mistral-safe-sft-full / trainer_state.json
AmberYifan's picture
Model save
ba9cc5e verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 4.0,
"eval_steps": 500,
"global_step": 740,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.005405405405405406,
"grad_norm": 61.001103964725324,
"learning_rate": 1.0526315789473685e-06,
"loss": 1.9097,
"step": 1
},
{
"epoch": 0.02702702702702703,
"grad_norm": 35.97997402423919,
"learning_rate": 5.263157894736842e-06,
"loss": 1.7102,
"step": 5
},
{
"epoch": 0.05405405405405406,
"grad_norm": 30.10156193198182,
"learning_rate": 1.0526315789473684e-05,
"loss": 1.5516,
"step": 10
},
{
"epoch": 0.08108108108108109,
"grad_norm": 6.532111146208257,
"learning_rate": 1.578947368421053e-05,
"loss": 1.4536,
"step": 15
},
{
"epoch": 0.10810810810810811,
"grad_norm": 4.983904851506608,
"learning_rate": 1.999820922669738e-05,
"loss": 1.4112,
"step": 20
},
{
"epoch": 0.13513513513513514,
"grad_norm": 3.8732866550483545,
"learning_rate": 1.993559947963185e-05,
"loss": 1.4056,
"step": 25
},
{
"epoch": 0.16216216216216217,
"grad_norm": 3.0620367608079846,
"learning_rate": 1.9784091409455728e-05,
"loss": 1.4016,
"step": 30
},
{
"epoch": 0.1891891891891892,
"grad_norm": 3.2352358324557167,
"learning_rate": 1.9545040627715554e-05,
"loss": 1.4128,
"step": 35
},
{
"epoch": 0.21621621621621623,
"grad_norm": 3.022703286965415,
"learning_rate": 1.9220586030376135e-05,
"loss": 1.4148,
"step": 40
},
{
"epoch": 0.24324324324324326,
"grad_norm": 2.8955611484982033,
"learning_rate": 1.881363066014649e-05,
"loss": 1.4159,
"step": 45
},
{
"epoch": 0.2702702702702703,
"grad_norm": 2.8660620524615688,
"learning_rate": 1.8327815731637612e-05,
"loss": 1.42,
"step": 50
},
{
"epoch": 0.2972972972972973,
"grad_norm": 2.632475843797219,
"learning_rate": 1.7767488051760858e-05,
"loss": 1.4078,
"step": 55
},
{
"epoch": 0.32432432432432434,
"grad_norm": 2.5436272914892872,
"learning_rate": 1.713766112687139e-05,
"loss": 1.4015,
"step": 60
},
{
"epoch": 0.35135135135135137,
"grad_norm": 2.431239149211266,
"learning_rate": 1.644397030464877e-05,
"loss": 1.3807,
"step": 65
},
{
"epoch": 0.3783783783783784,
"grad_norm": 2.328539582829742,
"learning_rate": 1.5692622352080662e-05,
"loss": 1.3639,
"step": 70
},
{
"epoch": 0.40540540540540543,
"grad_norm": 2.708625235728292,
"learning_rate": 1.4890339920698334e-05,
"loss": 1.3836,
"step": 75
},
{
"epoch": 0.43243243243243246,
"grad_norm": 2.4202434897618317,
"learning_rate": 1.404430139595877e-05,
"loss": 1.3823,
"step": 80
},
{
"epoch": 0.4594594594594595,
"grad_norm": 2.243346482513894,
"learning_rate": 1.316207666896824e-05,
"loss": 1.3538,
"step": 85
},
{
"epoch": 0.4864864864864865,
"grad_norm": 2.4109049967489433,
"learning_rate": 1.2251559405226943e-05,
"loss": 1.3503,
"step": 90
},
{
"epoch": 0.5135135135135135,
"grad_norm": 2.242883116591874,
"learning_rate": 1.1320896416417026e-05,
"loss": 1.3562,
"step": 95
},
{
"epoch": 0.5405405405405406,
"grad_norm": 2.1959662600808203,
"learning_rate": 1.0378414767176706e-05,
"loss": 1.3447,
"step": 100
},
{
"epoch": 0.5675675675675675,
"grad_norm": 2.2348646553535714,
"learning_rate": 9.43254726906926e-06,
"loss": 1.3398,
"step": 105
},
{
"epoch": 0.5945945945945946,
"grad_norm": 2.21405256524929,
"learning_rate": 8.491757028386262e-06,
"loss": 1.3383,
"step": 110
},
{
"epoch": 0.6216216216216216,
"grad_norm": 2.1710850419967263,
"learning_rate": 7.564461722890082e-06,
"loss": 1.3271,
"step": 115
},
{
"epoch": 0.6486486486486487,
"grad_norm": 2.1752417729097338,
"learning_rate": 6.6589582850261025e-06,
"loss": 1.3176,
"step": 120
},
{
"epoch": 0.6756756756756757,
"grad_norm": 2.1322248363189686,
"learning_rate": 5.78334866549816e-06,
"loss": 1.321,
"step": 125
},
{
"epoch": 0.7027027027027027,
"grad_norm": 2.1516864548066525,
"learning_rate": 4.9454673414341945e-06,
"loss": 1.3074,
"step": 130
},
{
"epoch": 0.7297297297297297,
"grad_norm": 2.04797883118049,
"learning_rate": 4.152811217759529e-06,
"loss": 1.3123,
"step": 135
},
{
"epoch": 0.7567567567567568,
"grad_norm": 2.1202023783270367,
"learning_rate": 3.4124725489820643e-06,
"loss": 1.3026,
"step": 140
},
{
"epoch": 0.7837837837837838,
"grad_norm": 2.1203746539007158,
"learning_rate": 2.7310754815685627e-06,
"loss": 1.2838,
"step": 145
},
{
"epoch": 0.8108108108108109,
"grad_norm": 2.058016060238832,
"learning_rate": 2.114716784696342e-06,
"loss": 1.2805,
"step": 150
},
{
"epoch": 0.8378378378378378,
"grad_norm": 2.072702697896577,
"learning_rate": 1.5689112996891576e-06,
"loss": 1.276,
"step": 155
},
{
"epoch": 0.8648648648648649,
"grad_norm": 2.070358297476048,
"learning_rate": 1.0985425962260342e-06,
"loss": 1.2768,
"step": 160
},
{
"epoch": 0.8918918918918919,
"grad_norm": 1.993421902652665,
"learning_rate": 7.078192768243486e-07,
"loss": 1.2755,
"step": 165
},
{
"epoch": 0.918918918918919,
"grad_norm": 2.02321205406663,
"learning_rate": 4.0023732056077235e-07,
"loss": 1.276,
"step": 170
},
{
"epoch": 0.9459459459459459,
"grad_norm": 2.0216958269361154,
"learning_rate": 1.7854880295797406e-07,
"loss": 1.2694,
"step": 175
},
{
"epoch": 0.972972972972973,
"grad_norm": 2.0002791761840157,
"learning_rate": 4.473727191441124e-08,
"loss": 1.2669,
"step": 180
},
{
"epoch": 1.0,
"grad_norm": 1.9847900492979156,
"learning_rate": 0.0,
"loss": 1.2583,
"step": 185
},
{
"epoch": 1.027027027027027,
"grad_norm": 5.410199324361583,
"learning_rate": 1.853992867931721e-05,
"loss": 0.944,
"step": 190
},
{
"epoch": 1.054054054054054,
"grad_norm": 5.152995409876598,
"learning_rate": 1.8414852973000503e-05,
"loss": 0.956,
"step": 195
},
{
"epoch": 1.0810810810810811,
"grad_norm": 3.933917659184313,
"learning_rate": 1.8285096492438424e-05,
"loss": 0.9283,
"step": 200
},
{
"epoch": 1.1081081081081081,
"grad_norm": 2.759086157792593,
"learning_rate": 1.8150731414862623e-05,
"loss": 0.9151,
"step": 205
},
{
"epoch": 1.135135135135135,
"grad_norm": 2.910077750970637,
"learning_rate": 1.8011832481043577e-05,
"loss": 0.9413,
"step": 210
},
{
"epoch": 1.1621621621621623,
"grad_norm": 3.1297209827761985,
"learning_rate": 1.78684769537159e-05,
"loss": 0.9243,
"step": 215
},
{
"epoch": 1.1891891891891893,
"grad_norm": 3.406559269295354,
"learning_rate": 1.7720744574600865e-05,
"loss": 0.9474,
"step": 220
},
{
"epoch": 1.2162162162162162,
"grad_norm": 3.2575862603317876,
"learning_rate": 1.756871752004992e-05,
"loss": 0.9461,
"step": 225
},
{
"epoch": 1.2432432432432432,
"grad_norm": 3.2529118198639644,
"learning_rate": 1.7412480355334006e-05,
"loss": 0.9385,
"step": 230
},
{
"epoch": 1.2702702702702702,
"grad_norm": 3.096374104973297,
"learning_rate": 1.7252119987603976e-05,
"loss": 0.9564,
"step": 235
},
{
"epoch": 1.2972972972972974,
"grad_norm": 2.9780724324068695,
"learning_rate": 1.7087725617548385e-05,
"loss": 0.9456,
"step": 240
},
{
"epoch": 1.3243243243243243,
"grad_norm": 2.887431987514818,
"learning_rate": 1.6919388689775463e-05,
"loss": 0.9608,
"step": 245
},
{
"epoch": 1.3513513513513513,
"grad_norm": 3.420843173177096,
"learning_rate": 1.6747202841946928e-05,
"loss": 0.9817,
"step": 250
},
{
"epoch": 1.3783783783783785,
"grad_norm": 2.817422226047505,
"learning_rate": 1.6571263852691887e-05,
"loss": 0.9794,
"step": 255
},
{
"epoch": 1.4054054054054055,
"grad_norm": 2.8475793868122135,
"learning_rate": 1.639166958832985e-05,
"loss": 0.9249,
"step": 260
},
{
"epoch": 1.4324324324324325,
"grad_norm": 3.084978828155213,
"learning_rate": 1.6208519948432438e-05,
"loss": 0.9694,
"step": 265
},
{
"epoch": 1.4594594594594594,
"grad_norm": 2.9443084757542923,
"learning_rate": 1.6021916810254096e-05,
"loss": 0.9646,
"step": 270
},
{
"epoch": 1.4864864864864864,
"grad_norm": 2.7978723842230555,
"learning_rate": 1.5831963972062734e-05,
"loss": 0.9614,
"step": 275
},
{
"epoch": 1.5135135135135136,
"grad_norm": 2.8201773773365155,
"learning_rate": 1.5638767095401778e-05,
"loss": 0.9785,
"step": 280
},
{
"epoch": 1.5405405405405406,
"grad_norm": 2.6511276981563907,
"learning_rate": 1.5442433646315792e-05,
"loss": 0.956,
"step": 285
},
{
"epoch": 1.5675675675675675,
"grad_norm": 2.907386340027783,
"learning_rate": 1.5243072835572319e-05,
"loss": 0.9386,
"step": 290
},
{
"epoch": 1.5945945945945947,
"grad_norm": 3.2379041081891455,
"learning_rate": 1.5040795557913246e-05,
"loss": 0.9891,
"step": 295
},
{
"epoch": 1.6216216216216215,
"grad_norm": 2.659956186011908,
"learning_rate": 1.4835714330369445e-05,
"loss": 0.9806,
"step": 300
},
{
"epoch": 1.6486486486486487,
"grad_norm": 2.794287463461555,
"learning_rate": 1.4627943229672992e-05,
"loss": 0.9856,
"step": 305
},
{
"epoch": 1.6756756756756757,
"grad_norm": 2.9825237321465647,
"learning_rate": 1.4417597828801833e-05,
"loss": 1.011,
"step": 310
},
{
"epoch": 1.7027027027027026,
"grad_norm": 2.6599834596055145,
"learning_rate": 1.4204795132692146e-05,
"loss": 0.9682,
"step": 315
},
{
"epoch": 1.7297297297297298,
"grad_norm": 2.63763723653628,
"learning_rate": 1.3989653513154165e-05,
"loss": 0.9703,
"step": 320
},
{
"epoch": 1.7567567567567568,
"grad_norm": 2.702492331093115,
"learning_rate": 1.37722926430277e-05,
"loss": 0.9695,
"step": 325
},
{
"epoch": 1.7837837837837838,
"grad_norm": 2.6668927523262234,
"learning_rate": 1.3552833429613939e-05,
"loss": 0.9621,
"step": 330
},
{
"epoch": 1.810810810810811,
"grad_norm": 2.830202295132874,
"learning_rate": 1.3331397947420578e-05,
"loss": 1.0006,
"step": 335
},
{
"epoch": 1.8378378378378377,
"grad_norm": 2.960495369045161,
"learning_rate": 1.3108109370257714e-05,
"loss": 0.9658,
"step": 340
},
{
"epoch": 1.864864864864865,
"grad_norm": 2.8089453201114005,
"learning_rate": 1.288309190272222e-05,
"loss": 0.978,
"step": 345
},
{
"epoch": 1.8918918918918919,
"grad_norm": 2.690571777156308,
"learning_rate": 1.2656470711108763e-05,
"loss": 0.9694,
"step": 350
},
{
"epoch": 1.9189189189189189,
"grad_norm": 3.072726161364189,
"learning_rate": 1.2428371853785872e-05,
"loss": 0.995,
"step": 355
},
{
"epoch": 1.945945945945946,
"grad_norm": 2.650492212953216,
"learning_rate": 1.2198922211075779e-05,
"loss": 0.9633,
"step": 360
},
{
"epoch": 1.972972972972973,
"grad_norm": 2.4600728300935524,
"learning_rate": 1.1968249414677055e-05,
"loss": 0.9518,
"step": 365
},
{
"epoch": 2.0,
"grad_norm": 2.660921417467904,
"learning_rate": 1.1736481776669307e-05,
"loss": 0.9697,
"step": 370
},
{
"epoch": 2.0,
"eval_loss": 1.2888436317443848,
"eval_runtime": 49.8535,
"eval_samples_per_second": 13.219,
"eval_steps_per_second": 0.421,
"step": 370
},
{
"epoch": 2.027027027027027,
"grad_norm": 4.650300715219466,
"learning_rate": 1.150374821813937e-05,
"loss": 0.4819,
"step": 375
},
{
"epoch": 2.054054054054054,
"grad_norm": 3.6035388443171055,
"learning_rate": 1.1270178197468788e-05,
"loss": 0.4496,
"step": 380
},
{
"epoch": 2.081081081081081,
"grad_norm": 3.2195791970064738,
"learning_rate": 1.1035901638322392e-05,
"loss": 0.4485,
"step": 385
},
{
"epoch": 2.108108108108108,
"grad_norm": 3.969479182342283,
"learning_rate": 1.080104885737807e-05,
"loss": 0.431,
"step": 390
},
{
"epoch": 2.135135135135135,
"grad_norm": 3.5220560829841108,
"learning_rate": 1.0565750491837925e-05,
"loss": 0.4358,
"step": 395
},
{
"epoch": 2.1621621621621623,
"grad_norm": 3.054414879545972,
"learning_rate": 1.0330137426761136e-05,
"loss": 0.4195,
"step": 400
},
{
"epoch": 2.189189189189189,
"grad_norm": 2.8407451960442844,
"learning_rate": 1.0094340722258969e-05,
"loss": 0.4139,
"step": 405
},
{
"epoch": 2.2162162162162162,
"grad_norm": 3.1380887590616835,
"learning_rate": 9.858491540592383e-06,
"loss": 0.4288,
"step": 410
},
{
"epoch": 2.2432432432432434,
"grad_norm": 3.0000811662471607,
"learning_rate": 9.622721073212831e-06,
"loss": 0.4307,
"step": 415
},
{
"epoch": 2.27027027027027,
"grad_norm": 2.9078632951242,
"learning_rate": 9.38716046778684e-06,
"loss": 0.4346,
"step": 420
},
{
"epoch": 2.2972972972972974,
"grad_norm": 3.0788575744315363,
"learning_rate": 9.151940755244912e-06,
"loss": 0.4354,
"step": 425
},
{
"epoch": 2.3243243243243246,
"grad_norm": 3.2452582293940764,
"learning_rate": 8.917192776895382e-06,
"loss": 0.43,
"step": 430
},
{
"epoch": 2.3513513513513513,
"grad_norm": 2.9959040185970345,
"learning_rate": 8.683047111643764e-06,
"loss": 0.4288,
"step": 435
},
{
"epoch": 2.3783783783783785,
"grad_norm": 2.9976789493802642,
"learning_rate": 8.449634003358022e-06,
"loss": 0.4154,
"step": 440
},
{
"epoch": 2.4054054054054053,
"grad_norm": 3.342720495579085,
"learning_rate": 8.217083288420241e-06,
"loss": 0.4184,
"step": 445
},
{
"epoch": 2.4324324324324325,
"grad_norm": 2.9954997310319817,
"learning_rate": 7.985524323504948e-06,
"loss": 0.4244,
"step": 450
},
{
"epoch": 2.4594594594594597,
"grad_norm": 3.1107476712409436,
"learning_rate": 7.755085913624274e-06,
"loss": 0.4201,
"step": 455
},
{
"epoch": 2.4864864864864864,
"grad_norm": 2.982814359917317,
"learning_rate": 7.525896240479977e-06,
"loss": 0.4223,
"step": 460
},
{
"epoch": 2.5135135135135136,
"grad_norm": 2.918536641642885,
"learning_rate": 7.29808279116218e-06,
"loss": 0.4355,
"step": 465
},
{
"epoch": 2.5405405405405403,
"grad_norm": 3.0794937831504914,
"learning_rate": 7.071772287234497e-06,
"loss": 0.4182,
"step": 470
},
{
"epoch": 2.5675675675675675,
"grad_norm": 3.350485105176484,
"learning_rate": 6.8470906142449764e-06,
"loss": 0.4173,
"step": 475
},
{
"epoch": 2.5945945945945947,
"grad_norm": 2.978358242778936,
"learning_rate": 6.624162751702077e-06,
"loss": 0.4257,
"step": 480
},
{
"epoch": 2.6216216216216215,
"grad_norm": 3.03990431149403,
"learning_rate": 6.403112703554643e-06,
"loss": 0.4198,
"step": 485
},
{
"epoch": 2.6486486486486487,
"grad_norm": 3.2537909886176797,
"learning_rate": 6.184063429214515e-06,
"loss": 0.4289,
"step": 490
},
{
"epoch": 2.6756756756756754,
"grad_norm": 2.9401645896134068,
"learning_rate": 5.967136775160188e-06,
"loss": 0.4229,
"step": 495
},
{
"epoch": 2.7027027027027026,
"grad_norm": 2.8796497097799385,
"learning_rate": 5.752453407159521e-06,
"loss": 0.4134,
"step": 500
},
{
"epoch": 2.72972972972973,
"grad_norm": 3.240398354047165,
"learning_rate": 5.5401327431492425e-06,
"loss": 0.4212,
"step": 505
},
{
"epoch": 2.756756756756757,
"grad_norm": 3.0875752772388556,
"learning_rate": 5.33029288680852e-06,
"loss": 0.4366,
"step": 510
},
{
"epoch": 2.7837837837837838,
"grad_norm": 2.950418599154064,
"learning_rate": 5.1230505618636575e-06,
"loss": 0.4132,
"step": 515
},
{
"epoch": 2.810810810810811,
"grad_norm": 2.9691405536481708,
"learning_rate": 4.918521047160309e-06,
"loss": 0.421,
"step": 520
},
{
"epoch": 2.8378378378378377,
"grad_norm": 3.288503082597523,
"learning_rate": 4.716818112539485e-06,
"loss": 0.4177,
"step": 525
},
{
"epoch": 2.864864864864865,
"grad_norm": 2.8513863896448304,
"learning_rate": 4.518053955552903e-06,
"loss": 0.4027,
"step": 530
},
{
"epoch": 2.891891891891892,
"grad_norm": 2.906451097506746,
"learning_rate": 4.322339139052922e-06,
"loss": 0.4104,
"step": 535
},
{
"epoch": 2.918918918918919,
"grad_norm": 2.8739958546058255,
"learning_rate": 4.1297825296918145e-06,
"loss": 0.4006,
"step": 540
},
{
"epoch": 2.945945945945946,
"grad_norm": 2.9696442817419113,
"learning_rate": 3.940491237364519e-06,
"loss": 0.4,
"step": 545
},
{
"epoch": 2.972972972972973,
"grad_norm": 3.0645448850809793,
"learning_rate": 3.754570555628613e-06,
"loss": 0.4043,
"step": 550
},
{
"epoch": 3.0,
"grad_norm": 2.9322335051873423,
"learning_rate": 3.5721239031346067e-06,
"loss": 0.4114,
"step": 555
},
{
"epoch": 3.0,
"eval_loss": 1.4739221334457397,
"eval_runtime": 47.6894,
"eval_samples_per_second": 13.819,
"eval_steps_per_second": 0.44,
"step": 555
},
{
"epoch": 3.027027027027027,
"grad_norm": 2.239642555101549,
"learning_rate": 3.3932527660991877e-06,
"loss": 0.1581,
"step": 560
},
{
"epoch": 3.054054054054054,
"grad_norm": 3.1082908400603464,
"learning_rate": 3.2180566418533365e-06,
"loss": 0.1515,
"step": 565
},
{
"epoch": 3.081081081081081,
"grad_norm": 2.152743576813358,
"learning_rate": 3.0466329834968234e-06,
"loss": 0.1438,
"step": 570
},
{
"epoch": 3.108108108108108,
"grad_norm": 2.232485171340327,
"learning_rate": 2.879077145689746e-06,
"loss": 0.1452,
"step": 575
},
{
"epoch": 3.135135135135135,
"grad_norm": 2.1315167934046464,
"learning_rate": 2.715482331611393e-06,
"loss": 0.1483,
"step": 580
},
{
"epoch": 3.1621621621621623,
"grad_norm": 1.9641583881749838,
"learning_rate": 2.5559395411158116e-06,
"loss": 0.1428,
"step": 585
},
{
"epoch": 3.189189189189189,
"grad_norm": 2.136285665522953,
"learning_rate": 2.4005375201130275e-06,
"loss": 0.1456,
"step": 590
},
{
"epoch": 3.2162162162162162,
"grad_norm": 2.3296203504449413,
"learning_rate": 2.249362711203985e-06,
"loss": 0.1451,
"step": 595
},
{
"epoch": 3.2432432432432434,
"grad_norm": 2.1953889356463105,
"learning_rate": 2.102499205596743e-06,
"loss": 0.1412,
"step": 600
},
{
"epoch": 3.27027027027027,
"grad_norm": 1.9846220715180103,
"learning_rate": 1.960028696330596e-06,
"loss": 0.139,
"step": 605
},
{
"epoch": 3.2972972972972974,
"grad_norm": 2.0955721946100345,
"learning_rate": 1.8220304328342253e-06,
"loss": 0.1379,
"step": 610
},
{
"epoch": 3.3243243243243246,
"grad_norm": 2.13188633069558,
"learning_rate": 1.688581176843066e-06,
"loss": 0.1401,
"step": 615
},
{
"epoch": 3.3513513513513513,
"grad_norm": 2.152817300804356,
"learning_rate": 1.5597551597004968e-06,
"loss": 0.1389,
"step": 620
},
{
"epoch": 3.3783783783783785,
"grad_norm": 2.0539946534038305,
"learning_rate": 1.4356240410665435e-06,
"loss": 0.141,
"step": 625
},
{
"epoch": 3.4054054054054053,
"grad_norm": 2.162669615931553,
"learning_rate": 1.3162568690570743e-06,
"loss": 0.1437,
"step": 630
},
{
"epoch": 3.4324324324324325,
"grad_norm": 2.153506447556338,
"learning_rate": 1.2017200418357077e-06,
"loss": 0.1417,
"step": 635
},
{
"epoch": 3.4594594594594597,
"grad_norm": 2.1701177657529964,
"learning_rate": 1.0920772706797166e-06,
"loss": 0.142,
"step": 640
},
{
"epoch": 3.4864864864864864,
"grad_norm": 2.0721670365744407,
"learning_rate": 9.873895445405523e-07,
"loss": 0.14,
"step": 645
},
{
"epoch": 3.5135135135135136,
"grad_norm": 2.0406246464567612,
"learning_rate": 8.87715096118642e-07,
"loss": 0.1398,
"step": 650
},
{
"epoch": 3.5405405405405403,
"grad_norm": 2.0428479721186035,
"learning_rate": 7.931093694713687e-07,
"loss": 0.1382,
"step": 655
},
{
"epoch": 3.5675675675675675,
"grad_norm": 2.234111473839529,
"learning_rate": 7.03624989172228e-07,
"loss": 0.1428,
"step": 660
},
{
"epoch": 3.5945945945945947,
"grad_norm": 1.9632293170603063,
"learning_rate": 6.193117310383412e-07,
"loss": 0.136,
"step": 665
},
{
"epoch": 3.6216216216216215,
"grad_norm": 2.0540773804658286,
"learning_rate": 5.402164944425758e-07,
"loss": 0.1355,
"step": 670
},
{
"epoch": 3.6486486486486487,
"grad_norm": 2.0789858869481144,
"learning_rate": 4.66383276225707e-07,
"loss": 0.1404,
"step": 675
},
{
"epoch": 3.6756756756756754,
"grad_norm": 2.2440192963067678,
"learning_rate": 3.97853146223105e-07,
"loss": 0.1384,
"step": 680
},
{
"epoch": 3.7027027027027026,
"grad_norm": 2.0732015708107703,
"learning_rate": 3.346642244195863e-07,
"loss": 0.1359,
"step": 685
},
{
"epoch": 3.72972972972973,
"grad_norm": 1.982186914698493,
"learning_rate": 2.7685165974510987e-07,
"loss": 0.1367,
"step": 690
},
{
"epoch": 3.756756756756757,
"grad_norm": 2.0636323678600306,
"learning_rate": 2.2444761052313857e-07,
"loss": 0.136,
"step": 695
},
{
"epoch": 3.7837837837837838,
"grad_norm": 2.133975894121271,
"learning_rate": 1.7748122658251877e-07,
"loss": 0.1398,
"step": 700
},
{
"epoch": 3.810810810810811,
"grad_norm": 1.9445931920281623,
"learning_rate": 1.3597863304285475e-07,
"loss": 0.1384,
"step": 705
},
{
"epoch": 3.8378378378378377,
"grad_norm": 2.1247798692529414,
"learning_rate": 9.996291578236228e-08,
"loss": 0.1361,
"step": 710
},
{
"epoch": 3.864864864864865,
"grad_norm": 2.128261177290376,
"learning_rate": 6.945410859632295e-08,
"loss": 0.1326,
"step": 715
},
{
"epoch": 3.891891891891892,
"grad_norm": 2.059927006960735,
"learning_rate": 4.44691820532539e-08,
"loss": 0.1355,
"step": 720
},
{
"epoch": 3.918918918918919,
"grad_norm": 1.942567955434284,
"learning_rate": 2.5022034055003363e-08,
"loss": 0.1415,
"step": 725
},
{
"epoch": 3.945945945945946,
"grad_norm": 1.9857313689273417,
"learning_rate": 1.1123482106021322e-08,
"loss": 0.1356,
"step": 730
},
{
"epoch": 3.972972972972973,
"grad_norm": 2.0940341606759816,
"learning_rate": 2.7812572961127825e-09,
"loss": 0.1399,
"step": 735
},
{
"epoch": 4.0,
"grad_norm": 1.9231586161743661,
"learning_rate": 0.0,
"loss": 0.1361,
"step": 740
},
{
"epoch": 4.0,
"eval_loss": 1.7972235679626465,
"eval_runtime": 46.1514,
"eval_samples_per_second": 14.279,
"eval_steps_per_second": 0.455,
"step": 740
},
{
"epoch": 4.0,
"step": 740,
"total_flos": 77470472601600.0,
"train_loss": 0.38161755220310106,
"train_runtime": 4944.5179,
"train_samples_per_second": 4.784,
"train_steps_per_second": 0.15
}
],
"logging_steps": 5,
"max_steps": 740,
"num_input_tokens_seen": 0,
"num_train_epochs": 4,
"save_steps": 800,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 77470472601600.0,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}