checkpoints / checkpoint-4908 /trainer_state.json
DazMashaly's picture
Upload folder
dbf497b verified
raw
history blame
No virus
37.1 kB
{
"best_metric": 1.0,
"best_model_checkpoint": "DazMashaly/output/checkpoint-409",
"epoch": 12.0,
"eval_steps": 500,
"global_step": 4908,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.061124694376528114,
"grad_norm": 1.064210295677185,
"learning_rate": 5.000000000000001e-07,
"loss": 6.0673,
"step": 25
},
{
"epoch": 0.12224938875305623,
"grad_norm": 0.982063889503479,
"learning_rate": 1.0000000000000002e-06,
"loss": 6.0659,
"step": 50
},
{
"epoch": 0.18337408312958436,
"grad_norm": 0.9887572526931763,
"learning_rate": 1.5e-06,
"loss": 6.069,
"step": 75
},
{
"epoch": 0.24449877750611246,
"grad_norm": 1.076440691947937,
"learning_rate": 2.0000000000000003e-06,
"loss": 6.0588,
"step": 100
},
{
"epoch": 0.3056234718826406,
"grad_norm": 1.1176767349243164,
"learning_rate": 2.5e-06,
"loss": 6.0642,
"step": 125
},
{
"epoch": 0.36674816625916873,
"grad_norm": 1.0234830379486084,
"learning_rate": 3e-06,
"loss": 6.0708,
"step": 150
},
{
"epoch": 0.4278728606356968,
"grad_norm": 1.090944528579712,
"learning_rate": 3.5e-06,
"loss": 6.0339,
"step": 175
},
{
"epoch": 0.4889975550122249,
"grad_norm": 1.1804206371307373,
"learning_rate": 4.000000000000001e-06,
"loss": 6.0542,
"step": 200
},
{
"epoch": 0.5501222493887531,
"grad_norm": 1.3007625341415405,
"learning_rate": 4.5e-06,
"loss": 6.0549,
"step": 225
},
{
"epoch": 0.6112469437652812,
"grad_norm": 1.2753345966339111,
"learning_rate": 5e-06,
"loss": 6.0535,
"step": 250
},
{
"epoch": 0.6723716381418093,
"grad_norm": 1.321979284286499,
"learning_rate": 5.500000000000001e-06,
"loss": 6.0401,
"step": 275
},
{
"epoch": 0.7334963325183375,
"grad_norm": 1.1843823194503784,
"learning_rate": 6e-06,
"loss": 6.0463,
"step": 300
},
{
"epoch": 0.7946210268948656,
"grad_norm": 1.2534185647964478,
"learning_rate": 6.5000000000000004e-06,
"loss": 6.0248,
"step": 325
},
{
"epoch": 0.8557457212713936,
"grad_norm": 1.2556980848312378,
"learning_rate": 7e-06,
"loss": 6.045,
"step": 350
},
{
"epoch": 0.9168704156479217,
"grad_norm": 1.263355016708374,
"learning_rate": 7.500000000000001e-06,
"loss": 6.034,
"step": 375
},
{
"epoch": 0.9779951100244498,
"grad_norm": 1.1825202703475952,
"learning_rate": 8.000000000000001e-06,
"loss": 5.9866,
"step": 400
},
{
"epoch": 1.0,
"eval_cer": 1.0,
"eval_loss": 5.615485191345215,
"eval_runtime": 703.3832,
"eval_samples_per_second": 3.126,
"eval_steps_per_second": 0.026,
"eval_wer": 1.0,
"step": 409
},
{
"epoch": 1.039119804400978,
"grad_norm": 1.388098955154419,
"learning_rate": 8.5e-06,
"loss": 6.0005,
"step": 425
},
{
"epoch": 1.1002444987775062,
"grad_norm": 1.1844673156738281,
"learning_rate": 9e-06,
"loss": 5.9922,
"step": 450
},
{
"epoch": 1.1613691931540342,
"grad_norm": 1.2709770202636719,
"learning_rate": 9.5e-06,
"loss": 5.9917,
"step": 475
},
{
"epoch": 1.2224938875305624,
"grad_norm": 1.331032395362854,
"learning_rate": 1e-05,
"loss": 5.9767,
"step": 500
},
{
"epoch": 1.2836185819070904,
"grad_norm": 1.436553955078125,
"learning_rate": 9.967447916666668e-06,
"loss": 5.984,
"step": 525
},
{
"epoch": 1.3447432762836184,
"grad_norm": 1.2195194959640503,
"learning_rate": 9.934895833333335e-06,
"loss": 5.9959,
"step": 550
},
{
"epoch": 1.4058679706601467,
"grad_norm": 1.265336036682129,
"learning_rate": 9.902343750000001e-06,
"loss": 5.9657,
"step": 575
},
{
"epoch": 1.466992665036675,
"grad_norm": 1.2491796016693115,
"learning_rate": 9.869791666666667e-06,
"loss": 5.9545,
"step": 600
},
{
"epoch": 1.528117359413203,
"grad_norm": 1.5214732885360718,
"learning_rate": 9.837239583333335e-06,
"loss": 5.9556,
"step": 625
},
{
"epoch": 1.589242053789731,
"grad_norm": 1.478154182434082,
"learning_rate": 9.8046875e-06,
"loss": 5.9557,
"step": 650
},
{
"epoch": 1.6503667481662592,
"grad_norm": 1.4325579404830933,
"learning_rate": 9.772135416666667e-06,
"loss": 5.9392,
"step": 675
},
{
"epoch": 1.7114914425427874,
"grad_norm": 1.258053183555603,
"learning_rate": 9.739583333333334e-06,
"loss": 5.9329,
"step": 700
},
{
"epoch": 1.7726161369193154,
"grad_norm": 1.439457893371582,
"learning_rate": 9.707031250000001e-06,
"loss": 5.9353,
"step": 725
},
{
"epoch": 1.8337408312958434,
"grad_norm": 2.1067497730255127,
"learning_rate": 9.674479166666668e-06,
"loss": 5.9253,
"step": 750
},
{
"epoch": 1.8948655256723717,
"grad_norm": 1.6002326011657715,
"learning_rate": 9.641927083333335e-06,
"loss": 5.8955,
"step": 775
},
{
"epoch": 1.9559902200488999,
"grad_norm": 1.249984622001648,
"learning_rate": 9.609375000000001e-06,
"loss": 5.8937,
"step": 800
},
{
"epoch": 2.0,
"eval_cer": 1.0,
"eval_loss": 5.5135722160339355,
"eval_runtime": 702.4432,
"eval_samples_per_second": 3.131,
"eval_steps_per_second": 0.026,
"eval_wer": 1.0,
"step": 818
},
{
"epoch": 2.0171149144254277,
"grad_norm": 1.5596375465393066,
"learning_rate": 9.576822916666667e-06,
"loss": 5.8783,
"step": 825
},
{
"epoch": 2.078239608801956,
"grad_norm": 1.5268152952194214,
"learning_rate": 9.544270833333335e-06,
"loss": 5.8633,
"step": 850
},
{
"epoch": 2.139364303178484,
"grad_norm": 1.8570572137832642,
"learning_rate": 9.51171875e-06,
"loss": 5.875,
"step": 875
},
{
"epoch": 2.2004889975550124,
"grad_norm": 1.7126903533935547,
"learning_rate": 9.479166666666667e-06,
"loss": 5.844,
"step": 900
},
{
"epoch": 2.26161369193154,
"grad_norm": 1.4625195264816284,
"learning_rate": 9.446614583333334e-06,
"loss": 5.8778,
"step": 925
},
{
"epoch": 2.3227383863080684,
"grad_norm": 1.4512864351272583,
"learning_rate": 9.414062500000001e-06,
"loss": 5.8592,
"step": 950
},
{
"epoch": 2.3838630806845966,
"grad_norm": 1.7913607358932495,
"learning_rate": 9.381510416666668e-06,
"loss": 5.8493,
"step": 975
},
{
"epoch": 2.444987775061125,
"grad_norm": 1.9679557085037231,
"learning_rate": 9.348958333333335e-06,
"loss": 5.8396,
"step": 1000
},
{
"epoch": 2.5061124694376526,
"grad_norm": 1.8194330930709839,
"learning_rate": 9.316406250000002e-06,
"loss": 5.8435,
"step": 1025
},
{
"epoch": 2.567237163814181,
"grad_norm": 1.9096548557281494,
"learning_rate": 9.283854166666667e-06,
"loss": 5.8206,
"step": 1050
},
{
"epoch": 2.628361858190709,
"grad_norm": 1.6974539756774902,
"learning_rate": 9.251302083333335e-06,
"loss": 5.8062,
"step": 1075
},
{
"epoch": 2.689486552567237,
"grad_norm": 1.7971349954605103,
"learning_rate": 9.21875e-06,
"loss": 5.8371,
"step": 1100
},
{
"epoch": 2.750611246943765,
"grad_norm": 1.5566753149032593,
"learning_rate": 9.186197916666667e-06,
"loss": 5.8337,
"step": 1125
},
{
"epoch": 2.8117359413202934,
"grad_norm": 1.7522369623184204,
"learning_rate": 9.153645833333334e-06,
"loss": 5.7644,
"step": 1150
},
{
"epoch": 2.8728606356968216,
"grad_norm": 1.545310378074646,
"learning_rate": 9.121093750000001e-06,
"loss": 5.7901,
"step": 1175
},
{
"epoch": 2.93398533007335,
"grad_norm": 1.4376916885375977,
"learning_rate": 9.088541666666666e-06,
"loss": 5.7892,
"step": 1200
},
{
"epoch": 2.9951100244498776,
"grad_norm": 1.7640162706375122,
"learning_rate": 9.055989583333335e-06,
"loss": 5.7924,
"step": 1225
},
{
"epoch": 3.0,
"eval_cer": 1.0,
"eval_loss": 5.430821895599365,
"eval_runtime": 955.8928,
"eval_samples_per_second": 2.3,
"eval_steps_per_second": 0.019,
"eval_wer": 1.0,
"step": 1227
},
{
"epoch": 3.056234718826406,
"grad_norm": 1.6651889085769653,
"learning_rate": 9.0234375e-06,
"loss": 5.778,
"step": 1250
},
{
"epoch": 3.117359413202934,
"grad_norm": 2.1398913860321045,
"learning_rate": 8.990885416666667e-06,
"loss": 5.75,
"step": 1275
},
{
"epoch": 3.178484107579462,
"grad_norm": 1.9647698402404785,
"learning_rate": 8.958333333333334e-06,
"loss": 5.7652,
"step": 1300
},
{
"epoch": 3.23960880195599,
"grad_norm": 1.7572517395019531,
"learning_rate": 8.92578125e-06,
"loss": 5.7267,
"step": 1325
},
{
"epoch": 3.3007334963325183,
"grad_norm": 2.1968727111816406,
"learning_rate": 8.893229166666667e-06,
"loss": 5.7432,
"step": 1350
},
{
"epoch": 3.3618581907090466,
"grad_norm": 1.857980728149414,
"learning_rate": 8.860677083333334e-06,
"loss": 5.7513,
"step": 1375
},
{
"epoch": 3.422982885085575,
"grad_norm": 1.9934026002883911,
"learning_rate": 8.828125000000001e-06,
"loss": 5.7348,
"step": 1400
},
{
"epoch": 3.4841075794621026,
"grad_norm": 2.0249619483947754,
"learning_rate": 8.795572916666666e-06,
"loss": 5.7324,
"step": 1425
},
{
"epoch": 3.545232273838631,
"grad_norm": 2.061025381088257,
"learning_rate": 8.763020833333335e-06,
"loss": 5.7322,
"step": 1450
},
{
"epoch": 3.606356968215159,
"grad_norm": 1.7257977724075317,
"learning_rate": 8.73046875e-06,
"loss": 5.7322,
"step": 1475
},
{
"epoch": 3.667481662591687,
"grad_norm": 1.7211699485778809,
"learning_rate": 8.697916666666667e-06,
"loss": 5.6969,
"step": 1500
},
{
"epoch": 3.728606356968215,
"grad_norm": 2.0788753032684326,
"learning_rate": 8.665364583333334e-06,
"loss": 5.7322,
"step": 1525
},
{
"epoch": 3.7897310513447433,
"grad_norm": 1.8665424585342407,
"learning_rate": 8.6328125e-06,
"loss": 5.7221,
"step": 1550
},
{
"epoch": 3.8508557457212715,
"grad_norm": 2.3112571239471436,
"learning_rate": 8.600260416666667e-06,
"loss": 5.6939,
"step": 1575
},
{
"epoch": 3.9119804400977998,
"grad_norm": 1.8274345397949219,
"learning_rate": 8.567708333333334e-06,
"loss": 5.6866,
"step": 1600
},
{
"epoch": 3.9731051344743276,
"grad_norm": 1.6363919973373413,
"learning_rate": 8.535156250000001e-06,
"loss": 5.7078,
"step": 1625
},
{
"epoch": 4.0,
"eval_cer": 1.0,
"eval_loss": 5.3543291091918945,
"eval_runtime": 973.8534,
"eval_samples_per_second": 2.258,
"eval_steps_per_second": 0.018,
"eval_wer": 1.0,
"step": 1636
},
{
"epoch": 4.034229828850855,
"grad_norm": 1.8639775514602661,
"learning_rate": 8.502604166666666e-06,
"loss": 5.684,
"step": 1650
},
{
"epoch": 4.095354523227384,
"grad_norm": 1.908738613128662,
"learning_rate": 8.470052083333335e-06,
"loss": 5.6709,
"step": 1675
},
{
"epoch": 4.156479217603912,
"grad_norm": 1.785302996635437,
"learning_rate": 8.4375e-06,
"loss": 5.6488,
"step": 1700
},
{
"epoch": 4.21760391198044,
"grad_norm": 2.2629177570343018,
"learning_rate": 8.404947916666667e-06,
"loss": 5.6633,
"step": 1725
},
{
"epoch": 4.278728606356968,
"grad_norm": 2.158043384552002,
"learning_rate": 8.372395833333334e-06,
"loss": 5.6558,
"step": 1750
},
{
"epoch": 4.3398533007334965,
"grad_norm": 2.032712936401367,
"learning_rate": 8.33984375e-06,
"loss": 5.661,
"step": 1775
},
{
"epoch": 4.400977995110025,
"grad_norm": 2.150649070739746,
"learning_rate": 8.307291666666668e-06,
"loss": 5.6587,
"step": 1800
},
{
"epoch": 4.462102689486553,
"grad_norm": 1.866034746170044,
"learning_rate": 8.274739583333334e-06,
"loss": 5.6306,
"step": 1825
},
{
"epoch": 4.52322738386308,
"grad_norm": 2.2522027492523193,
"learning_rate": 8.242187500000001e-06,
"loss": 5.6434,
"step": 1850
},
{
"epoch": 4.5843520782396086,
"grad_norm": 2.266977071762085,
"learning_rate": 8.209635416666666e-06,
"loss": 5.6378,
"step": 1875
},
{
"epoch": 4.645476772616137,
"grad_norm": 2.038745164871216,
"learning_rate": 8.177083333333335e-06,
"loss": 5.6404,
"step": 1900
},
{
"epoch": 4.706601466992665,
"grad_norm": 1.8621001243591309,
"learning_rate": 8.14453125e-06,
"loss": 5.651,
"step": 1925
},
{
"epoch": 4.767726161369193,
"grad_norm": 1.8880043029785156,
"learning_rate": 8.111979166666667e-06,
"loss": 5.6317,
"step": 1950
},
{
"epoch": 4.8288508557457215,
"grad_norm": 1.8351385593414307,
"learning_rate": 8.079427083333334e-06,
"loss": 5.5911,
"step": 1975
},
{
"epoch": 4.88997555012225,
"grad_norm": 1.9690709114074707,
"learning_rate": 8.046875e-06,
"loss": 5.6174,
"step": 2000
},
{
"epoch": 4.951100244498777,
"grad_norm": 1.8599187135696411,
"learning_rate": 8.014322916666668e-06,
"loss": 5.6252,
"step": 2025
},
{
"epoch": 5.0,
"eval_cer": 1.0,
"eval_loss": 5.2904744148254395,
"eval_runtime": 980.786,
"eval_samples_per_second": 2.242,
"eval_steps_per_second": 0.018,
"eval_wer": 1.0,
"step": 2045
},
{
"epoch": 5.012224938875305,
"grad_norm": 2.098679780960083,
"learning_rate": 7.981770833333335e-06,
"loss": 5.5864,
"step": 2050
},
{
"epoch": 5.0733496332518335,
"grad_norm": 2.3753552436828613,
"learning_rate": 7.949218750000001e-06,
"loss": 5.5792,
"step": 2075
},
{
"epoch": 5.134474327628362,
"grad_norm": 2.645547866821289,
"learning_rate": 7.916666666666667e-06,
"loss": 5.5911,
"step": 2100
},
{
"epoch": 5.19559902200489,
"grad_norm": 2.3270106315612793,
"learning_rate": 7.884114583333335e-06,
"loss": 5.6018,
"step": 2125
},
{
"epoch": 5.256723716381418,
"grad_norm": 1.961776852607727,
"learning_rate": 7.8515625e-06,
"loss": 5.5717,
"step": 2150
},
{
"epoch": 5.3178484107579465,
"grad_norm": 2.229034662246704,
"learning_rate": 7.819010416666667e-06,
"loss": 5.5855,
"step": 2175
},
{
"epoch": 5.378973105134475,
"grad_norm": 2.431147575378418,
"learning_rate": 7.786458333333334e-06,
"loss": 5.5863,
"step": 2200
},
{
"epoch": 5.440097799511002,
"grad_norm": 2.657665967941284,
"learning_rate": 7.753906250000001e-06,
"loss": 5.5782,
"step": 2225
},
{
"epoch": 5.50122249388753,
"grad_norm": 1.991417646408081,
"learning_rate": 7.721354166666668e-06,
"loss": 5.5418,
"step": 2250
},
{
"epoch": 5.5623471882640585,
"grad_norm": 2.1882779598236084,
"learning_rate": 7.688802083333335e-06,
"loss": 5.5776,
"step": 2275
},
{
"epoch": 5.623471882640587,
"grad_norm": 1.843143343925476,
"learning_rate": 7.656250000000001e-06,
"loss": 5.588,
"step": 2300
},
{
"epoch": 5.684596577017115,
"grad_norm": 2.3870656490325928,
"learning_rate": 7.623697916666667e-06,
"loss": 5.5632,
"step": 2325
},
{
"epoch": 5.745721271393643,
"grad_norm": 2.0239953994750977,
"learning_rate": 7.591145833333334e-06,
"loss": 5.5445,
"step": 2350
},
{
"epoch": 5.806845965770171,
"grad_norm": 2.3642804622650146,
"learning_rate": 7.55859375e-06,
"loss": 5.5608,
"step": 2375
},
{
"epoch": 5.8679706601467,
"grad_norm": 3.237014055252075,
"learning_rate": 7.526041666666667e-06,
"loss": 5.5308,
"step": 2400
},
{
"epoch": 5.929095354523227,
"grad_norm": 2.084507942199707,
"learning_rate": 7.493489583333334e-06,
"loss": 5.5418,
"step": 2425
},
{
"epoch": 5.990220048899755,
"grad_norm": 1.9642763137817383,
"learning_rate": 7.460937500000001e-06,
"loss": 5.5532,
"step": 2450
},
{
"epoch": 6.0,
"eval_cer": 1.0,
"eval_loss": 5.23870325088501,
"eval_runtime": 909.9853,
"eval_samples_per_second": 2.417,
"eval_steps_per_second": 0.02,
"eval_wer": 1.0,
"step": 2454
},
{
"epoch": 6.0513447432762835,
"grad_norm": 2.086639404296875,
"learning_rate": 7.428385416666667e-06,
"loss": 5.5231,
"step": 2475
},
{
"epoch": 6.112469437652812,
"grad_norm": 1.9924498796463013,
"learning_rate": 7.395833333333335e-06,
"loss": 5.5191,
"step": 2500
},
{
"epoch": 6.17359413202934,
"grad_norm": 2.331226348876953,
"learning_rate": 7.363281250000001e-06,
"loss": 5.5092,
"step": 2525
},
{
"epoch": 6.234718826405868,
"grad_norm": 2.276024580001831,
"learning_rate": 7.330729166666667e-06,
"loss": 5.5208,
"step": 2550
},
{
"epoch": 6.295843520782396,
"grad_norm": 2.2748773097991943,
"learning_rate": 7.298177083333334e-06,
"loss": 5.5241,
"step": 2575
},
{
"epoch": 6.356968215158924,
"grad_norm": 2.164454460144043,
"learning_rate": 7.265625e-06,
"loss": 5.4821,
"step": 2600
},
{
"epoch": 6.418092909535452,
"grad_norm": 2.0158824920654297,
"learning_rate": 7.233072916666667e-06,
"loss": 5.5143,
"step": 2625
},
{
"epoch": 6.47921760391198,
"grad_norm": 2.3213398456573486,
"learning_rate": 7.200520833333334e-06,
"loss": 5.5244,
"step": 2650
},
{
"epoch": 6.540342298288508,
"grad_norm": 3.5166070461273193,
"learning_rate": 7.16796875e-06,
"loss": 5.5251,
"step": 2675
},
{
"epoch": 6.601466992665037,
"grad_norm": 2.501091480255127,
"learning_rate": 7.135416666666667e-06,
"loss": 5.5072,
"step": 2700
},
{
"epoch": 6.662591687041565,
"grad_norm": 2.1940529346466064,
"learning_rate": 7.102864583333334e-06,
"loss": 5.5146,
"step": 2725
},
{
"epoch": 6.723716381418093,
"grad_norm": 2.563851833343506,
"learning_rate": 7.070312500000001e-06,
"loss": 5.4957,
"step": 2750
},
{
"epoch": 6.784841075794621,
"grad_norm": 2.2761764526367188,
"learning_rate": 7.037760416666667e-06,
"loss": 5.4873,
"step": 2775
},
{
"epoch": 6.84596577017115,
"grad_norm": 2.4268720149993896,
"learning_rate": 7.0052083333333345e-06,
"loss": 5.4818,
"step": 2800
},
{
"epoch": 6.907090464547677,
"grad_norm": 2.075167179107666,
"learning_rate": 6.9726562500000005e-06,
"loss": 5.517,
"step": 2825
},
{
"epoch": 6.968215158924205,
"grad_norm": 2.1719608306884766,
"learning_rate": 6.9401041666666665e-06,
"loss": 5.485,
"step": 2850
},
{
"epoch": 7.0,
"eval_cer": 1.0,
"eval_loss": 5.207675457000732,
"eval_runtime": 852.4911,
"eval_samples_per_second": 2.579,
"eval_steps_per_second": 0.021,
"eval_wer": 1.0,
"step": 2863
},
{
"epoch": 7.029339853300733,
"grad_norm": 1.9253339767456055,
"learning_rate": 6.907552083333334e-06,
"loss": 5.4902,
"step": 2875
},
{
"epoch": 7.090464547677262,
"grad_norm": 2.6965110301971436,
"learning_rate": 6.875e-06,
"loss": 5.4694,
"step": 2900
},
{
"epoch": 7.15158924205379,
"grad_norm": 3.292003870010376,
"learning_rate": 6.842447916666667e-06,
"loss": 5.457,
"step": 2925
},
{
"epoch": 7.212713936430318,
"grad_norm": 3.321614980697632,
"learning_rate": 6.809895833333334e-06,
"loss": 5.475,
"step": 2950
},
{
"epoch": 7.273838630806846,
"grad_norm": 2.9773731231689453,
"learning_rate": 6.777343750000001e-06,
"loss": 5.4533,
"step": 2975
},
{
"epoch": 7.334963325183374,
"grad_norm": 2.3799688816070557,
"learning_rate": 6.744791666666667e-06,
"loss": 5.4699,
"step": 3000
},
{
"epoch": 7.396088019559902,
"grad_norm": 2.3236491680145264,
"learning_rate": 6.7122395833333346e-06,
"loss": 5.4694,
"step": 3025
},
{
"epoch": 7.45721271393643,
"grad_norm": 2.295271635055542,
"learning_rate": 6.679687500000001e-06,
"loss": 5.4637,
"step": 3050
},
{
"epoch": 7.518337408312958,
"grad_norm": 2.130282402038574,
"learning_rate": 6.647135416666667e-06,
"loss": 5.4752,
"step": 3075
},
{
"epoch": 7.579462102689487,
"grad_norm": 2.148224115371704,
"learning_rate": 6.614583333333334e-06,
"loss": 5.4303,
"step": 3100
},
{
"epoch": 7.640586797066015,
"grad_norm": 2.275786876678467,
"learning_rate": 6.58203125e-06,
"loss": 5.4389,
"step": 3125
},
{
"epoch": 7.701711491442543,
"grad_norm": 2.203730583190918,
"learning_rate": 6.549479166666667e-06,
"loss": 5.4531,
"step": 3150
},
{
"epoch": 7.76283618581907,
"grad_norm": 2.1406664848327637,
"learning_rate": 6.516927083333334e-06,
"loss": 5.4607,
"step": 3175
},
{
"epoch": 7.823960880195599,
"grad_norm": 2.7617645263671875,
"learning_rate": 6.484375000000001e-06,
"loss": 5.4426,
"step": 3200
},
{
"epoch": 7.885085574572127,
"grad_norm": 2.337761878967285,
"learning_rate": 6.451822916666667e-06,
"loss": 5.453,
"step": 3225
},
{
"epoch": 7.946210268948655,
"grad_norm": 2.157966375350952,
"learning_rate": 6.419270833333335e-06,
"loss": 5.4253,
"step": 3250
},
{
"epoch": 8.0,
"eval_cer": 1.0,
"eval_loss": 5.176779747009277,
"eval_runtime": 850.0532,
"eval_samples_per_second": 2.587,
"eval_steps_per_second": 0.021,
"eval_wer": 1.0,
"step": 3272
},
{
"epoch": 8.007334963325183,
"grad_norm": 2.4567079544067383,
"learning_rate": 6.386718750000001e-06,
"loss": 5.4184,
"step": 3275
},
{
"epoch": 8.06845965770171,
"grad_norm": 2.5252790451049805,
"learning_rate": 6.354166666666667e-06,
"loss": 5.4112,
"step": 3300
},
{
"epoch": 8.12958435207824,
"grad_norm": 2.282362461090088,
"learning_rate": 6.321614583333334e-06,
"loss": 5.4263,
"step": 3325
},
{
"epoch": 8.190709046454767,
"grad_norm": 2.0922985076904297,
"learning_rate": 6.2890625e-06,
"loss": 5.4201,
"step": 3350
},
{
"epoch": 8.251833740831296,
"grad_norm": 2.4645814895629883,
"learning_rate": 6.256510416666666e-06,
"loss": 5.4055,
"step": 3375
},
{
"epoch": 8.312958435207824,
"grad_norm": 2.5172641277313232,
"learning_rate": 6.223958333333334e-06,
"loss": 5.4059,
"step": 3400
},
{
"epoch": 8.374083129584353,
"grad_norm": 2.4925782680511475,
"learning_rate": 6.19140625e-06,
"loss": 5.4282,
"step": 3425
},
{
"epoch": 8.43520782396088,
"grad_norm": 2.384561538696289,
"learning_rate": 6.158854166666667e-06,
"loss": 5.4223,
"step": 3450
},
{
"epoch": 8.496332518337407,
"grad_norm": 2.8028838634490967,
"learning_rate": 6.126302083333334e-06,
"loss": 5.404,
"step": 3475
},
{
"epoch": 8.557457212713937,
"grad_norm": 2.1153008937835693,
"learning_rate": 6.093750000000001e-06,
"loss": 5.4063,
"step": 3500
},
{
"epoch": 8.618581907090464,
"grad_norm": 2.722527265548706,
"learning_rate": 6.061197916666667e-06,
"loss": 5.3986,
"step": 3525
},
{
"epoch": 8.679706601466993,
"grad_norm": 2.9357943534851074,
"learning_rate": 6.0286458333333345e-06,
"loss": 5.3809,
"step": 3550
},
{
"epoch": 8.74083129584352,
"grad_norm": 2.2688732147216797,
"learning_rate": 5.9960937500000005e-06,
"loss": 5.4155,
"step": 3575
},
{
"epoch": 8.80195599022005,
"grad_norm": 2.988487958908081,
"learning_rate": 5.9635416666666665e-06,
"loss": 5.393,
"step": 3600
},
{
"epoch": 8.863080684596577,
"grad_norm": 2.2804296016693115,
"learning_rate": 5.930989583333334e-06,
"loss": 5.4099,
"step": 3625
},
{
"epoch": 8.924205378973106,
"grad_norm": 2.473313808441162,
"learning_rate": 5.8984375e-06,
"loss": 5.4092,
"step": 3650
},
{
"epoch": 8.985330073349633,
"grad_norm": 2.255056142807007,
"learning_rate": 5.865885416666667e-06,
"loss": 5.3916,
"step": 3675
},
{
"epoch": 9.0,
"eval_cer": 1.0,
"eval_loss": 5.145139217376709,
"eval_runtime": 883.7091,
"eval_samples_per_second": 2.488,
"eval_steps_per_second": 0.02,
"eval_wer": 1.0,
"step": 3681
},
{
"epoch": 9.04645476772616,
"grad_norm": 3.237083911895752,
"learning_rate": 5.833333333333334e-06,
"loss": 5.3803,
"step": 3700
},
{
"epoch": 9.10757946210269,
"grad_norm": 2.2703075408935547,
"learning_rate": 5.800781250000001e-06,
"loss": 5.3776,
"step": 3725
},
{
"epoch": 9.168704156479217,
"grad_norm": 2.647369861602783,
"learning_rate": 5.768229166666667e-06,
"loss": 5.3563,
"step": 3750
},
{
"epoch": 9.229828850855746,
"grad_norm": 2.5123939514160156,
"learning_rate": 5.7356770833333345e-06,
"loss": 5.3882,
"step": 3775
},
{
"epoch": 9.290953545232274,
"grad_norm": 2.516632318496704,
"learning_rate": 5.7031250000000006e-06,
"loss": 5.3812,
"step": 3800
},
{
"epoch": 9.352078239608803,
"grad_norm": 2.4889652729034424,
"learning_rate": 5.6705729166666666e-06,
"loss": 5.3907,
"step": 3825
},
{
"epoch": 9.41320293398533,
"grad_norm": 2.4762446880340576,
"learning_rate": 5.638020833333334e-06,
"loss": 5.359,
"step": 3850
},
{
"epoch": 9.474327628361857,
"grad_norm": 3.399854898452759,
"learning_rate": 5.60546875e-06,
"loss": 5.355,
"step": 3875
},
{
"epoch": 9.535452322738386,
"grad_norm": 2.9597182273864746,
"learning_rate": 5.572916666666667e-06,
"loss": 5.3519,
"step": 3900
},
{
"epoch": 9.596577017114914,
"grad_norm": 3.1187338829040527,
"learning_rate": 5.540364583333334e-06,
"loss": 5.3813,
"step": 3925
},
{
"epoch": 9.657701711491443,
"grad_norm": 2.4926772117614746,
"learning_rate": 5.5078125e-06,
"loss": 5.358,
"step": 3950
},
{
"epoch": 9.71882640586797,
"grad_norm": 3.7688283920288086,
"learning_rate": 5.475260416666667e-06,
"loss": 5.3735,
"step": 3975
},
{
"epoch": 9.7799511002445,
"grad_norm": 2.4709203243255615,
"learning_rate": 5.442708333333334e-06,
"loss": 5.3816,
"step": 4000
},
{
"epoch": 9.841075794621027,
"grad_norm": 2.425363540649414,
"learning_rate": 5.410156250000001e-06,
"loss": 5.3674,
"step": 4025
},
{
"epoch": 9.902200488997554,
"grad_norm": 2.380235195159912,
"learning_rate": 5.377604166666667e-06,
"loss": 5.3601,
"step": 4050
},
{
"epoch": 9.963325183374083,
"grad_norm": 2.629493474960327,
"learning_rate": 5.345052083333334e-06,
"loss": 5.3498,
"step": 4075
},
{
"epoch": 10.0,
"eval_cer": 1.0,
"eval_loss": 5.1247124671936035,
"eval_runtime": 818.961,
"eval_samples_per_second": 2.685,
"eval_steps_per_second": 0.022,
"eval_wer": 1.0,
"step": 4090
},
{
"epoch": 10.02444987775061,
"grad_norm": 2.362367630004883,
"learning_rate": 5.3125e-06,
"loss": 5.3588,
"step": 4100
},
{
"epoch": 10.08557457212714,
"grad_norm": 2.3369734287261963,
"learning_rate": 5.279947916666666e-06,
"loss": 5.3411,
"step": 4125
},
{
"epoch": 10.146699266503667,
"grad_norm": 2.3301732540130615,
"learning_rate": 5.247395833333334e-06,
"loss": 5.3375,
"step": 4150
},
{
"epoch": 10.207823960880196,
"grad_norm": 2.3181700706481934,
"learning_rate": 5.21484375e-06,
"loss": 5.3429,
"step": 4175
},
{
"epoch": 10.268948655256724,
"grad_norm": 2.468357563018799,
"learning_rate": 5.182291666666667e-06,
"loss": 5.3622,
"step": 4200
},
{
"epoch": 10.330073349633253,
"grad_norm": 3.00422739982605,
"learning_rate": 5.149739583333334e-06,
"loss": 5.3267,
"step": 4225
},
{
"epoch": 10.39119804400978,
"grad_norm": 2.2816388607025146,
"learning_rate": 5.117187500000001e-06,
"loss": 5.3053,
"step": 4250
},
{
"epoch": 10.452322738386307,
"grad_norm": 2.850501537322998,
"learning_rate": 5.084635416666667e-06,
"loss": 5.3401,
"step": 4275
},
{
"epoch": 10.513447432762836,
"grad_norm": 3.056809663772583,
"learning_rate": 5.0520833333333344e-06,
"loss": 5.323,
"step": 4300
},
{
"epoch": 10.574572127139364,
"grad_norm": 2.261385202407837,
"learning_rate": 5.0195312500000005e-06,
"loss": 5.346,
"step": 4325
},
{
"epoch": 10.635696821515893,
"grad_norm": 2.2135519981384277,
"learning_rate": 4.986979166666667e-06,
"loss": 5.3366,
"step": 4350
},
{
"epoch": 10.69682151589242,
"grad_norm": 2.3966352939605713,
"learning_rate": 4.954427083333333e-06,
"loss": 5.3449,
"step": 4375
},
{
"epoch": 10.75794621026895,
"grad_norm": 2.6069741249084473,
"learning_rate": 4.921875e-06,
"loss": 5.332,
"step": 4400
},
{
"epoch": 10.819070904645477,
"grad_norm": 2.4886317253112793,
"learning_rate": 4.889322916666667e-06,
"loss": 5.3004,
"step": 4425
},
{
"epoch": 10.880195599022004,
"grad_norm": 3.1709907054901123,
"learning_rate": 4.856770833333334e-06,
"loss": 5.3121,
"step": 4450
},
{
"epoch": 10.941320293398533,
"grad_norm": 2.6423654556274414,
"learning_rate": 4.824218750000001e-06,
"loss": 5.3289,
"step": 4475
},
{
"epoch": 11.0,
"eval_cer": 1.0,
"eval_loss": 5.10599422454834,
"eval_runtime": 817.0767,
"eval_samples_per_second": 2.691,
"eval_steps_per_second": 0.022,
"eval_wer": 1.0,
"step": 4499
},
{
"epoch": 11.00244498777506,
"grad_norm": 2.79960036277771,
"learning_rate": 4.791666666666668e-06,
"loss": 5.3274,
"step": 4500
},
{
"epoch": 11.06356968215159,
"grad_norm": 2.352649450302124,
"learning_rate": 4.759114583333334e-06,
"loss": 5.3093,
"step": 4525
},
{
"epoch": 11.124694376528117,
"grad_norm": 3.2299227714538574,
"learning_rate": 4.7265625000000005e-06,
"loss": 5.3201,
"step": 4550
},
{
"epoch": 11.185819070904646,
"grad_norm": 2.820113182067871,
"learning_rate": 4.694010416666667e-06,
"loss": 5.3219,
"step": 4575
},
{
"epoch": 11.246943765281173,
"grad_norm": 2.6876327991485596,
"learning_rate": 4.661458333333333e-06,
"loss": 5.3188,
"step": 4600
},
{
"epoch": 11.3080684596577,
"grad_norm": 2.5107908248901367,
"learning_rate": 4.62890625e-06,
"loss": 5.3094,
"step": 4625
},
{
"epoch": 11.36919315403423,
"grad_norm": 2.567784547805786,
"learning_rate": 4.596354166666667e-06,
"loss": 5.3122,
"step": 4650
},
{
"epoch": 11.430317848410757,
"grad_norm": 2.7552671432495117,
"learning_rate": 4.563802083333333e-06,
"loss": 5.2943,
"step": 4675
},
{
"epoch": 11.491442542787286,
"grad_norm": 3.333078145980835,
"learning_rate": 4.53125e-06,
"loss": 5.2979,
"step": 4700
},
{
"epoch": 11.552567237163814,
"grad_norm": 2.593428373336792,
"learning_rate": 4.498697916666667e-06,
"loss": 5.286,
"step": 4725
},
{
"epoch": 11.613691931540343,
"grad_norm": 2.9048354625701904,
"learning_rate": 4.466145833333334e-06,
"loss": 5.3054,
"step": 4750
},
{
"epoch": 11.67481662591687,
"grad_norm": 2.767232894897461,
"learning_rate": 4.433593750000001e-06,
"loss": 5.304,
"step": 4775
},
{
"epoch": 11.7359413202934,
"grad_norm": 2.6838302612304688,
"learning_rate": 4.4010416666666675e-06,
"loss": 5.2967,
"step": 4800
},
{
"epoch": 11.797066014669927,
"grad_norm": 2.392949104309082,
"learning_rate": 4.3684895833333335e-06,
"loss": 5.3095,
"step": 4825
},
{
"epoch": 11.858190709046454,
"grad_norm": 2.424485921859741,
"learning_rate": 4.3359375e-06,
"loss": 5.2737,
"step": 4850
},
{
"epoch": 11.919315403422983,
"grad_norm": 2.665613889694214,
"learning_rate": 4.303385416666667e-06,
"loss": 5.2863,
"step": 4875
},
{
"epoch": 11.98044009779951,
"grad_norm": 2.450711250305176,
"learning_rate": 4.270833333333333e-06,
"loss": 5.3016,
"step": 4900
},
{
"epoch": 12.0,
"eval_cer": 1.0,
"eval_loss": 5.094453811645508,
"eval_runtime": 808.2878,
"eval_samples_per_second": 2.721,
"eval_steps_per_second": 0.022,
"eval_wer": 1.0,
"step": 4908
}
],
"logging_steps": 25,
"max_steps": 8180,
"num_input_tokens_seen": 0,
"num_train_epochs": 20,
"save_steps": 500,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 3.94724964630528e+19,
"train_batch_size": 124,
"trial_name": null,
"trial_params": null
}