wav2vec2-large-xls-r-300m-welsh / trainer_state.json
infinitejoy's picture
End of training
079b1fd
raw
history blame
24.4 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 50.0,
"global_step": 18300,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.27,
"learning_rate": 2.2866666666666662e-06,
"loss": 16.3914,
"step": 100
},
{
"epoch": 0.55,
"learning_rate": 4.596666666666666e-06,
"loss": 10.3351,
"step": 200
},
{
"epoch": 0.82,
"learning_rate": 6.93e-06,
"loss": 5.7909,
"step": 300
},
{
"epoch": 1.09,
"learning_rate": 9.263333333333332e-06,
"loss": 4.7705,
"step": 400
},
{
"epoch": 1.37,
"learning_rate": 1.1596666666666665e-05,
"loss": 4.1586,
"step": 500
},
{
"epoch": 1.64,
"learning_rate": 1.3929999999999999e-05,
"loss": 3.7094,
"step": 600
},
{
"epoch": 1.91,
"learning_rate": 1.626333333333333e-05,
"loss": 3.3499,
"step": 700
},
{
"epoch": 2.19,
"learning_rate": 1.8596666666666664e-05,
"loss": 3.1161,
"step": 800
},
{
"epoch": 2.46,
"learning_rate": 2.0929999999999998e-05,
"loss": 2.9926,
"step": 900
},
{
"epoch": 2.73,
"learning_rate": 2.326333333333333e-05,
"loss": 2.9385,
"step": 1000
},
{
"epoch": 3.01,
"learning_rate": 2.559666666666666e-05,
"loss": 2.889,
"step": 1100
},
{
"epoch": 3.28,
"learning_rate": 2.793e-05,
"loss": 2.8583,
"step": 1200
},
{
"epoch": 3.55,
"learning_rate": 3.0263333333333332e-05,
"loss": 2.8445,
"step": 1300
},
{
"epoch": 3.83,
"learning_rate": 3.2596666666666666e-05,
"loss": 2.8281,
"step": 1400
},
{
"epoch": 4.1,
"learning_rate": 3.493e-05,
"loss": 2.8191,
"step": 1500
},
{
"epoch": 4.37,
"learning_rate": 3.7263333333333326e-05,
"loss": 2.7995,
"step": 1600
},
{
"epoch": 4.64,
"learning_rate": 3.959666666666666e-05,
"loss": 2.6081,
"step": 1700
},
{
"epoch": 4.92,
"learning_rate": 4.192999999999999e-05,
"loss": 2.0775,
"step": 1800
},
{
"epoch": 5.19,
"learning_rate": 4.426333333333333e-05,
"loss": 1.763,
"step": 1900
},
{
"epoch": 5.46,
"learning_rate": 4.659666666666666e-05,
"loss": 1.6346,
"step": 2000
},
{
"epoch": 5.74,
"learning_rate": 4.8929999999999994e-05,
"loss": 1.5687,
"step": 2100
},
{
"epoch": 6.01,
"learning_rate": 5.126333333333333e-05,
"loss": 1.5288,
"step": 2200
},
{
"epoch": 6.28,
"learning_rate": 5.359666666666667e-05,
"loss": 1.4756,
"step": 2300
},
{
"epoch": 6.56,
"learning_rate": 5.593e-05,
"loss": 1.4463,
"step": 2400
},
{
"epoch": 6.83,
"learning_rate": 5.826333333333333e-05,
"loss": 1.4291,
"step": 2500
},
{
"epoch": 7.1,
"learning_rate": 6.059666666666666e-05,
"loss": 1.4114,
"step": 2600
},
{
"epoch": 7.38,
"learning_rate": 6.293e-05,
"loss": 1.3955,
"step": 2700
},
{
"epoch": 7.65,
"learning_rate": 6.526333333333332e-05,
"loss": 1.3691,
"step": 2800
},
{
"epoch": 7.92,
"learning_rate": 6.759666666666666e-05,
"loss": 1.3556,
"step": 2900
},
{
"epoch": 8.2,
"learning_rate": 6.992999999999999e-05,
"loss": 1.3454,
"step": 3000
},
{
"epoch": 8.2,
"eval_loss": 0.49262478947639465,
"eval_runtime": 209.1038,
"eval_samples_per_second": 23.194,
"eval_steps_per_second": 1.454,
"eval_wer": 0.5703299486541871,
"step": 3000
},
{
"epoch": 8.47,
"learning_rate": 6.955620915032679e-05,
"loss": 1.326,
"step": 3100
},
{
"epoch": 8.74,
"learning_rate": 6.90986928104575e-05,
"loss": 1.318,
"step": 3200
},
{
"epoch": 9.02,
"learning_rate": 6.864117647058823e-05,
"loss": 1.3012,
"step": 3300
},
{
"epoch": 9.29,
"learning_rate": 6.818366013071895e-05,
"loss": 1.2888,
"step": 3400
},
{
"epoch": 9.56,
"learning_rate": 6.772614379084966e-05,
"loss": 1.2874,
"step": 3500
},
{
"epoch": 9.84,
"learning_rate": 6.726862745098038e-05,
"loss": 1.2699,
"step": 3600
},
{
"epoch": 10.11,
"learning_rate": 6.681111111111111e-05,
"loss": 1.2558,
"step": 3700
},
{
"epoch": 10.38,
"learning_rate": 6.635359477124183e-05,
"loss": 1.2512,
"step": 3800
},
{
"epoch": 10.66,
"learning_rate": 6.589607843137254e-05,
"loss": 1.2502,
"step": 3900
},
{
"epoch": 10.93,
"learning_rate": 6.543856209150327e-05,
"loss": 1.246,
"step": 4000
},
{
"epoch": 11.2,
"learning_rate": 6.498104575163397e-05,
"loss": 1.2205,
"step": 4100
},
{
"epoch": 11.48,
"learning_rate": 6.45235294117647e-05,
"loss": 1.2192,
"step": 4200
},
{
"epoch": 11.75,
"learning_rate": 6.406601307189542e-05,
"loss": 1.2146,
"step": 4300
},
{
"epoch": 12.02,
"learning_rate": 6.360849673202614e-05,
"loss": 1.2143,
"step": 4400
},
{
"epoch": 12.3,
"learning_rate": 6.315098039215685e-05,
"loss": 1.205,
"step": 4500
},
{
"epoch": 12.57,
"learning_rate": 6.269346405228757e-05,
"loss": 1.1777,
"step": 4600
},
{
"epoch": 12.84,
"learning_rate": 6.22359477124183e-05,
"loss": 1.1901,
"step": 4700
},
{
"epoch": 13.11,
"learning_rate": 6.177843137254901e-05,
"loss": 1.1729,
"step": 4800
},
{
"epoch": 13.39,
"learning_rate": 6.132091503267974e-05,
"loss": 1.1813,
"step": 4900
},
{
"epoch": 13.66,
"learning_rate": 6.086339869281045e-05,
"loss": 1.1773,
"step": 5000
},
{
"epoch": 13.93,
"learning_rate": 6.040588235294117e-05,
"loss": 1.1517,
"step": 5100
},
{
"epoch": 14.21,
"learning_rate": 5.994836601307189e-05,
"loss": 1.1568,
"step": 5200
},
{
"epoch": 14.48,
"learning_rate": 5.9490849673202613e-05,
"loss": 1.1591,
"step": 5300
},
{
"epoch": 14.75,
"learning_rate": 5.903333333333333e-05,
"loss": 1.146,
"step": 5400
},
{
"epoch": 15.03,
"learning_rate": 5.8575816993464045e-05,
"loss": 1.1561,
"step": 5500
},
{
"epoch": 15.3,
"learning_rate": 5.811830065359477e-05,
"loss": 1.1304,
"step": 5600
},
{
"epoch": 15.57,
"learning_rate": 5.7660784313725484e-05,
"loss": 1.1355,
"step": 5700
},
{
"epoch": 15.85,
"learning_rate": 5.720326797385621e-05,
"loss": 1.137,
"step": 5800
},
{
"epoch": 16.12,
"learning_rate": 5.674575163398693e-05,
"loss": 1.1401,
"step": 5900
},
{
"epoch": 16.39,
"learning_rate": 5.628823529411764e-05,
"loss": 1.1202,
"step": 6000
},
{
"epoch": 16.39,
"eval_loss": 0.35287711024284363,
"eval_runtime": 207.6346,
"eval_samples_per_second": 23.358,
"eval_steps_per_second": 1.464,
"eval_wer": 0.3944187331629481,
"step": 6000
},
{
"epoch": 16.67,
"learning_rate": 5.583071895424836e-05,
"loss": 1.1132,
"step": 6100
},
{
"epoch": 16.94,
"learning_rate": 5.537777777777777e-05,
"loss": 1.1199,
"step": 6200
},
{
"epoch": 17.21,
"learning_rate": 5.4920261437908495e-05,
"loss": 1.1089,
"step": 6300
},
{
"epoch": 17.49,
"learning_rate": 5.446274509803921e-05,
"loss": 1.1038,
"step": 6400
},
{
"epoch": 17.76,
"learning_rate": 5.400522875816993e-05,
"loss": 1.1022,
"step": 6500
},
{
"epoch": 18.03,
"learning_rate": 5.354771241830065e-05,
"loss": 1.1059,
"step": 6600
},
{
"epoch": 18.31,
"learning_rate": 5.3090196078431366e-05,
"loss": 1.097,
"step": 6700
},
{
"epoch": 18.58,
"learning_rate": 5.263267973856209e-05,
"loss": 1.0793,
"step": 6800
},
{
"epoch": 18.85,
"learning_rate": 5.217516339869281e-05,
"loss": 1.0814,
"step": 6900
},
{
"epoch": 19.13,
"learning_rate": 5.171764705882352e-05,
"loss": 1.087,
"step": 7000
},
{
"epoch": 19.4,
"learning_rate": 5.126013071895424e-05,
"loss": 1.0784,
"step": 7100
},
{
"epoch": 19.67,
"learning_rate": 5.0802614379084966e-05,
"loss": 1.0618,
"step": 7200
},
{
"epoch": 19.95,
"learning_rate": 5.034509803921568e-05,
"loss": 1.069,
"step": 7300
},
{
"epoch": 20.22,
"learning_rate": 4.9887581699346405e-05,
"loss": 1.0555,
"step": 7400
},
{
"epoch": 20.49,
"learning_rate": 4.9430065359477114e-05,
"loss": 1.0559,
"step": 7500
},
{
"epoch": 20.77,
"learning_rate": 4.897254901960784e-05,
"loss": 1.0545,
"step": 7600
},
{
"epoch": 21.04,
"learning_rate": 4.851503267973856e-05,
"loss": 1.0644,
"step": 7700
},
{
"epoch": 21.31,
"learning_rate": 4.8057516339869276e-05,
"loss": 1.0541,
"step": 7800
},
{
"epoch": 21.58,
"learning_rate": 4.76e-05,
"loss": 1.0467,
"step": 7900
},
{
"epoch": 21.86,
"learning_rate": 4.7142483660130714e-05,
"loss": 1.0309,
"step": 8000
},
{
"epoch": 22.13,
"learning_rate": 4.668496732026143e-05,
"loss": 1.0453,
"step": 8100
},
{
"epoch": 22.4,
"learning_rate": 4.622745098039215e-05,
"loss": 1.0321,
"step": 8200
},
{
"epoch": 22.68,
"learning_rate": 4.5769934640522876e-05,
"loss": 1.0439,
"step": 8300
},
{
"epoch": 22.95,
"learning_rate": 4.5316993464052286e-05,
"loss": 1.0403,
"step": 8400
},
{
"epoch": 23.22,
"learning_rate": 4.485947712418301e-05,
"loss": 1.0145,
"step": 8500
},
{
"epoch": 23.5,
"learning_rate": 4.440196078431372e-05,
"loss": 1.0228,
"step": 8600
},
{
"epoch": 23.77,
"learning_rate": 4.394444444444444e-05,
"loss": 1.0236,
"step": 8700
},
{
"epoch": 24.04,
"learning_rate": 4.348692810457516e-05,
"loss": 1.0185,
"step": 8800
},
{
"epoch": 24.32,
"learning_rate": 4.302941176470588e-05,
"loss": 1.0165,
"step": 8900
},
{
"epoch": 24.59,
"learning_rate": 4.2571895424836596e-05,
"loss": 1.0058,
"step": 9000
},
{
"epoch": 24.59,
"eval_loss": 0.31432652473449707,
"eval_runtime": 204.5409,
"eval_samples_per_second": 23.712,
"eval_steps_per_second": 1.486,
"eval_wer": 0.3340931592641201,
"step": 9000
},
{
"epoch": 24.86,
"learning_rate": 4.211437908496731e-05,
"loss": 1.0177,
"step": 9100
},
{
"epoch": 25.14,
"learning_rate": 4.1656862745098035e-05,
"loss": 1.0175,
"step": 9200
},
{
"epoch": 25.41,
"learning_rate": 4.119934640522876e-05,
"loss": 1.0046,
"step": 9300
},
{
"epoch": 25.68,
"learning_rate": 4.0741830065359473e-05,
"loss": 0.9922,
"step": 9400
},
{
"epoch": 25.96,
"learning_rate": 4.028431372549019e-05,
"loss": 1.0011,
"step": 9500
},
{
"epoch": 26.23,
"learning_rate": 3.982679738562091e-05,
"loss": 0.9852,
"step": 9600
},
{
"epoch": 26.5,
"learning_rate": 3.936928104575163e-05,
"loss": 0.9901,
"step": 9700
},
{
"epoch": 26.78,
"learning_rate": 3.891176470588235e-05,
"loss": 0.9848,
"step": 9800
},
{
"epoch": 27.05,
"learning_rate": 3.8454248366013074e-05,
"loss": 1.0035,
"step": 9900
},
{
"epoch": 27.32,
"learning_rate": 3.799673202614378e-05,
"loss": 0.9707,
"step": 10000
},
{
"epoch": 27.6,
"learning_rate": 3.7539215686274506e-05,
"loss": 0.9894,
"step": 10100
},
{
"epoch": 27.87,
"learning_rate": 3.708169934640523e-05,
"loss": 0.9873,
"step": 10200
},
{
"epoch": 28.14,
"learning_rate": 3.6624183006535944e-05,
"loss": 0.9902,
"step": 10300
},
{
"epoch": 28.42,
"learning_rate": 3.616666666666667e-05,
"loss": 0.97,
"step": 10400
},
{
"epoch": 28.69,
"learning_rate": 3.570915032679738e-05,
"loss": 0.9613,
"step": 10500
},
{
"epoch": 28.96,
"learning_rate": 3.52516339869281e-05,
"loss": 0.9734,
"step": 10600
},
{
"epoch": 29.23,
"learning_rate": 3.479411764705882e-05,
"loss": 0.9514,
"step": 10700
},
{
"epoch": 29.51,
"learning_rate": 3.434117647058823e-05,
"loss": 0.9566,
"step": 10800
},
{
"epoch": 29.78,
"learning_rate": 3.388366013071895e-05,
"loss": 0.9632,
"step": 10900
},
{
"epoch": 30.05,
"learning_rate": 3.343071895424836e-05,
"loss": 0.9635,
"step": 11000
},
{
"epoch": 30.33,
"learning_rate": 3.297320261437908e-05,
"loss": 0.9581,
"step": 11100
},
{
"epoch": 30.6,
"learning_rate": 3.2515686274509805e-05,
"loss": 0.9545,
"step": 11200
},
{
"epoch": 30.87,
"learning_rate": 3.205816993464052e-05,
"loss": 0.9501,
"step": 11300
},
{
"epoch": 31.15,
"learning_rate": 3.1600653594771244e-05,
"loss": 0.9506,
"step": 11400
},
{
"epoch": 31.42,
"learning_rate": 3.114313725490196e-05,
"loss": 0.9414,
"step": 11500
},
{
"epoch": 31.69,
"learning_rate": 3.0685620915032675e-05,
"loss": 0.9484,
"step": 11600
},
{
"epoch": 31.97,
"learning_rate": 3.0228104575163395e-05,
"loss": 0.9474,
"step": 11700
},
{
"epoch": 32.24,
"learning_rate": 2.9770588235294114e-05,
"loss": 0.9237,
"step": 11800
},
{
"epoch": 32.51,
"learning_rate": 2.9313071895424837e-05,
"loss": 0.9376,
"step": 11900
},
{
"epoch": 32.79,
"learning_rate": 2.8855555555555553e-05,
"loss": 0.9287,
"step": 12000
},
{
"epoch": 32.79,
"eval_loss": 0.28960883617401123,
"eval_runtime": 208.0073,
"eval_samples_per_second": 23.316,
"eval_steps_per_second": 1.461,
"eval_wer": 0.29796689000944027,
"step": 12000
},
{
"epoch": 33.06,
"learning_rate": 2.8398039215686272e-05,
"loss": 0.9325,
"step": 12100
},
{
"epoch": 33.33,
"learning_rate": 2.794052287581699e-05,
"loss": 0.9349,
"step": 12200
},
{
"epoch": 33.61,
"learning_rate": 2.748300653594771e-05,
"loss": 0.9342,
"step": 12300
},
{
"epoch": 33.88,
"learning_rate": 2.702549019607843e-05,
"loss": 0.9282,
"step": 12400
},
{
"epoch": 34.15,
"learning_rate": 2.6567973856209146e-05,
"loss": 0.9338,
"step": 12500
},
{
"epoch": 34.43,
"learning_rate": 2.611045751633987e-05,
"loss": 0.9195,
"step": 12600
},
{
"epoch": 34.7,
"learning_rate": 2.5652941176470585e-05,
"loss": 0.9146,
"step": 12700
},
{
"epoch": 34.97,
"learning_rate": 2.5195424836601305e-05,
"loss": 0.9249,
"step": 12800
},
{
"epoch": 35.25,
"learning_rate": 2.4737908496732024e-05,
"loss": 0.9137,
"step": 12900
},
{
"epoch": 35.52,
"learning_rate": 2.4280392156862743e-05,
"loss": 0.9147,
"step": 13000
},
{
"epoch": 35.79,
"learning_rate": 2.3822875816993463e-05,
"loss": 0.9171,
"step": 13100
},
{
"epoch": 36.07,
"learning_rate": 2.336535947712418e-05,
"loss": 0.9178,
"step": 13200
},
{
"epoch": 36.34,
"learning_rate": 2.29078431372549e-05,
"loss": 0.9164,
"step": 13300
},
{
"epoch": 36.61,
"learning_rate": 2.2450326797385617e-05,
"loss": 0.909,
"step": 13400
},
{
"epoch": 36.89,
"learning_rate": 2.1997385620915028e-05,
"loss": 0.9106,
"step": 13500
},
{
"epoch": 37.16,
"learning_rate": 2.153986928104575e-05,
"loss": 0.9059,
"step": 13600
},
{
"epoch": 37.43,
"learning_rate": 2.1082352941176467e-05,
"loss": 0.899,
"step": 13700
},
{
"epoch": 37.7,
"learning_rate": 2.0629411764705878e-05,
"loss": 0.9044,
"step": 13800
},
{
"epoch": 37.98,
"learning_rate": 2.01718954248366e-05,
"loss": 0.9037,
"step": 13900
},
{
"epoch": 38.25,
"learning_rate": 1.971437908496732e-05,
"loss": 0.9013,
"step": 14000
},
{
"epoch": 38.52,
"learning_rate": 1.9256862745098036e-05,
"loss": 0.8963,
"step": 14100
},
{
"epoch": 38.8,
"learning_rate": 1.879934640522876e-05,
"loss": 0.8998,
"step": 14200
},
{
"epoch": 39.07,
"learning_rate": 1.8341830065359474e-05,
"loss": 0.8995,
"step": 14300
},
{
"epoch": 39.34,
"learning_rate": 1.7884313725490194e-05,
"loss": 0.9125,
"step": 14400
},
{
"epoch": 39.62,
"learning_rate": 1.7426797385620913e-05,
"loss": 0.8982,
"step": 14500
},
{
"epoch": 39.89,
"learning_rate": 1.6969281045751633e-05,
"loss": 0.8959,
"step": 14600
},
{
"epoch": 40.16,
"learning_rate": 1.6511764705882352e-05,
"loss": 0.8826,
"step": 14700
},
{
"epoch": 40.44,
"learning_rate": 1.605424836601307e-05,
"loss": 0.8871,
"step": 14800
},
{
"epoch": 40.71,
"learning_rate": 1.559673202614379e-05,
"loss": 0.8815,
"step": 14900
},
{
"epoch": 40.98,
"learning_rate": 1.5139215686274508e-05,
"loss": 0.8849,
"step": 15000
},
{
"epoch": 40.98,
"eval_loss": 0.2726878821849823,
"eval_runtime": 205.4369,
"eval_samples_per_second": 23.608,
"eval_steps_per_second": 1.48,
"eval_wer": 0.27975409269876356,
"step": 15000
},
{
"epoch": 41.26,
"learning_rate": 1.4681699346405228e-05,
"loss": 0.8804,
"step": 15100
},
{
"epoch": 41.53,
"learning_rate": 1.4224183006535945e-05,
"loss": 0.8861,
"step": 15200
},
{
"epoch": 41.8,
"learning_rate": 1.3766666666666665e-05,
"loss": 0.8913,
"step": 15300
},
{
"epoch": 42.08,
"learning_rate": 1.3309150326797386e-05,
"loss": 0.8769,
"step": 15400
},
{
"epoch": 42.35,
"learning_rate": 1.2851633986928103e-05,
"loss": 0.885,
"step": 15500
},
{
"epoch": 42.62,
"learning_rate": 1.2394117647058823e-05,
"loss": 0.8788,
"step": 15600
},
{
"epoch": 42.9,
"learning_rate": 1.1936601307189542e-05,
"loss": 0.8752,
"step": 15700
},
{
"epoch": 43.17,
"learning_rate": 1.147908496732026e-05,
"loss": 0.869,
"step": 15800
},
{
"epoch": 43.44,
"learning_rate": 1.102156862745098e-05,
"loss": 0.8773,
"step": 15900
},
{
"epoch": 43.72,
"learning_rate": 1.0564052287581699e-05,
"loss": 0.8757,
"step": 16000
},
{
"epoch": 43.99,
"learning_rate": 1.0106535947712418e-05,
"loss": 0.8693,
"step": 16100
},
{
"epoch": 44.26,
"learning_rate": 9.649019607843137e-06,
"loss": 0.863,
"step": 16200
},
{
"epoch": 44.54,
"learning_rate": 9.191503267973855e-06,
"loss": 0.8634,
"step": 16300
},
{
"epoch": 44.81,
"learning_rate": 8.733986928104574e-06,
"loss": 0.8549,
"step": 16400
},
{
"epoch": 45.08,
"learning_rate": 8.276470588235294e-06,
"loss": 0.8528,
"step": 16500
},
{
"epoch": 45.36,
"learning_rate": 7.818954248366013e-06,
"loss": 0.861,
"step": 16600
},
{
"epoch": 45.63,
"learning_rate": 7.361437908496731e-06,
"loss": 0.8571,
"step": 16700
},
{
"epoch": 45.9,
"learning_rate": 6.903921568627451e-06,
"loss": 0.8595,
"step": 16800
},
{
"epoch": 46.17,
"learning_rate": 6.44640522875817e-06,
"loss": 0.8605,
"step": 16900
},
{
"epoch": 46.45,
"learning_rate": 5.988888888888888e-06,
"loss": 0.8635,
"step": 17000
},
{
"epoch": 46.72,
"learning_rate": 5.5313725490196075e-06,
"loss": 0.8594,
"step": 17100
},
{
"epoch": 46.99,
"learning_rate": 5.073856209150326e-06,
"loss": 0.8662,
"step": 17200
},
{
"epoch": 47.27,
"learning_rate": 4.616339869281045e-06,
"loss": 0.8569,
"step": 17300
},
{
"epoch": 47.54,
"learning_rate": 4.158823529411764e-06,
"loss": 0.8583,
"step": 17400
},
{
"epoch": 47.81,
"learning_rate": 3.7013071895424833e-06,
"loss": 0.8585,
"step": 17500
},
{
"epoch": 48.09,
"learning_rate": 3.2437908496732026e-06,
"loss": 0.8439,
"step": 17600
},
{
"epoch": 48.36,
"learning_rate": 2.786274509803921e-06,
"loss": 0.8487,
"step": 17700
},
{
"epoch": 48.63,
"learning_rate": 2.3287581699346405e-06,
"loss": 0.8459,
"step": 17800
},
{
"epoch": 48.91,
"learning_rate": 1.8712418300653593e-06,
"loss": 0.8576,
"step": 17900
},
{
"epoch": 49.18,
"learning_rate": 1.4137254901960782e-06,
"loss": 0.8665,
"step": 18000
},
{
"epoch": 49.18,
"eval_loss": 0.2661794126033783,
"eval_runtime": 208.8108,
"eval_samples_per_second": 23.227,
"eval_steps_per_second": 1.456,
"eval_wer": 0.269577030231862,
"step": 18000
},
{
"epoch": 49.45,
"learning_rate": 9.562091503267974e-07,
"loss": 0.8569,
"step": 18100
},
{
"epoch": 49.73,
"learning_rate": 4.986928104575163e-07,
"loss": 0.8505,
"step": 18200
},
{
"epoch": 50.0,
"learning_rate": 4.117647058823529e-08,
"loss": 0.863,
"step": 18300
},
{
"epoch": 50.0,
"step": 18300,
"total_flos": 9.169878518581522e+19,
"train_loss": 1.3658743798928183,
"train_runtime": 33116.2978,
"train_samples_per_second": 17.657,
"train_steps_per_second": 0.553
}
],
"max_steps": 18300,
"num_train_epochs": 50,
"total_flos": 9.169878518581522e+19,
"trial_name": null,
"trial_params": null
}