reach-vb's picture
reach-vb HF staff
Upload folder using huggingface_hub
4ce67bd verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.7575996219877146,
"eval_steps": 558,
"global_step": 11160,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.00015750511891636477,
"grad_norm": 8.1875,
"learning_rate": 2e-06,
"loss": 1.1045,
"step": 1
},
{
"epoch": 0.015750511891636478,
"grad_norm": 0.09521484375,
"learning_rate": 0.0002,
"loss": 0.7573,
"step": 100
},
{
"epoch": 0.031501023783272956,
"grad_norm": 0.1533203125,
"learning_rate": 0.0004,
"loss": 0.5534,
"step": 200
},
{
"epoch": 0.04725153567490943,
"grad_norm": 0.314453125,
"learning_rate": 0.0006,
"loss": 0.4982,
"step": 300
},
{
"epoch": 0.06300204756654591,
"grad_norm": 0.0986328125,
"learning_rate": 0.0008,
"loss": 0.223,
"step": 400
},
{
"epoch": 0.0787525594581824,
"grad_norm": 0.0830078125,
"learning_rate": 0.001,
"loss": 0.1493,
"step": 500
},
{
"epoch": 0.08788785635533154,
"eval_peoplespeech-clean-transcription_loss": 1.7001044750213623,
"eval_peoplespeech-clean-transcription_model_preparation_time": 0.0064,
"eval_peoplespeech-clean-transcription_runtime": 15.0028,
"eval_peoplespeech-clean-transcription_samples_per_second": 4.266,
"eval_peoplespeech-clean-transcription_steps_per_second": 0.067,
"step": 558
},
{
"epoch": 0.09450307134981886,
"grad_norm": 0.06103515625,
"learning_rate": 0.0012,
"loss": 0.1319,
"step": 600
},
{
"epoch": 0.11025358324145534,
"grad_norm": 0.054443359375,
"learning_rate": 0.0014,
"loss": 0.1219,
"step": 700
},
{
"epoch": 0.12600409513309183,
"grad_norm": 0.056396484375,
"learning_rate": 0.0016,
"loss": 0.1165,
"step": 800
},
{
"epoch": 0.1417546070247283,
"grad_norm": 0.05322265625,
"learning_rate": 0.0018000000000000002,
"loss": 0.1116,
"step": 900
},
{
"epoch": 0.1575051189163648,
"grad_norm": 0.05322265625,
"learning_rate": 0.002,
"loss": 0.1077,
"step": 1000
},
{
"epoch": 0.17325563080800127,
"grad_norm": 0.03955078125,
"learning_rate": 0.0019995697803527326,
"loss": 0.1049,
"step": 1100
},
{
"epoch": 0.17577571271066308,
"eval_peoplespeech-clean-transcription_loss": 1.5866798162460327,
"eval_peoplespeech-clean-transcription_model_preparation_time": 0.0064,
"eval_peoplespeech-clean-transcription_runtime": 15.6243,
"eval_peoplespeech-clean-transcription_samples_per_second": 4.096,
"eval_peoplespeech-clean-transcription_steps_per_second": 0.064,
"step": 1116
},
{
"epoch": 0.18900614269963772,
"grad_norm": 0.04150390625,
"learning_rate": 0.001998279532719697,
"loss": 0.1022,
"step": 1200
},
{
"epoch": 0.2047566545912742,
"grad_norm": 0.039794921875,
"learning_rate": 0.0019961304906339628,
"loss": 0.0999,
"step": 1300
},
{
"epoch": 0.2205071664829107,
"grad_norm": 0.0400390625,
"learning_rate": 0.0019931247086735927,
"loss": 0.0977,
"step": 1400
},
{
"epoch": 0.23625767837454717,
"grad_norm": 0.03515625,
"learning_rate": 0.0019892650604973754,
"loss": 0.097,
"step": 1500
},
{
"epoch": 0.25200819026618365,
"grad_norm": 0.039306640625,
"learning_rate": 0.0019845552360974817,
"loss": 0.0953,
"step": 1600
},
{
"epoch": 0.26366356906599464,
"eval_peoplespeech-clean-transcription_loss": 1.5846567153930664,
"eval_peoplespeech-clean-transcription_model_preparation_time": 0.0064,
"eval_peoplespeech-clean-transcription_runtime": 15.2276,
"eval_peoplespeech-clean-transcription_samples_per_second": 4.203,
"eval_peoplespeech-clean-transcription_steps_per_second": 0.066,
"step": 1674
},
{
"epoch": 0.2677587021578201,
"grad_norm": 0.03515625,
"learning_rate": 0.0019789997382716707,
"loss": 0.093,
"step": 1700
},
{
"epoch": 0.2835092140494566,
"grad_norm": 0.035888671875,
"learning_rate": 0.0019726038783184218,
"loss": 0.093,
"step": 1800
},
{
"epoch": 0.29925972594109307,
"grad_norm": 0.035888671875,
"learning_rate": 0.0019653737709590968,
"loss": 0.0919,
"step": 1900
},
{
"epoch": 0.3150102378327296,
"grad_norm": 0.0296630859375,
"learning_rate": 0.001957316328492001,
"loss": 0.0905,
"step": 2000
},
{
"epoch": 0.33076074972436603,
"grad_norm": 0.029296875,
"learning_rate": 0.001948439254183927,
"loss": 0.0904,
"step": 2100
},
{
"epoch": 0.34651126161600254,
"grad_norm": 0.031982421875,
"learning_rate": 0.001938751034905491,
"loss": 0.0892,
"step": 2200
},
{
"epoch": 0.35155142542132617,
"eval_peoplespeech-clean-transcription_loss": 1.5401419401168823,
"eval_peoplespeech-clean-transcription_model_preparation_time": 0.0064,
"eval_peoplespeech-clean-transcription_runtime": 15.0918,
"eval_peoplespeech-clean-transcription_samples_per_second": 4.241,
"eval_peoplespeech-clean-transcription_steps_per_second": 0.066,
"step": 2232
},
{
"epoch": 0.362261773507639,
"grad_norm": 0.0294189453125,
"learning_rate": 0.0019282609330173168,
"loss": 0.0894,
"step": 2300
},
{
"epoch": 0.37801228539927545,
"grad_norm": 0.0279541015625,
"learning_rate": 0.0019169789775148136,
"loss": 0.088,
"step": 2400
},
{
"epoch": 0.39376279729091196,
"grad_norm": 0.03125,
"learning_rate": 0.0019049159544400186,
"loss": 0.0873,
"step": 2500
},
{
"epoch": 0.4095133091825484,
"grad_norm": 0.0294189453125,
"learning_rate": 0.0018920833965696696,
"loss": 0.0858,
"step": 2600
},
{
"epoch": 0.4252638210741849,
"grad_norm": 0.027099609375,
"learning_rate": 0.0018784935723893685,
"loss": 0.0866,
"step": 2700
},
{
"epoch": 0.43943928177665775,
"eval_peoplespeech-clean-transcription_loss": 1.5469986200332642,
"eval_peoplespeech-clean-transcription_model_preparation_time": 0.0064,
"eval_peoplespeech-clean-transcription_runtime": 15.1359,
"eval_peoplespeech-clean-transcription_samples_per_second": 4.228,
"eval_peoplespeech-clean-transcription_steps_per_second": 0.066,
"step": 2790
},
{
"epoch": 0.4410143329658214,
"grad_norm": 0.0257568359375,
"learning_rate": 0.0018641594743643711,
"loss": 0.0857,
"step": 2800
},
{
"epoch": 0.4567648448574579,
"grad_norm": 0.0302734375,
"learning_rate": 0.0018490948065182245,
"loss": 0.0855,
"step": 2900
},
{
"epoch": 0.47251535674909434,
"grad_norm": 0.025390625,
"learning_rate": 0.001833313971331122,
"loss": 0.0847,
"step": 3000
},
{
"epoch": 0.48826586864073085,
"grad_norm": 0.024169921875,
"learning_rate": 0.0018168320559705034,
"loss": 0.0847,
"step": 3100
},
{
"epoch": 0.5040163805323673,
"grad_norm": 0.02783203125,
"learning_rate": 0.0017996648178670632,
"loss": 0.0847,
"step": 3200
},
{
"epoch": 0.5197668924240038,
"grad_norm": 0.0274658203125,
"learning_rate": 0.0017818286696499605,
"loss": 0.0841,
"step": 3300
},
{
"epoch": 0.5273271381319893,
"eval_peoplespeech-clean-transcription_loss": 1.5245492458343506,
"eval_peoplespeech-clean-transcription_model_preparation_time": 0.0064,
"eval_peoplespeech-clean-transcription_runtime": 14.9971,
"eval_peoplespeech-clean-transcription_samples_per_second": 4.267,
"eval_peoplespeech-clean-transcription_steps_per_second": 0.067,
"step": 3348
},
{
"epoch": 0.5355174043156402,
"grad_norm": 0.0242919921875,
"learning_rate": 0.0017633406634556276,
"loss": 0.0846,
"step": 3400
},
{
"epoch": 0.5512679162072768,
"grad_norm": 0.025634765625,
"learning_rate": 0.0017442184746251834,
"loss": 0.083,
"step": 3500
},
{
"epoch": 0.5670184280989132,
"grad_norm": 0.0289306640625,
"learning_rate": 0.0017244803848060362,
"loss": 0.0828,
"step": 3600
},
{
"epoch": 0.5827689399905497,
"grad_norm": 0.025390625,
"learning_rate": 0.0017041452644738296,
"loss": 0.0829,
"step": 3700
},
{
"epoch": 0.5985194518821861,
"grad_norm": 0.024658203125,
"learning_rate": 0.0016832325548914446,
"loss": 0.0825,
"step": 3800
},
{
"epoch": 0.6142699637738227,
"grad_norm": 0.0244140625,
"learning_rate": 0.0016617622495223036,
"loss": 0.0823,
"step": 3900
},
{
"epoch": 0.6152149944873209,
"eval_peoplespeech-clean-transcription_loss": 1.5190149545669556,
"eval_peoplespeech-clean-transcription_model_preparation_time": 0.0064,
"eval_peoplespeech-clean-transcription_runtime": 15.2829,
"eval_peoplespeech-clean-transcription_samples_per_second": 4.188,
"eval_peoplespeech-clean-transcription_steps_per_second": 0.065,
"step": 3906
},
{
"epoch": 0.6300204756654592,
"grad_norm": 0.0244140625,
"learning_rate": 0.001639754874915746,
"loss": 0.0814,
"step": 4000
},
{
"epoch": 0.6457709875570956,
"grad_norm": 0.025634765625,
"learning_rate": 0.0016172314710827496,
"loss": 0.0807,
"step": 4100
},
{
"epoch": 0.6615214994487321,
"grad_norm": 0.0260009765625,
"learning_rate": 0.0015942135713807655,
"loss": 0.0806,
"step": 4200
},
{
"epoch": 0.6772720113403685,
"grad_norm": 0.02734375,
"learning_rate": 0.0015707231819268826,
"loss": 0.0801,
"step": 4300
},
{
"epoch": 0.6930225232320051,
"grad_norm": 0.02392578125,
"learning_rate": 0.0015467827605590238,
"loss": 0.0803,
"step": 4400
},
{
"epoch": 0.7031028508426523,
"eval_peoplespeech-clean-transcription_loss": 1.5322189331054688,
"eval_peoplespeech-clean-transcription_model_preparation_time": 0.0064,
"eval_peoplespeech-clean-transcription_runtime": 15.1503,
"eval_peoplespeech-clean-transcription_samples_per_second": 4.224,
"eval_peoplespeech-clean-transcription_steps_per_second": 0.066,
"step": 4464
},
{
"epoch": 0.7087730351236415,
"grad_norm": 0.0245361328125,
"learning_rate": 0.0015224151953652696,
"loss": 0.0804,
"step": 4500
},
{
"epoch": 0.724523547015278,
"grad_norm": 0.02490234375,
"learning_rate": 0.0014976437828018482,
"loss": 0.0803,
"step": 4600
},
{
"epoch": 0.7402740589069144,
"grad_norm": 0.0250244140625,
"learning_rate": 0.0014724922054207043,
"loss": 0.0797,
"step": 4700
},
{
"epoch": 0.7560245707985509,
"grad_norm": 0.0242919921875,
"learning_rate": 0.0014469845092279468,
"loss": 0.0789,
"step": 4800
},
{
"epoch": 0.7717750826901875,
"grad_norm": 0.025390625,
"learning_rate": 0.0014211450806948175,
"loss": 0.0788,
"step": 4900
},
{
"epoch": 0.7875255945818239,
"grad_norm": 0.026123046875,
"learning_rate": 0.001394998623443159,
"loss": 0.0792,
"step": 5000
},
{
"epoch": 0.7909907071979839,
"eval_peoplespeech-clean-transcription_loss": 1.5153441429138184,
"eval_peoplespeech-clean-transcription_model_preparation_time": 0.0064,
"eval_peoplespeech-clean-transcription_runtime": 15.1311,
"eval_peoplespeech-clean-transcription_samples_per_second": 4.23,
"eval_peoplespeech-clean-transcription_steps_per_second": 0.066,
"step": 5022
},
{
"epoch": 0.8032761064734604,
"grad_norm": 0.02880859375,
"learning_rate": 0.001368570134627674,
"loss": 0.0785,
"step": 5100
},
{
"epoch": 0.8190266183650968,
"grad_norm": 0.026611328125,
"learning_rate": 0.0013418848810375528,
"loss": 0.0792,
"step": 5200
},
{
"epoch": 0.8347771302567334,
"grad_norm": 0.0269775390625,
"learning_rate": 0.0013149683749403222,
"loss": 0.0789,
"step": 5300
},
{
"epoch": 0.8505276421483698,
"grad_norm": 0.0260009765625,
"learning_rate": 0.0012878463496910017,
"loss": 0.0782,
"step": 5400
},
{
"epoch": 0.8662781540400063,
"grad_norm": 0.0235595703125,
"learning_rate": 0.0012605447351298933,
"loss": 0.0781,
"step": 5500
},
{
"epoch": 0.8788785635533155,
"eval_peoplespeech-clean-transcription_loss": 1.5053552389144897,
"eval_peoplespeech-clean-transcription_model_preparation_time": 0.0064,
"eval_peoplespeech-clean-transcription_runtime": 15.0764,
"eval_peoplespeech-clean-transcription_samples_per_second": 4.245,
"eval_peoplespeech-clean-transcription_steps_per_second": 0.066,
"step": 5580
},
{
"epoch": 0.8820286659316428,
"grad_norm": 0.0260009765625,
"learning_rate": 0.0012330896327925223,
"loss": 0.0774,
"step": 5600
},
{
"epoch": 0.8977791778232792,
"grad_norm": 0.02392578125,
"learning_rate": 0.001205507290955429,
"loss": 0.0774,
"step": 5700
},
{
"epoch": 0.9135296897149158,
"grad_norm": 0.0252685546875,
"learning_rate": 0.0011778240795416713,
"loss": 0.077,
"step": 5800
},
{
"epoch": 0.9292802016065522,
"grad_norm": 0.02587890625,
"learning_rate": 0.001150066464910026,
"loss": 0.0769,
"step": 5900
},
{
"epoch": 0.9450307134981887,
"grad_norm": 0.0244140625,
"learning_rate": 0.0011222609845519956,
"loss": 0.0775,
"step": 6000
},
{
"epoch": 0.9607812253898251,
"grad_norm": 0.02685546875,
"learning_rate": 0.0010944342217208054,
"loss": 0.0775,
"step": 6100
},
{
"epoch": 0.966766419908647,
"eval_peoplespeech-clean-transcription_loss": 1.5009715557098389,
"eval_peoplespeech-clean-transcription_model_preparation_time": 0.0064,
"eval_peoplespeech-clean-transcription_runtime": 14.7344,
"eval_peoplespeech-clean-transcription_samples_per_second": 4.344,
"eval_peoplespeech-clean-transcription_steps_per_second": 0.068,
"step": 6138
},
{
"epoch": 0.9765317372814617,
"grad_norm": 0.025146484375,
"learning_rate": 0.0010666127800166554,
"loss": 0.0772,
"step": 6200
},
{
"epoch": 0.9922822491730982,
"grad_norm": 0.032470703125,
"learning_rate": 0.0010388232579525154,
"loss": 0.0765,
"step": 6300
},
{
"epoch": 1.0078752559458182,
"grad_norm": 0.0274658203125,
"learning_rate": 0.0010110922235247852,
"loss": 0.0736,
"step": 6400
},
{
"epoch": 1.0236257678374547,
"grad_norm": 0.025146484375,
"learning_rate": 0.000983446188813132,
"loss": 0.0725,
"step": 6500
},
{
"epoch": 1.0393762797290913,
"grad_norm": 0.0255126953125,
"learning_rate": 0.0009559115846337823,
"loss": 0.0758,
"step": 6600
},
{
"epoch": 1.0544967711450621,
"eval_peoplespeech-clean-transcription_loss": 1.4964332580566406,
"eval_peoplespeech-clean-transcription_model_preparation_time": 0.0064,
"eval_peoplespeech-clean-transcription_runtime": 14.7734,
"eval_peoplespeech-clean-transcription_samples_per_second": 4.332,
"eval_peoplespeech-clean-transcription_steps_per_second": 0.068,
"step": 6696
},
{
"epoch": 1.0551267916207276,
"grad_norm": 0.0257568359375,
"learning_rate": 0.0009285147352705085,
"loss": 0.0759,
"step": 6700
},
{
"epoch": 1.0708773035123642,
"grad_norm": 0.0269775390625,
"learning_rate": 0.0009012818333074646,
"loss": 0.0757,
"step": 6800
},
{
"epoch": 1.0866278154040006,
"grad_norm": 0.0274658203125,
"learning_rate": 0.0008742389145879321,
"loss": 0.0747,
"step": 6900
},
{
"epoch": 1.1023783272956371,
"grad_norm": 0.025146484375,
"learning_rate": 0.0008474118333229166,
"loss": 0.075,
"step": 7000
},
{
"epoch": 1.1181288391872737,
"grad_norm": 0.02490234375,
"learning_rate": 0.0008208262373733935,
"loss": 0.0753,
"step": 7100
},
{
"epoch": 1.13387935107891,
"grad_norm": 0.0238037109375,
"learning_rate": 0.0007945075437298341,
"loss": 0.0752,
"step": 7200
},
{
"epoch": 1.1423846275003937,
"eval_peoplespeech-clean-transcription_loss": 1.4869170188903809,
"eval_peoplespeech-clean-transcription_model_preparation_time": 0.0064,
"eval_peoplespeech-clean-transcription_runtime": 14.9106,
"eval_peoplespeech-clean-transcription_samples_per_second": 4.292,
"eval_peoplespeech-clean-transcription_steps_per_second": 0.067,
"step": 7254
},
{
"epoch": 1.1496298629705466,
"grad_norm": 0.023193359375,
"learning_rate": 0.0007684809142124512,
"loss": 0.0746,
"step": 7300
},
{
"epoch": 1.165380374862183,
"grad_norm": 0.0252685546875,
"learning_rate": 0.0007427712314154018,
"loss": 0.0742,
"step": 7400
},
{
"epoch": 1.1811308867538195,
"grad_norm": 0.024658203125,
"learning_rate": 0.000717403074917939,
"loss": 0.074,
"step": 7500
},
{
"epoch": 1.196881398645456,
"grad_norm": 0.027099609375,
"learning_rate": 0.0006924006977852644,
"loss": 0.0743,
"step": 7600
},
{
"epoch": 1.2126319105370924,
"grad_norm": 0.0244140625,
"learning_rate": 0.0006677880033815343,
"loss": 0.0735,
"step": 7700
},
{
"epoch": 1.228382422428729,
"grad_norm": 0.025390625,
"learning_rate": 0.0006435885225172043,
"loss": 0.0734,
"step": 7800
},
{
"epoch": 1.2302724838557253,
"eval_peoplespeech-clean-transcription_loss": 1.469083309173584,
"eval_peoplespeech-clean-transcription_model_preparation_time": 0.0064,
"eval_peoplespeech-clean-transcription_runtime": 14.7401,
"eval_peoplespeech-clean-transcription_samples_per_second": 4.342,
"eval_peoplespeech-clean-transcription_steps_per_second": 0.068,
"step": 7812
},
{
"epoch": 1.2441329343203655,
"grad_norm": 0.0279541015625,
"learning_rate": 0.0006198253909525442,
"loss": 0.0735,
"step": 7900
},
{
"epoch": 1.2598834462120019,
"grad_norm": 0.0245361328125,
"learning_rate": 0.0005965213272788422,
"loss": 0.0728,
"step": 8000
},
{
"epoch": 1.2756339581036384,
"grad_norm": 0.025146484375,
"learning_rate": 0.0005736986111984399,
"loss": 0.0726,
"step": 8100
},
{
"epoch": 1.2913844699952748,
"grad_norm": 0.0252685546875,
"learning_rate": 0.0005513790622243631,
"loss": 0.0731,
"step": 8200
},
{
"epoch": 1.3071349818869114,
"grad_norm": 0.02392578125,
"learning_rate": 0.0005295840188199162,
"loss": 0.0729,
"step": 8300
},
{
"epoch": 1.3181603402110569,
"eval_peoplespeech-clean-transcription_loss": 1.4812222719192505,
"eval_peoplespeech-clean-transcription_model_preparation_time": 0.0064,
"eval_peoplespeech-clean-transcription_runtime": 15.2287,
"eval_peoplespeech-clean-transcription_samples_per_second": 4.203,
"eval_peoplespeech-clean-transcription_steps_per_second": 0.066,
"step": 8370
},
{
"epoch": 1.3228854937785477,
"grad_norm": 0.031982421875,
"learning_rate": 0.0005083343179981776,
"loss": 0.0728,
"step": 8400
},
{
"epoch": 1.3386360056701843,
"grad_norm": 0.02587890625,
"learning_rate": 0.00048765027540090847,
"loss": 0.0723,
"step": 8500
},
{
"epoch": 1.3543865175618208,
"grad_norm": 0.0250244140625,
"learning_rate": 0.000467551665875909,
"loss": 0.0726,
"step": 8600
},
{
"epoch": 1.3701370294534572,
"grad_norm": 0.023193359375,
"learning_rate": 0.00044805770457140217,
"loss": 0.0722,
"step": 8700
},
{
"epoch": 1.3858875413450937,
"grad_norm": 0.0250244140625,
"learning_rate": 0.0004291870285655132,
"loss": 0.072,
"step": 8800
},
{
"epoch": 1.4016380532367303,
"grad_norm": 0.0244140625,
"learning_rate": 0.0004109576790484084,
"loss": 0.0711,
"step": 8900
},
{
"epoch": 1.4060481965663885,
"eval_peoplespeech-clean-transcription_loss": 1.4718445539474487,
"eval_peoplespeech-clean-transcription_model_preparation_time": 0.0064,
"eval_peoplespeech-clean-transcription_runtime": 15.2569,
"eval_peoplespeech-clean-transcription_samples_per_second": 4.195,
"eval_peoplespeech-clean-transcription_steps_per_second": 0.066,
"step": 8928
},
{
"epoch": 1.4173885651283666,
"grad_norm": 0.0230712890625,
"learning_rate": 0.0003933870840741306,
"loss": 0.0718,
"step": 9000
},
{
"epoch": 1.4331390770200032,
"grad_norm": 0.0225830078125,
"learning_rate": 0.00037649204189861796,
"loss": 0.0714,
"step": 9100
},
{
"epoch": 1.4488895889116395,
"grad_norm": 0.025390625,
"learning_rate": 0.00036028870491983735,
"loss": 0.0713,
"step": 9200
},
{
"epoch": 1.4646401008032761,
"grad_norm": 0.0223388671875,
"learning_rate": 0.00034479256423538846,
"loss": 0.071,
"step": 9300
},
{
"epoch": 1.4803906126949125,
"grad_norm": 0.02685546875,
"learning_rate": 0.00033001843483233573,
"loss": 0.0713,
"step": 9400
},
{
"epoch": 1.49393605292172,
"eval_peoplespeech-clean-transcription_loss": 1.4718666076660156,
"eval_peoplespeech-clean-transcription_model_preparation_time": 0.0064,
"eval_peoplespeech-clean-transcription_runtime": 15.2112,
"eval_peoplespeech-clean-transcription_samples_per_second": 4.207,
"eval_peoplespeech-clean-transcription_steps_per_second": 0.066,
"step": 9486
},
{
"epoch": 1.496141124586549,
"grad_norm": 0.0238037109375,
"learning_rate": 0.0003159804414234361,
"loss": 0.0711,
"step": 9500
},
{
"epoch": 1.5118916364781856,
"grad_norm": 0.0242919921875,
"learning_rate": 0.0003026920049432955,
"loss": 0.0713,
"step": 9600
},
{
"epoch": 1.527642148369822,
"grad_norm": 0.0238037109375,
"learning_rate": 0.0002901658297173699,
"loss": 0.0717,
"step": 9700
},
{
"epoch": 1.5433926602614585,
"grad_norm": 0.02392578125,
"learning_rate": 0.0002784138913160758,
"loss": 0.0712,
"step": 9800
},
{
"epoch": 1.559143172153095,
"grad_norm": 0.027099609375,
"learning_rate": 0.00026744742510562114,
"loss": 0.0705,
"step": 9900
},
{
"epoch": 1.5748936840447314,
"grad_norm": 0.0277099609375,
"learning_rate": 0.00025727691550650575,
"loss": 0.0708,
"step": 10000
},
{
"epoch": 1.5818239092770514,
"eval_peoplespeech-clean-transcription_loss": 1.4717382192611694,
"eval_peoplespeech-clean-transcription_model_preparation_time": 0.0064,
"eval_peoplespeech-clean-transcription_runtime": 15.5016,
"eval_peoplespeech-clean-transcription_samples_per_second": 4.129,
"eval_peoplespeech-clean-transcription_steps_per_second": 0.065,
"step": 10044
},
{
"epoch": 1.590644195936368,
"grad_norm": 0.028564453125,
"learning_rate": 0.0002479120859699568,
"loss": 0.0709,
"step": 10100
},
{
"epoch": 1.6063947078280045,
"grad_norm": 0.0250244140625,
"learning_rate": 0.00023936188968188535,
"loss": 0.0706,
"step": 10200
},
{
"epoch": 1.6221452197196409,
"grad_norm": 0.024169921875,
"learning_rate": 0.00023163450100324932,
"loss": 0.0709,
"step": 10300
},
{
"epoch": 1.6378957316112772,
"grad_norm": 0.027099609375,
"learning_rate": 0.00022473730765500778,
"loss": 0.07,
"step": 10400
},
{
"epoch": 1.653646243502914,
"grad_norm": 0.022705078125,
"learning_rate": 0.0002186769036551376,
"loss": 0.0697,
"step": 10500
},
{
"epoch": 1.6693967553945503,
"grad_norm": 0.0308837890625,
"learning_rate": 0.00021345908301446308,
"loss": 0.0696,
"step": 10600
},
{
"epoch": 1.6697117656323832,
"eval_peoplespeech-clean-transcription_loss": 1.4700562953948975,
"eval_peoplespeech-clean-transcription_model_preparation_time": 0.0064,
"eval_peoplespeech-clean-transcription_runtime": 14.9638,
"eval_peoplespeech-clean-transcription_samples_per_second": 4.277,
"eval_peoplespeech-clean-transcription_steps_per_second": 0.067,
"step": 10602
},
{
"epoch": 1.6851472672861867,
"grad_norm": 0.0257568359375,
"learning_rate": 0.00020908883419733002,
"loss": 0.0694,
"step": 10700
},
{
"epoch": 1.7008977791778233,
"grad_norm": 0.026123046875,
"learning_rate": 0.00020557033535241527,
"loss": 0.0703,
"step": 10800
},
{
"epoch": 1.7166482910694598,
"grad_norm": 0.025146484375,
"learning_rate": 0.00020290695031823435,
"loss": 0.0701,
"step": 10900
},
{
"epoch": 1.7323988029610962,
"grad_norm": 0.0244140625,
"learning_rate": 0.00020110122540716485,
"loss": 0.0704,
"step": 11000
},
{
"epoch": 1.7481493148527327,
"grad_norm": 0.02490234375,
"learning_rate": 0.00020015488697106083,
"loss": 0.0698,
"step": 11100
},
{
"epoch": 1.7575996219877146,
"eval_peoplespeech-clean-transcription_loss": 1.466253638267517,
"eval_peoplespeech-clean-transcription_model_preparation_time": 0.0064,
"eval_peoplespeech-clean-transcription_runtime": 15.3045,
"eval_peoplespeech-clean-transcription_samples_per_second": 4.182,
"eval_peoplespeech-clean-transcription_steps_per_second": 0.065,
"step": 11160
}
],
"logging_steps": 100,
"max_steps": 11160,
"num_input_tokens_seen": 0,
"num_train_epochs": 2,
"save_steps": 2790,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 1.94317275896291e+19,
"train_batch_size": 672,
"trial_name": null,
"trial_params": null
}