qwen2-0.5b-sft / trainer_state.json
yinmingzhang's picture
Model save
664c53d verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.9993049349617714,
"eval_steps": 500,
"global_step": 1258,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0007943600436898023,
"grad_norm": 11.12145892458997,
"learning_rate": 1.5873015873015874e-07,
"loss": 1.8009,
"step": 1
},
{
"epoch": 0.003971800218449012,
"grad_norm": 10.234395651621842,
"learning_rate": 7.936507936507937e-07,
"loss": 1.7723,
"step": 5
},
{
"epoch": 0.007943600436898023,
"grad_norm": 2.755899780672782,
"learning_rate": 1.5873015873015873e-06,
"loss": 1.709,
"step": 10
},
{
"epoch": 0.011915400655347037,
"grad_norm": 1.5712351118319727,
"learning_rate": 2.380952380952381e-06,
"loss": 1.6524,
"step": 15
},
{
"epoch": 0.015887200873796047,
"grad_norm": 1.1660528045797167,
"learning_rate": 3.1746031746031746e-06,
"loss": 1.6423,
"step": 20
},
{
"epoch": 0.01985900109224506,
"grad_norm": 1.0216120526399777,
"learning_rate": 3.968253968253968e-06,
"loss": 1.6244,
"step": 25
},
{
"epoch": 0.023830801310694073,
"grad_norm": 0.9509263430769295,
"learning_rate": 4.761904761904762e-06,
"loss": 1.6241,
"step": 30
},
{
"epoch": 0.027802601529143083,
"grad_norm": 0.9135838434062775,
"learning_rate": 5.555555555555557e-06,
"loss": 1.6066,
"step": 35
},
{
"epoch": 0.03177440174759209,
"grad_norm": 0.9277594740139052,
"learning_rate": 6.349206349206349e-06,
"loss": 1.5847,
"step": 40
},
{
"epoch": 0.035746201966041107,
"grad_norm": 0.8992831827632032,
"learning_rate": 7.1428571428571436e-06,
"loss": 1.5999,
"step": 45
},
{
"epoch": 0.03971800218449012,
"grad_norm": 0.9718735971756692,
"learning_rate": 7.936507936507936e-06,
"loss": 1.6023,
"step": 50
},
{
"epoch": 0.04368980240293913,
"grad_norm": 0.947148151716203,
"learning_rate": 8.730158730158731e-06,
"loss": 1.5825,
"step": 55
},
{
"epoch": 0.04766160262138815,
"grad_norm": 0.9159573540177632,
"learning_rate": 9.523809523809525e-06,
"loss": 1.5978,
"step": 60
},
{
"epoch": 0.05163340283983715,
"grad_norm": 0.9226608146834196,
"learning_rate": 1.031746031746032e-05,
"loss": 1.6003,
"step": 65
},
{
"epoch": 0.055605203058286166,
"grad_norm": 0.9212318659674721,
"learning_rate": 1.1111111111111113e-05,
"loss": 1.5856,
"step": 70
},
{
"epoch": 0.05957700327673518,
"grad_norm": 0.8933451806059763,
"learning_rate": 1.1904761904761905e-05,
"loss": 1.5845,
"step": 75
},
{
"epoch": 0.06354880349518419,
"grad_norm": 0.9657915647359273,
"learning_rate": 1.2698412698412699e-05,
"loss": 1.6089,
"step": 80
},
{
"epoch": 0.0675206037136332,
"grad_norm": 0.9337955988860143,
"learning_rate": 1.3492063492063494e-05,
"loss": 1.5815,
"step": 85
},
{
"epoch": 0.07149240393208221,
"grad_norm": 0.9873409918119863,
"learning_rate": 1.4285714285714287e-05,
"loss": 1.5816,
"step": 90
},
{
"epoch": 0.07546420415053123,
"grad_norm": 0.9015172917838395,
"learning_rate": 1.507936507936508e-05,
"loss": 1.5958,
"step": 95
},
{
"epoch": 0.07943600436898024,
"grad_norm": 0.9772975467717536,
"learning_rate": 1.5873015873015872e-05,
"loss": 1.5911,
"step": 100
},
{
"epoch": 0.08340780458742925,
"grad_norm": 0.9472359700179834,
"learning_rate": 1.6666666666666667e-05,
"loss": 1.5865,
"step": 105
},
{
"epoch": 0.08737960480587827,
"grad_norm": 0.9247784591978276,
"learning_rate": 1.7460317460317463e-05,
"loss": 1.5793,
"step": 110
},
{
"epoch": 0.09135140502432727,
"grad_norm": 0.9231839444533193,
"learning_rate": 1.8253968253968254e-05,
"loss": 1.5822,
"step": 115
},
{
"epoch": 0.0953232052427763,
"grad_norm": 0.9573224627010042,
"learning_rate": 1.904761904761905e-05,
"loss": 1.5718,
"step": 120
},
{
"epoch": 0.0992950054612253,
"grad_norm": 1.0158728633269807,
"learning_rate": 1.9841269841269845e-05,
"loss": 1.5773,
"step": 125
},
{
"epoch": 0.1032668056796743,
"grad_norm": 1.0442473658793399,
"learning_rate": 1.999938384153589e-05,
"loss": 1.585,
"step": 130
},
{
"epoch": 0.10723860589812333,
"grad_norm": 0.9568516629501055,
"learning_rate": 1.999688082790923e-05,
"loss": 1.5868,
"step": 135
},
{
"epoch": 0.11121040611657233,
"grad_norm": 0.9665579055787764,
"learning_rate": 1.9992452930796544e-05,
"loss": 1.5776,
"step": 140
},
{
"epoch": 0.11518220633502135,
"grad_norm": 0.9323962932240065,
"learning_rate": 1.9986101002782376e-05,
"loss": 1.5789,
"step": 145
},
{
"epoch": 0.11915400655347036,
"grad_norm": 0.9477001351382732,
"learning_rate": 1.997782626692034e-05,
"loss": 1.5814,
"step": 150
},
{
"epoch": 0.12312580677191937,
"grad_norm": 0.9311645028761003,
"learning_rate": 1.9967630316497663e-05,
"loss": 1.5658,
"step": 155
},
{
"epoch": 0.12709760699036837,
"grad_norm": 0.9585549328757471,
"learning_rate": 1.995551511472836e-05,
"loss": 1.5843,
"step": 160
},
{
"epoch": 0.1310694072088174,
"grad_norm": 0.94350147619133,
"learning_rate": 1.994148299437524e-05,
"loss": 1.5589,
"step": 165
},
{
"epoch": 0.1350412074272664,
"grad_norm": 0.9702241501746585,
"learning_rate": 1.9925536657300734e-05,
"loss": 1.5783,
"step": 170
},
{
"epoch": 0.13901300764571542,
"grad_norm": 0.90878186459247,
"learning_rate": 1.990767917394666e-05,
"loss": 1.5716,
"step": 175
},
{
"epoch": 0.14298480786416443,
"grad_norm": 0.9926126694962801,
"learning_rate": 1.9887913982743e-05,
"loss": 1.5705,
"step": 180
},
{
"epoch": 0.14695660808261343,
"grad_norm": 0.8958939695984969,
"learning_rate": 1.986624488944585e-05,
"loss": 1.5738,
"step": 185
},
{
"epoch": 0.15092840830106247,
"grad_norm": 0.9194318242066775,
"learning_rate": 1.984267606640462e-05,
"loss": 1.5729,
"step": 190
},
{
"epoch": 0.15490020851951147,
"grad_norm": 0.9497866986358652,
"learning_rate": 1.9817212051758667e-05,
"loss": 1.5674,
"step": 195
},
{
"epoch": 0.15887200873796048,
"grad_norm": 0.8988871329419085,
"learning_rate": 1.978985774856346e-05,
"loss": 1.5683,
"step": 200
},
{
"epoch": 0.16284380895640949,
"grad_norm": 0.9722584116447944,
"learning_rate": 1.9760618423846526e-05,
"loss": 1.5737,
"step": 205
},
{
"epoch": 0.1668156091748585,
"grad_norm": 0.9803590939025236,
"learning_rate": 1.9729499707593284e-05,
"loss": 1.5826,
"step": 210
},
{
"epoch": 0.17078740939330753,
"grad_norm": 0.9344503381295605,
"learning_rate": 1.9696507591663003e-05,
"loss": 1.5565,
"step": 215
},
{
"epoch": 0.17475920961175653,
"grad_norm": 0.9071381723958235,
"learning_rate": 1.9661648428635066e-05,
"loss": 1.5621,
"step": 220
},
{
"epoch": 0.17873100983020554,
"grad_norm": 0.9288351049804889,
"learning_rate": 1.962492893058582e-05,
"loss": 1.5532,
"step": 225
},
{
"epoch": 0.18270281004865455,
"grad_norm": 0.9288525670728277,
"learning_rate": 1.9586356167796145e-05,
"loss": 1.5801,
"step": 230
},
{
"epoch": 0.18667461026710355,
"grad_norm": 0.9183899709915191,
"learning_rate": 1.954593756739009e-05,
"loss": 1.5801,
"step": 235
},
{
"epoch": 0.1906464104855526,
"grad_norm": 0.9877621229108748,
"learning_rate": 1.9512278901942467e-05,
"loss": 1.5817,
"step": 240
},
{
"epoch": 0.1946182107040016,
"grad_norm": 0.9364533708770005,
"learning_rate": 1.9468557643703262e-05,
"loss": 1.571,
"step": 245
},
{
"epoch": 0.1985900109224506,
"grad_norm": 0.9156871855852468,
"learning_rate": 1.942301322976593e-05,
"loss": 1.5693,
"step": 250
},
{
"epoch": 0.2025618111408996,
"grad_norm": 0.91239458392149,
"learning_rate": 1.9375654429634866e-05,
"loss": 1.556,
"step": 255
},
{
"epoch": 0.2065336113593486,
"grad_norm": 0.904505667076919,
"learning_rate": 1.9326490362171625e-05,
"loss": 1.5763,
"step": 260
},
{
"epoch": 0.21050541157779765,
"grad_norm": 0.901682231936739,
"learning_rate": 1.9275530493839118e-05,
"loss": 1.5706,
"step": 265
},
{
"epoch": 0.21447721179624665,
"grad_norm": 0.9063579769395875,
"learning_rate": 1.9222784636878853e-05,
"loss": 1.567,
"step": 270
},
{
"epoch": 0.21844901201469566,
"grad_norm": 0.9214787619294059,
"learning_rate": 1.91682629474216e-05,
"loss": 1.5742,
"step": 275
},
{
"epoch": 0.22242081223314467,
"grad_norm": 0.9524031793810314,
"learning_rate": 1.9111975923531858e-05,
"loss": 1.5653,
"step": 280
},
{
"epoch": 0.22639261245159367,
"grad_norm": 0.9437628509942092,
"learning_rate": 1.905393440318645e-05,
"loss": 1.555,
"step": 285
},
{
"epoch": 0.2303644126700427,
"grad_norm": 0.9556933869893008,
"learning_rate": 1.8994149562187702e-05,
"loss": 1.5625,
"step": 290
},
{
"epoch": 0.2343362128884917,
"grad_norm": 0.9430484585606292,
"learning_rate": 1.8932632912011565e-05,
"loss": 1.5567,
"step": 295
},
{
"epoch": 0.23830801310694072,
"grad_norm": 0.9741243595635902,
"learning_rate": 1.886939629759107e-05,
"loss": 1.5553,
"step": 300
},
{
"epoch": 0.24227981332538973,
"grad_norm": 0.9052975360645378,
"learning_rate": 1.8804451895035645e-05,
"loss": 1.5793,
"step": 305
},
{
"epoch": 0.24625161354383873,
"grad_norm": 0.9742325579737205,
"learning_rate": 1.873781220928659e-05,
"loss": 1.5605,
"step": 310
},
{
"epoch": 0.25022341376228774,
"grad_norm": 0.9031213199856675,
"learning_rate": 1.866949007170929e-05,
"loss": 1.5649,
"step": 315
},
{
"epoch": 0.25419521398073675,
"grad_norm": 0.8702720949465248,
"learning_rate": 1.859949863762256e-05,
"loss": 1.5749,
"step": 320
},
{
"epoch": 0.2581670141991858,
"grad_norm": 0.961271121572224,
"learning_rate": 1.852785138376558e-05,
"loss": 1.5752,
"step": 325
},
{
"epoch": 0.2621388144176348,
"grad_norm": 0.8718585392590509,
"learning_rate": 1.8454562105703e-05,
"loss": 1.5577,
"step": 330
},
{
"epoch": 0.2661106146360838,
"grad_norm": 0.8690540556968002,
"learning_rate": 1.8379644915168623e-05,
"loss": 1.55,
"step": 335
},
{
"epoch": 0.2700824148545328,
"grad_norm": 0.9097212286930987,
"learning_rate": 1.83031142373482e-05,
"loss": 1.5737,
"step": 340
},
{
"epoch": 0.27405421507298183,
"grad_norm": 0.8712230897009824,
"learning_rate": 1.822498480810189e-05,
"loss": 1.5739,
"step": 345
},
{
"epoch": 0.27802601529143084,
"grad_norm": 0.948552271738661,
"learning_rate": 1.8145271671126892e-05,
"loss": 1.5657,
"step": 350
},
{
"epoch": 0.28199781550987985,
"grad_norm": 0.9284269691863282,
"learning_rate": 1.8063990175060807e-05,
"loss": 1.5601,
"step": 355
},
{
"epoch": 0.28596961572832885,
"grad_norm": 0.8854698701217475,
"learning_rate": 1.798115597052629e-05,
"loss": 1.5595,
"step": 360
},
{
"epoch": 0.28994141594677786,
"grad_norm": 0.9222629630617156,
"learning_rate": 1.7896785007117526e-05,
"loss": 1.5682,
"step": 365
},
{
"epoch": 0.29391321616522686,
"grad_norm": 0.9259111426870978,
"learning_rate": 1.781089353032918e-05,
"loss": 1.5614,
"step": 370
},
{
"epoch": 0.2978850163836759,
"grad_norm": 0.9425926440581255,
"learning_rate": 1.7723498078428355e-05,
"loss": 1.5726,
"step": 375
},
{
"epoch": 0.30185681660212493,
"grad_norm": 0.8788362482569841,
"learning_rate": 1.7634615479270157e-05,
"loss": 1.542,
"step": 380
},
{
"epoch": 0.30582861682057394,
"grad_norm": 0.8993764823284865,
"learning_rate": 1.754426284705753e-05,
"loss": 1.5517,
"step": 385
},
{
"epoch": 0.30980041703902295,
"grad_norm": 0.9114957468679018,
"learning_rate": 1.7452457579045948e-05,
"loss": 1.5566,
"step": 390
},
{
"epoch": 0.31377221725747195,
"grad_norm": 0.9303896756930642,
"learning_rate": 1.7359217352193587e-05,
"loss": 1.5733,
"step": 395
},
{
"epoch": 0.31774401747592096,
"grad_norm": 0.909449512224505,
"learning_rate": 1.726456011975767e-05,
"loss": 1.5492,
"step": 400
},
{
"epoch": 0.32171581769436997,
"grad_norm": 0.9015167969914157,
"learning_rate": 1.716850410783758e-05,
"loss": 1.5696,
"step": 405
},
{
"epoch": 0.32568761791281897,
"grad_norm": 0.9132981601959252,
"learning_rate": 1.7071067811865477e-05,
"loss": 1.5507,
"step": 410
},
{
"epoch": 0.329659418131268,
"grad_norm": 0.8960449583705886,
"learning_rate": 1.6972269993045004e-05,
"loss": 1.5903,
"step": 415
},
{
"epoch": 0.333631218349717,
"grad_norm": 0.8844544873628968,
"learning_rate": 1.6872129674738866e-05,
"loss": 1.5593,
"step": 420
},
{
"epoch": 0.33760301856816605,
"grad_norm": 0.9059334124525383,
"learning_rate": 1.6770666138805904e-05,
"loss": 1.5829,
"step": 425
},
{
"epoch": 0.34157481878661505,
"grad_norm": 0.8586072355217647,
"learning_rate": 1.666789892188841e-05,
"loss": 1.5577,
"step": 430
},
{
"epoch": 0.34554661900506406,
"grad_norm": 0.929783164434847,
"learning_rate": 1.6563847811650376e-05,
"loss": 1.5684,
"step": 435
},
{
"epoch": 0.34951841922351307,
"grad_norm": 0.8725442050763763,
"learning_rate": 1.64585328429674e-05,
"loss": 1.5448,
"step": 440
},
{
"epoch": 0.3534902194419621,
"grad_norm": 0.8749100355256942,
"learning_rate": 1.635197429406901e-05,
"loss": 1.5726,
"step": 445
},
{
"epoch": 0.3574620196604111,
"grad_norm": 0.948191551262984,
"learning_rate": 1.6244192682634143e-05,
"loss": 1.5464,
"step": 450
},
{
"epoch": 0.3614338198788601,
"grad_norm": 0.9305834084339885,
"learning_rate": 1.6135208761840457e-05,
"loss": 1.559,
"step": 455
},
{
"epoch": 0.3654056200973091,
"grad_norm": 0.9397991327586414,
"learning_rate": 1.602504351636838e-05,
"loss": 1.5534,
"step": 460
},
{
"epoch": 0.3693774203157581,
"grad_norm": 0.9213324585480839,
"learning_rate": 1.591371815836051e-05,
"loss": 1.5542,
"step": 465
},
{
"epoch": 0.3733492205342071,
"grad_norm": 0.8853761110334394,
"learning_rate": 1.580125412333728e-05,
"loss": 1.5402,
"step": 470
},
{
"epoch": 0.37732102075265617,
"grad_norm": 0.8691498605561269,
"learning_rate": 1.5687673066069568e-05,
"loss": 1.552,
"step": 475
},
{
"epoch": 0.3812928209711052,
"grad_norm": 0.8998572635366405,
"learning_rate": 1.5572996856409094e-05,
"loss": 1.5638,
"step": 480
},
{
"epoch": 0.3852646211895542,
"grad_norm": 0.8941950276870779,
"learning_rate": 1.5457247575077445e-05,
"loss": 1.5406,
"step": 485
},
{
"epoch": 0.3892364214080032,
"grad_norm": 0.9097426424762942,
"learning_rate": 1.534044750941444e-05,
"loss": 1.5472,
"step": 490
},
{
"epoch": 0.3932082216264522,
"grad_norm": 0.8791345662658966,
"learning_rate": 1.5222619149086746e-05,
"loss": 1.5413,
"step": 495
},
{
"epoch": 0.3971800218449012,
"grad_norm": 0.8776919953132802,
"learning_rate": 1.5103785181757533e-05,
"loss": 1.5396,
"step": 500
},
{
"epoch": 0.4011518220633502,
"grad_norm": 0.8664994594605832,
"learning_rate": 1.4983968488718005e-05,
"loss": 1.5426,
"step": 505
},
{
"epoch": 0.4051236222817992,
"grad_norm": 0.8931546183056348,
"learning_rate": 1.4863192140481624e-05,
"loss": 1.5537,
"step": 510
},
{
"epoch": 0.4090954225002482,
"grad_norm": 0.8442200510246032,
"learning_rate": 1.4741479392341941e-05,
"loss": 1.5586,
"step": 515
},
{
"epoch": 0.4130672227186972,
"grad_norm": 0.9413240109314772,
"learning_rate": 1.4618853679894813e-05,
"loss": 1.5202,
"step": 520
},
{
"epoch": 0.4170390229371463,
"grad_norm": 0.8865446672373302,
"learning_rate": 1.4495338614525927e-05,
"loss": 1.5503,
"step": 525
},
{
"epoch": 0.4210108231555953,
"grad_norm": 0.9668394693388267,
"learning_rate": 1.437095797886445e-05,
"loss": 1.5488,
"step": 530
},
{
"epoch": 0.4249826233740443,
"grad_norm": 0.9214606309227755,
"learning_rate": 1.4245735722203736e-05,
"loss": 1.54,
"step": 535
},
{
"epoch": 0.4289544235924933,
"grad_norm": 0.9555320905995948,
"learning_rate": 1.4119695955889925e-05,
"loss": 1.5492,
"step": 540
},
{
"epoch": 0.4329262238109423,
"grad_norm": 0.8802626029124945,
"learning_rate": 1.3992862948679332e-05,
"loss": 1.549,
"step": 545
},
{
"epoch": 0.4368980240293913,
"grad_norm": 0.8902159443351422,
"learning_rate": 1.3865261122065551e-05,
"loss": 1.5482,
"step": 550
},
{
"epoch": 0.4408698242478403,
"grad_norm": 0.8427883741871806,
"learning_rate": 1.3736915045577122e-05,
"loss": 1.5487,
"step": 555
},
{
"epoch": 0.44484162446628933,
"grad_norm": 0.8251277406307889,
"learning_rate": 1.3607849432046717e-05,
"loss": 1.5478,
"step": 560
},
{
"epoch": 0.44881342468473834,
"grad_norm": 0.8371689103015839,
"learning_rate": 1.3478089132852717e-05,
"loss": 1.5597,
"step": 565
},
{
"epoch": 0.45278522490318734,
"grad_norm": 0.8217449927915415,
"learning_rate": 1.3347659133134118e-05,
"loss": 1.5141,
"step": 570
},
{
"epoch": 0.4567570251216364,
"grad_norm": 0.8372002721470211,
"learning_rate": 1.3216584546979702e-05,
"loss": 1.5338,
"step": 575
},
{
"epoch": 0.4607288253400854,
"grad_norm": 0.9501487795845303,
"learning_rate": 1.3084890612592325e-05,
"loss": 1.5633,
"step": 580
},
{
"epoch": 0.4647006255585344,
"grad_norm": 0.8558411230061426,
"learning_rate": 1.2952602687429364e-05,
"loss": 1.5623,
"step": 585
},
{
"epoch": 0.4686724257769834,
"grad_norm": 0.830027301501606,
"learning_rate": 1.2819746243320176e-05,
"loss": 1.5512,
"step": 590
},
{
"epoch": 0.47264422599543243,
"grad_norm": 0.903709450047426,
"learning_rate": 1.2686346861561538e-05,
"loss": 1.5608,
"step": 595
},
{
"epoch": 0.47661602621388144,
"grad_norm": 0.961449581540646,
"learning_rate": 1.2552430227992005e-05,
"loss": 1.542,
"step": 600
},
{
"epoch": 0.48058782643233044,
"grad_norm": 0.877769489004286,
"learning_rate": 1.2418022128046144e-05,
"loss": 1.539,
"step": 605
},
{
"epoch": 0.48455962665077945,
"grad_norm": 0.9130539885425439,
"learning_rate": 1.2283148441789586e-05,
"loss": 1.5276,
"step": 610
},
{
"epoch": 0.48853142686922846,
"grad_norm": 0.882980412032183,
"learning_rate": 1.2147835138935868e-05,
"loss": 1.5204,
"step": 615
},
{
"epoch": 0.49250322708767746,
"grad_norm": 0.8366128423326633,
"learning_rate": 1.2012108273846011e-05,
"loss": 1.5362,
"step": 620
},
{
"epoch": 0.4964750273061265,
"grad_norm": 0.8538967577380823,
"learning_rate": 1.1875993980511772e-05,
"loss": 1.5187,
"step": 625
},
{
"epoch": 0.5004468275245755,
"grad_norm": 0.8942558682507143,
"learning_rate": 1.1739518467523614e-05,
"loss": 1.5386,
"step": 630
},
{
"epoch": 0.5044186277430245,
"grad_norm": 0.9187709823388034,
"learning_rate": 1.1602708013024255e-05,
"loss": 1.5245,
"step": 635
},
{
"epoch": 0.5083904279614735,
"grad_norm": 0.8442877528307388,
"learning_rate": 1.146558895964888e-05,
"loss": 1.5596,
"step": 640
},
{
"epoch": 0.5123622281799225,
"grad_norm": 0.898548400584606,
"learning_rate": 1.1328187709452884e-05,
"loss": 1.5406,
"step": 645
},
{
"epoch": 0.5163340283983716,
"grad_norm": 0.8811961105698817,
"learning_rate": 1.119053071882822e-05,
"loss": 1.5356,
"step": 650
},
{
"epoch": 0.5203058286168206,
"grad_norm": 0.8804203720770784,
"learning_rate": 1.1052644493409255e-05,
"loss": 1.5528,
"step": 655
},
{
"epoch": 0.5242776288352696,
"grad_norm": 0.8797158963008079,
"learning_rate": 1.091455558296914e-05,
"loss": 1.5372,
"step": 660
},
{
"epoch": 0.5282494290537186,
"grad_norm": 0.8350754715847598,
"learning_rate": 1.077629057630771e-05,
"loss": 1.5452,
"step": 665
},
{
"epoch": 0.5322212292721676,
"grad_norm": 0.8903077066951119,
"learning_rate": 1.0637876096131852e-05,
"loss": 1.5283,
"step": 670
},
{
"epoch": 0.5361930294906166,
"grad_norm": 0.8576752803021083,
"learning_rate": 1.049933879392933e-05,
"loss": 1.5433,
"step": 675
},
{
"epoch": 0.5401648297090657,
"grad_norm": 0.8467022082112134,
"learning_rate": 1.0360705344837105e-05,
"loss": 1.5133,
"step": 680
},
{
"epoch": 0.5441366299275147,
"grad_norm": 0.8817844142301194,
"learning_rate": 1.0222002442505057e-05,
"loss": 1.5411,
"step": 685
},
{
"epoch": 0.5481084301459637,
"grad_norm": 0.8151311907804857,
"learning_rate": 1.0083256793956186e-05,
"loss": 1.5326,
"step": 690
},
{
"epoch": 0.5520802303644127,
"grad_norm": 0.8251603017546312,
"learning_rate": 9.944495114444205e-06,
"loss": 1.5082,
"step": 695
},
{
"epoch": 0.5560520305828617,
"grad_norm": 0.8869207810981268,
"learning_rate": 9.805744122309557e-06,
"loss": 1.5481,
"step": 700
},
{
"epoch": 0.5600238308013107,
"grad_norm": 0.8719890884427858,
"learning_rate": 9.667030533834862e-06,
"loss": 1.5223,
"step": 705
},
{
"epoch": 0.5639956310197597,
"grad_norm": 0.8664009171759752,
"learning_rate": 9.528381058100707e-06,
"loss": 1.5289,
"step": 710
},
{
"epoch": 0.5679674312382087,
"grad_norm": 0.8819223345245485,
"learning_rate": 9.389822391842886e-06,
"loss": 1.5248,
"step": 715
},
{
"epoch": 0.5719392314566577,
"grad_norm": 0.8462371220192982,
"learning_rate": 9.251381214311974e-06,
"loss": 1.5263,
"step": 720
},
{
"epoch": 0.5759110316751067,
"grad_norm": 0.8556000311877207,
"learning_rate": 9.113084182136267e-06,
"loss": 1.5464,
"step": 725
},
{
"epoch": 0.5798828318935557,
"grad_norm": 0.8854678303214264,
"learning_rate": 8.974957924189108e-06,
"loss": 1.5388,
"step": 730
},
{
"epoch": 0.5838546321120047,
"grad_norm": 0.8605998763470911,
"learning_rate": 8.837029036461537e-06,
"loss": 1.5391,
"step": 735
},
{
"epoch": 0.5878264323304537,
"grad_norm": 0.8176529738699428,
"learning_rate": 8.699324076941278e-06,
"loss": 1.5578,
"step": 740
},
{
"epoch": 0.5917982325489027,
"grad_norm": 0.8500747154468798,
"learning_rate": 8.561869560499017e-06,
"loss": 1.5387,
"step": 745
},
{
"epoch": 0.5957700327673519,
"grad_norm": 0.8074215452026124,
"learning_rate": 8.42469195378304e-06,
"loss": 1.5312,
"step": 750
},
{
"epoch": 0.5997418329858009,
"grad_norm": 0.8452784103185206,
"learning_rate": 8.287817670123101e-06,
"loss": 1.5258,
"step": 755
},
{
"epoch": 0.6037136332042499,
"grad_norm": 0.8486586156355476,
"learning_rate": 8.151273064444583e-06,
"loss": 1.5192,
"step": 760
},
{
"epoch": 0.6076854334226989,
"grad_norm": 0.8146588435190393,
"learning_rate": 8.015084428193895e-06,
"loss": 1.5045,
"step": 765
},
{
"epoch": 0.6116572336411479,
"grad_norm": 0.8320002856800587,
"learning_rate": 7.879277984276106e-06,
"loss": 1.5371,
"step": 770
},
{
"epoch": 0.6156290338595969,
"grad_norm": 0.8314805091840901,
"learning_rate": 7.74387988200576e-06,
"loss": 1.5364,
"step": 775
},
{
"epoch": 0.6196008340780459,
"grad_norm": 0.8340274339801351,
"learning_rate": 7.608916192071856e-06,
"loss": 1.5326,
"step": 780
},
{
"epoch": 0.6235726342964949,
"grad_norm": 0.8445705617901014,
"learning_rate": 7.474412901517998e-06,
"loss": 1.543,
"step": 785
},
{
"epoch": 0.6275444345149439,
"grad_norm": 0.8631402798545373,
"learning_rate": 7.340395908738622e-06,
"loss": 1.5258,
"step": 790
},
{
"epoch": 0.6315162347333929,
"grad_norm": 0.839562146660397,
"learning_rate": 7.206891018492308e-06,
"loss": 1.5464,
"step": 795
},
{
"epoch": 0.6354880349518419,
"grad_norm": 0.879806527044698,
"learning_rate": 7.073923936933091e-06,
"loss": 1.541,
"step": 800
},
{
"epoch": 0.6394598351702909,
"grad_norm": 0.8536606816656509,
"learning_rate": 6.941520266660819e-06,
"loss": 1.5206,
"step": 805
},
{
"epoch": 0.6434316353887399,
"grad_norm": 0.8318733759254066,
"learning_rate": 6.809705501791379e-06,
"loss": 1.5407,
"step": 810
},
{
"epoch": 0.6474034356071889,
"grad_norm": 0.8459590871404428,
"learning_rate": 6.678505023047871e-06,
"loss": 1.5216,
"step": 815
},
{
"epoch": 0.6513752358256379,
"grad_norm": 0.8295606496141018,
"learning_rate": 6.5479440928735695e-06,
"loss": 1.5388,
"step": 820
},
{
"epoch": 0.655347036044087,
"grad_norm": 0.8553791665518854,
"learning_rate": 6.418047850567699e-06,
"loss": 1.5045,
"step": 825
},
{
"epoch": 0.659318836262536,
"grad_norm": 0.8249727222480653,
"learning_rate": 6.2888413074448954e-06,
"loss": 1.5061,
"step": 830
},
{
"epoch": 0.663290636480985,
"grad_norm": 0.8486443716521056,
"learning_rate": 6.160349342019299e-06,
"loss": 1.5202,
"step": 835
},
{
"epoch": 0.667262436699434,
"grad_norm": 0.8425578204636239,
"learning_rate": 6.032596695214272e-06,
"loss": 1.524,
"step": 840
},
{
"epoch": 0.671234236917883,
"grad_norm": 0.8607932133751491,
"learning_rate": 5.905607965598542e-06,
"loss": 1.5054,
"step": 845
},
{
"epoch": 0.6752060371363321,
"grad_norm": 0.87671463806252,
"learning_rate": 5.779407604649794e-06,
"loss": 1.5256,
"step": 850
},
{
"epoch": 0.6791778373547811,
"grad_norm": 0.8240359232861163,
"learning_rate": 5.6540199120465985e-06,
"loss": 1.5311,
"step": 855
},
{
"epoch": 0.6831496375732301,
"grad_norm": 0.8880311823283904,
"learning_rate": 5.529469030989511e-06,
"loss": 1.5073,
"step": 860
},
{
"epoch": 0.6871214377916791,
"grad_norm": 0.8623990682016829,
"learning_rate": 5.405778943552373e-06,
"loss": 1.5379,
"step": 865
},
{
"epoch": 0.6910932380101281,
"grad_norm": 0.8262496371808098,
"learning_rate": 5.282973466064567e-06,
"loss": 1.5145,
"step": 870
},
{
"epoch": 0.6950650382285771,
"grad_norm": 0.8482342550959973,
"learning_rate": 5.161076244525254e-06,
"loss": 1.5275,
"step": 875
},
{
"epoch": 0.6990368384470261,
"grad_norm": 0.8328878311342953,
"learning_rate": 5.040110750050356e-06,
"loss": 1.528,
"step": 880
},
{
"epoch": 0.7030086386654751,
"grad_norm": 0.8164657341322779,
"learning_rate": 4.9201002743532175e-06,
"loss": 1.5092,
"step": 885
},
{
"epoch": 0.7069804388839241,
"grad_norm": 0.8490381133150996,
"learning_rate": 4.801067925259857e-06,
"loss": 1.5454,
"step": 890
},
{
"epoch": 0.7109522391023732,
"grad_norm": 0.8440725903075691,
"learning_rate": 4.683036622259562e-06,
"loss": 1.5224,
"step": 895
},
{
"epoch": 0.7149240393208222,
"grad_norm": 0.8366180827607501,
"learning_rate": 4.566029092091798e-06,
"loss": 1.5413,
"step": 900
},
{
"epoch": 0.7188958395392712,
"grad_norm": 0.8427766816018408,
"learning_rate": 4.450067864370187e-06,
"loss": 1.5173,
"step": 905
},
{
"epoch": 0.7228676397577202,
"grad_norm": 0.8028067037799445,
"learning_rate": 4.335175267244495e-06,
"loss": 1.5275,
"step": 910
},
{
"epoch": 0.7268394399761692,
"grad_norm": 0.795885144548523,
"learning_rate": 4.221373423101356e-06,
"loss": 1.5166,
"step": 915
},
{
"epoch": 0.7308112401946182,
"grad_norm": 0.8378592839978902,
"learning_rate": 4.108684244304668e-06,
"loss": 1.5444,
"step": 920
},
{
"epoch": 0.7347830404130672,
"grad_norm": 0.8299609184798702,
"learning_rate": 3.997129428976374e-06,
"loss": 1.5373,
"step": 925
},
{
"epoch": 0.7387548406315162,
"grad_norm": 0.8200382060222382,
"learning_rate": 3.886730456818546e-06,
"loss": 1.5203,
"step": 930
},
{
"epoch": 0.7427266408499652,
"grad_norm": 0.8198010231353143,
"learning_rate": 3.7775085849774973e-06,
"loss": 1.5291,
"step": 935
},
{
"epoch": 0.7466984410684142,
"grad_norm": 0.8232628031861067,
"learning_rate": 3.6694848439507347e-06,
"loss": 1.5334,
"step": 940
},
{
"epoch": 0.7506702412868632,
"grad_norm": 0.81406141716096,
"learning_rate": 3.562680033537598e-06,
"loss": 1.5091,
"step": 945
},
{
"epoch": 0.7546420415053123,
"grad_norm": 0.8087303425914779,
"learning_rate": 3.457114718834278e-06,
"loss": 1.5313,
"step": 950
},
{
"epoch": 0.7586138417237613,
"grad_norm": 0.8083510625412864,
"learning_rate": 3.3528092262740284e-06,
"loss": 1.5234,
"step": 955
},
{
"epoch": 0.7625856419422103,
"grad_norm": 0.8366676453174993,
"learning_rate": 3.2497836397133663e-06,
"loss": 1.538,
"step": 960
},
{
"epoch": 0.7665574421606594,
"grad_norm": 0.8393538795255264,
"learning_rate": 3.148057796564944e-06,
"loss": 1.5216,
"step": 965
},
{
"epoch": 0.7705292423791084,
"grad_norm": 0.8520167906125237,
"learning_rate": 3.0476512839778894e-06,
"loss": 1.5293,
"step": 970
},
{
"epoch": 0.7745010425975574,
"grad_norm": 0.8371019471875751,
"learning_rate": 2.948583435066329e-06,
"loss": 1.513,
"step": 975
},
{
"epoch": 0.7784728428160064,
"grad_norm": 0.8298639126272734,
"learning_rate": 2.850873325186826e-06,
"loss": 1.5268,
"step": 980
},
{
"epoch": 0.7824446430344554,
"grad_norm": 0.8374513555969967,
"learning_rate": 2.7545397682654693e-06,
"loss": 1.5097,
"step": 985
},
{
"epoch": 0.7864164432529044,
"grad_norm": 0.8200846092055774,
"learning_rate": 2.6596013131752498e-06,
"loss": 1.543,
"step": 990
},
{
"epoch": 0.7903882434713534,
"grad_norm": 0.8148429888845868,
"learning_rate": 2.566076240164536e-06,
"loss": 1.5522,
"step": 995
},
{
"epoch": 0.7943600436898024,
"grad_norm": 0.8131435969188188,
"learning_rate": 2.47398255733722e-06,
"loss": 1.5245,
"step": 1000
},
{
"epoch": 0.7983318439082514,
"grad_norm": 0.8284789559376622,
"learning_rate": 2.383337997185299e-06,
"loss": 1.5427,
"step": 1005
},
{
"epoch": 0.8023036441267004,
"grad_norm": 0.8321861504525199,
"learning_rate": 2.2941600131744978e-06,
"loss": 1.5072,
"step": 1010
},
{
"epoch": 0.8062754443451494,
"grad_norm": 0.8065047141388554,
"learning_rate": 2.2064657763836474e-06,
"loss": 1.5179,
"step": 1015
},
{
"epoch": 0.8102472445635984,
"grad_norm": 0.7992276385422232,
"learning_rate": 2.120272172198412e-06,
"loss": 1.5272,
"step": 1020
},
{
"epoch": 0.8142190447820474,
"grad_norm": 0.8229954286323785,
"learning_rate": 2.0355957970600406e-06,
"loss": 1.5027,
"step": 1025
},
{
"epoch": 0.8181908450004964,
"grad_norm": 0.8275865130237602,
"learning_rate": 1.952452955269738e-06,
"loss": 1.5267,
"step": 1030
},
{
"epoch": 0.8221626452189454,
"grad_norm": 0.8081049582664315,
"learning_rate": 1.870859655849304e-06,
"loss": 1.5426,
"step": 1035
},
{
"epoch": 0.8261344454373944,
"grad_norm": 0.823108756311465,
"learning_rate": 1.7908316094586343e-06,
"loss": 1.5388,
"step": 1040
},
{
"epoch": 0.8301062456558436,
"grad_norm": 0.8355851750630425,
"learning_rate": 1.7123842253706302e-06,
"loss": 1.5014,
"step": 1045
},
{
"epoch": 0.8340780458742926,
"grad_norm": 0.8374873929351686,
"learning_rate": 1.6355326085041944e-06,
"loss": 1.5008,
"step": 1050
},
{
"epoch": 0.8380498460927416,
"grad_norm": 0.8230242479079336,
"learning_rate": 1.5602915565157895e-06,
"loss": 1.5249,
"step": 1055
},
{
"epoch": 0.8420216463111906,
"grad_norm": 0.8028112832237307,
"learning_rate": 1.4866755569501757e-06,
"loss": 1.5134,
"step": 1060
},
{
"epoch": 0.8459934465296396,
"grad_norm": 0.7982837438136364,
"learning_rate": 1.4146987844508509e-06,
"loss": 1.5127,
"step": 1065
},
{
"epoch": 0.8499652467480886,
"grad_norm": 0.799526922604763,
"learning_rate": 1.344375098030759e-06,
"loss": 1.5145,
"step": 1070
},
{
"epoch": 0.8539370469665376,
"grad_norm": 0.8028491606704994,
"learning_rate": 1.2757180384037505e-06,
"loss": 1.5601,
"step": 1075
},
{
"epoch": 0.8579088471849866,
"grad_norm": 0.8062739981966481,
"learning_rate": 1.2087408253773326e-06,
"loss": 1.5104,
"step": 1080
},
{
"epoch": 0.8618806474034356,
"grad_norm": 0.810426514136181,
"learning_rate": 1.1434563553072332e-06,
"loss": 1.5271,
"step": 1085
},
{
"epoch": 0.8658524476218846,
"grad_norm": 0.8106217030883788,
"learning_rate": 1.0798771986142186e-06,
"loss": 1.5275,
"step": 1090
},
{
"epoch": 0.8698242478403336,
"grad_norm": 0.8239831789081276,
"learning_rate": 1.018015597363673e-06,
"loss": 1.5435,
"step": 1095
},
{
"epoch": 0.8737960480587826,
"grad_norm": 0.8032988356076526,
"learning_rate": 9.578834629084199e-07,
"loss": 1.5006,
"step": 1100
},
{
"epoch": 0.8777678482772316,
"grad_norm": 0.8107500288685499,
"learning_rate": 8.994923735952033e-07,
"loss": 1.5175,
"step": 1105
},
{
"epoch": 0.8817396484956807,
"grad_norm": 0.7947044768084438,
"learning_rate": 8.428535725353016e-07,
"loss": 1.5134,
"step": 1110
},
{
"epoch": 0.8857114487141297,
"grad_norm": 0.8184180092012469,
"learning_rate": 7.879779654396724e-07,
"loss": 1.5308,
"step": 1115
},
{
"epoch": 0.8896832489325787,
"grad_norm": 0.7812672670695122,
"learning_rate": 7.348761185190889e-07,
"loss": 1.5286,
"step": 1120
},
{
"epoch": 0.8936550491510277,
"grad_norm": 0.8031429398431277,
"learning_rate": 6.835582564496257e-07,
"loss": 1.5189,
"step": 1125
},
{
"epoch": 0.8976268493694767,
"grad_norm": 0.8262986271960238,
"learning_rate": 6.340342604039151e-07,
"loss": 1.5137,
"step": 1130
},
{
"epoch": 0.9015986495879257,
"grad_norm": 0.8091884297906192,
"learning_rate": 5.86313666148538e-07,
"loss": 1.5287,
"step": 1135
},
{
"epoch": 0.9055704498063747,
"grad_norm": 0.8083089870588875,
"learning_rate": 5.40405662207939e-07,
"loss": 1.5211,
"step": 1140
},
{
"epoch": 0.9095422500248238,
"grad_norm": 0.8080847975437143,
"learning_rate": 4.963190880951807e-07,
"loss": 1.4942,
"step": 1145
},
{
"epoch": 0.9135140502432728,
"grad_norm": 0.8112755205962424,
"learning_rate": 4.540624326099108e-07,
"loss": 1.5361,
"step": 1150
},
{
"epoch": 0.9174858504617218,
"grad_norm": 0.8137159093665213,
"learning_rate": 4.136438322038594e-07,
"loss": 1.5305,
"step": 1155
},
{
"epoch": 0.9214576506801708,
"grad_norm": 0.8099723012257717,
"learning_rate": 3.7507106941418127e-07,
"loss": 1.5136,
"step": 1160
},
{
"epoch": 0.9254294508986198,
"grad_norm": 0.7874754581462157,
"learning_rate": 3.3835157136493434e-07,
"loss": 1.5192,
"step": 1165
},
{
"epoch": 0.9294012511170688,
"grad_norm": 0.8097635833636108,
"learning_rate": 3.0349240833699944e-07,
"loss": 1.53,
"step": 1170
},
{
"epoch": 0.9333730513355178,
"grad_norm": 0.7839597922407979,
"learning_rate": 2.7050029240671573e-07,
"loss": 1.5406,
"step": 1175
},
{
"epoch": 0.9373448515539669,
"grad_norm": 0.8036152094174389,
"learning_rate": 2.3938157615347455e-07,
"loss": 1.5142,
"step": 1180
},
{
"epoch": 0.9413166517724159,
"grad_norm": 0.7969437729501767,
"learning_rate": 2.1014225143654145e-07,
"loss": 1.5168,
"step": 1185
},
{
"epoch": 0.9452884519908649,
"grad_norm": 0.7986549378127612,
"learning_rate": 1.8278794824133417e-07,
"loss": 1.5233,
"step": 1190
},
{
"epoch": 0.9492602522093139,
"grad_norm": 0.807465267893701,
"learning_rate": 1.5732393359537912e-07,
"loss": 1.5193,
"step": 1195
},
{
"epoch": 0.9532320524277629,
"grad_norm": 0.8289613150037851,
"learning_rate": 1.3375511055415346e-07,
"loss": 1.5306,
"step": 1200
},
{
"epoch": 0.9572038526462119,
"grad_norm": 0.8200430426978503,
"learning_rate": 1.1208601725700374e-07,
"loss": 1.5185,
"step": 1205
},
{
"epoch": 0.9611756528646609,
"grad_norm": 0.8207956989586505,
"learning_rate": 9.232082605334369e-08,
"loss": 1.5261,
"step": 1210
},
{
"epoch": 0.9651474530831099,
"grad_norm": 0.8115966572361262,
"learning_rate": 7.446334269926803e-08,
"loss": 1.5243,
"step": 1215
},
{
"epoch": 0.9691192533015589,
"grad_norm": 0.7861487020593475,
"learning_rate": 5.851700562476304e-08,
"loss": 1.5168,
"step": 1220
},
{
"epoch": 0.9730910535200079,
"grad_norm": 0.7965492838019634,
"learning_rate": 4.448488527164174e-08,
"loss": 1.5238,
"step": 1225
},
{
"epoch": 0.9770628537384569,
"grad_norm": 0.7940697296490887,
"learning_rate": 3.236968350233793e-08,
"loss": 1.5131,
"step": 1230
},
{
"epoch": 0.9810346539569059,
"grad_norm": 0.8173719106280355,
"learning_rate": 2.217373307966009e-08,
"loss": 1.5142,
"step": 1235
},
{
"epoch": 0.9850064541753549,
"grad_norm": 0.7895572135274385,
"learning_rate": 1.389899721762844e-08,
"loss": 1.5156,
"step": 1240
},
{
"epoch": 0.988978254393804,
"grad_norm": 0.7929243131364883,
"learning_rate": 7.54706920345738e-09,
"loss": 1.5024,
"step": 1245
},
{
"epoch": 0.992950054612253,
"grad_norm": 0.7966487036007772,
"learning_rate": 3.119172090771949e-09,
"loss": 1.5061,
"step": 1250
},
{
"epoch": 0.9969218548307021,
"grad_norm": 0.779081898680904,
"learning_rate": 6.161584641128926e-10,
"loss": 1.526,
"step": 1255
},
{
"epoch": 0.9993049349617714,
"eval_loss": 1.5326507091522217,
"eval_runtime": 268.2952,
"eval_samples_per_second": 99.681,
"eval_steps_per_second": 4.156,
"step": 1258
},
{
"epoch": 0.9993049349617714,
"step": 1258,
"total_flos": 106161864966144.0,
"train_loss": 1.5477431688475496,
"train_runtime": 10888.6684,
"train_samples_per_second": 22.196,
"train_steps_per_second": 0.116
}
],
"logging_steps": 5,
"max_steps": 1258,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 500,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": false,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 106161864966144.0,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}