gemma7b-summarize-gpt4o-80k / trainer_state.json
chansung's picture
Model save
0cbb00e verified
raw
history blame
93.2 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 9.98185117967332,
"eval_steps": 500,
"global_step": 2750,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.003629764065335753,
"grad_norm": 516.0,
"learning_rate": 7.272727272727273e-07,
"loss": 51.5982,
"step": 1
},
{
"epoch": 0.018148820326678767,
"grad_norm": 454.0,
"learning_rate": 3.636363636363636e-06,
"loss": 52.3483,
"step": 5
},
{
"epoch": 0.036297640653357534,
"grad_norm": 536.0,
"learning_rate": 7.272727272727272e-06,
"loss": 53.1669,
"step": 10
},
{
"epoch": 0.0544464609800363,
"grad_norm": 440.0,
"learning_rate": 1.0909090909090909e-05,
"loss": 46.8624,
"step": 15
},
{
"epoch": 0.07259528130671507,
"grad_norm": 306.0,
"learning_rate": 1.4545454545454545e-05,
"loss": 35.5942,
"step": 20
},
{
"epoch": 0.09074410163339383,
"grad_norm": 41.5,
"learning_rate": 1.8181818181818182e-05,
"loss": 25.8916,
"step": 25
},
{
"epoch": 0.1088929219600726,
"grad_norm": 44.0,
"learning_rate": 2.1818181818181818e-05,
"loss": 24.4549,
"step": 30
},
{
"epoch": 0.12704174228675136,
"grad_norm": 24.375,
"learning_rate": 2.5454545454545454e-05,
"loss": 22.3596,
"step": 35
},
{
"epoch": 0.14519056261343014,
"grad_norm": 20.0,
"learning_rate": 2.909090909090909e-05,
"loss": 21.008,
"step": 40
},
{
"epoch": 0.16333938294010888,
"grad_norm": 7.4375,
"learning_rate": 3.272727272727273e-05,
"loss": 20.0885,
"step": 45
},
{
"epoch": 0.18148820326678766,
"grad_norm": 7.28125,
"learning_rate": 3.6363636363636364e-05,
"loss": 19.1877,
"step": 50
},
{
"epoch": 0.1996370235934664,
"grad_norm": 10.25,
"learning_rate": 4e-05,
"loss": 18.5964,
"step": 55
},
{
"epoch": 0.2177858439201452,
"grad_norm": 14.5,
"learning_rate": 4.3636363636363636e-05,
"loss": 17.6581,
"step": 60
},
{
"epoch": 0.23593466424682397,
"grad_norm": 26.5,
"learning_rate": 4.7272727272727275e-05,
"loss": 16.1181,
"step": 65
},
{
"epoch": 0.2540834845735027,
"grad_norm": 49.75,
"learning_rate": 5.090909090909091e-05,
"loss": 13.6493,
"step": 70
},
{
"epoch": 0.27223230490018147,
"grad_norm": 50.75,
"learning_rate": 5.4545454545454546e-05,
"loss": 7.9128,
"step": 75
},
{
"epoch": 0.29038112522686027,
"grad_norm": 6.625,
"learning_rate": 5.818181818181818e-05,
"loss": 2.7543,
"step": 80
},
{
"epoch": 0.308529945553539,
"grad_norm": 12.375,
"learning_rate": 6.181818181818182e-05,
"loss": 2.0995,
"step": 85
},
{
"epoch": 0.32667876588021777,
"grad_norm": 2.53125,
"learning_rate": 6.545454545454546e-05,
"loss": 1.8007,
"step": 90
},
{
"epoch": 0.3448275862068966,
"grad_norm": 1.984375,
"learning_rate": 6.90909090909091e-05,
"loss": 1.6358,
"step": 95
},
{
"epoch": 0.3629764065335753,
"grad_norm": 2.828125,
"learning_rate": 7.272727272727273e-05,
"loss": 1.509,
"step": 100
},
{
"epoch": 0.3811252268602541,
"grad_norm": 4.0,
"learning_rate": 7.636363636363637e-05,
"loss": 1.4215,
"step": 105
},
{
"epoch": 0.3992740471869328,
"grad_norm": 3.21875,
"learning_rate": 8e-05,
"loss": 1.3173,
"step": 110
},
{
"epoch": 0.41742286751361163,
"grad_norm": 2.625,
"learning_rate": 8.363636363636364e-05,
"loss": 1.2786,
"step": 115
},
{
"epoch": 0.4355716878402904,
"grad_norm": 2.8125,
"learning_rate": 8.727272727272727e-05,
"loss": 1.2548,
"step": 120
},
{
"epoch": 0.4537205081669691,
"grad_norm": 2.96875,
"learning_rate": 9.090909090909092e-05,
"loss": 1.1995,
"step": 125
},
{
"epoch": 0.47186932849364793,
"grad_norm": 1.390625,
"learning_rate": 9.454545454545455e-05,
"loss": 1.1818,
"step": 130
},
{
"epoch": 0.4900181488203267,
"grad_norm": 5.78125,
"learning_rate": 9.818181818181818e-05,
"loss": 1.1444,
"step": 135
},
{
"epoch": 0.5081669691470054,
"grad_norm": 7.34375,
"learning_rate": 0.00010181818181818181,
"loss": 1.0977,
"step": 140
},
{
"epoch": 0.5263157894736842,
"grad_norm": 5.1875,
"learning_rate": 0.00010545454545454545,
"loss": 1.0818,
"step": 145
},
{
"epoch": 0.5444646098003629,
"grad_norm": 1.984375,
"learning_rate": 0.00010909090909090909,
"loss": 1.0524,
"step": 150
},
{
"epoch": 0.5626134301270418,
"grad_norm": 20.5,
"learning_rate": 0.00011272727272727272,
"loss": 1.0397,
"step": 155
},
{
"epoch": 0.5807622504537205,
"grad_norm": 7.71875,
"learning_rate": 0.00011636363636363636,
"loss": 1.0695,
"step": 160
},
{
"epoch": 0.5989110707803993,
"grad_norm": 5.28125,
"learning_rate": 0.00012,
"loss": 1.0426,
"step": 165
},
{
"epoch": 0.617059891107078,
"grad_norm": 1.453125,
"learning_rate": 0.00012363636363636364,
"loss": 1.0295,
"step": 170
},
{
"epoch": 0.6352087114337568,
"grad_norm": 17.75,
"learning_rate": 0.00012727272727272728,
"loss": 0.9987,
"step": 175
},
{
"epoch": 0.6533575317604355,
"grad_norm": 1.3515625,
"learning_rate": 0.00013090909090909093,
"loss": 0.9895,
"step": 180
},
{
"epoch": 0.6715063520871143,
"grad_norm": 1.2421875,
"learning_rate": 0.00013454545454545455,
"loss": 1.0001,
"step": 185
},
{
"epoch": 0.6896551724137931,
"grad_norm": 1.3828125,
"learning_rate": 0.0001381818181818182,
"loss": 0.9797,
"step": 190
},
{
"epoch": 0.7078039927404719,
"grad_norm": 0.84375,
"learning_rate": 0.00014181818181818184,
"loss": 0.9846,
"step": 195
},
{
"epoch": 0.7259528130671506,
"grad_norm": 1.8671875,
"learning_rate": 0.00014545454545454546,
"loss": 0.9978,
"step": 200
},
{
"epoch": 0.7441016333938294,
"grad_norm": 1.1640625,
"learning_rate": 0.0001490909090909091,
"loss": 0.9639,
"step": 205
},
{
"epoch": 0.7622504537205081,
"grad_norm": 0.78125,
"learning_rate": 0.00015272727272727275,
"loss": 0.9302,
"step": 210
},
{
"epoch": 0.7803992740471869,
"grad_norm": 2.390625,
"learning_rate": 0.00015636363636363637,
"loss": 0.9798,
"step": 215
},
{
"epoch": 0.7985480943738656,
"grad_norm": 2.390625,
"learning_rate": 0.00016,
"loss": 0.9477,
"step": 220
},
{
"epoch": 0.8166969147005445,
"grad_norm": 3.265625,
"learning_rate": 0.00016363636363636366,
"loss": 0.9237,
"step": 225
},
{
"epoch": 0.8348457350272233,
"grad_norm": 4.34375,
"learning_rate": 0.00016727272727272728,
"loss": 0.9335,
"step": 230
},
{
"epoch": 0.852994555353902,
"grad_norm": 3.640625,
"learning_rate": 0.0001709090909090909,
"loss": 0.9452,
"step": 235
},
{
"epoch": 0.8711433756805808,
"grad_norm": 3.265625,
"learning_rate": 0.00017454545454545454,
"loss": 0.9112,
"step": 240
},
{
"epoch": 0.8892921960072595,
"grad_norm": 3.796875,
"learning_rate": 0.0001781818181818182,
"loss": 0.9211,
"step": 245
},
{
"epoch": 0.9074410163339383,
"grad_norm": 6.125,
"learning_rate": 0.00018181818181818183,
"loss": 0.9188,
"step": 250
},
{
"epoch": 0.925589836660617,
"grad_norm": 2.609375,
"learning_rate": 0.00018545454545454545,
"loss": 0.937,
"step": 255
},
{
"epoch": 0.9437386569872959,
"grad_norm": 3.03125,
"learning_rate": 0.0001890909090909091,
"loss": 0.9138,
"step": 260
},
{
"epoch": 0.9618874773139746,
"grad_norm": 1.875,
"learning_rate": 0.00019272727272727274,
"loss": 0.9222,
"step": 265
},
{
"epoch": 0.9800362976406534,
"grad_norm": 3.84375,
"learning_rate": 0.00019636363636363636,
"loss": 0.9238,
"step": 270
},
{
"epoch": 0.9981851179673321,
"grad_norm": 2.6875,
"learning_rate": 0.0002,
"loss": 0.9152,
"step": 275
},
{
"epoch": 0.9981851179673321,
"eval_loss": 2.195019245147705,
"eval_runtime": 1.0238,
"eval_samples_per_second": 4.884,
"eval_steps_per_second": 1.953,
"step": 275
},
{
"epoch": 1.0163339382940109,
"grad_norm": 1.875,
"learning_rate": 0.00019999798600729064,
"loss": 0.8799,
"step": 280
},
{
"epoch": 1.0344827586206897,
"grad_norm": 4.28125,
"learning_rate": 0.00019999194411028594,
"loss": 0.87,
"step": 285
},
{
"epoch": 1.0526315789473684,
"grad_norm": 3.28125,
"learning_rate": 0.0001999818745523526,
"loss": 0.8781,
"step": 290
},
{
"epoch": 1.0707803992740472,
"grad_norm": 0.87109375,
"learning_rate": 0.00019996777773909093,
"loss": 0.8886,
"step": 295
},
{
"epoch": 1.0889292196007259,
"grad_norm": 4.375,
"learning_rate": 0.00019994965423831854,
"loss": 0.9042,
"step": 300
},
{
"epoch": 1.1070780399274047,
"grad_norm": 1.2265625,
"learning_rate": 0.00019992750478004738,
"loss": 0.8569,
"step": 305
},
{
"epoch": 1.1252268602540836,
"grad_norm": 1.828125,
"learning_rate": 0.0001999013302564544,
"loss": 0.8641,
"step": 310
},
{
"epoch": 1.1433756805807622,
"grad_norm": 1.03125,
"learning_rate": 0.00019987113172184563,
"loss": 0.8521,
"step": 315
},
{
"epoch": 1.161524500907441,
"grad_norm": 1.15625,
"learning_rate": 0.00019983691039261357,
"loss": 0.8658,
"step": 320
},
{
"epoch": 1.1796733212341197,
"grad_norm": 12.5,
"learning_rate": 0.00019979866764718843,
"loss": 0.8702,
"step": 325
},
{
"epoch": 1.1978221415607986,
"grad_norm": 1.8203125,
"learning_rate": 0.00019975640502598244,
"loss": 0.874,
"step": 330
},
{
"epoch": 1.2159709618874772,
"grad_norm": 2.109375,
"learning_rate": 0.00019971012423132775,
"loss": 0.8552,
"step": 335
},
{
"epoch": 1.234119782214156,
"grad_norm": 4.5625,
"learning_rate": 0.00019965982712740808,
"loss": 0.8459,
"step": 340
},
{
"epoch": 1.252268602540835,
"grad_norm": 2.046875,
"learning_rate": 0.0001996055157401834,
"loss": 0.8528,
"step": 345
},
{
"epoch": 1.2704174228675136,
"grad_norm": 1.3203125,
"learning_rate": 0.00019954719225730847,
"loss": 0.8687,
"step": 350
},
{
"epoch": 1.2885662431941924,
"grad_norm": 1.3828125,
"learning_rate": 0.0001994848590280447,
"loss": 0.8539,
"step": 355
},
{
"epoch": 1.306715063520871,
"grad_norm": 1.515625,
"learning_rate": 0.00019941851856316548,
"loss": 0.8335,
"step": 360
},
{
"epoch": 1.32486388384755,
"grad_norm": 12.75,
"learning_rate": 0.00019934817353485501,
"loss": 0.8529,
"step": 365
},
{
"epoch": 1.3430127041742286,
"grad_norm": 0.94140625,
"learning_rate": 0.00019927382677660088,
"loss": 0.861,
"step": 370
},
{
"epoch": 1.3611615245009074,
"grad_norm": 2.4375,
"learning_rate": 0.00019919548128307954,
"loss": 0.8304,
"step": 375
},
{
"epoch": 1.3793103448275863,
"grad_norm": 0.93359375,
"learning_rate": 0.00019911314021003613,
"loss": 0.8179,
"step": 380
},
{
"epoch": 1.397459165154265,
"grad_norm": 1.5234375,
"learning_rate": 0.00019902680687415705,
"loss": 0.8295,
"step": 385
},
{
"epoch": 1.4156079854809438,
"grad_norm": 1.703125,
"learning_rate": 0.00019893648475293648,
"loss": 0.8607,
"step": 390
},
{
"epoch": 1.4337568058076224,
"grad_norm": 1.1015625,
"learning_rate": 0.00019884217748453623,
"loss": 0.8223,
"step": 395
},
{
"epoch": 1.4519056261343013,
"grad_norm": 0.796875,
"learning_rate": 0.00019874388886763944,
"loss": 0.8603,
"step": 400
},
{
"epoch": 1.47005444646098,
"grad_norm": 0.640625,
"learning_rate": 0.0001986416228612972,
"loss": 0.8393,
"step": 405
},
{
"epoch": 1.4882032667876588,
"grad_norm": 1.5390625,
"learning_rate": 0.00019853538358476932,
"loss": 0.8363,
"step": 410
},
{
"epoch": 1.5063520871143377,
"grad_norm": 1.09375,
"learning_rate": 0.00019842517531735838,
"loss": 0.8403,
"step": 415
},
{
"epoch": 1.5245009074410163,
"grad_norm": 1.296875,
"learning_rate": 0.00019831100249823733,
"loss": 0.8198,
"step": 420
},
{
"epoch": 1.542649727767695,
"grad_norm": 17.5,
"learning_rate": 0.00019819286972627066,
"loss": 0.8203,
"step": 425
},
{
"epoch": 1.560798548094374,
"grad_norm": 4.3125,
"learning_rate": 0.00019807078175982924,
"loss": 0.8595,
"step": 430
},
{
"epoch": 1.5789473684210527,
"grad_norm": 1.890625,
"learning_rate": 0.00019794474351659852,
"loss": 0.8302,
"step": 435
},
{
"epoch": 1.5970961887477313,
"grad_norm": 1.125,
"learning_rate": 0.00019781476007338058,
"loss": 0.8248,
"step": 440
},
{
"epoch": 1.6152450090744102,
"grad_norm": 1.0234375,
"learning_rate": 0.00019768083666588953,
"loss": 0.8099,
"step": 445
},
{
"epoch": 1.633393829401089,
"grad_norm": 1.4921875,
"learning_rate": 0.00019754297868854073,
"loss": 0.8127,
"step": 450
},
{
"epoch": 1.6515426497277677,
"grad_norm": 1.5078125,
"learning_rate": 0.00019740119169423337,
"loss": 0.8199,
"step": 455
},
{
"epoch": 1.6696914700544465,
"grad_norm": 0.6640625,
"learning_rate": 0.00019725548139412692,
"loss": 0.8185,
"step": 460
},
{
"epoch": 1.6878402903811254,
"grad_norm": 0.71875,
"learning_rate": 0.00019710585365741103,
"loss": 0.8347,
"step": 465
},
{
"epoch": 1.705989110707804,
"grad_norm": 1.046875,
"learning_rate": 0.00019695231451106912,
"loss": 0.8217,
"step": 470
},
{
"epoch": 1.7241379310344827,
"grad_norm": 0.87109375,
"learning_rate": 0.00019679487013963564,
"loss": 0.8093,
"step": 475
},
{
"epoch": 1.7422867513611615,
"grad_norm": 0.6953125,
"learning_rate": 0.00019663352688494684,
"loss": 0.8139,
"step": 480
},
{
"epoch": 1.7604355716878404,
"grad_norm": 0.6796875,
"learning_rate": 0.0001964682912458856,
"loss": 0.8232,
"step": 485
},
{
"epoch": 1.778584392014519,
"grad_norm": 0.75,
"learning_rate": 0.00019629916987811926,
"loss": 0.8143,
"step": 490
},
{
"epoch": 1.7967332123411979,
"grad_norm": 1.046875,
"learning_rate": 0.0001961261695938319,
"loss": 0.8022,
"step": 495
},
{
"epoch": 1.8148820326678767,
"grad_norm": 0.88671875,
"learning_rate": 0.00019594929736144976,
"loss": 0.8233,
"step": 500
},
{
"epoch": 1.8330308529945554,
"grad_norm": 0.71875,
"learning_rate": 0.00019576856030536054,
"loss": 0.8185,
"step": 505
},
{
"epoch": 1.851179673321234,
"grad_norm": 1.0234375,
"learning_rate": 0.0001955839657056265,
"loss": 0.8208,
"step": 510
},
{
"epoch": 1.8693284936479129,
"grad_norm": 1.546875,
"learning_rate": 0.00019539552099769126,
"loss": 0.8178,
"step": 515
},
{
"epoch": 1.8874773139745917,
"grad_norm": 2.5625,
"learning_rate": 0.00019520323377208017,
"loss": 0.8086,
"step": 520
},
{
"epoch": 1.9056261343012704,
"grad_norm": 0.9140625,
"learning_rate": 0.00019500711177409454,
"loss": 0.8057,
"step": 525
},
{
"epoch": 1.9237749546279492,
"grad_norm": 2.15625,
"learning_rate": 0.00019480716290349995,
"loss": 0.8139,
"step": 530
},
{
"epoch": 1.941923774954628,
"grad_norm": 0.86328125,
"learning_rate": 0.00019460339521420772,
"loss": 0.8028,
"step": 535
},
{
"epoch": 1.9600725952813067,
"grad_norm": 1.5546875,
"learning_rate": 0.00019439581691395067,
"loss": 0.8212,
"step": 540
},
{
"epoch": 1.9782214156079854,
"grad_norm": 0.984375,
"learning_rate": 0.00019418443636395248,
"loss": 0.8107,
"step": 545
},
{
"epoch": 1.9963702359346642,
"grad_norm": 0.546875,
"learning_rate": 0.00019396926207859084,
"loss": 0.8104,
"step": 550
},
{
"epoch": 2.0,
"eval_loss": 2.1405208110809326,
"eval_runtime": 1.0236,
"eval_samples_per_second": 4.885,
"eval_steps_per_second": 1.954,
"step": 551
},
{
"epoch": 2.014519056261343,
"grad_norm": 0.85546875,
"learning_rate": 0.00019375030272505463,
"loss": 0.7501,
"step": 555
},
{
"epoch": 2.0326678765880217,
"grad_norm": 0.69921875,
"learning_rate": 0.00019352756712299468,
"loss": 0.7576,
"step": 560
},
{
"epoch": 2.0508166969147004,
"grad_norm": 1.8828125,
"learning_rate": 0.00019330106424416852,
"loss": 0.7446,
"step": 565
},
{
"epoch": 2.0689655172413794,
"grad_norm": 5.9375,
"learning_rate": 0.00019307080321207912,
"loss": 0.7617,
"step": 570
},
{
"epoch": 2.087114337568058,
"grad_norm": 0.734375,
"learning_rate": 0.00019283679330160726,
"loss": 0.7408,
"step": 575
},
{
"epoch": 2.1052631578947367,
"grad_norm": 1.2734375,
"learning_rate": 0.00019259904393863802,
"loss": 0.7426,
"step": 580
},
{
"epoch": 2.123411978221416,
"grad_norm": 1.25,
"learning_rate": 0.0001923575646996811,
"loss": 0.7396,
"step": 585
},
{
"epoch": 2.1415607985480944,
"grad_norm": 2.03125,
"learning_rate": 0.000192112365311485,
"loss": 0.7581,
"step": 590
},
{
"epoch": 2.159709618874773,
"grad_norm": 0.921875,
"learning_rate": 0.00019186345565064535,
"loss": 0.7494,
"step": 595
},
{
"epoch": 2.1778584392014517,
"grad_norm": 0.73046875,
"learning_rate": 0.00019161084574320696,
"loss": 0.7356,
"step": 600
},
{
"epoch": 2.196007259528131,
"grad_norm": 1.3125,
"learning_rate": 0.0001913545457642601,
"loss": 0.7502,
"step": 605
},
{
"epoch": 2.2141560798548094,
"grad_norm": 2.078125,
"learning_rate": 0.0001910945660375305,
"loss": 0.7566,
"step": 610
},
{
"epoch": 2.232304900181488,
"grad_norm": 1.109375,
"learning_rate": 0.0001908309170349637,
"loss": 0.7429,
"step": 615
},
{
"epoch": 2.250453720508167,
"grad_norm": 0.87109375,
"learning_rate": 0.0001905636093763031,
"loss": 0.7449,
"step": 620
},
{
"epoch": 2.268602540834846,
"grad_norm": 0.78515625,
"learning_rate": 0.00019029265382866214,
"loss": 0.735,
"step": 625
},
{
"epoch": 2.2867513611615244,
"grad_norm": 1.5546875,
"learning_rate": 0.0001900180613060908,
"loss": 0.7585,
"step": 630
},
{
"epoch": 2.304900181488203,
"grad_norm": 2.015625,
"learning_rate": 0.00018973984286913584,
"loss": 0.7429,
"step": 635
},
{
"epoch": 2.323049001814882,
"grad_norm": 0.7890625,
"learning_rate": 0.00018945800972439538,
"loss": 0.758,
"step": 640
},
{
"epoch": 2.341197822141561,
"grad_norm": 1.1015625,
"learning_rate": 0.00018917257322406734,
"loss": 0.757,
"step": 645
},
{
"epoch": 2.3593466424682394,
"grad_norm": 0.9296875,
"learning_rate": 0.00018888354486549237,
"loss": 0.7506,
"step": 650
},
{
"epoch": 2.3774954627949185,
"grad_norm": 1.0625,
"learning_rate": 0.00018859093629069058,
"loss": 0.756,
"step": 655
},
{
"epoch": 2.395644283121597,
"grad_norm": 0.609375,
"learning_rate": 0.00018829475928589271,
"loss": 0.7411,
"step": 660
},
{
"epoch": 2.413793103448276,
"grad_norm": 0.58203125,
"learning_rate": 0.00018799502578106534,
"loss": 0.7426,
"step": 665
},
{
"epoch": 2.4319419237749544,
"grad_norm": 0.8359375,
"learning_rate": 0.0001876917478494303,
"loss": 0.7359,
"step": 670
},
{
"epoch": 2.4500907441016335,
"grad_norm": 1.1640625,
"learning_rate": 0.00018738493770697852,
"loss": 0.7516,
"step": 675
},
{
"epoch": 2.468239564428312,
"grad_norm": 0.82421875,
"learning_rate": 0.00018707460771197774,
"loss": 0.7524,
"step": 680
},
{
"epoch": 2.486388384754991,
"grad_norm": 1.0078125,
"learning_rate": 0.00018676077036447494,
"loss": 0.7655,
"step": 685
},
{
"epoch": 2.50453720508167,
"grad_norm": 0.546875,
"learning_rate": 0.0001864434383057927,
"loss": 0.7457,
"step": 690
},
{
"epoch": 2.5226860254083485,
"grad_norm": 0.921875,
"learning_rate": 0.00018612262431802007,
"loss": 0.7608,
"step": 695
},
{
"epoch": 2.540834845735027,
"grad_norm": 2.390625,
"learning_rate": 0.00018579834132349772,
"loss": 0.7437,
"step": 700
},
{
"epoch": 2.558983666061706,
"grad_norm": 1.7890625,
"learning_rate": 0.00018547060238429736,
"loss": 0.7586,
"step": 705
},
{
"epoch": 2.577132486388385,
"grad_norm": 3.3125,
"learning_rate": 0.0001851394207016957,
"loss": 0.752,
"step": 710
},
{
"epoch": 2.5952813067150635,
"grad_norm": 0.58984375,
"learning_rate": 0.0001848048096156426,
"loss": 0.7484,
"step": 715
},
{
"epoch": 2.613430127041742,
"grad_norm": 6.3125,
"learning_rate": 0.00018446678260422385,
"loss": 0.7554,
"step": 720
},
{
"epoch": 2.6315789473684212,
"grad_norm": 0.62109375,
"learning_rate": 0.00018412535328311814,
"loss": 0.7442,
"step": 725
},
{
"epoch": 2.6497277676951,
"grad_norm": 0.8828125,
"learning_rate": 0.00018378053540504873,
"loss": 0.7581,
"step": 730
},
{
"epoch": 2.6678765880217785,
"grad_norm": 1.7265625,
"learning_rate": 0.00018343234285922953,
"loss": 0.7727,
"step": 735
},
{
"epoch": 2.686025408348457,
"grad_norm": 0.58203125,
"learning_rate": 0.00018308078967080546,
"loss": 0.7525,
"step": 740
},
{
"epoch": 2.7041742286751362,
"grad_norm": 1.0703125,
"learning_rate": 0.00018272589000028772,
"loss": 0.737,
"step": 745
},
{
"epoch": 2.722323049001815,
"grad_norm": 0.7578125,
"learning_rate": 0.0001823676581429833,
"loss": 0.7724,
"step": 750
},
{
"epoch": 2.7404718693284935,
"grad_norm": 1.640625,
"learning_rate": 0.00018200610852841913,
"loss": 0.746,
"step": 755
},
{
"epoch": 2.7586206896551726,
"grad_norm": 0.7265625,
"learning_rate": 0.00018164125571976098,
"loss": 0.7609,
"step": 760
},
{
"epoch": 2.7767695099818512,
"grad_norm": 1.1640625,
"learning_rate": 0.0001812731144132268,
"loss": 0.7623,
"step": 765
},
{
"epoch": 2.79491833030853,
"grad_norm": 0.6328125,
"learning_rate": 0.00018090169943749476,
"loss": 0.7668,
"step": 770
},
{
"epoch": 2.8130671506352085,
"grad_norm": 0.546875,
"learning_rate": 0.00018052702575310588,
"loss": 0.747,
"step": 775
},
{
"epoch": 2.8312159709618876,
"grad_norm": 1.09375,
"learning_rate": 0.00018014910845186153,
"loss": 0.7533,
"step": 780
},
{
"epoch": 2.8493647912885662,
"grad_norm": 0.83984375,
"learning_rate": 0.00017976796275621555,
"loss": 0.7435,
"step": 785
},
{
"epoch": 2.867513611615245,
"grad_norm": 0.73046875,
"learning_rate": 0.00017938360401866093,
"loss": 0.7424,
"step": 790
},
{
"epoch": 2.885662431941924,
"grad_norm": 0.75,
"learning_rate": 0.00017899604772111163,
"loss": 0.7321,
"step": 795
},
{
"epoch": 2.9038112522686026,
"grad_norm": 0.703125,
"learning_rate": 0.00017860530947427875,
"loss": 0.7474,
"step": 800
},
{
"epoch": 2.9219600725952812,
"grad_norm": 0.63671875,
"learning_rate": 0.00017821140501704194,
"loss": 0.7502,
"step": 805
},
{
"epoch": 2.94010889292196,
"grad_norm": 1.1328125,
"learning_rate": 0.00017781435021581527,
"loss": 0.7432,
"step": 810
},
{
"epoch": 2.958257713248639,
"grad_norm": 1.046875,
"learning_rate": 0.00017741416106390826,
"loss": 0.7529,
"step": 815
},
{
"epoch": 2.9764065335753176,
"grad_norm": 0.921875,
"learning_rate": 0.00017701085368088156,
"loss": 0.7633,
"step": 820
},
{
"epoch": 2.9945553539019962,
"grad_norm": 0.5859375,
"learning_rate": 0.0001766044443118978,
"loss": 0.7914,
"step": 825
},
{
"epoch": 2.9981851179673322,
"eval_loss": 2.1592297554016113,
"eval_runtime": 1.0245,
"eval_samples_per_second": 4.881,
"eval_steps_per_second": 1.952,
"step": 826
},
{
"epoch": 3.0127041742286753,
"grad_norm": 0.53515625,
"learning_rate": 0.0001761949493270671,
"loss": 0.6962,
"step": 830
},
{
"epoch": 3.030852994555354,
"grad_norm": 0.61328125,
"learning_rate": 0.0001757823852207877,
"loss": 0.6788,
"step": 835
},
{
"epoch": 3.0490018148820326,
"grad_norm": 0.6953125,
"learning_rate": 0.00017536676861108164,
"loss": 0.6785,
"step": 840
},
{
"epoch": 3.0671506352087112,
"grad_norm": 0.796875,
"learning_rate": 0.0001749481162389254,
"loss": 0.6784,
"step": 845
},
{
"epoch": 3.0852994555353903,
"grad_norm": 1.3515625,
"learning_rate": 0.0001745264449675755,
"loss": 0.6796,
"step": 850
},
{
"epoch": 3.103448275862069,
"grad_norm": 1.7109375,
"learning_rate": 0.00017410177178188918,
"loss": 0.6682,
"step": 855
},
{
"epoch": 3.1215970961887476,
"grad_norm": 0.76953125,
"learning_rate": 0.0001736741137876405,
"loss": 0.6941,
"step": 860
},
{
"epoch": 3.1397459165154267,
"grad_norm": 1.1015625,
"learning_rate": 0.0001732434882108311,
"loss": 0.6757,
"step": 865
},
{
"epoch": 3.1578947368421053,
"grad_norm": 1.0859375,
"learning_rate": 0.00017280991239699642,
"loss": 0.6772,
"step": 870
},
{
"epoch": 3.176043557168784,
"grad_norm": 0.79296875,
"learning_rate": 0.00017237340381050703,
"loss": 0.6951,
"step": 875
},
{
"epoch": 3.1941923774954626,
"grad_norm": 1.390625,
"learning_rate": 0.0001719339800338651,
"loss": 0.7005,
"step": 880
},
{
"epoch": 3.2123411978221417,
"grad_norm": 1.46875,
"learning_rate": 0.00017149165876699635,
"loss": 0.6948,
"step": 885
},
{
"epoch": 3.2304900181488203,
"grad_norm": 0.80078125,
"learning_rate": 0.0001710464578265369,
"loss": 0.6809,
"step": 890
},
{
"epoch": 3.248638838475499,
"grad_norm": 0.79296875,
"learning_rate": 0.00017059839514511565,
"loss": 0.6848,
"step": 895
},
{
"epoch": 3.266787658802178,
"grad_norm": 0.66015625,
"learning_rate": 0.00017014748877063214,
"loss": 0.6695,
"step": 900
},
{
"epoch": 3.2849364791288567,
"grad_norm": 0.5859375,
"learning_rate": 0.00016969375686552937,
"loss": 0.6939,
"step": 905
},
{
"epoch": 3.3030852994555353,
"grad_norm": 0.83984375,
"learning_rate": 0.00016923721770606228,
"loss": 0.6792,
"step": 910
},
{
"epoch": 3.321234119782214,
"grad_norm": 0.81640625,
"learning_rate": 0.0001687778896815617,
"loss": 0.7015,
"step": 915
},
{
"epoch": 3.339382940108893,
"grad_norm": 0.64453125,
"learning_rate": 0.00016831579129369346,
"loss": 0.6828,
"step": 920
},
{
"epoch": 3.3575317604355717,
"grad_norm": 0.6015625,
"learning_rate": 0.00016785094115571322,
"loss": 0.6704,
"step": 925
},
{
"epoch": 3.3756805807622503,
"grad_norm": 2.15625,
"learning_rate": 0.00016738335799171682,
"loss": 0.6796,
"step": 930
},
{
"epoch": 3.3938294010889294,
"grad_norm": 0.671875,
"learning_rate": 0.00016691306063588583,
"loss": 0.6866,
"step": 935
},
{
"epoch": 3.411978221415608,
"grad_norm": 1.609375,
"learning_rate": 0.00016644006803172924,
"loss": 0.6788,
"step": 940
},
{
"epoch": 3.4301270417422867,
"grad_norm": 0.6484375,
"learning_rate": 0.00016596439923132017,
"loss": 0.7052,
"step": 945
},
{
"epoch": 3.4482758620689653,
"grad_norm": 0.859375,
"learning_rate": 0.00016548607339452853,
"loss": 0.672,
"step": 950
},
{
"epoch": 3.4664246823956444,
"grad_norm": 0.56640625,
"learning_rate": 0.00016500510978824926,
"loss": 0.6981,
"step": 955
},
{
"epoch": 3.484573502722323,
"grad_norm": 0.609375,
"learning_rate": 0.0001645215277856263,
"loss": 0.6958,
"step": 960
},
{
"epoch": 3.5027223230490017,
"grad_norm": 0.7890625,
"learning_rate": 0.00016403534686527225,
"loss": 0.6779,
"step": 965
},
{
"epoch": 3.5208711433756807,
"grad_norm": 0.5703125,
"learning_rate": 0.00016354658661048364,
"loss": 0.6773,
"step": 970
},
{
"epoch": 3.5390199637023594,
"grad_norm": 0.703125,
"learning_rate": 0.00016305526670845226,
"loss": 0.6883,
"step": 975
},
{
"epoch": 3.557168784029038,
"grad_norm": 0.765625,
"learning_rate": 0.00016256140694947217,
"loss": 0.6962,
"step": 980
},
{
"epoch": 3.5753176043557167,
"grad_norm": 0.703125,
"learning_rate": 0.00016206502722614238,
"loss": 0.6783,
"step": 985
},
{
"epoch": 3.5934664246823957,
"grad_norm": 0.53515625,
"learning_rate": 0.0001615661475325658,
"loss": 0.6933,
"step": 990
},
{
"epoch": 3.6116152450090744,
"grad_norm": 0.67578125,
"learning_rate": 0.00016106478796354382,
"loss": 0.6791,
"step": 995
},
{
"epoch": 3.629764065335753,
"grad_norm": 0.859375,
"learning_rate": 0.00016056096871376667,
"loss": 0.7038,
"step": 1000
},
{
"epoch": 3.647912885662432,
"grad_norm": 0.6796875,
"learning_rate": 0.00016005471007700031,
"loss": 0.6966,
"step": 1005
},
{
"epoch": 3.6660617059891107,
"grad_norm": 1.1875,
"learning_rate": 0.0001595460324452688,
"loss": 0.6856,
"step": 1010
},
{
"epoch": 3.6842105263157894,
"grad_norm": 0.9609375,
"learning_rate": 0.000159034956308033,
"loss": 0.6949,
"step": 1015
},
{
"epoch": 3.702359346642468,
"grad_norm": 0.6171875,
"learning_rate": 0.00015852150225136518,
"loss": 0.6962,
"step": 1020
},
{
"epoch": 3.720508166969147,
"grad_norm": 0.50390625,
"learning_rate": 0.00015800569095711982,
"loss": 0.6948,
"step": 1025
},
{
"epoch": 3.7386569872958257,
"grad_norm": 0.6328125,
"learning_rate": 0.00015748754320210072,
"loss": 0.695,
"step": 1030
},
{
"epoch": 3.7568058076225044,
"grad_norm": 0.71484375,
"learning_rate": 0.0001569670798572239,
"loss": 0.6778,
"step": 1035
},
{
"epoch": 3.7749546279491835,
"grad_norm": 0.87109375,
"learning_rate": 0.00015644432188667695,
"loss": 0.6932,
"step": 1040
},
{
"epoch": 3.793103448275862,
"grad_norm": 1.0703125,
"learning_rate": 0.0001559192903470747,
"loss": 0.7006,
"step": 1045
},
{
"epoch": 3.8112522686025407,
"grad_norm": 0.796875,
"learning_rate": 0.00015539200638661104,
"loss": 0.6905,
"step": 1050
},
{
"epoch": 3.8294010889292194,
"grad_norm": 0.6328125,
"learning_rate": 0.000154862491244207,
"loss": 0.684,
"step": 1055
},
{
"epoch": 3.8475499092558985,
"grad_norm": 0.9765625,
"learning_rate": 0.00015433076624865531,
"loss": 0.6838,
"step": 1060
},
{
"epoch": 3.865698729582577,
"grad_norm": 0.8046875,
"learning_rate": 0.00015379685281776125,
"loss": 0.6807,
"step": 1065
},
{
"epoch": 3.8838475499092557,
"grad_norm": 0.59375,
"learning_rate": 0.00015326077245747999,
"loss": 0.6966,
"step": 1070
},
{
"epoch": 3.901996370235935,
"grad_norm": 0.6875,
"learning_rate": 0.00015272254676105025,
"loss": 0.6782,
"step": 1075
},
{
"epoch": 3.9201451905626135,
"grad_norm": 0.8984375,
"learning_rate": 0.0001521821974081246,
"loss": 0.6908,
"step": 1080
},
{
"epoch": 3.938294010889292,
"grad_norm": 0.58203125,
"learning_rate": 0.0001516397461638962,
"loss": 0.7119,
"step": 1085
},
{
"epoch": 3.9564428312159707,
"grad_norm": 0.86328125,
"learning_rate": 0.00015109521487822206,
"loss": 0.6926,
"step": 1090
},
{
"epoch": 3.97459165154265,
"grad_norm": 0.62109375,
"learning_rate": 0.000150548625484743,
"loss": 0.6885,
"step": 1095
},
{
"epoch": 3.9927404718693285,
"grad_norm": 0.640625,
"learning_rate": 0.00015000000000000001,
"loss": 0.6978,
"step": 1100
},
{
"epoch": 4.0,
"eval_loss": 2.2176449298858643,
"eval_runtime": 1.0239,
"eval_samples_per_second": 4.883,
"eval_steps_per_second": 1.953,
"step": 1102
},
{
"epoch": 4.0108892921960075,
"grad_norm": 0.6015625,
"learning_rate": 0.0001494493605225477,
"loss": 0.6515,
"step": 1105
},
{
"epoch": 4.029038112522686,
"grad_norm": 0.7421875,
"learning_rate": 0.0001488967292320639,
"loss": 0.6052,
"step": 1110
},
{
"epoch": 4.047186932849365,
"grad_norm": 0.640625,
"learning_rate": 0.00014834212838845637,
"loss": 0.6075,
"step": 1115
},
{
"epoch": 4.0653357531760435,
"grad_norm": 0.65234375,
"learning_rate": 0.00014778558033096633,
"loss": 0.6193,
"step": 1120
},
{
"epoch": 4.083484573502722,
"grad_norm": 1.125,
"learning_rate": 0.0001472271074772683,
"loss": 0.6051,
"step": 1125
},
{
"epoch": 4.101633393829401,
"grad_norm": 0.9765625,
"learning_rate": 0.00014666673232256738,
"loss": 0.6112,
"step": 1130
},
{
"epoch": 4.11978221415608,
"grad_norm": 0.68359375,
"learning_rate": 0.00014610447743869314,
"loss": 0.6208,
"step": 1135
},
{
"epoch": 4.137931034482759,
"grad_norm": 0.89453125,
"learning_rate": 0.00014554036547319033,
"loss": 0.6185,
"step": 1140
},
{
"epoch": 4.1560798548094375,
"grad_norm": 1.09375,
"learning_rate": 0.0001449744191484066,
"loss": 0.6209,
"step": 1145
},
{
"epoch": 4.174228675136116,
"grad_norm": 0.87109375,
"learning_rate": 0.00014440666126057744,
"loss": 0.6134,
"step": 1150
},
{
"epoch": 4.192377495462795,
"grad_norm": 0.76171875,
"learning_rate": 0.00014383711467890774,
"loss": 0.6234,
"step": 1155
},
{
"epoch": 4.2105263157894735,
"grad_norm": 0.640625,
"learning_rate": 0.00014326580234465085,
"loss": 0.6104,
"step": 1160
},
{
"epoch": 4.228675136116152,
"grad_norm": 0.9296875,
"learning_rate": 0.0001426927472701842,
"loss": 0.6183,
"step": 1165
},
{
"epoch": 4.246823956442832,
"grad_norm": 0.88671875,
"learning_rate": 0.00014211797253808268,
"loss": 0.6201,
"step": 1170
},
{
"epoch": 4.26497277676951,
"grad_norm": 0.89453125,
"learning_rate": 0.00014154150130018866,
"loss": 0.6291,
"step": 1175
},
{
"epoch": 4.283121597096189,
"grad_norm": 1.046875,
"learning_rate": 0.00014096335677667954,
"loss": 0.6248,
"step": 1180
},
{
"epoch": 4.3012704174228675,
"grad_norm": 0.8984375,
"learning_rate": 0.00014038356225513248,
"loss": 0.6321,
"step": 1185
},
{
"epoch": 4.319419237749546,
"grad_norm": 0.61328125,
"learning_rate": 0.00013980214108958624,
"loss": 0.6284,
"step": 1190
},
{
"epoch": 4.337568058076225,
"grad_norm": 0.79296875,
"learning_rate": 0.00013921911669960055,
"loss": 0.6294,
"step": 1195
},
{
"epoch": 4.3557168784029034,
"grad_norm": 0.80859375,
"learning_rate": 0.00013863451256931287,
"loss": 0.6166,
"step": 1200
},
{
"epoch": 4.373865698729583,
"grad_norm": 0.60546875,
"learning_rate": 0.0001380483522464923,
"loss": 0.6254,
"step": 1205
},
{
"epoch": 4.392014519056262,
"grad_norm": 0.875,
"learning_rate": 0.00013746065934159123,
"loss": 0.6194,
"step": 1210
},
{
"epoch": 4.41016333938294,
"grad_norm": 0.69921875,
"learning_rate": 0.0001368714575267941,
"loss": 0.6287,
"step": 1215
},
{
"epoch": 4.428312159709619,
"grad_norm": 0.55859375,
"learning_rate": 0.0001362807705350641,
"loss": 0.6306,
"step": 1220
},
{
"epoch": 4.4464609800362975,
"grad_norm": 0.7421875,
"learning_rate": 0.00013568862215918717,
"loss": 0.6298,
"step": 1225
},
{
"epoch": 4.464609800362976,
"grad_norm": 0.62109375,
"learning_rate": 0.00013509503625081358,
"loss": 0.6162,
"step": 1230
},
{
"epoch": 4.482758620689655,
"grad_norm": 0.6484375,
"learning_rate": 0.00013450003671949706,
"loss": 0.6235,
"step": 1235
},
{
"epoch": 4.500907441016334,
"grad_norm": 0.7578125,
"learning_rate": 0.00013390364753173206,
"loss": 0.6217,
"step": 1240
},
{
"epoch": 4.519056261343013,
"grad_norm": 0.69921875,
"learning_rate": 0.00013330589270998808,
"loss": 0.6119,
"step": 1245
},
{
"epoch": 4.537205081669692,
"grad_norm": 0.6484375,
"learning_rate": 0.00013270679633174218,
"loss": 0.6272,
"step": 1250
},
{
"epoch": 4.55535390199637,
"grad_norm": 0.60546875,
"learning_rate": 0.00013210638252850908,
"loss": 0.6244,
"step": 1255
},
{
"epoch": 4.573502722323049,
"grad_norm": 0.609375,
"learning_rate": 0.0001315046754848693,
"loss": 0.6335,
"step": 1260
},
{
"epoch": 4.5916515426497275,
"grad_norm": 0.59375,
"learning_rate": 0.00013090169943749476,
"loss": 0.6282,
"step": 1265
},
{
"epoch": 4.609800362976406,
"grad_norm": 0.59375,
"learning_rate": 0.00013029747867417276,
"loss": 0.628,
"step": 1270
},
{
"epoch": 4.627949183303086,
"grad_norm": 0.6171875,
"learning_rate": 0.0001296920375328275,
"loss": 0.6257,
"step": 1275
},
{
"epoch": 4.646098003629764,
"grad_norm": 0.57421875,
"learning_rate": 0.0001290854004005399,
"loss": 0.6229,
"step": 1280
},
{
"epoch": 4.664246823956443,
"grad_norm": 0.640625,
"learning_rate": 0.00012847759171256523,
"loss": 0.6401,
"step": 1285
},
{
"epoch": 4.682395644283122,
"grad_norm": 0.5859375,
"learning_rate": 0.0001278686359513488,
"loss": 0.627,
"step": 1290
},
{
"epoch": 4.7005444646098,
"grad_norm": 0.6171875,
"learning_rate": 0.0001272585576455398,
"loss": 0.6368,
"step": 1295
},
{
"epoch": 4.718693284936479,
"grad_norm": 0.65234375,
"learning_rate": 0.00012664738136900348,
"loss": 0.6378,
"step": 1300
},
{
"epoch": 4.7368421052631575,
"grad_norm": 0.609375,
"learning_rate": 0.0001260351317398312,
"loss": 0.6318,
"step": 1305
},
{
"epoch": 4.754990925589837,
"grad_norm": 0.59375,
"learning_rate": 0.00012542183341934872,
"loss": 0.6277,
"step": 1310
},
{
"epoch": 4.773139745916516,
"grad_norm": 0.59375,
"learning_rate": 0.0001248075111111229,
"loss": 0.6407,
"step": 1315
},
{
"epoch": 4.791288566243194,
"grad_norm": 0.73828125,
"learning_rate": 0.00012419218955996676,
"loss": 0.6421,
"step": 1320
},
{
"epoch": 4.809437386569873,
"grad_norm": 0.80859375,
"learning_rate": 0.00012357589355094275,
"loss": 0.6323,
"step": 1325
},
{
"epoch": 4.827586206896552,
"grad_norm": 0.59765625,
"learning_rate": 0.0001229586479083641,
"loss": 0.6437,
"step": 1330
},
{
"epoch": 4.84573502722323,
"grad_norm": 0.5625,
"learning_rate": 0.00012234047749479544,
"loss": 0.6415,
"step": 1335
},
{
"epoch": 4.863883847549909,
"grad_norm": 0.59375,
"learning_rate": 0.00012172140721005079,
"loss": 0.6293,
"step": 1340
},
{
"epoch": 4.882032667876588,
"grad_norm": 0.609375,
"learning_rate": 0.000121101461990191,
"loss": 0.6245,
"step": 1345
},
{
"epoch": 4.900181488203267,
"grad_norm": 0.5703125,
"learning_rate": 0.00012048066680651908,
"loss": 0.6146,
"step": 1350
},
{
"epoch": 4.918330308529946,
"grad_norm": 0.65625,
"learning_rate": 0.00011985904666457455,
"loss": 0.6425,
"step": 1355
},
{
"epoch": 4.936479128856624,
"grad_norm": 0.80078125,
"learning_rate": 0.00011923662660312611,
"loss": 0.64,
"step": 1360
},
{
"epoch": 4.954627949183303,
"grad_norm": 0.54296875,
"learning_rate": 0.00011861343169316301,
"loss": 0.6316,
"step": 1365
},
{
"epoch": 4.972776769509982,
"grad_norm": 0.76171875,
"learning_rate": 0.00011798948703688539,
"loss": 0.629,
"step": 1370
},
{
"epoch": 4.99092558983666,
"grad_norm": 0.625,
"learning_rate": 0.00011736481776669306,
"loss": 0.6386,
"step": 1375
},
{
"epoch": 4.998185117967332,
"eval_loss": 2.3272275924682617,
"eval_runtime": 1.0255,
"eval_samples_per_second": 4.876,
"eval_steps_per_second": 1.95,
"step": 1377
},
{
"epoch": 5.00907441016334,
"grad_norm": 0.77734375,
"learning_rate": 0.00011673944904417308,
"loss": 0.5952,
"step": 1380
},
{
"epoch": 5.027223230490018,
"grad_norm": 0.79296875,
"learning_rate": 0.00011611340605908642,
"loss": 0.5449,
"step": 1385
},
{
"epoch": 5.045372050816697,
"grad_norm": 0.65234375,
"learning_rate": 0.00011548671402835325,
"loss": 0.5476,
"step": 1390
},
{
"epoch": 5.063520871143376,
"grad_norm": 0.72265625,
"learning_rate": 0.00011485939819503717,
"loss": 0.5626,
"step": 1395
},
{
"epoch": 5.081669691470054,
"grad_norm": 0.5703125,
"learning_rate": 0.00011423148382732853,
"loss": 0.5616,
"step": 1400
},
{
"epoch": 5.099818511796733,
"grad_norm": 0.61328125,
"learning_rate": 0.00011360299621752644,
"loss": 0.569,
"step": 1405
},
{
"epoch": 5.117967332123412,
"grad_norm": 0.6953125,
"learning_rate": 0.00011297396068102017,
"loss": 0.5644,
"step": 1410
},
{
"epoch": 5.136116152450091,
"grad_norm": 0.64453125,
"learning_rate": 0.00011234440255526948,
"loss": 0.5577,
"step": 1415
},
{
"epoch": 5.15426497277677,
"grad_norm": 0.66015625,
"learning_rate": 0.00011171434719878384,
"loss": 0.5484,
"step": 1420
},
{
"epoch": 5.172413793103448,
"grad_norm": 0.671875,
"learning_rate": 0.00011108381999010111,
"loss": 0.5599,
"step": 1425
},
{
"epoch": 5.190562613430127,
"grad_norm": 0.62109375,
"learning_rate": 0.00011045284632676536,
"loss": 0.5588,
"step": 1430
},
{
"epoch": 5.208711433756806,
"grad_norm": 0.625,
"learning_rate": 0.00010982145162430373,
"loss": 0.5562,
"step": 1435
},
{
"epoch": 5.226860254083484,
"grad_norm": 0.59375,
"learning_rate": 0.00010918966131520277,
"loss": 0.549,
"step": 1440
},
{
"epoch": 5.245009074410163,
"grad_norm": 0.640625,
"learning_rate": 0.00010855750084788398,
"loss": 0.5667,
"step": 1445
},
{
"epoch": 5.2631578947368425,
"grad_norm": 0.64453125,
"learning_rate": 0.00010792499568567884,
"loss": 0.5611,
"step": 1450
},
{
"epoch": 5.281306715063521,
"grad_norm": 0.7421875,
"learning_rate": 0.0001072921713058031,
"loss": 0.5577,
"step": 1455
},
{
"epoch": 5.2994555353902,
"grad_norm": 0.65234375,
"learning_rate": 0.00010665905319833041,
"loss": 0.559,
"step": 1460
},
{
"epoch": 5.317604355716878,
"grad_norm": 0.6953125,
"learning_rate": 0.00010602566686516586,
"loss": 0.552,
"step": 1465
},
{
"epoch": 5.335753176043557,
"grad_norm": 1.0546875,
"learning_rate": 0.00010539203781901861,
"loss": 0.5516,
"step": 1470
},
{
"epoch": 5.353901996370236,
"grad_norm": 0.88671875,
"learning_rate": 0.00010475819158237425,
"loss": 0.5646,
"step": 1475
},
{
"epoch": 5.372050816696914,
"grad_norm": 0.73046875,
"learning_rate": 0.00010412415368646673,
"loss": 0.5614,
"step": 1480
},
{
"epoch": 5.390199637023594,
"grad_norm": 0.75,
"learning_rate": 0.00010348994967025012,
"loss": 0.5532,
"step": 1485
},
{
"epoch": 5.4083484573502725,
"grad_norm": 0.91796875,
"learning_rate": 0.00010285560507936961,
"loss": 0.5631,
"step": 1490
},
{
"epoch": 5.426497277676951,
"grad_norm": 0.80078125,
"learning_rate": 0.00010222114546513295,
"loss": 0.581,
"step": 1495
},
{
"epoch": 5.44464609800363,
"grad_norm": 0.61328125,
"learning_rate": 0.00010158659638348081,
"loss": 0.5503,
"step": 1500
},
{
"epoch": 5.462794918330308,
"grad_norm": 0.69921875,
"learning_rate": 0.00010095198339395769,
"loss": 0.5645,
"step": 1505
},
{
"epoch": 5.480943738656987,
"grad_norm": 0.65234375,
"learning_rate": 0.00010031733205868224,
"loss": 0.565,
"step": 1510
},
{
"epoch": 5.499092558983666,
"grad_norm": 0.64453125,
"learning_rate": 9.968266794131777e-05,
"loss": 0.5616,
"step": 1515
},
{
"epoch": 5.517241379310345,
"grad_norm": 0.640625,
"learning_rate": 9.904801660604234e-05,
"loss": 0.5601,
"step": 1520
},
{
"epoch": 5.535390199637024,
"grad_norm": 0.7109375,
"learning_rate": 9.84134036165192e-05,
"loss": 0.5661,
"step": 1525
},
{
"epoch": 5.5535390199637025,
"grad_norm": 0.65234375,
"learning_rate": 9.777885453486706e-05,
"loss": 0.5756,
"step": 1530
},
{
"epoch": 5.571687840290381,
"grad_norm": 0.625,
"learning_rate": 9.71443949206304e-05,
"loss": 0.5702,
"step": 1535
},
{
"epoch": 5.58983666061706,
"grad_norm": 0.5859375,
"learning_rate": 9.651005032974994e-05,
"loss": 0.561,
"step": 1540
},
{
"epoch": 5.607985480943738,
"grad_norm": 0.6953125,
"learning_rate": 9.587584631353329e-05,
"loss": 0.5747,
"step": 1545
},
{
"epoch": 5.626134301270417,
"grad_norm": 0.6875,
"learning_rate": 9.524180841762577e-05,
"loss": 0.562,
"step": 1550
},
{
"epoch": 5.6442831215970966,
"grad_norm": 0.76953125,
"learning_rate": 9.460796218098143e-05,
"loss": 0.5608,
"step": 1555
},
{
"epoch": 5.662431941923775,
"grad_norm": 1.03125,
"learning_rate": 9.397433313483416e-05,
"loss": 0.5677,
"step": 1560
},
{
"epoch": 5.680580762250454,
"grad_norm": 0.734375,
"learning_rate": 9.334094680166962e-05,
"loss": 0.575,
"step": 1565
},
{
"epoch": 5.6987295825771325,
"grad_norm": 0.9296875,
"learning_rate": 9.270782869419694e-05,
"loss": 0.5709,
"step": 1570
},
{
"epoch": 5.716878402903811,
"grad_norm": 0.86328125,
"learning_rate": 9.207500431432115e-05,
"loss": 0.5556,
"step": 1575
},
{
"epoch": 5.73502722323049,
"grad_norm": 0.8828125,
"learning_rate": 9.144249915211605e-05,
"loss": 0.5694,
"step": 1580
},
{
"epoch": 5.753176043557168,
"grad_norm": 0.65625,
"learning_rate": 9.081033868479727e-05,
"loss": 0.5652,
"step": 1585
},
{
"epoch": 5.771324863883848,
"grad_norm": 0.8125,
"learning_rate": 9.01785483756963e-05,
"loss": 0.568,
"step": 1590
},
{
"epoch": 5.7894736842105265,
"grad_norm": 0.74609375,
"learning_rate": 8.954715367323468e-05,
"loss": 0.5638,
"step": 1595
},
{
"epoch": 5.807622504537205,
"grad_norm": 0.671875,
"learning_rate": 8.891618000989891e-05,
"loss": 0.5668,
"step": 1600
},
{
"epoch": 5.825771324863884,
"grad_norm": 0.640625,
"learning_rate": 8.828565280121617e-05,
"loss": 0.5714,
"step": 1605
},
{
"epoch": 5.8439201451905625,
"grad_norm": 0.69921875,
"learning_rate": 8.765559744473053e-05,
"loss": 0.5709,
"step": 1610
},
{
"epoch": 5.862068965517241,
"grad_norm": 0.69140625,
"learning_rate": 8.702603931897982e-05,
"loss": 0.5834,
"step": 1615
},
{
"epoch": 5.88021778584392,
"grad_norm": 0.64453125,
"learning_rate": 8.639700378247361e-05,
"loss": 0.5673,
"step": 1620
},
{
"epoch": 5.898366606170599,
"grad_norm": 0.84765625,
"learning_rate": 8.57685161726715e-05,
"loss": 0.5702,
"step": 1625
},
{
"epoch": 5.916515426497278,
"grad_norm": 0.6328125,
"learning_rate": 8.514060180496285e-05,
"loss": 0.5671,
"step": 1630
},
{
"epoch": 5.9346642468239565,
"grad_norm": 0.640625,
"learning_rate": 8.451328597164679e-05,
"loss": 0.5686,
"step": 1635
},
{
"epoch": 5.952813067150635,
"grad_norm": 0.70703125,
"learning_rate": 8.38865939409136e-05,
"loss": 0.5686,
"step": 1640
},
{
"epoch": 5.970961887477314,
"grad_norm": 0.6796875,
"learning_rate": 8.326055095582694e-05,
"loss": 0.5685,
"step": 1645
},
{
"epoch": 5.9891107078039925,
"grad_norm": 0.625,
"learning_rate": 8.263518223330697e-05,
"loss": 0.5725,
"step": 1650
},
{
"epoch": 6.0,
"eval_loss": 2.4713406562805176,
"eval_runtime": 1.0255,
"eval_samples_per_second": 4.876,
"eval_steps_per_second": 1.95,
"step": 1653
},
{
"epoch": 6.007259528130671,
"grad_norm": 0.640625,
"learning_rate": 8.201051296311462e-05,
"loss": 0.5327,
"step": 1655
},
{
"epoch": 6.025408348457351,
"grad_norm": 0.7265625,
"learning_rate": 8.1386568306837e-05,
"loss": 0.5063,
"step": 1660
},
{
"epoch": 6.043557168784029,
"grad_norm": 0.703125,
"learning_rate": 8.076337339687394e-05,
"loss": 0.4988,
"step": 1665
},
{
"epoch": 6.061705989110708,
"grad_norm": 0.7421875,
"learning_rate": 8.014095333542548e-05,
"loss": 0.4931,
"step": 1670
},
{
"epoch": 6.0798548094373865,
"grad_norm": 0.62890625,
"learning_rate": 7.951933319348095e-05,
"loss": 0.4928,
"step": 1675
},
{
"epoch": 6.098003629764065,
"grad_norm": 0.67578125,
"learning_rate": 7.889853800980904e-05,
"loss": 0.4987,
"step": 1680
},
{
"epoch": 6.116152450090744,
"grad_norm": 0.734375,
"learning_rate": 7.827859278994925e-05,
"loss": 0.502,
"step": 1685
},
{
"epoch": 6.1343012704174225,
"grad_norm": 0.73828125,
"learning_rate": 7.765952250520459e-05,
"loss": 0.494,
"step": 1690
},
{
"epoch": 6.152450090744102,
"grad_norm": 0.71875,
"learning_rate": 7.704135209163589e-05,
"loss": 0.4876,
"step": 1695
},
{
"epoch": 6.170598911070781,
"grad_norm": 0.6875,
"learning_rate": 7.642410644905726e-05,
"loss": 0.51,
"step": 1700
},
{
"epoch": 6.188747731397459,
"grad_norm": 0.64453125,
"learning_rate": 7.580781044003324e-05,
"loss": 0.4999,
"step": 1705
},
{
"epoch": 6.206896551724138,
"grad_norm": 0.69140625,
"learning_rate": 7.519248888887716e-05,
"loss": 0.5019,
"step": 1710
},
{
"epoch": 6.2250453720508165,
"grad_norm": 0.65234375,
"learning_rate": 7.457816658065134e-05,
"loss": 0.4964,
"step": 1715
},
{
"epoch": 6.243194192377495,
"grad_norm": 0.6953125,
"learning_rate": 7.39648682601688e-05,
"loss": 0.4976,
"step": 1720
},
{
"epoch": 6.261343012704174,
"grad_norm": 0.62890625,
"learning_rate": 7.335261863099651e-05,
"loss": 0.5046,
"step": 1725
},
{
"epoch": 6.279491833030853,
"grad_norm": 0.69140625,
"learning_rate": 7.274144235446023e-05,
"loss": 0.4968,
"step": 1730
},
{
"epoch": 6.297640653357532,
"grad_norm": 0.703125,
"learning_rate": 7.213136404865124e-05,
"loss": 0.508,
"step": 1735
},
{
"epoch": 6.315789473684211,
"grad_norm": 0.68359375,
"learning_rate": 7.152240828743477e-05,
"loss": 0.4968,
"step": 1740
},
{
"epoch": 6.333938294010889,
"grad_norm": 0.83984375,
"learning_rate": 7.09145995994601e-05,
"loss": 0.5009,
"step": 1745
},
{
"epoch": 6.352087114337568,
"grad_norm": 0.671875,
"learning_rate": 7.030796246717255e-05,
"loss": 0.4925,
"step": 1750
},
{
"epoch": 6.3702359346642465,
"grad_norm": 0.6953125,
"learning_rate": 6.970252132582728e-05,
"loss": 0.5057,
"step": 1755
},
{
"epoch": 6.388384754990925,
"grad_norm": 0.6875,
"learning_rate": 6.909830056250527e-05,
"loss": 0.505,
"step": 1760
},
{
"epoch": 6.406533575317605,
"grad_norm": 0.69140625,
"learning_rate": 6.849532451513074e-05,
"loss": 0.4962,
"step": 1765
},
{
"epoch": 6.424682395644283,
"grad_norm": 0.66796875,
"learning_rate": 6.789361747149093e-05,
"loss": 0.5082,
"step": 1770
},
{
"epoch": 6.442831215970962,
"grad_norm": 0.640625,
"learning_rate": 6.729320366825784e-05,
"loss": 0.5006,
"step": 1775
},
{
"epoch": 6.460980036297641,
"grad_norm": 0.78515625,
"learning_rate": 6.669410729001193e-05,
"loss": 0.5089,
"step": 1780
},
{
"epoch": 6.479128856624319,
"grad_norm": 0.76171875,
"learning_rate": 6.609635246826794e-05,
"loss": 0.5005,
"step": 1785
},
{
"epoch": 6.497277676950998,
"grad_norm": 0.7890625,
"learning_rate": 6.549996328050296e-05,
"loss": 0.5045,
"step": 1790
},
{
"epoch": 6.5154264972776765,
"grad_norm": 0.78125,
"learning_rate": 6.490496374918647e-05,
"loss": 0.5117,
"step": 1795
},
{
"epoch": 6.533575317604356,
"grad_norm": 0.69921875,
"learning_rate": 6.431137784081282e-05,
"loss": 0.5074,
"step": 1800
},
{
"epoch": 6.551724137931035,
"grad_norm": 0.65625,
"learning_rate": 6.371922946493591e-05,
"loss": 0.5078,
"step": 1805
},
{
"epoch": 6.569872958257713,
"grad_norm": 0.69140625,
"learning_rate": 6.312854247320595e-05,
"loss": 0.5032,
"step": 1810
},
{
"epoch": 6.588021778584392,
"grad_norm": 0.7109375,
"learning_rate": 6.25393406584088e-05,
"loss": 0.507,
"step": 1815
},
{
"epoch": 6.606170598911071,
"grad_norm": 0.71875,
"learning_rate": 6.19516477535077e-05,
"loss": 0.5106,
"step": 1820
},
{
"epoch": 6.624319419237749,
"grad_norm": 0.66796875,
"learning_rate": 6.136548743068713e-05,
"loss": 0.4985,
"step": 1825
},
{
"epoch": 6.642468239564428,
"grad_norm": 0.66796875,
"learning_rate": 6.078088330039945e-05,
"loss": 0.5001,
"step": 1830
},
{
"epoch": 6.660617059891107,
"grad_norm": 0.68359375,
"learning_rate": 6.019785891041381e-05,
"loss": 0.5002,
"step": 1835
},
{
"epoch": 6.678765880217786,
"grad_norm": 0.66015625,
"learning_rate": 5.9616437744867535e-05,
"loss": 0.5107,
"step": 1840
},
{
"epoch": 6.696914700544465,
"grad_norm": 0.7890625,
"learning_rate": 5.9036643223320475e-05,
"loss": 0.4967,
"step": 1845
},
{
"epoch": 6.715063520871143,
"grad_norm": 0.765625,
"learning_rate": 5.845849869981137e-05,
"loss": 0.4988,
"step": 1850
},
{
"epoch": 6.733212341197822,
"grad_norm": 0.7265625,
"learning_rate": 5.788202746191734e-05,
"loss": 0.5003,
"step": 1855
},
{
"epoch": 6.751361161524501,
"grad_norm": 0.76171875,
"learning_rate": 5.7307252729815833e-05,
"loss": 0.5062,
"step": 1860
},
{
"epoch": 6.769509981851179,
"grad_norm": 0.80859375,
"learning_rate": 5.6734197655349156e-05,
"loss": 0.5056,
"step": 1865
},
{
"epoch": 6.787658802177859,
"grad_norm": 0.72265625,
"learning_rate": 5.616288532109225e-05,
"loss": 0.5081,
"step": 1870
},
{
"epoch": 6.805807622504537,
"grad_norm": 0.6640625,
"learning_rate": 5.559333873942259e-05,
"loss": 0.502,
"step": 1875
},
{
"epoch": 6.823956442831216,
"grad_norm": 0.76171875,
"learning_rate": 5.5025580851593436e-05,
"loss": 0.5091,
"step": 1880
},
{
"epoch": 6.842105263157895,
"grad_norm": 0.7265625,
"learning_rate": 5.445963452680973e-05,
"loss": 0.5017,
"step": 1885
},
{
"epoch": 6.860254083484573,
"grad_norm": 0.7890625,
"learning_rate": 5.38955225613069e-05,
"loss": 0.5148,
"step": 1890
},
{
"epoch": 6.878402903811252,
"grad_norm": 0.6484375,
"learning_rate": 5.333326767743263e-05,
"loss": 0.5044,
"step": 1895
},
{
"epoch": 6.896551724137931,
"grad_norm": 0.66015625,
"learning_rate": 5.277289252273174e-05,
"loss": 0.5094,
"step": 1900
},
{
"epoch": 6.91470054446461,
"grad_norm": 0.69140625,
"learning_rate": 5.221441966903371e-05,
"loss": 0.5056,
"step": 1905
},
{
"epoch": 6.932849364791289,
"grad_norm": 0.78515625,
"learning_rate": 5.1657871611543605e-05,
"loss": 0.5108,
"step": 1910
},
{
"epoch": 6.950998185117967,
"grad_norm": 0.796875,
"learning_rate": 5.110327076793613e-05,
"loss": 0.5068,
"step": 1915
},
{
"epoch": 6.969147005444646,
"grad_norm": 0.72265625,
"learning_rate": 5.055063947745233e-05,
"loss": 0.5038,
"step": 1920
},
{
"epoch": 6.987295825771325,
"grad_norm": 0.67578125,
"learning_rate": 5.000000000000002e-05,
"loss": 0.5089,
"step": 1925
},
{
"epoch": 6.998185117967332,
"eval_loss": 2.6490731239318848,
"eval_runtime": 1.0248,
"eval_samples_per_second": 4.879,
"eval_steps_per_second": 1.952,
"step": 1928
},
{
"epoch": 7.005444646098003,
"grad_norm": 0.65625,
"learning_rate": 4.945137451525707e-05,
"loss": 0.4914,
"step": 1930
},
{
"epoch": 7.023593466424682,
"grad_norm": 0.87109375,
"learning_rate": 4.890478512177795e-05,
"loss": 0.4567,
"step": 1935
},
{
"epoch": 7.0417422867513615,
"grad_norm": 0.7109375,
"learning_rate": 4.836025383610382e-05,
"loss": 0.4459,
"step": 1940
},
{
"epoch": 7.05989110707804,
"grad_norm": 0.6796875,
"learning_rate": 4.7817802591875426e-05,
"loss": 0.4523,
"step": 1945
},
{
"epoch": 7.078039927404719,
"grad_norm": 0.71484375,
"learning_rate": 4.727745323894976e-05,
"loss": 0.4551,
"step": 1950
},
{
"epoch": 7.096188747731397,
"grad_norm": 0.72265625,
"learning_rate": 4.673922754252002e-05,
"loss": 0.4435,
"step": 1955
},
{
"epoch": 7.114337568058076,
"grad_norm": 0.74609375,
"learning_rate": 4.620314718223876e-05,
"loss": 0.4572,
"step": 1960
},
{
"epoch": 7.132486388384755,
"grad_norm": 0.78515625,
"learning_rate": 4.566923375134472e-05,
"loss": 0.4564,
"step": 1965
},
{
"epoch": 7.150635208711433,
"grad_norm": 0.73828125,
"learning_rate": 4.513750875579303e-05,
"loss": 0.4482,
"step": 1970
},
{
"epoch": 7.168784029038113,
"grad_norm": 0.6875,
"learning_rate": 4.4607993613388976e-05,
"loss": 0.4536,
"step": 1975
},
{
"epoch": 7.1869328493647915,
"grad_norm": 0.7578125,
"learning_rate": 4.4080709652925336e-05,
"loss": 0.4571,
"step": 1980
},
{
"epoch": 7.20508166969147,
"grad_norm": 0.69921875,
"learning_rate": 4.355567811332311e-05,
"loss": 0.4548,
"step": 1985
},
{
"epoch": 7.223230490018149,
"grad_norm": 0.6796875,
"learning_rate": 4.3032920142776125e-05,
"loss": 0.4538,
"step": 1990
},
{
"epoch": 7.241379310344827,
"grad_norm": 0.71875,
"learning_rate": 4.251245679789928e-05,
"loss": 0.4497,
"step": 1995
},
{
"epoch": 7.259528130671506,
"grad_norm": 0.74609375,
"learning_rate": 4.19943090428802e-05,
"loss": 0.4546,
"step": 2000
},
{
"epoch": 7.277676950998185,
"grad_norm": 0.73828125,
"learning_rate": 4.147849774863488e-05,
"loss": 0.4612,
"step": 2005
},
{
"epoch": 7.295825771324864,
"grad_norm": 0.7265625,
"learning_rate": 4.096504369196704e-05,
"loss": 0.4437,
"step": 2010
},
{
"epoch": 7.313974591651543,
"grad_norm": 0.765625,
"learning_rate": 4.045396755473121e-05,
"loss": 0.4468,
"step": 2015
},
{
"epoch": 7.3321234119782215,
"grad_norm": 0.73828125,
"learning_rate": 3.994528992299971e-05,
"loss": 0.4456,
"step": 2020
},
{
"epoch": 7.3502722323049,
"grad_norm": 0.671875,
"learning_rate": 3.943903128623335e-05,
"loss": 0.4458,
"step": 2025
},
{
"epoch": 7.368421052631579,
"grad_norm": 0.7265625,
"learning_rate": 3.893521203645618e-05,
"loss": 0.4524,
"step": 2030
},
{
"epoch": 7.386569872958257,
"grad_norm": 0.73828125,
"learning_rate": 3.843385246743417e-05,
"loss": 0.4534,
"step": 2035
},
{
"epoch": 7.404718693284936,
"grad_norm": 0.76171875,
"learning_rate": 3.7934972773857634e-05,
"loss": 0.4577,
"step": 2040
},
{
"epoch": 7.422867513611616,
"grad_norm": 0.78515625,
"learning_rate": 3.7438593050527845e-05,
"loss": 0.455,
"step": 2045
},
{
"epoch": 7.441016333938294,
"grad_norm": 0.703125,
"learning_rate": 3.694473329154778e-05,
"loss": 0.462,
"step": 2050
},
{
"epoch": 7.459165154264973,
"grad_norm": 0.6796875,
"learning_rate": 3.645341338951639e-05,
"loss": 0.4538,
"step": 2055
},
{
"epoch": 7.4773139745916515,
"grad_norm": 0.72265625,
"learning_rate": 3.5964653134727776e-05,
"loss": 0.455,
"step": 2060
},
{
"epoch": 7.49546279491833,
"grad_norm": 0.7265625,
"learning_rate": 3.547847221437372e-05,
"loss": 0.459,
"step": 2065
},
{
"epoch": 7.513611615245009,
"grad_norm": 0.765625,
"learning_rate": 3.4994890211750754e-05,
"loss": 0.4491,
"step": 2070
},
{
"epoch": 7.531760435571687,
"grad_norm": 0.7421875,
"learning_rate": 3.45139266054715e-05,
"loss": 0.4535,
"step": 2075
},
{
"epoch": 7.549909255898367,
"grad_norm": 0.79296875,
"learning_rate": 3.4035600768679855e-05,
"loss": 0.4517,
"step": 2080
},
{
"epoch": 7.568058076225046,
"grad_norm": 0.67578125,
"learning_rate": 3.3559931968270753e-05,
"loss": 0.4535,
"step": 2085
},
{
"epoch": 7.586206896551724,
"grad_norm": 0.6875,
"learning_rate": 3.308693936411421e-05,
"loss": 0.4486,
"step": 2090
},
{
"epoch": 7.604355716878403,
"grad_norm": 0.7265625,
"learning_rate": 3.2616642008283213e-05,
"loss": 0.459,
"step": 2095
},
{
"epoch": 7.6225045372050815,
"grad_norm": 0.76953125,
"learning_rate": 3.21490588442868e-05,
"loss": 0.4579,
"step": 2100
},
{
"epoch": 7.64065335753176,
"grad_norm": 0.81640625,
"learning_rate": 3.1684208706306574e-05,
"loss": 0.4431,
"step": 2105
},
{
"epoch": 7.658802177858439,
"grad_norm": 0.734375,
"learning_rate": 3.1222110318438304e-05,
"loss": 0.4554,
"step": 2110
},
{
"epoch": 7.676950998185118,
"grad_norm": 0.734375,
"learning_rate": 3.076278229393773e-05,
"loss": 0.4574,
"step": 2115
},
{
"epoch": 7.695099818511797,
"grad_norm": 0.75,
"learning_rate": 3.030624313447067e-05,
"loss": 0.466,
"step": 2120
},
{
"epoch": 7.713248638838476,
"grad_norm": 0.72265625,
"learning_rate": 2.9852511229367865e-05,
"loss": 0.4592,
"step": 2125
},
{
"epoch": 7.731397459165154,
"grad_norm": 0.67578125,
"learning_rate": 2.9401604854884357e-05,
"loss": 0.4622,
"step": 2130
},
{
"epoch": 7.749546279491833,
"grad_norm": 0.7890625,
"learning_rate": 2.8953542173463133e-05,
"loss": 0.4527,
"step": 2135
},
{
"epoch": 7.7676950998185115,
"grad_norm": 0.73828125,
"learning_rate": 2.8508341233003654e-05,
"loss": 0.4565,
"step": 2140
},
{
"epoch": 7.78584392014519,
"grad_norm": 0.7734375,
"learning_rate": 2.8066019966134904e-05,
"loss": 0.4548,
"step": 2145
},
{
"epoch": 7.80399274047187,
"grad_norm": 0.73046875,
"learning_rate": 2.7626596189492983e-05,
"loss": 0.4584,
"step": 2150
},
{
"epoch": 7.822141560798548,
"grad_norm": 0.79296875,
"learning_rate": 2.719008760300359e-05,
"loss": 0.4523,
"step": 2155
},
{
"epoch": 7.840290381125227,
"grad_norm": 0.70703125,
"learning_rate": 2.6756511789168925e-05,
"loss": 0.4503,
"step": 2160
},
{
"epoch": 7.8584392014519056,
"grad_norm": 0.73046875,
"learning_rate": 2.6325886212359498e-05,
"loss": 0.451,
"step": 2165
},
{
"epoch": 7.876588021778584,
"grad_norm": 0.75390625,
"learning_rate": 2.589822821811083e-05,
"loss": 0.4574,
"step": 2170
},
{
"epoch": 7.894736842105263,
"grad_norm": 0.7109375,
"learning_rate": 2.5473555032424533e-05,
"loss": 0.4692,
"step": 2175
},
{
"epoch": 7.9128856624319415,
"grad_norm": 0.76953125,
"learning_rate": 2.5051883761074614e-05,
"loss": 0.4595,
"step": 2180
},
{
"epoch": 7.931034482758621,
"grad_norm": 0.66796875,
"learning_rate": 2.4633231388918378e-05,
"loss": 0.4543,
"step": 2185
},
{
"epoch": 7.9491833030853,
"grad_norm": 0.7265625,
"learning_rate": 2.4217614779212315e-05,
"loss": 0.4552,
"step": 2190
},
{
"epoch": 7.967332123411978,
"grad_norm": 0.8203125,
"learning_rate": 2.3805050672932928e-05,
"loss": 0.4542,
"step": 2195
},
{
"epoch": 7.985480943738657,
"grad_norm": 0.72265625,
"learning_rate": 2.339555568810221e-05,
"loss": 0.4678,
"step": 2200
},
{
"epoch": 8.0,
"eval_loss": 2.8434367179870605,
"eval_runtime": 1.027,
"eval_samples_per_second": 4.869,
"eval_steps_per_second": 1.947,
"step": 2204
},
{
"epoch": 8.003629764065336,
"grad_norm": 0.6875,
"learning_rate": 2.2989146319118425e-05,
"loss": 0.4538,
"step": 2205
},
{
"epoch": 8.021778584392015,
"grad_norm": 0.6484375,
"learning_rate": 2.2585838936091754e-05,
"loss": 0.4327,
"step": 2210
},
{
"epoch": 8.039927404718693,
"grad_norm": 0.71875,
"learning_rate": 2.2185649784184746e-05,
"loss": 0.4241,
"step": 2215
},
{
"epoch": 8.058076225045372,
"grad_norm": 0.76171875,
"learning_rate": 2.178859498295809e-05,
"loss": 0.4201,
"step": 2220
},
{
"epoch": 8.07622504537205,
"grad_norm": 0.77734375,
"learning_rate": 2.139469052572127e-05,
"loss": 0.4283,
"step": 2225
},
{
"epoch": 8.09437386569873,
"grad_norm": 0.703125,
"learning_rate": 2.1003952278888382e-05,
"loss": 0.4324,
"step": 2230
},
{
"epoch": 8.11252268602541,
"grad_norm": 0.6953125,
"learning_rate": 2.0616395981339075e-05,
"loss": 0.4288,
"step": 2235
},
{
"epoch": 8.130671506352087,
"grad_norm": 0.7421875,
"learning_rate": 2.0232037243784475e-05,
"loss": 0.4235,
"step": 2240
},
{
"epoch": 8.148820326678766,
"grad_norm": 0.7734375,
"learning_rate": 1.985089154813846e-05,
"loss": 0.4291,
"step": 2245
},
{
"epoch": 8.166969147005444,
"grad_norm": 0.71875,
"learning_rate": 1.947297424689414e-05,
"loss": 0.4272,
"step": 2250
},
{
"epoch": 8.185117967332124,
"grad_norm": 0.69921875,
"learning_rate": 1.9098300562505266e-05,
"loss": 0.4274,
"step": 2255
},
{
"epoch": 8.203266787658801,
"grad_norm": 0.74609375,
"learning_rate": 1.8726885586773212e-05,
"loss": 0.4336,
"step": 2260
},
{
"epoch": 8.221415607985481,
"grad_norm": 0.6953125,
"learning_rate": 1.835874428023905e-05,
"loss": 0.4359,
"step": 2265
},
{
"epoch": 8.23956442831216,
"grad_norm": 0.703125,
"learning_rate": 1.7993891471580893e-05,
"loss": 0.4275,
"step": 2270
},
{
"epoch": 8.257713248638838,
"grad_norm": 0.6796875,
"learning_rate": 1.763234185701673e-05,
"loss": 0.4315,
"step": 2275
},
{
"epoch": 8.275862068965518,
"grad_norm": 0.70703125,
"learning_rate": 1.7274109999712295e-05,
"loss": 0.4249,
"step": 2280
},
{
"epoch": 8.294010889292196,
"grad_norm": 0.734375,
"learning_rate": 1.6919210329194533e-05,
"loss": 0.4276,
"step": 2285
},
{
"epoch": 8.312159709618875,
"grad_norm": 0.765625,
"learning_rate": 1.6567657140770475e-05,
"loss": 0.4315,
"step": 2290
},
{
"epoch": 8.330308529945553,
"grad_norm": 0.71875,
"learning_rate": 1.621946459495127e-05,
"loss": 0.4247,
"step": 2295
},
{
"epoch": 8.348457350272232,
"grad_norm": 0.76171875,
"learning_rate": 1.587464671688187e-05,
"loss": 0.4441,
"step": 2300
},
{
"epoch": 8.366606170598912,
"grad_norm": 0.6875,
"learning_rate": 1.553321739577619e-05,
"loss": 0.4267,
"step": 2305
},
{
"epoch": 8.38475499092559,
"grad_norm": 0.66796875,
"learning_rate": 1.5195190384357404e-05,
"loss": 0.4224,
"step": 2310
},
{
"epoch": 8.40290381125227,
"grad_norm": 0.73828125,
"learning_rate": 1.4860579298304312e-05,
"loss": 0.4348,
"step": 2315
},
{
"epoch": 8.421052631578947,
"grad_norm": 0.703125,
"learning_rate": 1.4529397615702656e-05,
"loss": 0.4321,
"step": 2320
},
{
"epoch": 8.439201451905626,
"grad_norm": 0.73828125,
"learning_rate": 1.4201658676502294e-05,
"loss": 0.4386,
"step": 2325
},
{
"epoch": 8.457350272232304,
"grad_norm": 0.77734375,
"learning_rate": 1.3877375681979943e-05,
"loss": 0.4366,
"step": 2330
},
{
"epoch": 8.475499092558984,
"grad_norm": 0.72265625,
"learning_rate": 1.3556561694207338e-05,
"loss": 0.4336,
"step": 2335
},
{
"epoch": 8.493647912885663,
"grad_norm": 0.75,
"learning_rate": 1.3239229635525074e-05,
"loss": 0.4274,
"step": 2340
},
{
"epoch": 8.511796733212341,
"grad_norm": 0.7109375,
"learning_rate": 1.2925392288022298e-05,
"loss": 0.4259,
"step": 2345
},
{
"epoch": 8.52994555353902,
"grad_norm": 0.66796875,
"learning_rate": 1.2615062293021507e-05,
"loss": 0.427,
"step": 2350
},
{
"epoch": 8.548094373865698,
"grad_norm": 0.67578125,
"learning_rate": 1.230825215056971e-05,
"loss": 0.4226,
"step": 2355
},
{
"epoch": 8.566243194192378,
"grad_norm": 0.7421875,
"learning_rate": 1.2004974218934695e-05,
"loss": 0.4361,
"step": 2360
},
{
"epoch": 8.584392014519056,
"grad_norm": 0.65625,
"learning_rate": 1.1705240714107302e-05,
"loss": 0.4291,
"step": 2365
},
{
"epoch": 8.602540834845735,
"grad_norm": 1.171875,
"learning_rate": 1.1409063709309442e-05,
"loss": 0.4416,
"step": 2370
},
{
"epoch": 8.620689655172415,
"grad_norm": 0.69921875,
"learning_rate": 1.1116455134507664e-05,
"loss": 0.425,
"step": 2375
},
{
"epoch": 8.638838475499092,
"grad_norm": 0.74609375,
"learning_rate": 1.0827426775932658e-05,
"loss": 0.4316,
"step": 2380
},
{
"epoch": 8.656987295825772,
"grad_norm": 0.703125,
"learning_rate": 1.054199027560463e-05,
"loss": 0.425,
"step": 2385
},
{
"epoch": 8.67513611615245,
"grad_norm": 0.6953125,
"learning_rate": 1.026015713086418e-05,
"loss": 0.4344,
"step": 2390
},
{
"epoch": 8.69328493647913,
"grad_norm": 0.76171875,
"learning_rate": 9.98193869390922e-06,
"loss": 0.4261,
"step": 2395
},
{
"epoch": 8.711433756805807,
"grad_norm": 0.75,
"learning_rate": 9.707346171337894e-06,
"loss": 0.4323,
"step": 2400
},
{
"epoch": 8.729582577132486,
"grad_norm": 0.71484375,
"learning_rate": 9.436390623696911e-06,
"loss": 0.4309,
"step": 2405
},
{
"epoch": 8.747731397459166,
"grad_norm": 0.76171875,
"learning_rate": 9.16908296503628e-06,
"loss": 0.4278,
"step": 2410
},
{
"epoch": 8.765880217785844,
"grad_norm": 0.765625,
"learning_rate": 8.905433962469489e-06,
"loss": 0.4289,
"step": 2415
},
{
"epoch": 8.784029038112523,
"grad_norm": 0.76171875,
"learning_rate": 8.645454235739903e-06,
"loss": 0.4323,
"step": 2420
},
{
"epoch": 8.802177858439201,
"grad_norm": 0.66796875,
"learning_rate": 8.38915425679304e-06,
"loss": 0.4286,
"step": 2425
},
{
"epoch": 8.82032667876588,
"grad_norm": 0.69140625,
"learning_rate": 8.13654434935467e-06,
"loss": 0.4267,
"step": 2430
},
{
"epoch": 8.838475499092558,
"grad_norm": 0.75390625,
"learning_rate": 7.887634688515e-06,
"loss": 0.4351,
"step": 2435
},
{
"epoch": 8.856624319419238,
"grad_norm": 0.7265625,
"learning_rate": 7.642435300318907e-06,
"loss": 0.4294,
"step": 2440
},
{
"epoch": 8.874773139745917,
"grad_norm": 0.73828125,
"learning_rate": 7.400956061361974e-06,
"loss": 0.4301,
"step": 2445
},
{
"epoch": 8.892921960072595,
"grad_norm": 0.69921875,
"learning_rate": 7.163206698392744e-06,
"loss": 0.4306,
"step": 2450
},
{
"epoch": 8.911070780399275,
"grad_norm": 0.7265625,
"learning_rate": 6.929196787920899e-06,
"loss": 0.433,
"step": 2455
},
{
"epoch": 8.929219600725952,
"grad_norm": 0.7265625,
"learning_rate": 6.698935755831492e-06,
"loss": 0.4332,
"step": 2460
},
{
"epoch": 8.947368421052632,
"grad_norm": 0.70703125,
"learning_rate": 6.472432877005341e-06,
"loss": 0.4364,
"step": 2465
},
{
"epoch": 8.96551724137931,
"grad_norm": 0.72265625,
"learning_rate": 6.2496972749453766e-06,
"loss": 0.434,
"step": 2470
},
{
"epoch": 8.98366606170599,
"grad_norm": 0.65625,
"learning_rate": 6.030737921409169e-06,
"loss": 0.433,
"step": 2475
},
{
"epoch": 8.998185117967331,
"eval_loss": 2.9603512287139893,
"eval_runtime": 1.0237,
"eval_samples_per_second": 4.884,
"eval_steps_per_second": 1.954,
"step": 2479
},
{
"epoch": 9.001814882032669,
"grad_norm": 0.68359375,
"learning_rate": 5.8155636360475385e-06,
"loss": 0.4369,
"step": 2480
},
{
"epoch": 9.019963702359346,
"grad_norm": 0.68359375,
"learning_rate": 5.604183086049342e-06,
"loss": 0.4232,
"step": 2485
},
{
"epoch": 9.038112522686026,
"grad_norm": 0.7109375,
"learning_rate": 5.396604785792281e-06,
"loss": 0.4265,
"step": 2490
},
{
"epoch": 9.056261343012704,
"grad_norm": 0.73828125,
"learning_rate": 5.192837096500058e-06,
"loss": 0.4255,
"step": 2495
},
{
"epoch": 9.074410163339383,
"grad_norm": 0.7265625,
"learning_rate": 4.992888225905468e-06,
"loss": 0.4207,
"step": 2500
},
{
"epoch": 9.092558983666061,
"grad_norm": 0.6875,
"learning_rate": 4.796766227919857e-06,
"loss": 0.4223,
"step": 2505
},
{
"epoch": 9.11070780399274,
"grad_norm": 0.72265625,
"learning_rate": 4.604479002308737e-06,
"loss": 0.4237,
"step": 2510
},
{
"epoch": 9.12885662431942,
"grad_norm": 0.7421875,
"learning_rate": 4.416034294373472e-06,
"loss": 0.4417,
"step": 2515
},
{
"epoch": 9.147005444646098,
"grad_norm": 0.69140625,
"learning_rate": 4.231439694639483e-06,
"loss": 0.4247,
"step": 2520
},
{
"epoch": 9.165154264972777,
"grad_norm": 0.70703125,
"learning_rate": 4.050702638550275e-06,
"loss": 0.4244,
"step": 2525
},
{
"epoch": 9.183303085299455,
"grad_norm": 0.69921875,
"learning_rate": 3.873830406168111e-06,
"loss": 0.4309,
"step": 2530
},
{
"epoch": 9.201451905626135,
"grad_norm": 0.71875,
"learning_rate": 3.7008301218807716e-06,
"loss": 0.4264,
"step": 2535
},
{
"epoch": 9.219600725952812,
"grad_norm": 0.73046875,
"learning_rate": 3.5317087541144377e-06,
"loss": 0.4226,
"step": 2540
},
{
"epoch": 9.237749546279492,
"grad_norm": 0.6953125,
"learning_rate": 3.3664731150531482e-06,
"loss": 0.4263,
"step": 2545
},
{
"epoch": 9.255898366606171,
"grad_norm": 0.7578125,
"learning_rate": 3.2051298603643753e-06,
"loss": 0.4272,
"step": 2550
},
{
"epoch": 9.27404718693285,
"grad_norm": 0.84375,
"learning_rate": 3.047685488930874e-06,
"loss": 0.4207,
"step": 2555
},
{
"epoch": 9.292196007259529,
"grad_norm": 0.73046875,
"learning_rate": 2.894146342588977e-06,
"loss": 0.426,
"step": 2560
},
{
"epoch": 9.310344827586206,
"grad_norm": 0.75390625,
"learning_rate": 2.744518605873092e-06,
"loss": 0.4308,
"step": 2565
},
{
"epoch": 9.328493647912886,
"grad_norm": 0.76953125,
"learning_rate": 2.5988083057666533e-06,
"loss": 0.4262,
"step": 2570
},
{
"epoch": 9.346642468239564,
"grad_norm": 0.75,
"learning_rate": 2.4570213114592954e-06,
"loss": 0.4225,
"step": 2575
},
{
"epoch": 9.364791288566243,
"grad_norm": 0.73046875,
"learning_rate": 2.3191633341104856e-06,
"loss": 0.4198,
"step": 2580
},
{
"epoch": 9.382940108892923,
"grad_norm": 0.69921875,
"learning_rate": 2.1852399266194314e-06,
"loss": 0.4168,
"step": 2585
},
{
"epoch": 9.4010889292196,
"grad_norm": 0.69921875,
"learning_rate": 2.05525648340148e-06,
"loss": 0.4234,
"step": 2590
},
{
"epoch": 9.41923774954628,
"grad_norm": 0.75,
"learning_rate": 1.9292182401707603e-06,
"loss": 0.4383,
"step": 2595
},
{
"epoch": 9.437386569872958,
"grad_norm": 0.68359375,
"learning_rate": 1.8071302737293295e-06,
"loss": 0.4203,
"step": 2600
},
{
"epoch": 9.455535390199637,
"grad_norm": 0.7265625,
"learning_rate": 1.6889975017626903e-06,
"loss": 0.4152,
"step": 2605
},
{
"epoch": 9.473684210526315,
"grad_norm": 0.6875,
"learning_rate": 1.574824682641629e-06,
"loss": 0.4275,
"step": 2610
},
{
"epoch": 9.491833030852995,
"grad_norm": 0.6640625,
"learning_rate": 1.4646164152307018e-06,
"loss": 0.4209,
"step": 2615
},
{
"epoch": 9.509981851179674,
"grad_norm": 0.734375,
"learning_rate": 1.3583771387028265e-06,
"loss": 0.429,
"step": 2620
},
{
"epoch": 9.528130671506352,
"grad_norm": 0.69140625,
"learning_rate": 1.2561111323605712e-06,
"loss": 0.4182,
"step": 2625
},
{
"epoch": 9.546279491833031,
"grad_norm": 0.7109375,
"learning_rate": 1.157822515463758e-06,
"loss": 0.424,
"step": 2630
},
{
"epoch": 9.56442831215971,
"grad_norm": 0.6875,
"learning_rate": 1.0635152470635512e-06,
"loss": 0.4211,
"step": 2635
},
{
"epoch": 9.582577132486389,
"grad_norm": 0.69140625,
"learning_rate": 9.731931258429638e-07,
"loss": 0.4202,
"step": 2640
},
{
"epoch": 9.600725952813066,
"grad_norm": 0.6953125,
"learning_rate": 8.868597899638898e-07,
"loss": 0.4183,
"step": 2645
},
{
"epoch": 9.618874773139746,
"grad_norm": 0.6796875,
"learning_rate": 8.04518716920466e-07,
"loss": 0.4258,
"step": 2650
},
{
"epoch": 9.637023593466425,
"grad_norm": 0.6953125,
"learning_rate": 7.261732233991513e-07,
"loss": 0.426,
"step": 2655
},
{
"epoch": 9.655172413793103,
"grad_norm": 0.69921875,
"learning_rate": 6.518264651449779e-07,
"loss": 0.4341,
"step": 2660
},
{
"epoch": 9.673321234119783,
"grad_norm": 0.7265625,
"learning_rate": 5.814814368345412e-07,
"loss": 0.4311,
"step": 2665
},
{
"epoch": 9.69147005444646,
"grad_norm": 0.72265625,
"learning_rate": 5.151409719553079e-07,
"loss": 0.4219,
"step": 2670
},
{
"epoch": 9.70961887477314,
"grad_norm": 0.6796875,
"learning_rate": 4.5280774269154115e-07,
"loss": 0.4205,
"step": 2675
},
{
"epoch": 9.727767695099818,
"grad_norm": 0.74609375,
"learning_rate": 3.9448425981661876e-07,
"loss": 0.4268,
"step": 2680
},
{
"epoch": 9.745916515426497,
"grad_norm": 0.69921875,
"learning_rate": 3.401728725919373e-07,
"loss": 0.4258,
"step": 2685
},
{
"epoch": 9.764065335753177,
"grad_norm": 0.73046875,
"learning_rate": 2.898757686722542e-07,
"loss": 0.4191,
"step": 2690
},
{
"epoch": 9.782214156079855,
"grad_norm": 0.78125,
"learning_rate": 2.4359497401758024e-07,
"loss": 0.4277,
"step": 2695
},
{
"epoch": 9.800362976406534,
"grad_norm": 0.68359375,
"learning_rate": 2.0133235281156736e-07,
"loss": 0.434,
"step": 2700
},
{
"epoch": 9.818511796733212,
"grad_norm": 0.69921875,
"learning_rate": 1.630896073864352e-07,
"loss": 0.4307,
"step": 2705
},
{
"epoch": 9.836660617059891,
"grad_norm": 0.71484375,
"learning_rate": 1.2886827815440372e-07,
"loss": 0.4291,
"step": 2710
},
{
"epoch": 9.85480943738657,
"grad_norm": 0.73046875,
"learning_rate": 9.866974354560965e-08,
"loss": 0.4284,
"step": 2715
},
{
"epoch": 9.872958257713249,
"grad_norm": 0.796875,
"learning_rate": 7.249521995263964e-08,
"loss": 0.4319,
"step": 2720
},
{
"epoch": 9.891107078039928,
"grad_norm": 0.71484375,
"learning_rate": 5.0345761681491746e-08,
"loss": 0.4329,
"step": 2725
},
{
"epoch": 9.909255898366606,
"grad_norm": 0.7421875,
"learning_rate": 3.2222260909087196e-08,
"loss": 0.4218,
"step": 2730
},
{
"epoch": 9.927404718693285,
"grad_norm": 1.0,
"learning_rate": 1.81254476474213e-08,
"loss": 0.4236,
"step": 2735
},
{
"epoch": 9.945553539019963,
"grad_norm": 0.69921875,
"learning_rate": 8.055889714064791e-09,
"loss": 0.424,
"step": 2740
},
{
"epoch": 9.963702359346643,
"grad_norm": 0.71875,
"learning_rate": 2.0139927093487664e-09,
"loss": 0.4362,
"step": 2745
},
{
"epoch": 9.98185117967332,
"grad_norm": 0.69140625,
"learning_rate": 0.0,
"loss": 0.4229,
"step": 2750
},
{
"epoch": 9.98185117967332,
"eval_loss": 2.9801077842712402,
"eval_runtime": 1.0256,
"eval_samples_per_second": 4.875,
"eval_steps_per_second": 1.95,
"step": 2750
},
{
"epoch": 9.98185117967332,
"step": 2750,
"total_flos": 4.2044012259841147e+18,
"train_loss": 1.3294648733139038,
"train_runtime": 22350.6599,
"train_samples_per_second": 1.971,
"train_steps_per_second": 0.123
}
],
"logging_steps": 5,
"max_steps": 2750,
"num_input_tokens_seen": 0,
"num_train_epochs": 10,
"save_steps": 100,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 4.2044012259841147e+18,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}