whisper-base-ft-cv-cy-en / trainer_state.json
DewiBrynJones's picture
End of training
3701e18 verified
raw
history blame
15.2 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.4149274849663955,
"eval_steps": 1000,
"global_step": 2000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.017686593562079942,
"grad_norm": 28.968021392822266,
"learning_rate": 4.6000000000000004e-07,
"loss": 3.698,
"step": 25
},
{
"epoch": 0.035373187124159884,
"grad_norm": 21.002214431762695,
"learning_rate": 9.600000000000001e-07,
"loss": 3.5124,
"step": 50
},
{
"epoch": 0.05305978068623983,
"grad_norm": 21.64297866821289,
"learning_rate": 1.46e-06,
"loss": 3.2335,
"step": 75
},
{
"epoch": 0.07074637424831977,
"grad_norm": 19.800973892211914,
"learning_rate": 1.9600000000000003e-06,
"loss": 2.9933,
"step": 100
},
{
"epoch": 0.08843296781039972,
"grad_norm": 19.281307220458984,
"learning_rate": 2.46e-06,
"loss": 2.5097,
"step": 125
},
{
"epoch": 0.10611956137247966,
"grad_norm": 19.187774658203125,
"learning_rate": 2.96e-06,
"loss": 2.3572,
"step": 150
},
{
"epoch": 0.1238061549345596,
"grad_norm": 17.483562469482422,
"learning_rate": 3.46e-06,
"loss": 2.1479,
"step": 175
},
{
"epoch": 0.14149274849663954,
"grad_norm": 17.524105072021484,
"learning_rate": 3.96e-06,
"loss": 2.2237,
"step": 200
},
{
"epoch": 0.1591793420587195,
"grad_norm": 17.354618072509766,
"learning_rate": 4.4600000000000005e-06,
"loss": 2.0436,
"step": 225
},
{
"epoch": 0.17686593562079944,
"grad_norm": 18.539562225341797,
"learning_rate": 4.960000000000001e-06,
"loss": 2.0212,
"step": 250
},
{
"epoch": 0.19455252918287938,
"grad_norm": 18.09463882446289,
"learning_rate": 5.460000000000001e-06,
"loss": 1.9828,
"step": 275
},
{
"epoch": 0.21223912274495932,
"grad_norm": 16.73234748840332,
"learning_rate": 5.9600000000000005e-06,
"loss": 1.9199,
"step": 300
},
{
"epoch": 0.22992571630703926,
"grad_norm": 17.19611930847168,
"learning_rate": 6.460000000000001e-06,
"loss": 1.8943,
"step": 325
},
{
"epoch": 0.2476123098691192,
"grad_norm": 18.489797592163086,
"learning_rate": 6.96e-06,
"loss": 1.7673,
"step": 350
},
{
"epoch": 0.26529890343119916,
"grad_norm": 17.019506454467773,
"learning_rate": 7.4600000000000006e-06,
"loss": 1.6931,
"step": 375
},
{
"epoch": 0.2829854969932791,
"grad_norm": 16.345083236694336,
"learning_rate": 7.960000000000002e-06,
"loss": 1.6946,
"step": 400
},
{
"epoch": 0.30067209055535904,
"grad_norm": 16.138513565063477,
"learning_rate": 8.46e-06,
"loss": 1.6849,
"step": 425
},
{
"epoch": 0.318358684117439,
"grad_norm": 17.647178649902344,
"learning_rate": 8.96e-06,
"loss": 1.677,
"step": 450
},
{
"epoch": 0.3360452776795189,
"grad_norm": 17.51073455810547,
"learning_rate": 9.460000000000001e-06,
"loss": 1.608,
"step": 475
},
{
"epoch": 0.3537318712415989,
"grad_norm": 15.542696952819824,
"learning_rate": 9.960000000000001e-06,
"loss": 1.5617,
"step": 500
},
{
"epoch": 0.3714184648036788,
"grad_norm": 14.3616361618042,
"learning_rate": 9.846666666666668e-06,
"loss": 1.4832,
"step": 525
},
{
"epoch": 0.38910505836575876,
"grad_norm": 14.655681610107422,
"learning_rate": 9.68e-06,
"loss": 1.521,
"step": 550
},
{
"epoch": 0.40679165192783867,
"grad_norm": 15.943355560302734,
"learning_rate": 9.513333333333334e-06,
"loss": 1.4946,
"step": 575
},
{
"epoch": 0.42447824548991864,
"grad_norm": 17.380294799804688,
"learning_rate": 9.346666666666666e-06,
"loss": 1.4358,
"step": 600
},
{
"epoch": 0.4421648390519986,
"grad_norm": 16.160900115966797,
"learning_rate": 9.180000000000002e-06,
"loss": 1.4093,
"step": 625
},
{
"epoch": 0.4598514326140785,
"grad_norm": 14.763204574584961,
"learning_rate": 9.013333333333334e-06,
"loss": 1.4101,
"step": 650
},
{
"epoch": 0.4775380261761585,
"grad_norm": 15.164654731750488,
"learning_rate": 8.846666666666668e-06,
"loss": 1.3897,
"step": 675
},
{
"epoch": 0.4952246197382384,
"grad_norm": 15.769131660461426,
"learning_rate": 8.68e-06,
"loss": 1.3422,
"step": 700
},
{
"epoch": 0.5129112133003184,
"grad_norm": 16.859813690185547,
"learning_rate": 8.513333333333335e-06,
"loss": 1.3451,
"step": 725
},
{
"epoch": 0.5305978068623983,
"grad_norm": 14.078543663024902,
"learning_rate": 8.346666666666668e-06,
"loss": 1.3333,
"step": 750
},
{
"epoch": 0.5482844004244782,
"grad_norm": 13.418024063110352,
"learning_rate": 8.18e-06,
"loss": 1.3618,
"step": 775
},
{
"epoch": 0.5659709939865581,
"grad_norm": 14.887019157409668,
"learning_rate": 8.013333333333333e-06,
"loss": 1.2983,
"step": 800
},
{
"epoch": 0.5836575875486382,
"grad_norm": 13.075471878051758,
"learning_rate": 7.846666666666667e-06,
"loss": 1.3021,
"step": 825
},
{
"epoch": 0.6013441811107181,
"grad_norm": 14.660274505615234,
"learning_rate": 7.680000000000001e-06,
"loss": 1.2375,
"step": 850
},
{
"epoch": 0.619030774672798,
"grad_norm": 16.32628631591797,
"learning_rate": 7.513333333333334e-06,
"loss": 1.3038,
"step": 875
},
{
"epoch": 0.636717368234878,
"grad_norm": 17.701143264770508,
"learning_rate": 7.346666666666668e-06,
"loss": 1.2082,
"step": 900
},
{
"epoch": 0.6544039617969579,
"grad_norm": 14.465027809143066,
"learning_rate": 7.180000000000001e-06,
"loss": 1.2543,
"step": 925
},
{
"epoch": 0.6720905553590378,
"grad_norm": 12.923417091369629,
"learning_rate": 7.0133333333333345e-06,
"loss": 1.2544,
"step": 950
},
{
"epoch": 0.6897771489211177,
"grad_norm": 15.675958633422852,
"learning_rate": 6.846666666666667e-06,
"loss": 1.2279,
"step": 975
},
{
"epoch": 0.7074637424831978,
"grad_norm": 15.041681289672852,
"learning_rate": 6.680000000000001e-06,
"loss": 1.2368,
"step": 1000
},
{
"epoch": 0.7074637424831978,
"eval_loss": 0.6446413397789001,
"eval_runtime": 2060.0729,
"eval_samples_per_second": 5.225,
"eval_steps_per_second": 0.327,
"eval_wer": 0.965980208088868,
"step": 1000
},
{
"epoch": 0.7251503360452777,
"grad_norm": 15.67192554473877,
"learning_rate": 6.513333333333333e-06,
"loss": 1.1917,
"step": 1025
},
{
"epoch": 0.7428369296073576,
"grad_norm": 14.699666023254395,
"learning_rate": 6.346666666666668e-06,
"loss": 1.1812,
"step": 1050
},
{
"epoch": 0.7605235231694376,
"grad_norm": 14.321735382080078,
"learning_rate": 6.18e-06,
"loss": 1.2046,
"step": 1075
},
{
"epoch": 0.7782101167315175,
"grad_norm": 14.692147254943848,
"learning_rate": 6.013333333333335e-06,
"loss": 1.2071,
"step": 1100
},
{
"epoch": 0.7958967102935974,
"grad_norm": 13.778362274169922,
"learning_rate": 5.846666666666667e-06,
"loss": 1.1476,
"step": 1125
},
{
"epoch": 0.8135833038556773,
"grad_norm": 13.074742317199707,
"learning_rate": 5.68e-06,
"loss": 1.1543,
"step": 1150
},
{
"epoch": 0.8312698974177574,
"grad_norm": 14.026896476745605,
"learning_rate": 5.513333333333334e-06,
"loss": 1.1841,
"step": 1175
},
{
"epoch": 0.8489564909798373,
"grad_norm": 13.244227409362793,
"learning_rate": 5.346666666666667e-06,
"loss": 1.1698,
"step": 1200
},
{
"epoch": 0.8666430845419172,
"grad_norm": 15.449938774108887,
"learning_rate": 5.18e-06,
"loss": 1.1213,
"step": 1225
},
{
"epoch": 0.8843296781039972,
"grad_norm": 13.706199645996094,
"learning_rate": 5.013333333333333e-06,
"loss": 1.1935,
"step": 1250
},
{
"epoch": 0.9020162716660771,
"grad_norm": 14.426807403564453,
"learning_rate": 4.846666666666667e-06,
"loss": 1.0659,
"step": 1275
},
{
"epoch": 0.919702865228157,
"grad_norm": 14.945075035095215,
"learning_rate": 4.680000000000001e-06,
"loss": 1.1169,
"step": 1300
},
{
"epoch": 0.937389458790237,
"grad_norm": 13.335543632507324,
"learning_rate": 4.513333333333333e-06,
"loss": 1.0949,
"step": 1325
},
{
"epoch": 0.955076052352317,
"grad_norm": 15.83086109161377,
"learning_rate": 4.346666666666667e-06,
"loss": 1.0845,
"step": 1350
},
{
"epoch": 0.9727626459143969,
"grad_norm": 12.246453285217285,
"learning_rate": 4.18e-06,
"loss": 1.148,
"step": 1375
},
{
"epoch": 0.9904492394764768,
"grad_norm": 16.202903747558594,
"learning_rate": 4.013333333333334e-06,
"loss": 1.1118,
"step": 1400
},
{
"epoch": 1.0081358330385568,
"grad_norm": 12.500658988952637,
"learning_rate": 3.8466666666666665e-06,
"loss": 1.0352,
"step": 1425
},
{
"epoch": 1.0258224266006368,
"grad_norm": 11.863042831420898,
"learning_rate": 3.6800000000000003e-06,
"loss": 1.0224,
"step": 1450
},
{
"epoch": 1.0435090201627166,
"grad_norm": 13.093315124511719,
"learning_rate": 3.5133333333333337e-06,
"loss": 0.966,
"step": 1475
},
{
"epoch": 1.0611956137247966,
"grad_norm": 11.886918067932129,
"learning_rate": 3.346666666666667e-06,
"loss": 0.9643,
"step": 1500
},
{
"epoch": 1.0788822072868764,
"grad_norm": 15.925506591796875,
"learning_rate": 3.1800000000000005e-06,
"loss": 0.8792,
"step": 1525
},
{
"epoch": 1.0965688008489565,
"grad_norm": 12.430258750915527,
"learning_rate": 3.013333333333334e-06,
"loss": 0.9462,
"step": 1550
},
{
"epoch": 1.1142553944110365,
"grad_norm": 13.365354537963867,
"learning_rate": 2.8466666666666672e-06,
"loss": 0.9246,
"step": 1575
},
{
"epoch": 1.1319419879731163,
"grad_norm": 13.153193473815918,
"learning_rate": 2.68e-06,
"loss": 0.8996,
"step": 1600
},
{
"epoch": 1.1496285815351963,
"grad_norm": 13.20574951171875,
"learning_rate": 2.5133333333333336e-06,
"loss": 0.943,
"step": 1625
},
{
"epoch": 1.1673151750972763,
"grad_norm": 10.716314315795898,
"learning_rate": 2.346666666666667e-06,
"loss": 0.9071,
"step": 1650
},
{
"epoch": 1.1850017686593561,
"grad_norm": 15.43217658996582,
"learning_rate": 2.1800000000000003e-06,
"loss": 0.9232,
"step": 1675
},
{
"epoch": 1.2026883622214362,
"grad_norm": 14.106558799743652,
"learning_rate": 2.0133333333333337e-06,
"loss": 0.8998,
"step": 1700
},
{
"epoch": 1.2203749557835162,
"grad_norm": 12.580400466918945,
"learning_rate": 1.8466666666666668e-06,
"loss": 0.928,
"step": 1725
},
{
"epoch": 1.238061549345596,
"grad_norm": 12.888578414916992,
"learning_rate": 1.6800000000000002e-06,
"loss": 0.9354,
"step": 1750
},
{
"epoch": 1.255748142907676,
"grad_norm": 11.692194938659668,
"learning_rate": 1.5133333333333334e-06,
"loss": 0.8441,
"step": 1775
},
{
"epoch": 1.2734347364697558,
"grad_norm": 14.173469543457031,
"learning_rate": 1.3466666666666668e-06,
"loss": 0.925,
"step": 1800
},
{
"epoch": 1.2911213300318358,
"grad_norm": 12.65889835357666,
"learning_rate": 1.1800000000000001e-06,
"loss": 0.905,
"step": 1825
},
{
"epoch": 1.3088079235939158,
"grad_norm": 12.559049606323242,
"learning_rate": 1.0133333333333333e-06,
"loss": 0.8771,
"step": 1850
},
{
"epoch": 1.3264945171559956,
"grad_norm": 11.348992347717285,
"learning_rate": 8.466666666666668e-07,
"loss": 0.9033,
"step": 1875
},
{
"epoch": 1.3441811107180757,
"grad_norm": 11.88202953338623,
"learning_rate": 6.800000000000001e-07,
"loss": 0.8451,
"step": 1900
},
{
"epoch": 1.3618677042801557,
"grad_norm": 11.244508743286133,
"learning_rate": 5.133333333333334e-07,
"loss": 0.8839,
"step": 1925
},
{
"epoch": 1.3795542978422355,
"grad_norm": 12.49364185333252,
"learning_rate": 3.466666666666667e-07,
"loss": 0.8928,
"step": 1950
},
{
"epoch": 1.3972408914043155,
"grad_norm": 11.834848403930664,
"learning_rate": 1.8e-07,
"loss": 0.9105,
"step": 1975
},
{
"epoch": 1.4149274849663955,
"grad_norm": 12.054742813110352,
"learning_rate": 1.3333333333333334e-08,
"loss": 0.8994,
"step": 2000
},
{
"epoch": 1.4149274849663955,
"eval_loss": 0.5812551379203796,
"eval_runtime": 1719.3277,
"eval_samples_per_second": 6.26,
"eval_steps_per_second": 0.391,
"eval_wer": 1.254253696315419,
"step": 2000
},
{
"epoch": 1.4149274849663955,
"step": 2000,
"total_flos": 4.15045601427456e+18,
"train_loss": 1.393887762069702,
"train_runtime": 12787.317,
"train_samples_per_second": 5.005,
"train_steps_per_second": 0.156
}
],
"logging_steps": 25,
"max_steps": 2000,
"num_input_tokens_seen": 0,
"num_train_epochs": 2,
"save_steps": 1000,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 4.15045601427456e+18,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}