zhko_xlsr_run1 / checkpoint-65930 /trainer_state.json
yesj1234's picture
Upload folder using huggingface_hub
af43350
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 10.0,
"eval_steps": 500,
"global_step": 65930,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.08,
"learning_rate": 0.0003,
"loss": 27.6845,
"step": 500
},
{
"epoch": 0.15,
"learning_rate": 0.0002977074736359468,
"loss": 5.9239,
"step": 1000
},
{
"epoch": 0.23,
"learning_rate": 0.0002954149472718936,
"loss": 5.8508,
"step": 1500
},
{
"epoch": 0.3,
"learning_rate": 0.0002931224209078404,
"loss": 5.8132,
"step": 2000
},
{
"epoch": 0.38,
"learning_rate": 0.00029082989454378723,
"loss": 5.7398,
"step": 2500
},
{
"epoch": 0.46,
"learning_rate": 0.00028853736817973405,
"loss": 5.6899,
"step": 3000
},
{
"epoch": 0.53,
"learning_rate": 0.00028624484181568086,
"loss": 5.669,
"step": 3500
},
{
"epoch": 0.61,
"learning_rate": 0.0002839523154516277,
"loss": 5.6871,
"step": 4000
},
{
"epoch": 0.68,
"learning_rate": 0.0002816597890875745,
"loss": 5.6503,
"step": 4500
},
{
"epoch": 0.76,
"learning_rate": 0.0002793672627235213,
"loss": 5.571,
"step": 5000
},
{
"epoch": 0.83,
"learning_rate": 0.0002770747363594681,
"loss": 5.4537,
"step": 5500
},
{
"epoch": 0.91,
"learning_rate": 0.00027478220999541494,
"loss": 5.3316,
"step": 6000
},
{
"epoch": 0.99,
"learning_rate": 0.0002724896836313617,
"loss": 5.0444,
"step": 6500
},
{
"epoch": 1.0,
"eval_cer": 0.8763969974979149,
"eval_loss": 4.669518947601318,
"eval_runtime": 55.1634,
"eval_samples_per_second": 5.819,
"eval_steps_per_second": 2.919,
"step": 6593
},
{
"epoch": 1.06,
"learning_rate": 0.00027019715726730857,
"loss": 4.7276,
"step": 7000
},
{
"epoch": 1.14,
"learning_rate": 0.0002679046309032554,
"loss": 4.5233,
"step": 7500
},
{
"epoch": 1.21,
"learning_rate": 0.0002656121045392022,
"loss": 4.3549,
"step": 8000
},
{
"epoch": 1.29,
"learning_rate": 0.000263319578175149,
"loss": 4.181,
"step": 8500
},
{
"epoch": 1.37,
"learning_rate": 0.0002610270518110958,
"loss": 4.0707,
"step": 9000
},
{
"epoch": 1.44,
"learning_rate": 0.00025873452544704264,
"loss": 3.8964,
"step": 9500
},
{
"epoch": 1.52,
"learning_rate": 0.0002564419990829894,
"loss": 3.7554,
"step": 10000
},
{
"epoch": 1.59,
"learning_rate": 0.00025414947271893627,
"loss": 3.673,
"step": 10500
},
{
"epoch": 1.67,
"learning_rate": 0.00025185694635488303,
"loss": 3.5734,
"step": 11000
},
{
"epoch": 1.74,
"learning_rate": 0.0002495644199908299,
"loss": 3.5483,
"step": 11500
},
{
"epoch": 1.82,
"learning_rate": 0.00024727189362677666,
"loss": 3.4484,
"step": 12000
},
{
"epoch": 1.9,
"learning_rate": 0.00024497936726272353,
"loss": 3.3388,
"step": 12500
},
{
"epoch": 1.97,
"learning_rate": 0.0002426868408986703,
"loss": 3.2619,
"step": 13000
},
{
"epoch": 2.0,
"eval_cer": 0.6248540450375313,
"eval_loss": 2.9759151935577393,
"eval_runtime": 54.7232,
"eval_samples_per_second": 5.866,
"eval_steps_per_second": 2.942,
"step": 13186
},
{
"epoch": 2.05,
"learning_rate": 0.00024039431453461713,
"loss": 3.1258,
"step": 13500
},
{
"epoch": 2.12,
"learning_rate": 0.00023810178817056395,
"loss": 3.0259,
"step": 14000
},
{
"epoch": 2.2,
"learning_rate": 0.00023580926180651076,
"loss": 3.0577,
"step": 14500
},
{
"epoch": 2.28,
"learning_rate": 0.00023351673544245758,
"loss": 2.9761,
"step": 15000
},
{
"epoch": 2.35,
"learning_rate": 0.00023122420907840436,
"loss": 2.8932,
"step": 15500
},
{
"epoch": 2.43,
"learning_rate": 0.0002289316827143512,
"loss": 2.9507,
"step": 16000
},
{
"epoch": 2.5,
"learning_rate": 0.000226639156350298,
"loss": 2.9112,
"step": 16500
},
{
"epoch": 2.58,
"learning_rate": 0.00022434662998624484,
"loss": 2.9235,
"step": 17000
},
{
"epoch": 2.65,
"learning_rate": 0.00022205410362219162,
"loss": 2.8323,
"step": 17500
},
{
"epoch": 2.73,
"learning_rate": 0.00021976157725813846,
"loss": 2.8452,
"step": 18000
},
{
"epoch": 2.81,
"learning_rate": 0.00021746905089408525,
"loss": 2.8166,
"step": 18500
},
{
"epoch": 2.88,
"learning_rate": 0.00021517652453003207,
"loss": 2.807,
"step": 19000
},
{
"epoch": 2.96,
"learning_rate": 0.00021288399816597888,
"loss": 2.7859,
"step": 19500
},
{
"epoch": 3.0,
"eval_cer": 0.5421184320266889,
"eval_loss": 2.5369625091552734,
"eval_runtime": 56.2807,
"eval_samples_per_second": 5.704,
"eval_steps_per_second": 2.861,
"step": 19779
},
{
"epoch": 3.03,
"learning_rate": 0.0002105914718019257,
"loss": 2.664,
"step": 20000
},
{
"epoch": 3.11,
"learning_rate": 0.00020829894543787254,
"loss": 2.4556,
"step": 20500
},
{
"epoch": 3.19,
"learning_rate": 0.00020600641907381933,
"loss": 2.4617,
"step": 21000
},
{
"epoch": 3.26,
"learning_rate": 0.00020371389270976617,
"loss": 2.5162,
"step": 21500
},
{
"epoch": 3.34,
"learning_rate": 0.00020142136634571296,
"loss": 2.5092,
"step": 22000
},
{
"epoch": 3.41,
"learning_rate": 0.00019912883998165977,
"loss": 2.4437,
"step": 22500
},
{
"epoch": 3.49,
"learning_rate": 0.00019683631361760659,
"loss": 2.4205,
"step": 23000
},
{
"epoch": 3.56,
"learning_rate": 0.0001945437872535534,
"loss": 2.45,
"step": 23500
},
{
"epoch": 3.64,
"learning_rate": 0.00019225126088950022,
"loss": 2.4553,
"step": 24000
},
{
"epoch": 3.72,
"learning_rate": 0.00018995873452544703,
"loss": 2.3963,
"step": 24500
},
{
"epoch": 3.79,
"learning_rate": 0.00018766620816139382,
"loss": 2.3739,
"step": 25000
},
{
"epoch": 3.87,
"learning_rate": 0.00018537368179734066,
"loss": 2.463,
"step": 25500
},
{
"epoch": 3.94,
"learning_rate": 0.00018308115543328745,
"loss": 2.3947,
"step": 26000
},
{
"epoch": 4.0,
"eval_cer": 0.515929941618015,
"eval_loss": 2.358245611190796,
"eval_runtime": 55.2411,
"eval_samples_per_second": 5.811,
"eval_steps_per_second": 2.914,
"step": 26372
},
{
"epoch": 4.02,
"learning_rate": 0.0001807886290692343,
"loss": 2.3494,
"step": 26500
},
{
"epoch": 4.1,
"learning_rate": 0.0001784961027051811,
"loss": 2.1528,
"step": 27000
},
{
"epoch": 4.17,
"learning_rate": 0.00017620357634112792,
"loss": 2.1103,
"step": 27500
},
{
"epoch": 4.25,
"learning_rate": 0.00017391104997707473,
"loss": 2.0831,
"step": 28000
},
{
"epoch": 4.32,
"learning_rate": 0.00017161852361302152,
"loss": 2.124,
"step": 28500
},
{
"epoch": 4.4,
"learning_rate": 0.00016932599724896836,
"loss": 2.0801,
"step": 29000
},
{
"epoch": 4.47,
"learning_rate": 0.00016703347088491515,
"loss": 2.1513,
"step": 29500
},
{
"epoch": 4.55,
"learning_rate": 0.000164740944520862,
"loss": 2.0648,
"step": 30000
},
{
"epoch": 4.63,
"learning_rate": 0.00016244841815680878,
"loss": 2.0649,
"step": 30500
},
{
"epoch": 4.7,
"learning_rate": 0.0001601558917927556,
"loss": 2.1093,
"step": 31000
},
{
"epoch": 4.78,
"learning_rate": 0.0001578633654287024,
"loss": 2.1226,
"step": 31500
},
{
"epoch": 4.85,
"learning_rate": 0.00015557083906464923,
"loss": 2.066,
"step": 32000
},
{
"epoch": 4.93,
"learning_rate": 0.00015327831270059604,
"loss": 2.1313,
"step": 32500
},
{
"epoch": 5.0,
"eval_cer": 0.4872393661384487,
"eval_loss": 2.246213436126709,
"eval_runtime": 56.125,
"eval_samples_per_second": 5.719,
"eval_steps_per_second": 2.869,
"step": 32965
},
{
"epoch": 5.01,
"learning_rate": 0.00015098578633654285,
"loss": 2.0466,
"step": 33000
},
{
"epoch": 5.08,
"learning_rate": 0.00014869325997248967,
"loss": 1.7844,
"step": 33500
},
{
"epoch": 5.16,
"learning_rate": 0.00014640073360843648,
"loss": 1.8717,
"step": 34000
},
{
"epoch": 5.23,
"learning_rate": 0.0001441082072443833,
"loss": 1.8059,
"step": 34500
},
{
"epoch": 5.31,
"learning_rate": 0.00014181568088033011,
"loss": 1.818,
"step": 35000
},
{
"epoch": 5.38,
"learning_rate": 0.00013952315451627693,
"loss": 1.8108,
"step": 35500
},
{
"epoch": 5.46,
"learning_rate": 0.00013723062815222374,
"loss": 1.8391,
"step": 36000
},
{
"epoch": 5.54,
"learning_rate": 0.00013493810178817053,
"loss": 1.8801,
"step": 36500
},
{
"epoch": 5.61,
"learning_rate": 0.00013264557542411737,
"loss": 1.8382,
"step": 37000
},
{
"epoch": 5.69,
"learning_rate": 0.0001303530490600642,
"loss": 1.8418,
"step": 37500
},
{
"epoch": 5.76,
"learning_rate": 0.000128060522696011,
"loss": 1.7916,
"step": 38000
},
{
"epoch": 5.84,
"learning_rate": 0.00012576799633195782,
"loss": 1.7707,
"step": 38500
},
{
"epoch": 5.92,
"learning_rate": 0.00012347546996790463,
"loss": 1.807,
"step": 39000
},
{
"epoch": 5.99,
"learning_rate": 0.00012118294360385143,
"loss": 1.8083,
"step": 39500
},
{
"epoch": 6.0,
"eval_cer": 0.469557964970809,
"eval_loss": 2.210385799407959,
"eval_runtime": 57.4991,
"eval_samples_per_second": 5.583,
"eval_steps_per_second": 2.8,
"step": 39558
},
{
"epoch": 6.07,
"learning_rate": 0.00011889041723979825,
"loss": 1.5902,
"step": 40000
},
{
"epoch": 6.14,
"learning_rate": 0.00011659789087574506,
"loss": 1.5428,
"step": 40500
},
{
"epoch": 6.22,
"learning_rate": 0.00011430536451169188,
"loss": 1.5965,
"step": 41000
},
{
"epoch": 6.29,
"learning_rate": 0.00011201283814763868,
"loss": 1.5892,
"step": 41500
},
{
"epoch": 6.37,
"learning_rate": 0.0001097203117835855,
"loss": 1.5869,
"step": 42000
},
{
"epoch": 6.45,
"learning_rate": 0.00010742778541953231,
"loss": 1.5466,
"step": 42500
},
{
"epoch": 6.52,
"learning_rate": 0.00010513525905547912,
"loss": 1.6088,
"step": 43000
},
{
"epoch": 6.6,
"learning_rate": 0.00010284273269142595,
"loss": 1.5707,
"step": 43500
},
{
"epoch": 6.67,
"learning_rate": 0.00010055020632737277,
"loss": 1.5875,
"step": 44000
},
{
"epoch": 6.75,
"learning_rate": 9.825767996331957e-05,
"loss": 1.544,
"step": 44500
},
{
"epoch": 6.83,
"learning_rate": 9.596515359926638e-05,
"loss": 1.56,
"step": 45000
},
{
"epoch": 6.9,
"learning_rate": 9.36726272352132e-05,
"loss": 1.6172,
"step": 45500
},
{
"epoch": 6.98,
"learning_rate": 9.138010087116001e-05,
"loss": 1.5209,
"step": 46000
},
{
"epoch": 7.0,
"eval_cer": 0.46605504587155966,
"eval_loss": 2.162141799926758,
"eval_runtime": 55.7456,
"eval_samples_per_second": 5.758,
"eval_steps_per_second": 2.888,
"step": 46151
},
{
"epoch": 7.05,
"learning_rate": 8.908757450710683e-05,
"loss": 1.3888,
"step": 46500
},
{
"epoch": 7.13,
"learning_rate": 8.679504814305364e-05,
"loss": 1.3561,
"step": 47000
},
{
"epoch": 7.2,
"learning_rate": 8.450252177900044e-05,
"loss": 1.3643,
"step": 47500
},
{
"epoch": 7.28,
"learning_rate": 8.220999541494726e-05,
"loss": 1.3816,
"step": 48000
},
{
"epoch": 7.36,
"learning_rate": 7.991746905089407e-05,
"loss": 1.3173,
"step": 48500
},
{
"epoch": 7.43,
"learning_rate": 7.762494268684089e-05,
"loss": 1.4088,
"step": 49000
},
{
"epoch": 7.51,
"learning_rate": 7.53324163227877e-05,
"loss": 1.3434,
"step": 49500
},
{
"epoch": 7.58,
"learning_rate": 7.303988995873452e-05,
"loss": 1.379,
"step": 50000
},
{
"epoch": 7.66,
"learning_rate": 7.074736359468133e-05,
"loss": 1.3693,
"step": 50500
},
{
"epoch": 7.74,
"learning_rate": 6.845483723062815e-05,
"loss": 1.3075,
"step": 51000
},
{
"epoch": 7.81,
"learning_rate": 6.616231086657496e-05,
"loss": 1.3585,
"step": 51500
},
{
"epoch": 7.89,
"learning_rate": 6.386978450252178e-05,
"loss": 1.4037,
"step": 52000
},
{
"epoch": 7.96,
"learning_rate": 6.157725813846859e-05,
"loss": 1.3735,
"step": 52500
},
{
"epoch": 8.0,
"eval_cer": 0.45821517931609673,
"eval_loss": 2.2031326293945312,
"eval_runtime": 55.4705,
"eval_samples_per_second": 5.787,
"eval_steps_per_second": 2.902,
"step": 52744
},
{
"epoch": 8.04,
"learning_rate": 5.92847317744154e-05,
"loss": 1.288,
"step": 53000
},
{
"epoch": 8.11,
"learning_rate": 5.6992205410362215e-05,
"loss": 1.2032,
"step": 53500
},
{
"epoch": 8.19,
"learning_rate": 5.469967904630902e-05,
"loss": 1.2034,
"step": 54000
},
{
"epoch": 8.27,
"learning_rate": 5.240715268225584e-05,
"loss": 1.2098,
"step": 54500
},
{
"epoch": 8.34,
"learning_rate": 5.011462631820266e-05,
"loss": 1.196,
"step": 55000
},
{
"epoch": 8.42,
"learning_rate": 4.7822099954149474e-05,
"loss": 1.1629,
"step": 55500
},
{
"epoch": 8.49,
"learning_rate": 4.552957359009628e-05,
"loss": 1.1932,
"step": 56000
},
{
"epoch": 8.57,
"learning_rate": 4.3237047226043097e-05,
"loss": 1.1548,
"step": 56500
},
{
"epoch": 8.65,
"learning_rate": 4.0944520861989905e-05,
"loss": 1.1835,
"step": 57000
},
{
"epoch": 8.72,
"learning_rate": 3.865199449793672e-05,
"loss": 1.2334,
"step": 57500
},
{
"epoch": 8.8,
"learning_rate": 3.6359468133883534e-05,
"loss": 1.2385,
"step": 58000
},
{
"epoch": 8.87,
"learning_rate": 3.406694176983035e-05,
"loss": 1.1762,
"step": 58500
},
{
"epoch": 8.95,
"learning_rate": 3.1774415405777164e-05,
"loss": 1.1354,
"step": 59000
},
{
"epoch": 9.0,
"eval_cer": 0.44920767306088405,
"eval_loss": 2.1740152835845947,
"eval_runtime": 55.8382,
"eval_samples_per_second": 5.749,
"eval_steps_per_second": 2.883,
"step": 59337
},
{
"epoch": 9.02,
"learning_rate": 2.948188904172398e-05,
"loss": 1.1439,
"step": 59500
},
{
"epoch": 9.1,
"learning_rate": 2.718936267767079e-05,
"loss": 1.1242,
"step": 60000
},
{
"epoch": 9.18,
"learning_rate": 2.4896836313617605e-05,
"loss": 1.0779,
"step": 60500
},
{
"epoch": 9.25,
"learning_rate": 2.260430994956442e-05,
"loss": 1.1212,
"step": 61000
},
{
"epoch": 9.33,
"learning_rate": 2.031178358551123e-05,
"loss": 1.0518,
"step": 61500
},
{
"epoch": 9.4,
"learning_rate": 1.8019257221458046e-05,
"loss": 1.05,
"step": 62000
},
{
"epoch": 9.48,
"learning_rate": 1.572673085740486e-05,
"loss": 1.0233,
"step": 62500
},
{
"epoch": 9.56,
"learning_rate": 1.3434204493351672e-05,
"loss": 1.0439,
"step": 63000
},
{
"epoch": 9.63,
"learning_rate": 1.1141678129298487e-05,
"loss": 1.0929,
"step": 63500
},
{
"epoch": 9.71,
"learning_rate": 8.8491517652453e-06,
"loss": 1.0614,
"step": 64000
},
{
"epoch": 9.78,
"learning_rate": 6.556625401192113e-06,
"loss": 1.0696,
"step": 64500
},
{
"epoch": 9.86,
"learning_rate": 4.264099037138927e-06,
"loss": 1.0251,
"step": 65000
},
{
"epoch": 9.93,
"learning_rate": 1.9715726730857402e-06,
"loss": 1.0788,
"step": 65500
},
{
"epoch": 10.0,
"eval_cer": 0.44653878231859884,
"eval_loss": 2.1948180198669434,
"eval_runtime": 56.1921,
"eval_samples_per_second": 5.713,
"eval_steps_per_second": 2.865,
"step": 65930
}
],
"logging_steps": 500,
"max_steps": 65930,
"num_train_epochs": 10,
"save_steps": 500,
"total_flos": 3.725663511627642e+19,
"trial_name": null,
"trial_params": null
}