},
{
"epoch": 0.5256570713391739,
"grad_norm": 0.9607769250869751,
"learning_rate": 9.217609054776995e-06,
"loss": 0.5154,
"step": 1050
},
{
"epoch": 0.5306633291614519,
"grad_norm": 0.9243285655975342,
"learning_rate": 9.1939871471303e-06,
"loss": 0.5281,
"step": 1060
},
{
"epoch": 0.5356695869837297,
"grad_norm": 0.9070591926574707,
"learning_rate": 9.170045140192611e-06,
"loss": 0.5171,
"step": 1070
},
{
"epoch": 0.5406758448060075,
"grad_norm": 0.997380256652832,
"learning_rate": 9.14578486129891e-06,
"loss": 0.5398,
"step": 1080
},
{
"epoch": 0.5456821026282853,
"grad_norm": 0.8227772116661072,
"learning_rate": 9.121208162075777e-06,
"loss": 0.5183,
"step": 1090
},
{
"epoch": 0.5506883604505632,
"grad_norm": 0.8431420922279358,
"learning_rate": 9.096316918300062e-06,
"loss": 0.5244,
"step": 1100
},
{
"epoch": 0.5556946182728411,
"grad_norm": 0.8711059093475342,
"learning_rate": 9.071113029755716e-06,
"loss": 0.535,
"step": 1110
},
{
"epoch": 0.5607008760951189,
"grad_norm": 0.8576318025588989,
"learning_rate": 9.045598420088804e-06,
"loss": 0.5216,
"step": 1120
},
{
"epoch": 0.5657071339173968,
"grad_norm": 1.0832278728485107,
"learning_rate": 9.01977503666067e-06,
"loss": 0.5318,
"step": 1130
},
{
"epoch": 0.5707133917396746,
"grad_norm": 0.9785774350166321,
"learning_rate": 8.99364485039932e-06,
"loss": 0.533,
"step": 1140
},
{
"epoch": 0.5757196495619524,
"grad_norm": 0.9052590131759644,
"learning_rate": 8.967209855648996e-06,
"loss": 0.5305,
"step": 1150
},
{
"epoch": 0.5807259073842302,
"grad_norm": 0.8600693941116333,
"learning_rate": 8.940472070017944e-06,
"loss": 0.5204,
"step": 1160
},
{
"epoch": 0.5857321652065082,
"grad_norm": 0.8548076152801514,
"learning_rate": 8.913433534224448e-06,
"loss": 0.534,
"step": 1170
},
{
"epoch": 0.590738423028786,
"grad_norm": 0.8594138026237488,
"learning_rate": 8.886096311941045e-06,
"loss": 0.5327,
"step": 1180
},
{
"epoch": 0.5957446808510638,
"grad_norm": 0.795961320400238,
"learning_rate": 8.858462489637052e-06,
"loss": 0.5407,
"step": 1190
},
{
"epoch": 0.6007509386733417,
"grad_norm": 0.8019248247146606,
"learning_rate": 8.830534176419296e-06,
"loss": 0.527,
"step": 1200
},
{
"epoch": 0.6057571964956195,
"grad_norm": 0.8673723340034485,
"learning_rate": 8.802313503871148e-06,
"loss": 0.523,
"step": 1210
},
{
"epoch": 0.6107634543178974,
"grad_norm": 0.9038131833076477,
"learning_rate": 8.773802625889829e-06,
"loss": 0.5275,
"step": 1220
},
{
"epoch": 0.6157697121401752,
"grad_norm": 0.9576230049133301,
"learning_rate": 8.745003718522027e-06,
"loss": 0.5351,
"step": 1230
},
{
"epoch": 0.6207759699624531,
"grad_norm": 0.8945112228393555,
"learning_rate": 8.715918979797803e-06,
"loss": 0.5405,
"step": 1240
},
{
"epoch": 0.6257822277847309,
"grad_norm": 0.7857946753501892,
"learning_rate": 8.686550629562836e-06,
"loss": 0.5249,
"step": 1250
},
{
"epoch": 0.6307884856070087,
"grad_norm": 0.8336262702941895,
"learning_rate": 8.656900909308992e-06,
"loss": 0.5135,
"step": 1260
},
{
"epoch": 0.6357947434292867,
"grad_norm": 0.8685853481292725,
"learning_rate": 8.626972082003256e-06,
"loss": 0.5303,
"step": 1270
},
{
"epoch": 0.6408010012515645,
"grad_norm": 0.923499584197998,
"learning_rate": 8.596766431915001e-06,
"loss": 0.528,
"step": 1280
},
{
"epoch": 0.6458072590738423,
"grad_norm": 0.8822771906852722,
"learning_rate": 8.566286264441655e-06,
"loss": 0.5257,
"step": 1290
},
{
"epoch": 0.6508135168961201,
"grad_norm": 0.9381466507911682,
"learning_rate": 8.535533905932739e-06,
"loss": 0.534,
"step": 1300
},
{
"epoch": 0.655819774718398,
"grad_norm": 0.8062151670455933,
"learning_rate": 8.504511703512316e-06,
"loss": 0.5287,
"step": 1310
},
{
"epoch": 0.6608260325406758,
"grad_norm": 0.76947021484375,
"learning_rate": 8.473222024899854e-06,
"loss": 0.5313,
"step": 1320
},
{
"epoch": 0.6658322903629537,
"grad_norm": 0.8771793842315674,
"learning_rate": 8.441667258229502e-06,
"loss": 0.5297,
"step": 1330
},
{
"epoch": 0.6708385481852316,
"grad_norm": 0.8643835186958313,
"learning_rate": 8.40984981186783e-06,
"loss": 0.5192,
"step": 1340
},
{
"epoch": 0.6758448060075094,
"grad_norm": 0.9325259923934937,
"learning_rate": 8.377772114230011e-06,
"loss": 0.5332,
"step": 1350
},
{
"epoch": 0.6808510638297872,
"grad_norm": 0.8208771347999573,
"learning_rate": 8.345436613594475e-06,
"loss": 0.5384,
"step": 1360
},
{
"epoch": 0.685857321652065,
"grad_norm": 0.955217182636261,
"learning_rate": 8.31284577791605e-06,
"loss": 0.5298,
"step": 1370
},
{
"epoch": 0.690863579474343,
"grad_norm": 0.8638603687286377,
"learning_rate": 8.2800020946376e-06,
"loss": 0.5105,
"step": 1380
},
{
"epoch": 0.6958698372966208,
"grad_norm": 0.8174817562103271,
"learning_rate": 8.246908070500167e-06,
"loss": 0.5274,
"step": 1390
},
{
"epoch": 0.7008760951188986,
"grad_norm": 0.8428131937980652,
"learning_rate": 8.213566231351667e-06,
"loss": 0.5324,
"step": 1400
},
{
"epoch": 0.7058823529411765,
"grad_norm": 0.8647835850715637,
"learning_rate": 8.17997912195408e-06,
"loss": 0.5262,
"step": 1410
},
{
"epoch": 0.7108886107634543,
"grad_norm": 0.8310461640357971,
"learning_rate": 8.146149305789255e-06,
"loss": 0.5176,
"step": 1420
},
{
"epoch": 0.7158948685857321,
"grad_norm": 0.8275470733642578,
"learning_rate": 8.112079364863237e-06,
"loss": 0.5428,
"step": 1430
},
{
"epoch": 0.72090112640801,
"grad_norm": 0.8270696997642517,
"learning_rate": 8.077771899509206e-06,
"loss": 0.5149,
"step": 1440
},
{
"epoch": 0.7259073842302879,
"grad_norm": 0.8921800851821899,
"learning_rate": 8.043229528189008e-06,
"loss": 0.5319,
"step": 1450
},
{
"epoch": 0.7309136420525657,
"grad_norm": 0.8520259857177734,
"learning_rate": 8.008454887293308e-06,
"loss": 0.5164,
"step": 1460
},
{
"epoch": 0.7359198998748435,
"grad_norm": 0.9430526494979858,
"learning_rate": 7.973450630940377e-06,
"loss": 0.5197,
"step": 1470
},
{
"epoch": 0.7409261576971214,
"grad_norm": 0.7611558437347412,
"learning_rate": 7.938219430773508e-06,
"loss": 0.5224,
"step": 1480
},
{
"epoch": 0.7459324155193993,
"grad_norm": 0.9272931218147278,
"learning_rate": 7.902763975757112e-06,
"loss": 0.5269,
"step": 1490
},
{
"epoch": 0.7509386733416771,
"grad_norm": 0.8615291714668274,
"learning_rate": 7.867086971971501e-06,
"loss": 0.5317,
"step": 1500
},
{
"epoch": 0.7559449311639549,
"grad_norm": 0.8272076845169067,
"learning_rate": 7.83119114240632e-06,
"loss": 0.5281,
"step": 1510
},
{
"epoch": 0.7609511889862328,
"grad_norm": 0.7818087935447693,
"learning_rate": 7.795079226752756e-06,
"loss": 0.5277,
"step": 1520
},
{
"epoch": 0.7659574468085106,
"grad_norm": 0.8615156412124634,
"learning_rate": 7.758753981194408e-06,
"loss": 0.5281,
"step": 1530
},
{
"epoch": 0.7709637046307884,
"grad_norm": 0.9051347374916077,
"learning_rate": 7.722218178196936e-06,
"loss": 0.5096,
"step": 1540
},
{
"epoch": 0.7759699624530664,
"grad_norm": 0.9328134059906006,
"learning_rate": 7.68547460629646e-06,
"loss": 0.5155,
"step": 1550
},
{
"epoch": 0.7809762202753442,
"grad_norm": 0.8778051733970642,
"learning_rate": 7.648526069886717e-06,
"loss": 0.5088,
"step": 1560
},
{
"epoch": 0.785982478097622,
"grad_norm": 0.9166744947433472,
"learning_rate": 7.6113753890050425e-06,
"loss": 0.5264,
"step": 1570
},
{
"epoch": 0.7909887359198998,
"grad_norm": 0.8010239601135254,
"learning_rate": 7.574025399117111e-06,
"loss": 0.5272,
"step": 1580
},
{
"epoch": 0.7959949937421777,
"grad_norm": 0.8555699586868286,
"learning_rate": 7.536478950900537e-06,
"loss": 0.516,
"step": 1590
},
{
"epoch": 0.8010012515644556,
"grad_norm": 0.778422474861145,
"learning_rate": 7.498738910027303e-06,
"loss": 0.5206,
"step": 1600
},
{
"epoch": 0.8060075093867334,
"grad_norm": 0.7858911156654358,
"learning_rate": 7.4608081569450365e-06,
"loss": 0.5324,
"step": 1610
},
{
"epoch": 0.8110137672090113,
"grad_norm": 0.9215904474258423,
"learning_rate": 7.422689586657166e-06,
"loss": 0.5181,
"step": 1620
},
{
"epoch": 0.8160200250312891,
"grad_norm": 0.8352310657501221,
"learning_rate": 7.384386108501961e-06,
"loss": 0.5269,
"step": 1630
},
{
"epoch": 0.8210262828535669,
"grad_norm": 0.7926855683326721,
"learning_rate": 7.345900645930493e-06,
"loss": 0.5188,
"step": 1640
},
{
"epoch": 0.8260325406758448,
"grad_norm": 0.7346550226211548,
"learning_rate": 7.307236136283491e-06,
"loss": 0.5244,
"step": 1650
},
{
"epoch": 0.8310387984981227,
"grad_norm": 0.8933218121528625,
"learning_rate": 7.268395530567169e-06,
"loss": 0.5238,
"step": 1660
},
{
"epoch": 0.8360450563204005,
"grad_norm": 0.8537905216217041,
"learning_rate": 7.2293817932279875e-06,
"loss": 0.52,
"step": 1670
},
{
"epoch": 0.8410513141426783,
"grad_norm": 0.8203076124191284,
"learning_rate": 7.190197901926399e-06,
"loss": 0.5219,
"step": 1680
},
{
"epoch": 0.8460575719649562,
"grad_norm": 0.8255800008773804,
"learning_rate": 7.150846847309578e-06,
"loss": 0.5286,
"step": 1690
},
{
"epoch": 0.851063829787234,
"grad_norm": 0.8393645882606506,
"learning_rate": 7.111331632783175e-06,
"loss": 0.5164,
"step": 1700
},
{
"epoch": 0.8560700876095119,
"grad_norm": 0.7761767506599426,
"learning_rate": 7.071655274282074e-06,
"loss": 0.5293,
"step": 1710
},
{
"epoch": 0.8610763454317898,
"grad_norm": 0.9195387959480286,
"learning_rate": 7.031820800040217e-06,
"loss": 0.5376,
"step": 1720
},
{
"epoch": 0.8660826032540676,
"grad_norm": 0.8202856779098511,
"learning_rate": 6.991831250359475e-06,
"loss": 0.5066,
"step": 1730
},
{
"epoch": 0.8710888610763454,
"grad_norm": 0.947016716003418,
"learning_rate": 6.9516896773775976e-06,
"loss": 0.5234,
"step": 1740
},
{
"epoch": 0.8760951188986232,
"grad_norm": 0.9114251136779785,
"learning_rate": 6.911399144835269e-06,
"loss": 0.5181,
"step": 1750
},
{
"epoch": 0.8811013767209012,
"grad_norm": 0.8131252527236938,
"learning_rate": 6.87096272784227e-06,
"loss": 0.5077,
"step": 1760
},
{
"epoch": 0.886107634543179,
"grad_norm": 0.8690458536148071,
"learning_rate": 6.830383512642779e-06,
"loss": 0.5226,
"step": 1770
},
{
"epoch": 0.8911138923654568,
"grad_norm": 0.7998018860816956,
"learning_rate": 6.789664596379812e-06,
"loss": 0.5396,
"step": 1780
},
{
"epoch": 0.8961201501877347,
"grad_norm": 0.8682878017425537,
"learning_rate": 6.7488090868588505e-06,
"loss": 0.5295,
"step": 1790
},
{
"epoch": 0.9011264080100125,
"grad_norm": 0.7566678524017334,
"learning_rate": 6.707820102310627e-06,
"loss": 0.5181,
"step": 1800
},
{
"epoch": 0.9061326658322904,
"grad_norm": 0.9245033264160156,
"learning_rate": 6.6667007711531465e-06,
"loss": 0.5282,
"step": 1810
},
{
"epoch": 0.9111389236545682,
"grad_norm": 0.7727677226066589,
"learning_rate": 6.625454231752904e-06,
"loss": 0.5206,
"step": 1820
},
{
"epoch": 0.9161451814768461,
"grad_norm": 0.8721101880073547,
"learning_rate": 6.5840836321853606e-06,
"loss": 0.507,
"step": 1830
},
{
"epoch": 0.9211514392991239,
"grad_norm": 0.9498814344406128,
"learning_rate": 6.542592129994668e-06,
"loss": 0.5253,
"step": 1840
},
{
"epoch": 0.9261576971214017,
"grad_norm": 0.871817409992218,
"learning_rate": 6.500982891952675e-06,
"loss": 0.5156,
"step": 1850
},
{
"epoch": 0.9311639549436797,
"grad_norm": 0.8299517631530762,
"learning_rate": 6.459259093817229e-06,
"loss": 0.5344,
"step": 1860
},
{
"epoch": 0.9361702127659575,
"grad_norm": 1.0336198806762695,
"learning_rate": 6.417423920089793e-06,
"loss": 0.5034,
"step": 1870
},
{
"epoch": 0.9411764705882353,
"grad_norm": 0.8470144867897034,
"learning_rate": 6.375480563772391e-06,
"loss": 0.5367,
"step": 1880
},
{
"epoch": 0.9461827284105131,
"grad_norm": 0.8725905418395996,
"learning_rate": 6.333432226123908e-06,
"loss": 0.5259,
"step": 1890
},
{
"epoch": 0.951188986232791,
"grad_norm": 0.8204965591430664,
"learning_rate": 6.291282116415759e-06,
"loss": 0.499,
"step": 1900
},
{
"epoch": 0.9561952440550688,
"grad_norm": 0.8008022904396057,
"learning_rate": 6.249033451686946e-06,
"loss": 0.5111,
"step": 1910
},
{
"epoch": 0.9612015018773467,
"grad_norm": 0.856339156627655,
"learning_rate": 6.206689456498529e-06,
"loss": 0.5211,
"step": 1920
},
{
"epoch": 0.9662077596996246,
"grad_norm": 0.8754382729530334,
"learning_rate": 6.164253362687505e-06,
"loss": 0.5032,
"step": 1930
},
{
"epoch": 0.9712140175219024,
"grad_norm": 0.8912584781646729,
"learning_rate": 6.1217284091201535e-06,
"loss": 0.5377,
"step": 1940
},
{
"epoch": 0.9762202753441802,
"grad_norm": 0.8825300931930542,
"learning_rate": 6.079117841444829e-06,
"loss": 0.512,
"step": 1950
},
{
"epoch": 0.981226533166458,
"grad_norm": 0.9316754341125488,
"learning_rate": 6.036424911844245e-06,
"loss": 0.5013,
"step": 1960
},
{
"epoch": 0.986232790988736,
"grad_norm": 0.838261604309082,
"learning_rate": 5.993652878787258e-06,
"loss": 0.5098,
"step": 1970
},
{
"epoch": 0.9912390488110138,
"grad_norm": 0.8722130060195923,
"learning_rate": 5.9508050067801635e-06,
"loss": 0.5063,
"step": 1980
},
{
"epoch": 0.9962453066332916,
"grad_norm": 0.8907979726791382,
"learning_rate": 5.907884566117548e-06,
"loss": 0.4955,
"step": 1990
},
{
"epoch": 1.0010012515644555,
"grad_norm": 0.9894972443580627,
"learning_rate": 5.864894832632677e-06,
"loss": 0.4864,
"step": 2000
},
{
"epoch": 1.0060075093867333,
"grad_norm": 0.9434089660644531,
"learning_rate": 5.821839087447484e-06,
"loss": 0.4714,
"step": 2010
},
{
"epoch": 1.0110137672090114,
"grad_norm": 0.9867864847183228,
"learning_rate": 5.778720616722135e-06,
"loss": 0.4671,
"step": 2020
},
{
"epoch": 1.0160200250312892,
"grad_norm": 0.9105616211891174,
"learning_rate": 5.735542711404226e-06,
"loss": 0.4627,
"step": 2030
},
{
"epoch": 1.021026282853567,
"grad_norm": 0.8702636957168579,
"learning_rate": 5.692308666977598e-06,
"loss": 0.4813,
"step": 2040
},
{
"epoch": 1.0260325406758448,
"grad_norm": 0.882110595703125,
"learning_rate": 5.649021783210822e-06,
"loss": 0.4679,
"step": 2050
},
{
"epoch": 1.0310387984981226,
"grad_norm": 0.9106126427650452,
"learning_rate": 5.60568536390534e-06,
"loss": 0.4765,
"step": 2060
},
{
"epoch": 1.0360450563204004,
"grad_norm": 0.8356176614761353,
"learning_rate": 5.562302716643322e-06,
"loss": 0.4637,
"step": 2070
},
{
"epoch": 1.0410513141426783,
"grad_norm": 0.8483179211616516,
"learning_rate": 5.518877152535207e-06,
"loss": 0.4594,
"step": 2080
},
{
"epoch": 1.0460575719649563,
"grad_norm": 0.8073035478591919,
"learning_rate": 5.475411985966996e-06,
"loss": 0.471,
"step": 2090
},
{
"epoch": 1.0510638297872341,
"grad_norm": 0.8349577784538269,
"learning_rate": 5.431910534347281e-06,
"loss": 0.479,
"step": 2100
},
{
"epoch": 1.056070087609512,
"grad_norm": 0.891203761100769,
"learning_rate": 5.388376117854055e-06,
"loss": 0.4773,
"step": 2110
},
{
"epoch": 1.0610763454317897,
"grad_norm": 1.0728936195373535,
"learning_rate": 5.344812059181295e-06,
"loss": 0.4912,
"step": 2120
},
{
"epoch": 1.0660826032540676,
"grad_norm": 0.8275092244148254,
"learning_rate": 5.30122168328538e-06,
"loss": 0.4646,
"step": 2130
},
{
"epoch": 1.0710888610763454,
"grad_norm": 0.8136346936225891,
"learning_rate": 5.2576083171313e-06,
"loss": 0.459,
"step": 2140
},
{
"epoch": 1.0760951188986232,
"grad_norm": 0.8759229779243469,
"learning_rate": 5.213975289438741e-06,
"loss": 0.4554,
"step": 2150
},
{
"epoch": 1.0811013767209012,
"grad_norm": 0.92725670337677,
"learning_rate": 5.170325930428026e-06,
"loss": 0.4605,
"step": 2160
},
{
"epoch": 1.086107634543179,
"grad_norm": 0.8855956196784973,
"learning_rate": 5.12666357156594e-06,
"loss": 0.4629,
"step": 2170
},
{
"epoch": 1.0911138923654569,
"grad_norm": 0.9248740673065186,
"learning_rate": 5.082991545311459e-06,
"loss": 0.4531,
"step": 2180
},
{
"epoch": 1.0961201501877347,
"grad_norm": 0.9310891628265381,
"learning_rate": 5.039313184861406e-06,
"loss": 0.4672,
"step": 2190
},
{
"epoch": 1.1011264080100125,
"grad_norm": 0.9317619204521179,
"learning_rate": 4.995631823896054e-06,
"loss": 0.4577,
"step": 2200
},
{
"epoch": 1.1061326658322903,
"grad_norm": 0.7926850318908691,
"learning_rate": 4.951950796324681e-06,
"loss": 0.462,
"step": 2210
},
{
"epoch": 1.1111389236545681,
"grad_norm": 0.8007037043571472,
"learning_rate": 4.908273436031128e-06,
"loss": 0.4835,
"step": 2220
},
{
"epoch": 1.1161451814768462,
"grad_norm": 0.7736724615097046,
"learning_rate": 4.864603076619329e-06,
"loss": 0.469,
"step": 2230
},
{
"epoch": 1.121151439299124,
"grad_norm": 0.8057718276977539,
"learning_rate": 4.820943051158886e-06,
"loss": 0.4713,
"step": 2240
},
{
"epoch": 1.1261576971214018,
"grad_norm": 0.871327817440033,
"learning_rate": 4.777296691930686e-06,
"loss": 0.4875,
"step": 2250
},
{
"epoch": 1.1311639549436796,
"grad_norm": 0.8148676753044128,
"learning_rate": 4.7336673301725575e-06,
"loss": 0.456,
"step": 2260
},
{
"epoch": 1.1361702127659574,
"grad_norm": 0.7448599934577942,
"learning_rate": 4.690058295825028e-06,
"loss": 0.4623,
"step": 2270
},
{
"epoch": 1.1411764705882352,
"grad_norm": 0.8734057545661926,
"learning_rate": 4.6464729172771655e-06,
"loss": 0.454,
"step": 2280
},
{
"epoch": 1.146182728410513,
"grad_norm": 0.8993595242500305,
"learning_rate": 4.602914521112545e-06,
"loss": 0.4632,
"step": 2290
},
{
"epoch": 1.1511889862327909,
"grad_norm": 0.8533328771591187,
"learning_rate": 4.559386431855364e-06,
"loss": 0.4674,
"step": 2300
},
{
"epoch": 1.156195244055069,
"grad_norm": 0.7647966146469116,
"learning_rate": 4.515891971716683e-06,
"loss": 0.4739,
"step": 2310
},
{
"epoch": 1.1612015018773467,
"grad_norm": 0.9543892741203308,
"learning_rate": 4.472434460340882e-06,
"loss": 0.4581,
"step": 2320
},
{
"epoch": 1.1662077596996245,
"grad_norm": 0.8194736838340759,
"learning_rate": 4.429017214552285e-06,
"loss": 0.4581,
"step": 2330
},
{
"epoch": 1.1712140175219024,
"grad_norm": 0.8992741107940674,
"learning_rate": 4.385643548102006e-06,
"loss": 0.4588,
"step": 2340
},
{
"epoch": 1.1762202753441802,
"grad_norm": 0.7426264882087708,
"learning_rate": 4.342316771415042e-06,
"loss": 0.4505,
"step": 2350
},
{
"epoch": 1.181226533166458,
"grad_norm": 0.926711916923523,
"learning_rate": 4.2990401913376e-06,
"loss": 0.4564,
"step": 2360
},
{
"epoch": 1.1862327909887358,
"grad_norm": 0.8293261528015137,
"learning_rate": 4.255817110884717e-06,
"loss": 0.4604,
"step": 2370
},
{
"epoch": 1.1912390488110138,
"grad_norm": 0.9200506806373596,
"learning_rate": 4.212650828988153e-06,
"loss": 0.4696,
"step": 2380
},
{
"epoch": 1.1962453066332916,
"grad_norm": 0.8590342402458191,
"learning_rate": 4.169544640244612e-06,
"loss": 0.4754,
"step": 2390
},
{
"epoch": 1.2012515644555695,
"grad_norm": 0.8935486674308777,
"learning_rate": 4.126501834664285e-06,
"loss": 0.4519,
"step": 2400
},
{
"epoch": 1.2062578222778473,
"grad_norm": 0.854565441608429,
"learning_rate": 4.083525697419743e-06,
"loss": 0.4714,
"step": 2410
},
{
"epoch": 1.211264080100125,
"grad_norm": 0.8682575821876526,
"learning_rate": 4.040619508595211e-06,
"loss": 0.4707,
"step": 2420
},
{
"epoch": 1.216270337922403,
"grad_norm": 0.7884920239448547,
"learning_rate": 3.997786542936205e-06,
"loss": 0.4798,
"step": 2430
},
{
"epoch": 1.2212765957446807,
"grad_norm": 0.8475179672241211,
"learning_rate": 3.955030069599606e-06,
"loss": 0.4499,
"step": 2440
},
{
"epoch": 1.2262828535669588,
"grad_norm": 0.7469643354415894,
"learning_rate": 3.912353351904145e-06,
"loss": 0.4753,
"step": 2450
},
{
"epoch": 1.2312891113892366,
"grad_norm": 0.9615768790245056,
"learning_rate": 3.869759647081326e-06,
"loss": 0.4765,
"step": 2460
},
{
"epoch": 1.2362953692115144,
"grad_norm": 0.8711065053939819,
"learning_rate": 3.827252206026835e-06,
"loss": 0.45,
"step": 2470
},
{
"epoch": 1.2413016270337922,
"grad_norm": 0.8308599591255188,
"learning_rate": 3.7848342730524113e-06,
"loss": 0.4632,
"step": 2480
},
{
"epoch": 1.24630788485607,
"grad_norm": 0.9149040579795837,
"learning_rate": 3.742509085638235e-06,
"loss": 0.4684,
"step": 2490
},
{
"epoch": 1.251314142678348,
"grad_norm": 0.8095912337303162,
"learning_rate": 3.700279874185838e-06,
"loss": 0.4828,
"step": 2500
},
{
"epoch": 1.2563204005006257,
"grad_norm": 0.7164251804351807,
"learning_rate": 3.6581498617715306e-06,
"loss": 0.4548,
"step": 2510
},
{
"epoch": 1.2613266583229037,
"grad_norm": 0.857050895690918,
"learning_rate": 3.6161222639004277e-06,
"loss": 0.4706,
"step": 2520
},
{
"epoch": 1.2663329161451815,
"grad_norm": 0.8233107328414917,
"learning_rate": 3.574200288261014e-06,
"loss": 0.4843,
"step": 2530
},
{
"epoch": 1.2713391739674593,
"grad_norm": 0.7947669625282288,
"learning_rate": 3.532387134480327e-06,
"loss": 0.458,
"step": 2540
},
{
"epoch": 1.2763454317897371,
"grad_norm": 0.8212054371833801,
"learning_rate": 3.4906859938797553e-06,
"loss": 0.461,
"step": 2550
},
{
"epoch": 1.281351689612015,
"grad_norm": 0.9112369418144226,
"learning_rate": 3.449100049231459e-06,
"loss": 0.4585,
"step": 2560
},
{
"epoch": 1.286357947434293,
"grad_norm": 0.8596389889717102,
"learning_rate": 3.4076324745154586e-06,
"loss": 0.459,
"step": 2570
},
{
"epoch": 1.2913642052565706,
"grad_norm": 0.7858295440673828,
"learning_rate": 3.3662864346773765e-06,
"loss": 0.477,
"step": 2580
},
{
"epoch": 1.2963704630788486,
"grad_norm": 1.047404408454895,
"learning_rate": 3.3250650853868798e-06,
"loss": 0.4584,
"step": 2590
},
{
"epoch": 1.3013767209011264,
"grad_norm": 0.8477622270584106,
"learning_rate": 3.283971572796839e-06,
"loss": 0.4512,
"step": 2600
},
{
"epoch": 1.3063829787234043,
"grad_norm": 0.8000221848487854,
"learning_rate": 3.2430090333031867e-06,
"loss": 0.4619,
"step": 2610
},
{
"epoch": 1.311389236545682,
"grad_norm": 0.8963533639907837,
"learning_rate": 3.2021805933055517e-06,
"loss": 0.482,
"step": 2620
},
{
"epoch": 1.3163954943679599,
"grad_norm": 0.8562923669815063,
"learning_rate": 3.1614893689686314e-06,
"loss": 0.4616,
"step": 2630
},
{
"epoch": 1.321401752190238,
"grad_norm": 0.9439570903778076,
"learning_rate": 3.12093846598436e-06,
"loss": 0.4699,
"step": 2640
},
{
"epoch": 1.3264080100125155,
"grad_norm": 1.0059826374053955,
"learning_rate": 3.080530979334876e-06,
"loss": 0.4683,
"step": 2650
},
{
"epoch": 1.3314142678347936,
"grad_norm": 0.9622668027877808,
"learning_rate": 3.0402699930562907e-06,
"loss": 0.4686,
"step": 2660
},
{
"epoch": 1.3364205256570714,
"grad_norm": 0.9298732280731201,
"learning_rate": 3.0001585800033184e-06,
"loss": 0.4751,
"step": 2670
},
{
"epoch": 1.3414267834793492,
"grad_norm": 0.8137292265892029,
"learning_rate": 2.9601998016147317e-06,
"loss": 0.4608,
"step": 2680
},
{
"epoch": 1.346433041301627,
"grad_norm": 0.7804547548294067,
"learning_rate": 2.9203967076797156e-06,
"loss": 0.4649,
"step": 2690
},
{
"epoch": 1.3514392991239048,
"grad_norm": 0.8991029262542725,
"learning_rate": 2.880752336105087e-06,
"loss": 0.4593,
"step": 2700
},
{
"epoch": 1.3564455569461829,
"grad_norm": 0.8930824398994446,
"learning_rate": 2.8412697126834345e-06,
"loss": 0.4636,
"step": 2710
},
{
"epoch": 1.3614518147684604,
"grad_norm": 0.8249405026435852,
"learning_rate": 2.8019518508621856e-06,
"loss": 0.4533,
"step": 2720
},
{
"epoch": 1.3664580725907385,
"grad_norm": 0.8421344757080078,
"learning_rate": 2.7628017515135986e-06,
"loss": 0.4629,
"step": 2730
},
{
"epoch": 1.3714643304130163,
"grad_norm": 0.8764386177062988,
"learning_rate": 2.7238224027057337e-06,
"loss": 0.4606,
"step": 2740
},
{
"epoch": 1.3764705882352941,
"grad_norm": 0.9850201606750488,
"learning_rate": 2.6850167794743966e-06,
"loss": 0.4489,
"step": 2750
},
{
"epoch": 1.381476846057572,
"grad_norm": 0.800796627998352,
"learning_rate": 2.6463878435960616e-06,
"loss": 0.4711,
"step": 2760
},
{
"epoch": 1.3864831038798497,
"grad_norm": 0.7282071709632874,
"learning_rate": 2.607938543361835e-06,
"loss": 0.4711,
"step": 2770
},
{
"epoch": 1.3914893617021278,
"grad_norm": 0.8977017402648926,
"learning_rate": 2.5696718133524155e-06,
"loss": 0.4663,
"step": 2780
},
{
"epoch": 1.3964956195244054,
"grad_norm": 0.8676449656486511,
"learning_rate": 2.5315905742141277e-06,
"loss": 0.4617,
"step": 2790
},
{
"epoch": 1.4015018773466834,
"grad_norm": 0.9089524745941162,
"learning_rate": 2.4936977324360053e-06,
"loss": 0.4636,
"step": 2800
},
{
"epoch": 1.4065081351689612,
"grad_norm": 1.0042393207550049,
"learning_rate": 2.455996180127954e-06,
"loss": 0.4658,
"step": 2810
},
{
"epoch": 1.411514392991239,
"grad_norm": 0.8231027722358704,
"learning_rate": 2.418488794800029e-06,
"loss": 0.4442,
"step": 2820
},
{
"epoch": 1.4165206508135169,
"grad_norm": 0.8723143339157104,
"learning_rate": 2.381178439142796e-06,
"loss": 0.4507,
"step": 2830
},
{
"epoch": 1.4215269086357947,
"grad_norm": 0.8935348987579346,
"learning_rate": 2.34406796080885e-06,
"loss": 0.4691,
"step": 2840
},
{
"epoch": 1.4265331664580727,
"grad_norm": 0.812369167804718,
"learning_rate": 2.3071601921954797e-06,
"loss": 0.4556,
"step": 2850
},
{
"epoch": 1.4315394242803503,
"grad_norm": 0.8182671070098877,
"learning_rate": 2.2704579502284713e-06,
"loss": 0.4468,
"step": 2860
},
{
"epoch": 1.4365456821026283,
"grad_norm": 0.7253329157829285,
"learning_rate": 2.233964036147133e-06,
"loss": 0.4481,
"step": 2870
},
{
"epoch": 1.4415519399249062,
"grad_norm": 0.8144270181655884,
"learning_rate": 2.197681235290476e-06,
"loss": 0.4668,
"step": 2880
},
{
"epoch": 1.446558197747184,
"grad_norm": 0.8446046710014343,
"learning_rate": 2.161612316884636e-06,
"loss": 0.4681,
"step": 2890
},
{
"epoch": 1.4515644555694618,
"grad_norm": 0.9348142743110657,
"learning_rate": 2.125760033831522e-06,
"loss": 0.4641,
"step": 2900
},
{
"epoch": 1.4565707133917396,
"grad_norm": 0.8982442617416382,
"learning_rate": 2.0901271224986953e-06,
"loss": 0.4594,
"step": 2910
},
{
"epoch": 1.4615769712140176,
"grad_norm": 0.9543260931968689,
"learning_rate": 2.054716302510525e-06,
"loss": 0.4661,
"step": 2920
},
{
"epoch": 1.4665832290362952,
"grad_norm": 0.8085075616836548,
"learning_rate": 2.019530276540619e-06,
"loss": 0.4507,
"step": 2930
},
{
"epoch": 1.4715894868585733,
"grad_norm": 0.8637529611587524,
"learning_rate": 1.984571730105542e-06,
"loss": 0.4752,
"step": 2940
},
{
"epoch": 1.476595744680851,
"grad_norm": 0.8390467762947083,
"learning_rate": 1.9498433313598587e-06,
"loss": 0.457,
"step": 2950
},
{
"epoch": 1.481602002503129,
"grad_norm": 0.7390202283859253,
"learning_rate": 1.9153477308924723e-06,
"loss": 0.4534,
"step": 2960
},
{
"epoch": 1.4866082603254067,
"grad_norm": 1.034041404724121,
"learning_rate": 1.8810875615243434e-06,
"loss": 0.4586,
"step": 2970
},
{
"epoch": 1.4916145181476845,
"grad_norm": 0.9007934331893921,
"learning_rate": 1.8470654381075288e-06,
"loss": 0.4703,
"step": 2980
},
{
"epoch": 1.4966207759699626,
"grad_norm": 0.8302955031394958,
"learning_rate": 1.8132839573256128e-06,
"loss": 0.4629,
"step": 2990
},
{
"epoch": 1.5016270337922402,
"grad_norm": 0.8993987441062927,
"learning_rate": 1.779745697495524e-06,
"loss": 0.4656,
"step": 3000
},
{
"epoch": 1.5066332916145182,
"grad_norm": 0.8692158460617065,
"learning_rate": 1.7464532183707383e-06,
"loss": 0.4749,
"step": 3010
},
{
"epoch": 1.511639549436796,
"grad_norm": 0.8215277194976807,
"learning_rate": 1.7134090609459243e-06,
"loss": 0.4603,
"step": 3020
},
{
"epoch": 1.5166458072590738,
"grad_norm": 1.0028440952301025,
"learning_rate": 1.6806157472629913e-06,
"loss": 0.4642,
"step": 3030
},
{
"epoch": 1.5216520650813516,
"grad_norm": 0.9103449583053589,
"learning_rate": 1.648075780218607e-06,
"loss": 0.4655,
"step": 3040
},
{
"epoch": 1.5266583229036295,
"grad_norm": 0.8517196178436279,
"learning_rate": 1.6157916433731707e-06,
"loss": 0.4585,
"step": 3050
},
{
"epoch": 1.5316645807259075,
"grad_norm": 0.7916694283485413,
"learning_rate": 1.58376580076125e-06,
"loss": 0.4517,
"step": 3060
},
{
"epoch": 1.536670838548185,
"grad_norm": 0.9230484962463379,
"learning_rate": 1.5520006967035245e-06,
"loss": 0.4431,
"step": 3070
},
{
"epoch": 1.5416770963704631,
"grad_norm": 0.8834196925163269,
"learning_rate": 1.5204987556202256e-06,
"loss": 0.4578,
"step": 3080
},
{
"epoch": 1.546683354192741,
"grad_norm": 0.9575162529945374,
"learning_rate": 1.4892623818460926e-06,
"loss": 0.4604,
"step": 3090
},
{
"epoch": 1.5516896120150188,
"grad_norm": 0.8710171580314636,
"learning_rate": 1.4582939594468765e-06,
"loss": 0.467,
"step": 3100
},
{
"epoch": 1.5566958698372966,
"grad_norm": 0.8777572512626648,
"learning_rate": 1.427595852037364e-06,
"loss": 0.4639,
"step": 3110
},
{
"epoch": 1.5617021276595744,
"grad_norm": 0.9340876936912537,
"learning_rate": 1.3971704026009946e-06,
"loss": 0.4701,
"step": 3120
},
{
"epoch": 1.5667083854818524,
"grad_norm": 0.7884568572044373,
"learning_rate": 1.367019933311024e-06,
"loss": 0.4676,
"step": 3130
},
{
"epoch": 1.57171464330413,
"grad_norm": 0.7757740020751953,
"learning_rate": 1.3371467453532916e-06,
"loss": 0.4626,
"step": 3140
},
{
"epoch": 1.576720901126408,
"grad_norm": 0.7005230188369751,
"learning_rate": 1.307553118750593e-06,
"loss": 0.4645,
"step": 3150
},
{
"epoch": 1.5817271589486859,
"grad_norm": 0.8018795251846313,
"learning_rate": 1.2782413121886483e-06,
"loss": 0.4665,
"step": 3160
},
{
"epoch": 1.5867334167709637,
"grad_norm": 0.9680582880973816,
"learning_rate": 1.2492135628437252e-06,
"loss": 0.4459,
"step": 3170
},
{
"epoch": 1.5917396745932415,
"grad_norm": 0.8594993352890015,
"learning_rate": 1.220472086211878e-06,
"loss": 0.4642,
"step": 3180
},
{
"epoch": 1.5967459324155193,
"grad_norm": 0.7460825443267822,
"learning_rate": 1.1920190759398591e-06,
"loss": 0.4487,
"step": 3190
},
{
"epoch": 1.6017521902377974,
"grad_norm": 0.9065592885017395,
"learning_rate": 1.1638567036576982e-06,
"loss": 0.4723,
"step": 3200
},
{
"epoch": 1.606758448060075,
"grad_norm": 0.8521935939788818,
"learning_rate": 1.135987118812943e-06,
"loss": 0.4793,
"step": 3210
},
{
"epoch": 1.611764705882353,
"grad_norm": 0.9237802028656006,
"learning_rate": 1.1084124485066166e-06,
"loss": 0.4536,
"step": 3220
},
{
"epoch": 1.6167709637046308,
"grad_norm": 0.802192211151123,
"learning_rate": 1.0811347973308644e-06,
"loss": 0.4669,
"step": 3230
},
{
"epoch": 1.6217772215269086,
"grad_norm": 0.8151496648788452,
"learning_rate": 1.0541562472083316e-06,
"loss": 0.4722,
"step": 3240
},
{
"epoch": 1.6267834793491864,
"grad_norm": 0.8261227607727051,
"learning_rate": 1.0274788572332534e-06,
"loss": 0.4695,
"step": 3250
},
{
"epoch": 1.6317897371714642,
"grad_norm": 0.828486979007721,
"learning_rate": 1.0011046635143045e-06,
"loss": 0.4563,
"step": 3260
},
{
"epoch": 1.6367959949937423,
"grad_norm": 0.8518195152282715,
"learning_rate": 9.750356790192e-07,
"loss": 0.4602,
"step": 3270
},
{
"epoch": 1.6418022528160199,
"grad_norm": 0.8259122371673584,
"learning_rate": 9.492738934210472e-07,
"loss": 0.4692,
"step": 3280
},
{
"epoch": 1.646808510638298,
"grad_norm": 0.8876679539680481,
"learning_rate": 9.238212729465002e-07,
"loss": 0.4702,
"step": 3290
},
{
"epoch": 1.6518147684605757,
"grad_norm": 0.9750533699989319,
"learning_rate": 8.986797602256819e-07,
"loss": 0.4702,
"step": 3300
},
{
"epoch": 1.6568210262828535,
"grad_norm": 0.9484962821006775,
"learning_rate": 8.73851274143916e-07,
"loss": 0.464,
"step": 3310
},
{
"epoch": 1.6618272841051314,
"grad_norm": 0.7608646154403687,
"learning_rate": 8.493377096952809e-07,
"loss": 0.4624,
"step": 3320
},
{
"epoch": 1.6668335419274092,
"grad_norm": 0.8601351976394653,
"learning_rate": 8.251409378379638e-07,
"loss": 0.4637,
"step": 3330
},
{
"epoch": 1.6718397997496872,
"grad_norm": 0.7652588486671448,
"learning_rate": 8.012628053514715e-07,
"loss": 0.4476,
"step": 3340
},
{
"epoch": 1.6768460575719648,
"grad_norm": 0.886549174785614,
"learning_rate": 7.777051346956771e-07,
"loss": 0.4526,
"step": 3350
},
{
"epoch": 1.6818523153942428,
"grad_norm": 0.8611428141593933,
"learning_rate": 7.544697238717191e-07,
"loss": 0.4599,
"step": 3360
},
{
"epoch": 1.6868585732165207,
"grad_norm": 0.766806423664093,
"learning_rate": 7.315583462847808e-07,
"loss": 0.4584,
"step": 3370
},
{
"epoch": 1.6918648310387985,
"grad_norm": 0.9067396521568298,
"learning_rate": 7.089727506087251e-07,
"loss": 0.4519,
"step": 3380
},
{
"epoch": 1.6968710888610763,
"grad_norm": 0.8935049772262573,
"learning_rate": 6.867146606526442e-07,
"loss": 0.4504,
"step": 3390
},
{
"epoch": 1.701877346683354,
"grad_norm": 0.7634616494178772,
"learning_rate": 6.647857752292802e-07,
"loss": 0.4516,
"step": 3400
},
{
"epoch": 1.7068836045056321,
"grad_norm": 0.6964467763900757,
"learning_rate": 6.431877680253729e-07,
"loss": 0.4706,
"step": 3410
},
{
"epoch": 1.7118898623279097,
"grad_norm": 0.8620052933692932,
"learning_rate": 6.219222874739189e-07,
"loss": 0.4474,
"step": 3420
},
{
"epoch": 1.7168961201501878,
"grad_norm": 0.8637686967849731,
"learning_rate": 6.009909566283511e-07,
"loss": 0.4459,
"step": 3430
},
{
"epoch": 1.7219023779724656,
"grad_norm": 0.930946409702301,
"learning_rate": 5.803953730386713e-07,
"loss": 0.4695,
"step": 3440
},
{
"epoch": 1.7269086357947434,
"grad_norm": 0.8325541019439697,
"learning_rate": 5.601371086295109e-07,
"loss": 0.4515,
"step": 3450
},
{
"epoch": 1.7319148936170212,
"grad_norm": 0.7173551917076111,
"learning_rate": 5.402177095801598e-07,
"loss": 0.4606,
"step": 3460
},
{
"epoch": 1.736921151439299,
"grad_norm": 0.9236282706260681,
"learning_rate": 5.206386962065601e-07,
"loss": 0.4587,
"step": 3470
},
{
"epoch": 1.741927409261577,
"grad_norm": 0.892206072807312,
"learning_rate": 5.014015628452646e-07,
"loss": 0.4521,
"step": 3480
},
{
"epoch": 1.7469336670838547,
"grad_norm": 0.8856966495513916,
"learning_rate": 4.825077777393877e-07,
"loss": 0.457,
"step": 3490
},
{
"epoch": 1.7519399249061327,
"grad_norm": 0.8615520596504211,
"learning_rate": 4.639587829265424e-07,
"loss": 0.4473,
"step": 3500
},
{
"epoch": 1.7569461827284105,
"grad_norm": 0.7531428337097168,
"learning_rate": 4.4575599412877925e-07,
"loss": 0.4609,
"step": 3510
},
{
"epoch": 1.7619524405506883,
"grad_norm": 0.7816721796989441,
"learning_rate": 4.2790080064453787e-07,
"loss": 0.4524,
"step": 3520
},
{
"epoch": 1.7669586983729664,
"grad_norm": 0.804458498954773,
"learning_rate": 4.1039456524260225e-07,
"loss": 0.4775,
"step": 3530
},
{
"epoch": 1.771964956195244,
"grad_norm": 0.8876965045928955,
"learning_rate": 3.932386240580993e-07,
"loss": 0.4547,
"step": 3540
},
{
"epoch": 1.776971214017522,
"grad_norm": 0.8259099721908569,
"learning_rate": 3.7643428649051294e-07,
"loss": 0.4646,
"step": 3550
},
{
"epoch": 1.7819774718397996,
"grad_norm": 0.8717869520187378,
"learning_rate": 3.5998283510374876e-07,
"loss": 0.4677,
"step": 3560
},
{
"epoch": 1.7869837296620776,
"grad_norm": 1.0038467645645142,
"learning_rate": 3.438855255282486e-07,
"loss": 0.4486,
"step": 3570
},
{
"epoch": 1.7919899874843555,
"grad_norm": 0.7714816927909851,
"learning_rate": 3.281435863651483e-07,
"loss": 0.4607,
"step": 3580
},
{
"epoch": 1.7969962453066333,
"grad_norm": 0.8915358781814575,
"learning_rate": 3.127582190925155e-07,
"loss": 0.4529,
"step": 3590
},
{
"epoch": 1.8020025031289113,
"grad_norm": 0.8906078338623047,
"learning_rate": 2.9773059797364225e-07,
"loss": 0.4418,
"step": 3600
},
{
"epoch": 1.807008760951189,
"grad_norm": 0.8507105112075806,
"learning_rate": 2.83061869967422e-07,
"loss": 0.4654,
"step": 3610
},
{
"epoch": 1.812015018773467,
"grad_norm": 0.8758144378662109,
"learning_rate": 2.6875315464081566e-07,
"loss": 0.4537,
"step": 3620
},
{
"epoch": 1.8170212765957445,
"grad_norm": 0.8995975255966187,
"learning_rate": 2.548055440833952e-07,
"loss": 0.4518,
"step": 3630
},
{
"epoch": 1.8220275344180226,
"grad_norm": 0.9077524542808533,
"learning_rate": 2.4122010282399464e-07,
"loss": 0.4482,
"step": 3640
},
{
"epoch": 1.8270337922403004,
"grad_norm": 0.7204959392547607,
"learning_rate": 2.279978677494632e-07,
"loss": 0.4591,
"step": 3650
},
{
"epoch": 1.8320400500625782,
"grad_norm": 0.8752183318138123,
"learning_rate": 2.151398480255229e-07,
"loss": 0.4672,
"step": 3660
},
{
"epoch": 1.8370463078848562,
"grad_norm": 2.632391929626465,
"learning_rate": 2.026470250197504e-07,
"loss": 0.4695,
"step": 3670
},
{
"epoch": 1.8420525657071338,
"grad_norm": 0.8110368847846985,
"learning_rate": 1.9052035222667075e-07,
"loss": 0.4495,
"step": 3680
},
{
"epoch": 1.8470588235294119,
"grad_norm": 0.8178061246871948,
"learning_rate": 1.7876075519498715e-07,
"loss": 0.4606,
"step": 3690
},
{
"epoch": 1.8520650813516895,
"grad_norm": 0.8804686665534973,
"learning_rate": 1.673691314569381e-07,
"loss": 0.4664,
"step": 3700
},
{
"epoch": 1.8570713391739675,
"grad_norm": 0.8992244601249695,
"learning_rate": 1.563463504597945e-07,
"loss": 0.4779,
"step": 3710
},
{
"epoch": 1.8620775969962453,
"grad_norm": 0.9412287473678589,
"learning_rate": 1.4569325349950314e-07,
"loss": 0.4781,
"step": 3720
},
{
"epoch": 1.8670838548185231,
"grad_norm": 0.7709740996360779,
"learning_rate": 1.3541065365647087e-07,
"loss": 0.4613,
"step": 3730
},
{
"epoch": 1.8720901126408012,
"grad_norm": 0.8421194553375244,
"learning_rate": 1.254993357335138e-07,
"loss": 0.477,
"step": 3740
},
{
"epoch": 1.8770963704630788,
"grad_norm": 1.5245757102966309,
"learning_rate": 1.1596005619595507e-07,
"loss": 0.4534,
"step": 3750
},
{
"epoch": 1.8821026282853568,
"grad_norm": 0.8000764846801758,
"learning_rate": 1.0679354311388778e-07,
"loss": 0.4534,
"step": 3760
},
{
"epoch": 1.8871088861076344,
"grad_norm": 0.9021337628364563,
"learning_rate": 9.80004961066089e-08,
"loss": 0.4419,
"step": 3770
},
{
"epoch": 1.8921151439299124,
"grad_norm": 0.865584135055542,
"learning_rate": 8.95815862892202e-08,
"loss": 0.4638,
"step": 3780
},
{
"epoch": 1.8971214017521902,
"grad_norm": 0.9198890924453735,
"learning_rate": 8.15374562214083e-08,
"loss": 0.4442,
"step": 3790
},
{
"epoch": 1.902127659574468,
"grad_norm": 0.7920567393302917,
"learning_rate": 7.386871985839982e-08,
"loss": 0.4517,
"step": 3800
},
{
"epoch": 1.907133917396746,
"grad_norm": 0.919378399848938,
"learning_rate": 6.657596250410403e-08,
"loss": 0.4425,
"step": 3810
},
{
"epoch": 1.9121401752190237,
"grad_norm": 0.9356231093406677,
"learning_rate": 5.96597407664401e-08,
"loss": 0.4582,
"step": 3820
},
{
"epoch": 1.9171464330413017,
"grad_norm": 0.9223675727844238,
"learning_rate": 5.3120582514854527e-08,
"loss": 0.4462,
"step": 3830
},
{
"epoch": 1.9221526908635793,
"grad_norm": 0.8920125961303711,
"learning_rate": 4.6958986840033305e-08,
"loss": 0.4428,
"step": 3840
},
{
"epoch": 1.9271589486858574,
"grad_norm": 0.7659106850624084,
"learning_rate": 4.117542401580798e-08,
"loss": 0.4454,
"step": 3850
},
{
"epoch": 1.9321652065081352,
"grad_norm": 0.9413115382194519,
"learning_rate": 3.5770335463263805e-08,
"loss": 0.4525,
"step": 3860
},
{
"epoch": 1.937171464330413,
"grad_norm": 0.9268300533294678,
"learning_rate": 3.0744133717049984e-08,
"loss": 0.4596,
"step": 3870
},
{
"epoch": 1.942177722152691,
"grad_norm": 0.9277012944221497,
"learning_rate": 2.609720239389102e-08,
"loss": 0.4508,
"step": 3880
},
{
"epoch": 1.9471839799749686,
"grad_norm": 0.879331111907959,
"learning_rate": 2.1829896163311215e-08,
"loss": 0.4506,
"step": 3890
},
{
"epoch": 1.9521902377972467,
"grad_norm": 0.8949812650680542,
"learning_rate": 1.7942540720560787e-08,
"loss": 0.4731,
"step": 3900
},
{
"epoch": 1.9571964956195242,
"grad_norm": 0.7495149970054626,
"learning_rate": 1.4435432761762958e-08,
"loss": 0.4606,
"step": 3910
},
{
"epoch": 1.9622027534418023,
"grad_norm": 0.9262397885322571,
"learning_rate": 1.1308839961263196e-08,
"loss": 0.4652,
"step": 3920
},
{
"epoch": 1.96720901126408,
"grad_norm": 0.7670968174934387,
"learning_rate": 8.563000951204992e-09,
"loss": 0.4701,
"step": 3930
},
{
"epoch": 1.972215269086358,
"grad_norm": 0.9294739365577698,
"learning_rate": 6.198125303313318e-09,
"loss": 0.4759,
"step": 3940
},
{
"epoch": 1.977221526908636,
"grad_norm": 0.8033120036125183,
"learning_rate": 4.21439351289965e-09,
"loss": 0.4561,
"step": 3950
},
{
"epoch": 1.9822277847309135,
"grad_norm": 0.85311359167099,
"learning_rate": 2.611956985086317e-09,
"loss": 0.47,
"step": 3960
},
{
"epoch": 1.9872340425531916,
"grad_norm": 0.8222494125366211,
"learning_rate": 1.3909380232529635e-09,
"loss": 0.4707,
"step": 3970
},
{
"epoch": 1.9922403003754692,
"grad_norm": 0.7678888440132141,
"learning_rate": 5.514298196973577e-10,
"loss": 0.4574,
"step": 3980
},
{
"epoch": 1.9972465581977472,
"grad_norm": 0.8612265586853027,
"learning_rate": 9.34964485271861e-11,
"loss": 0.4748,
"step": 3990
},
{
"epoch": 2.0,
"step": 3996,
"total_flos": 267353655738368.0,
"train_loss": 0.5035881258703925,
"train_runtime": 67851.6206,
"train_samples_per_second": 0.942,
"train_steps_per_second": 0.059
}
],
"logging_steps": 10,
"max_steps": 3996,
"num_input_tokens_seen": 0,
"num_train_epochs": 2,
"save_steps": 500,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 267353655738368.0,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}