reach-vb's picture
reach-vb HF staff
Upload folder using huggingface_hub
7ad8883 verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.4792637465051257,
"eval_steps": 318,
"global_step": 6350,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.00023299161230195712,
"grad_norm": 6.78125,
"learning_rate": 2e-06,
"loss": 0.7236,
"step": 1
},
{
"epoch": 0.023299161230195712,
"grad_norm": 0.09033203125,
"learning_rate": 0.0002,
"loss": 0.3799,
"step": 100
},
{
"epoch": 0.046598322460391424,
"grad_norm": 0.11865234375,
"learning_rate": 0.0004,
"loss": 0.2452,
"step": 200
},
{
"epoch": 0.06989748369058714,
"grad_norm": 0.1396484375,
"learning_rate": 0.0006,
"loss": 0.2131,
"step": 300
},
{
"epoch": 0.07409133271202237,
"eval_peoplespeech-clean-transcription_loss": 3.0843491554260254,
"eval_peoplespeech-clean-transcription_model_preparation_time": 0.0063,
"eval_peoplespeech-clean-transcription_runtime": 14.6247,
"eval_peoplespeech-clean-transcription_samples_per_second": 4.376,
"eval_peoplespeech-clean-transcription_steps_per_second": 0.068,
"step": 318
},
{
"epoch": 0.09319664492078285,
"grad_norm": 0.07421875,
"learning_rate": 0.0008,
"loss": 0.155,
"step": 400
},
{
"epoch": 0.11649580615097857,
"grad_norm": 0.0576171875,
"learning_rate": 0.001,
"loss": 0.1083,
"step": 500
},
{
"epoch": 0.13979496738117428,
"grad_norm": 0.04736328125,
"learning_rate": 0.0012,
"loss": 0.0926,
"step": 600
},
{
"epoch": 0.14818266542404473,
"eval_peoplespeech-clean-transcription_loss": 1.7947263717651367,
"eval_peoplespeech-clean-transcription_model_preparation_time": 0.0063,
"eval_peoplespeech-clean-transcription_runtime": 13.9513,
"eval_peoplespeech-clean-transcription_samples_per_second": 4.587,
"eval_peoplespeech-clean-transcription_steps_per_second": 0.072,
"step": 636
},
{
"epoch": 0.16309412861137,
"grad_norm": 0.04150390625,
"learning_rate": 0.0014,
"loss": 0.0844,
"step": 700
},
{
"epoch": 0.1863932898415657,
"grad_norm": 0.034912109375,
"learning_rate": 0.0016,
"loss": 0.0793,
"step": 800
},
{
"epoch": 0.2096924510717614,
"grad_norm": 0.032958984375,
"learning_rate": 0.0018000000000000002,
"loss": 0.0758,
"step": 900
},
{
"epoch": 0.2222739981360671,
"eval_peoplespeech-clean-transcription_loss": 1.7458150386810303,
"eval_peoplespeech-clean-transcription_model_preparation_time": 0.0063,
"eval_peoplespeech-clean-transcription_runtime": 14.2202,
"eval_peoplespeech-clean-transcription_samples_per_second": 4.501,
"eval_peoplespeech-clean-transcription_steps_per_second": 0.07,
"step": 954
},
{
"epoch": 0.23299161230195714,
"grad_norm": 0.03125,
"learning_rate": 0.002,
"loss": 0.0732,
"step": 1000
},
{
"epoch": 0.25629077353215285,
"grad_norm": 0.02783203125,
"learning_rate": 0.0019984487567773325,
"loss": 0.0714,
"step": 1100
},
{
"epoch": 0.27958993476234856,
"grad_norm": 0.0289306640625,
"learning_rate": 0.0019938003745660765,
"loss": 0.0691,
"step": 1200
},
{
"epoch": 0.29636533084808947,
"eval_peoplespeech-clean-transcription_loss": 1.7118018865585327,
"eval_peoplespeech-clean-transcription_model_preparation_time": 0.0063,
"eval_peoplespeech-clean-transcription_runtime": 14.7608,
"eval_peoplespeech-clean-transcription_samples_per_second": 4.336,
"eval_peoplespeech-clean-transcription_steps_per_second": 0.068,
"step": 1272
},
{
"epoch": 0.30288909599254427,
"grad_norm": 0.027587890625,
"learning_rate": 0.0019860708773026797,
"loss": 0.0676,
"step": 1300
},
{
"epoch": 0.32618825722274,
"grad_norm": 0.0257568359375,
"learning_rate": 0.001975286910165463,
"loss": 0.0657,
"step": 1400
},
{
"epoch": 0.3494874184529357,
"grad_norm": 0.0244140625,
"learning_rate": 0.0019614856477231713,
"loss": 0.0648,
"step": 1500
},
{
"epoch": 0.3704566635601118,
"eval_peoplespeech-clean-transcription_loss": 1.6898815631866455,
"eval_peoplespeech-clean-transcription_model_preparation_time": 0.0063,
"eval_peoplespeech-clean-transcription_runtime": 14.5539,
"eval_peoplespeech-clean-transcription_samples_per_second": 4.397,
"eval_peoplespeech-clean-transcription_steps_per_second": 0.069,
"step": 1590
},
{
"epoch": 0.3727865796831314,
"grad_norm": 0.0208740234375,
"learning_rate": 0.0019447146657865335,
"loss": 0.0641,
"step": 1600
},
{
"epoch": 0.3960857409133271,
"grad_norm": 0.02587890625,
"learning_rate": 0.001925031777404586,
"loss": 0.0629,
"step": 1700
},
{
"epoch": 0.4193849021435228,
"grad_norm": 0.0269775390625,
"learning_rate": 0.0019025048335711181,
"loss": 0.0623,
"step": 1800
},
{
"epoch": 0.4426840633737186,
"grad_norm": 0.0208740234375,
"learning_rate": 0.0018772114893282391,
"loss": 0.0616,
"step": 1900
},
{
"epoch": 0.4445479962721342,
"eval_peoplespeech-clean-transcription_loss": 1.6862739324569702,
"eval_peoplespeech-clean-transcription_model_preparation_time": 0.0063,
"eval_peoplespeech-clean-transcription_runtime": 14.4561,
"eval_peoplespeech-clean-transcription_samples_per_second": 4.427,
"eval_peoplespeech-clean-transcription_steps_per_second": 0.069,
"step": 1908
},
{
"epoch": 0.4659832246039143,
"grad_norm": 0.02099609375,
"learning_rate": 0.0018492389360733603,
"loss": 0.0612,
"step": 2000
},
{
"epoch": 0.48928238583411,
"grad_norm": 0.0245361328125,
"learning_rate": 0.0018186836009923876,
"loss": 0.0605,
"step": 2100
},
{
"epoch": 0.5125815470643057,
"grad_norm": 0.0206298828125,
"learning_rate": 0.001785650814655233,
"loss": 0.0603,
"step": 2200
},
{
"epoch": 0.5186393289841565,
"eval_peoplespeech-clean-transcription_loss": 1.6496751308441162,
"eval_peoplespeech-clean-transcription_model_preparation_time": 0.0063,
"eval_peoplespeech-clean-transcription_runtime": 14.7041,
"eval_peoplespeech-clean-transcription_samples_per_second": 4.353,
"eval_peoplespeech-clean-transcription_steps_per_second": 0.068,
"step": 2226
},
{
"epoch": 0.5358807082945014,
"grad_norm": 0.022705078125,
"learning_rate": 0.0017502544479195221,
"loss": 0.0599,
"step": 2300
},
{
"epoch": 0.5591798695246971,
"grad_norm": 0.0189208984375,
"learning_rate": 0.0017126165193941567,
"loss": 0.0591,
"step": 2400
},
{
"epoch": 0.5824790307548928,
"grad_norm": 0.0208740234375,
"learning_rate": 0.0016728667748158923,
"loss": 0.0591,
"step": 2500
},
{
"epoch": 0.5927306616961789,
"eval_peoplespeech-clean-transcription_loss": 1.6663990020751953,
"eval_peoplespeech-clean-transcription_model_preparation_time": 0.0063,
"eval_peoplespeech-clean-transcription_runtime": 14.7316,
"eval_peoplespeech-clean-transcription_samples_per_second": 4.344,
"eval_peoplespeech-clean-transcription_steps_per_second": 0.068,
"step": 2544
},
{
"epoch": 0.6057781919850885,
"grad_norm": 0.021728515625,
"learning_rate": 0.0016311422397889072,
"loss": 0.0585,
"step": 2600
},
{
"epoch": 0.6290773532152842,
"grad_norm": 0.022216796875,
"learning_rate": 0.0015875867474291552,
"loss": 0.0582,
"step": 2700
},
{
"epoch": 0.65237651444548,
"grad_norm": 0.0189208984375,
"learning_rate": 0.0015423504425418212,
"loss": 0.0574,
"step": 2800
},
{
"epoch": 0.6668219944082013,
"eval_peoplespeech-clean-transcription_loss": 1.619414210319519,
"eval_peoplespeech-clean-transcription_model_preparation_time": 0.0063,
"eval_peoplespeech-clean-transcription_runtime": 14.4777,
"eval_peoplespeech-clean-transcription_samples_per_second": 4.421,
"eval_peoplespeech-clean-transcription_steps_per_second": 0.069,
"step": 2862
},
{
"epoch": 0.6756756756756757,
"grad_norm": 0.0191650390625,
"learning_rate": 0.0014955892640410716,
"loss": 0.057,
"step": 2900
},
{
"epoch": 0.6989748369058714,
"grad_norm": 0.017822265625,
"learning_rate": 0.0014474644073963134,
"loss": 0.0573,
"step": 3000
},
{
"epoch": 0.7222739981360671,
"grad_norm": 0.02001953125,
"learning_rate": 0.0013981417689580185,
"loss": 0.0572,
"step": 3100
},
{
"epoch": 0.7409133271202236,
"eval_peoplespeech-clean-transcription_loss": 1.6249196529388428,
"eval_peoplespeech-clean-transcription_model_preparation_time": 0.0063,
"eval_peoplespeech-clean-transcription_runtime": 14.2587,
"eval_peoplespeech-clean-transcription_samples_per_second": 4.488,
"eval_peoplespeech-clean-transcription_steps_per_second": 0.07,
"step": 3180
},
{
"epoch": 0.7455731593662628,
"grad_norm": 0.0196533203125,
"learning_rate": 0.0013477913740786338,
"loss": 0.0565,
"step": 3200
},
{
"epoch": 0.7688723205964585,
"grad_norm": 0.0196533203125,
"learning_rate": 0.0012965867909999716,
"loss": 0.0559,
"step": 3300
},
{
"epoch": 0.7921714818266542,
"grad_norm": 0.020263671875,
"learning_rate": 0.0012447045325275216,
"loss": 0.056,
"step": 3400
},
{
"epoch": 0.815004659832246,
"eval_peoplespeech-clean-transcription_loss": 1.623028039932251,
"eval_peoplespeech-clean-transcription_model_preparation_time": 0.0063,
"eval_peoplespeech-clean-transcription_runtime": 14.4831,
"eval_peoplespeech-clean-transcription_samples_per_second": 4.419,
"eval_peoplespeech-clean-transcription_steps_per_second": 0.069,
"step": 3498
},
{
"epoch": 0.8154706430568499,
"grad_norm": 0.01806640625,
"learning_rate": 0.0011923234475542434,
"loss": 0.0556,
"step": 3500
},
{
"epoch": 0.8387698042870456,
"grad_norm": 0.020751953125,
"learning_rate": 0.001139624104531383,
"loss": 0.056,
"step": 3600
},
{
"epoch": 0.8620689655172413,
"grad_norm": 0.0164794921875,
"learning_rate": 0.0010867881690116152,
"loss": 0.0552,
"step": 3700
},
{
"epoch": 0.8853681267474371,
"grad_norm": 0.0191650390625,
"learning_rate": 0.0010339977774102467,
"loss": 0.0548,
"step": 3800
},
{
"epoch": 0.8890959925442684,
"eval_peoplespeech-clean-transcription_loss": 1.6235145330429077,
"eval_peoplespeech-clean-transcription_model_preparation_time": 0.0063,
"eval_peoplespeech-clean-transcription_runtime": 14.6909,
"eval_peoplespeech-clean-transcription_samples_per_second": 4.356,
"eval_peoplespeech-clean-transcription_steps_per_second": 0.068,
"step": 3816
},
{
"epoch": 0.9086672879776329,
"grad_norm": 0.0299072265625,
"learning_rate": 0.0009814349091432633,
"loss": 0.0546,
"step": 3900
},
{
"epoch": 0.9319664492078286,
"grad_norm": 0.017578125,
"learning_rate": 0.0009292807593065831,
"loss": 0.0545,
"step": 4000
},
{
"epoch": 0.9552656104380243,
"grad_norm": 0.0166015625,
"learning_rate": 0.0008777151140590245,
"loss": 0.0547,
"step": 4100
},
{
"epoch": 0.9631873252562908,
"eval_peoplespeech-clean-transcription_loss": 1.598435640335083,
"eval_peoplespeech-clean-transcription_model_preparation_time": 0.0063,
"eval_peoplespeech-clean-transcription_runtime": 14.2932,
"eval_peoplespeech-clean-transcription_samples_per_second": 4.478,
"eval_peoplespeech-clean-transcription_steps_per_second": 0.07,
"step": 4134
},
{
"epoch": 0.97856477166822,
"grad_norm": 0.020751953125,
"learning_rate": 0.0008269157308621709,
"loss": 0.0546,
"step": 4200
},
{
"epoch": 1.0016309412861137,
"grad_norm": 0.0174560546875,
"learning_rate": 0.0007770577257135778,
"loss": 0.0539,
"step": 4300
},
{
"epoch": 1.0249301025163093,
"grad_norm": 0.02197265625,
"learning_rate": 0.0007283129694856508,
"loss": 0.0523,
"step": 4400
},
{
"epoch": 1.0370456663560113,
"eval_peoplespeech-clean-transcription_loss": 1.6100133657455444,
"eval_peoplespeech-clean-transcription_model_preparation_time": 0.0063,
"eval_peoplespeech-clean-transcription_runtime": 14.1344,
"eval_peoplespeech-clean-transcription_samples_per_second": 4.528,
"eval_peoplespeech-clean-transcription_steps_per_second": 0.071,
"step": 4452
},
{
"epoch": 1.0482292637465052,
"grad_norm": 0.0174560546875,
"learning_rate": 0.0006808494954511429,
"loss": 0.0538,
"step": 4500
},
{
"epoch": 1.0715284249767008,
"grad_norm": 0.0184326171875,
"learning_rate": 0.0006348309200376554,
"loss": 0.0538,
"step": 4600
},
{
"epoch": 1.0948275862068966,
"grad_norm": 0.0177001953125,
"learning_rate": 0.0005904158788079147,
"loss": 0.0533,
"step": 4700
},
{
"epoch": 1.1111369990680335,
"eval_peoplespeech-clean-transcription_loss": 1.6077880859375,
"eval_peoplespeech-clean-transcription_model_preparation_time": 0.0063,
"eval_peoplespeech-clean-transcription_runtime": 14.3377,
"eval_peoplespeech-clean-transcription_samples_per_second": 4.464,
"eval_peoplespeech-clean-transcription_steps_per_second": 0.07,
"step": 4770
},
{
"epoch": 1.1181267474370922,
"grad_norm": 0.01806640625,
"learning_rate": 0.0005477574796101257,
"loss": 0.0533,
"step": 4800
},
{
"epoch": 1.141425908667288,
"grad_norm": 0.0184326171875,
"learning_rate": 0.0005070027747835002,
"loss": 0.0535,
"step": 4900
},
{
"epoch": 1.1647250698974836,
"grad_norm": 0.016845703125,
"learning_rate": 0.0004682922542383806,
"loss": 0.0532,
"step": 5000
},
{
"epoch": 1.1852283317800558,
"eval_peoplespeech-clean-transcription_loss": 1.5890922546386719,
"eval_peoplespeech-clean-transcription_model_preparation_time": 0.0063,
"eval_peoplespeech-clean-transcription_runtime": 14.3703,
"eval_peoplespeech-clean-transcription_samples_per_second": 4.454,
"eval_peoplespeech-clean-transcription_steps_per_second": 0.07,
"step": 5088
},
{
"epoch": 1.1880242311276794,
"grad_norm": 0.023193359375,
"learning_rate": 0.00043175936115841517,
"loss": 0.0531,
"step": 5100
},
{
"epoch": 1.2113233923578752,
"grad_norm": 0.01904296875,
"learning_rate": 0.00039753003199425864,
"loss": 0.0526,
"step": 5200
},
{
"epoch": 1.2346225535880708,
"grad_norm": 0.018798828125,
"learning_rate": 0.00036572226233453855,
"loss": 0.0524,
"step": 5300
},
{
"epoch": 1.2579217148182664,
"grad_norm": 0.0185546875,
"learning_rate": 0.00033644570015061657,
"loss": 0.0521,
"step": 5400
},
{
"epoch": 1.2593196644920783,
"eval_peoplespeech-clean-transcription_loss": 1.5955970287322998,
"eval_peoplespeech-clean-transcription_model_preparation_time": 0.0063,
"eval_peoplespeech-clean-transcription_runtime": 14.105,
"eval_peoplespeech-clean-transcription_samples_per_second": 4.537,
"eval_peoplespeech-clean-transcription_steps_per_second": 0.071,
"step": 5406
},
{
"epoch": 1.2812208760484622,
"grad_norm": 0.0240478515625,
"learning_rate": 0.0003098012678173169,
"loss": 0.0519,
"step": 5500
},
{
"epoch": 1.304520037278658,
"grad_norm": 0.0174560546875,
"learning_rate": 0.00028588081421259233,
"loss": 0.0519,
"step": 5600
},
{
"epoch": 1.3278191985088537,
"grad_norm": 0.0177001953125,
"learning_rate": 0.0002647667980954152,
"loss": 0.0515,
"step": 5700
},
{
"epoch": 1.3334109972041006,
"eval_peoplespeech-clean-transcription_loss": 1.593904972076416,
"eval_peoplespeech-clean-transcription_model_preparation_time": 0.0063,
"eval_peoplespeech-clean-transcription_runtime": 14.3513,
"eval_peoplespeech-clean-transcription_samples_per_second": 4.46,
"eval_peoplespeech-clean-transcription_steps_per_second": 0.07,
"step": 5724
},
{
"epoch": 1.3511183597390493,
"grad_norm": 0.0162353515625,
"learning_rate": 0.00024653200385335225,
"loss": 0.0516,
"step": 5800
},
{
"epoch": 1.374417520969245,
"grad_norm": 0.0234375,
"learning_rate": 0.00023123929059970283,
"loss": 0.0515,
"step": 5900
},
{
"epoch": 1.397716682199441,
"grad_norm": 0.0184326171875,
"learning_rate": 0.0002189413754851135,
"loss": 0.0511,
"step": 6000
},
{
"epoch": 1.407502329916123,
"eval_peoplespeech-clean-transcription_loss": 1.5913946628570557,
"eval_peoplespeech-clean-transcription_model_preparation_time": 0.0063,
"eval_peoplespeech-clean-transcription_runtime": 14.3873,
"eval_peoplespeech-clean-transcription_samples_per_second": 4.448,
"eval_peoplespeech-clean-transcription_steps_per_second": 0.07,
"step": 6042
},
{
"epoch": 1.4210158434296365,
"grad_norm": 0.0164794921875,
"learning_rate": 0.00020968065197063865,
"loss": 0.0512,
"step": 6100
},
{
"epoch": 1.4443150046598323,
"grad_norm": 0.0169677734375,
"learning_rate": 0.00020348904368869776,
"loss": 0.0511,
"step": 6200
},
{
"epoch": 1.467614165890028,
"grad_norm": 0.01611328125,
"learning_rate": 0.00020038789439570147,
"loss": 0.051,
"step": 6300
}
],
"logging_steps": 100,
"max_steps": 6350,
"num_input_tokens_seen": 0,
"num_train_epochs": 2,
"save_steps": 1588,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 1.1071489029735186e+19,
"train_batch_size": 672,
"trial_name": null,
"trial_params": null
}