whisper-tiny-en / checkpoint-4000 /trainer_state.json
lfurman's picture
Upload folder using huggingface_hub
02625b8 verified
raw
history blame
29.1 kB
{
"best_metric": 93.11239193083574,
"best_model_checkpoint": "./whisper-tiny-en/checkpoint-3000",
"epoch": 97.5609756097561,
"eval_steps": 1000,
"global_step": 4000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.6097560975609756,
"grad_norm": 76.21949768066406,
"learning_rate": 4.2000000000000006e-07,
"loss": 7.5702,
"step": 25
},
{
"epoch": 1.2195121951219512,
"grad_norm": 72.27240753173828,
"learning_rate": 9.200000000000001e-07,
"loss": 7.1361,
"step": 50
},
{
"epoch": 1.8292682926829267,
"grad_norm": 27.645973205566406,
"learning_rate": 1.42e-06,
"loss": 6.4445,
"step": 75
},
{
"epoch": 2.4390243902439024,
"grad_norm": 21.712514877319336,
"learning_rate": 1.9200000000000003e-06,
"loss": 5.8148,
"step": 100
},
{
"epoch": 3.048780487804878,
"grad_norm": 19.066011428833008,
"learning_rate": 2.42e-06,
"loss": 5.1377,
"step": 125
},
{
"epoch": 3.658536585365854,
"grad_norm": 14.726506233215332,
"learning_rate": 2.92e-06,
"loss": 4.6,
"step": 150
},
{
"epoch": 4.2682926829268295,
"grad_norm": 13.597981452941895,
"learning_rate": 3.4200000000000007e-06,
"loss": 4.3595,
"step": 175
},
{
"epoch": 4.878048780487805,
"grad_norm": 16.399568557739258,
"learning_rate": 3.920000000000001e-06,
"loss": 4.171,
"step": 200
},
{
"epoch": 5.487804878048781,
"grad_norm": 15.264267921447754,
"learning_rate": 4.42e-06,
"loss": 3.9727,
"step": 225
},
{
"epoch": 6.097560975609756,
"grad_norm": 14.83893871307373,
"learning_rate": 4.92e-06,
"loss": 3.816,
"step": 250
},
{
"epoch": 6.7073170731707314,
"grad_norm": 14.460186004638672,
"learning_rate": 5.420000000000001e-06,
"loss": 3.6683,
"step": 275
},
{
"epoch": 7.317073170731708,
"grad_norm": 13.663064002990723,
"learning_rate": 5.92e-06,
"loss": 3.5392,
"step": 300
},
{
"epoch": 7.926829268292683,
"grad_norm": 14.359073638916016,
"learning_rate": 6.42e-06,
"loss": 3.4807,
"step": 325
},
{
"epoch": 8.536585365853659,
"grad_norm": 12.90373706817627,
"learning_rate": 6.92e-06,
"loss": 3.335,
"step": 350
},
{
"epoch": 9.146341463414634,
"grad_norm": 15.809270858764648,
"learning_rate": 7.420000000000001e-06,
"loss": 3.2196,
"step": 375
},
{
"epoch": 9.75609756097561,
"grad_norm": 14.483990669250488,
"learning_rate": 7.92e-06,
"loss": 3.1224,
"step": 400
},
{
"epoch": 10.365853658536585,
"grad_norm": 15.841504096984863,
"learning_rate": 8.42e-06,
"loss": 2.9858,
"step": 425
},
{
"epoch": 10.975609756097562,
"grad_norm": 16.420425415039062,
"learning_rate": 8.920000000000001e-06,
"loss": 2.9395,
"step": 450
},
{
"epoch": 11.585365853658537,
"grad_norm": 14.952699661254883,
"learning_rate": 9.42e-06,
"loss": 2.7501,
"step": 475
},
{
"epoch": 12.195121951219512,
"grad_norm": 15.765605926513672,
"learning_rate": 9.920000000000002e-06,
"loss": 2.7014,
"step": 500
},
{
"epoch": 12.804878048780488,
"grad_norm": 17.627931594848633,
"learning_rate": 9.967692307692308e-06,
"loss": 2.601,
"step": 525
},
{
"epoch": 13.414634146341463,
"grad_norm": 16.36449432373047,
"learning_rate": 9.92923076923077e-06,
"loss": 2.4699,
"step": 550
},
{
"epoch": 14.024390243902438,
"grad_norm": 15.8140869140625,
"learning_rate": 9.890769230769231e-06,
"loss": 2.3859,
"step": 575
},
{
"epoch": 14.634146341463415,
"grad_norm": 15.983495712280273,
"learning_rate": 9.852307692307693e-06,
"loss": 2.2235,
"step": 600
},
{
"epoch": 15.24390243902439,
"grad_norm": 15.25550651550293,
"learning_rate": 9.813846153846155e-06,
"loss": 2.1583,
"step": 625
},
{
"epoch": 15.853658536585366,
"grad_norm": 18.167879104614258,
"learning_rate": 9.775384615384616e-06,
"loss": 2.0671,
"step": 650
},
{
"epoch": 16.463414634146343,
"grad_norm": 16.86029624938965,
"learning_rate": 9.736923076923078e-06,
"loss": 1.9547,
"step": 675
},
{
"epoch": 17.073170731707318,
"grad_norm": 16.195270538330078,
"learning_rate": 9.698461538461539e-06,
"loss": 1.8877,
"step": 700
},
{
"epoch": 17.682926829268293,
"grad_norm": 17.331113815307617,
"learning_rate": 9.66e-06,
"loss": 1.7336,
"step": 725
},
{
"epoch": 18.29268292682927,
"grad_norm": 18.00438690185547,
"learning_rate": 9.621538461538463e-06,
"loss": 1.6845,
"step": 750
},
{
"epoch": 18.902439024390244,
"grad_norm": 17.1490478515625,
"learning_rate": 9.583076923076923e-06,
"loss": 1.6027,
"step": 775
},
{
"epoch": 19.51219512195122,
"grad_norm": 17.924156188964844,
"learning_rate": 9.544615384615385e-06,
"loss": 1.4647,
"step": 800
},
{
"epoch": 20.121951219512194,
"grad_norm": 17.920028686523438,
"learning_rate": 9.506153846153848e-06,
"loss": 1.4291,
"step": 825
},
{
"epoch": 20.73170731707317,
"grad_norm": 17.502954483032227,
"learning_rate": 9.467692307692308e-06,
"loss": 1.3113,
"step": 850
},
{
"epoch": 21.341463414634145,
"grad_norm": 18.125986099243164,
"learning_rate": 9.42923076923077e-06,
"loss": 1.232,
"step": 875
},
{
"epoch": 21.951219512195124,
"grad_norm": 16.026338577270508,
"learning_rate": 9.39076923076923e-06,
"loss": 1.1903,
"step": 900
},
{
"epoch": 22.5609756097561,
"grad_norm": 16.795467376708984,
"learning_rate": 9.352307692307693e-06,
"loss": 1.0742,
"step": 925
},
{
"epoch": 23.170731707317074,
"grad_norm": 19.86144256591797,
"learning_rate": 9.313846153846155e-06,
"loss": 1.0346,
"step": 950
},
{
"epoch": 23.78048780487805,
"grad_norm": 17.94236183166504,
"learning_rate": 9.275384615384616e-06,
"loss": 0.9589,
"step": 975
},
{
"epoch": 24.390243902439025,
"grad_norm": 17.72712516784668,
"learning_rate": 9.236923076923078e-06,
"loss": 0.8757,
"step": 1000
},
{
"epoch": 24.390243902439025,
"eval_loss": 4.123460292816162,
"eval_runtime": 64.6053,
"eval_samples_per_second": 2.492,
"eval_steps_per_second": 0.325,
"eval_wer": 97.89625360230548,
"step": 1000
},
{
"epoch": 25.0,
"grad_norm": 31.632688522338867,
"learning_rate": 9.19846153846154e-06,
"loss": 0.8421,
"step": 1025
},
{
"epoch": 25.609756097560975,
"grad_norm": 18.91576385498047,
"learning_rate": 9.16e-06,
"loss": 0.7294,
"step": 1050
},
{
"epoch": 26.21951219512195,
"grad_norm": 20.973003387451172,
"learning_rate": 9.121538461538463e-06,
"loss": 0.7139,
"step": 1075
},
{
"epoch": 26.829268292682926,
"grad_norm": 18.86784553527832,
"learning_rate": 9.083076923076923e-06,
"loss": 0.6638,
"step": 1100
},
{
"epoch": 27.4390243902439,
"grad_norm": 14.628901481628418,
"learning_rate": 9.044615384615385e-06,
"loss": 0.605,
"step": 1125
},
{
"epoch": 28.048780487804876,
"grad_norm": 13.873974800109863,
"learning_rate": 9.006153846153847e-06,
"loss": 0.5613,
"step": 1150
},
{
"epoch": 28.658536585365855,
"grad_norm": 15.853562355041504,
"learning_rate": 8.967692307692308e-06,
"loss": 0.4886,
"step": 1175
},
{
"epoch": 29.26829268292683,
"grad_norm": 12.60745906829834,
"learning_rate": 8.92923076923077e-06,
"loss": 0.4652,
"step": 1200
},
{
"epoch": 29.878048780487806,
"grad_norm": 17.16914176940918,
"learning_rate": 8.890769230769232e-06,
"loss": 0.4324,
"step": 1225
},
{
"epoch": 30.48780487804878,
"grad_norm": 11.67048168182373,
"learning_rate": 8.852307692307693e-06,
"loss": 0.3793,
"step": 1250
},
{
"epoch": 31.097560975609756,
"grad_norm": 10.136366844177246,
"learning_rate": 8.813846153846155e-06,
"loss": 0.3689,
"step": 1275
},
{
"epoch": 31.70731707317073,
"grad_norm": 15.549346923828125,
"learning_rate": 8.775384615384615e-06,
"loss": 0.3165,
"step": 1300
},
{
"epoch": 32.31707317073171,
"grad_norm": 10.75722599029541,
"learning_rate": 8.736923076923077e-06,
"loss": 0.2891,
"step": 1325
},
{
"epoch": 32.926829268292686,
"grad_norm": 10.687921524047852,
"learning_rate": 8.69846153846154e-06,
"loss": 0.2786,
"step": 1350
},
{
"epoch": 33.53658536585366,
"grad_norm": 15.220375061035156,
"learning_rate": 8.66e-06,
"loss": 0.2349,
"step": 1375
},
{
"epoch": 34.146341463414636,
"grad_norm": 9.226957321166992,
"learning_rate": 8.621538461538462e-06,
"loss": 0.2279,
"step": 1400
},
{
"epoch": 34.75609756097561,
"grad_norm": 10.428820610046387,
"learning_rate": 8.583076923076924e-06,
"loss": 0.2022,
"step": 1425
},
{
"epoch": 35.36585365853659,
"grad_norm": 10.70584487915039,
"learning_rate": 8.544615384615385e-06,
"loss": 0.186,
"step": 1450
},
{
"epoch": 35.97560975609756,
"grad_norm": 11.436528205871582,
"learning_rate": 8.506153846153847e-06,
"loss": 0.1763,
"step": 1475
},
{
"epoch": 36.58536585365854,
"grad_norm": 8.337956428527832,
"learning_rate": 8.467692307692308e-06,
"loss": 0.1482,
"step": 1500
},
{
"epoch": 37.19512195121951,
"grad_norm": 8.083637237548828,
"learning_rate": 8.42923076923077e-06,
"loss": 0.1515,
"step": 1525
},
{
"epoch": 37.80487804878049,
"grad_norm": 8.621879577636719,
"learning_rate": 8.390769230769232e-06,
"loss": 0.1345,
"step": 1550
},
{
"epoch": 38.41463414634146,
"grad_norm": 8.559672355651855,
"learning_rate": 8.352307692307692e-06,
"loss": 0.1204,
"step": 1575
},
{
"epoch": 39.02439024390244,
"grad_norm": 6.850173473358154,
"learning_rate": 8.313846153846155e-06,
"loss": 0.1208,
"step": 1600
},
{
"epoch": 39.63414634146341,
"grad_norm": 8.43338680267334,
"learning_rate": 8.275384615384617e-06,
"loss": 0.1007,
"step": 1625
},
{
"epoch": 40.24390243902439,
"grad_norm": 9.348348617553711,
"learning_rate": 8.236923076923077e-06,
"loss": 0.0995,
"step": 1650
},
{
"epoch": 40.853658536585364,
"grad_norm": 6.383693695068359,
"learning_rate": 8.19846153846154e-06,
"loss": 0.0956,
"step": 1675
},
{
"epoch": 41.46341463414634,
"grad_norm": 6.797173976898193,
"learning_rate": 8.16e-06,
"loss": 0.0863,
"step": 1700
},
{
"epoch": 42.073170731707314,
"grad_norm": 5.753118515014648,
"learning_rate": 8.121538461538462e-06,
"loss": 0.0874,
"step": 1725
},
{
"epoch": 42.68292682926829,
"grad_norm": 8.442388534545898,
"learning_rate": 8.083076923076924e-06,
"loss": 0.0737,
"step": 1750
},
{
"epoch": 43.292682926829265,
"grad_norm": 4.439198017120361,
"learning_rate": 8.044615384615385e-06,
"loss": 0.0733,
"step": 1775
},
{
"epoch": 43.90243902439025,
"grad_norm": 6.787204742431641,
"learning_rate": 8.006153846153847e-06,
"loss": 0.0714,
"step": 1800
},
{
"epoch": 44.51219512195122,
"grad_norm": 5.47157621383667,
"learning_rate": 7.967692307692309e-06,
"loss": 0.0627,
"step": 1825
},
{
"epoch": 45.1219512195122,
"grad_norm": 5.718803882598877,
"learning_rate": 7.92923076923077e-06,
"loss": 0.0666,
"step": 1850
},
{
"epoch": 45.73170731707317,
"grad_norm": 4.490105628967285,
"learning_rate": 7.890769230769232e-06,
"loss": 0.055,
"step": 1875
},
{
"epoch": 46.34146341463415,
"grad_norm": 5.873692035675049,
"learning_rate": 7.852307692307692e-06,
"loss": 0.0569,
"step": 1900
},
{
"epoch": 46.951219512195124,
"grad_norm": 10.3510103225708,
"learning_rate": 7.813846153846154e-06,
"loss": 0.0601,
"step": 1925
},
{
"epoch": 47.5609756097561,
"grad_norm": 5.004169464111328,
"learning_rate": 7.775384615384616e-06,
"loss": 0.0524,
"step": 1950
},
{
"epoch": 48.170731707317074,
"grad_norm": 3.3900997638702393,
"learning_rate": 7.736923076923077e-06,
"loss": 0.0508,
"step": 1975
},
{
"epoch": 48.78048780487805,
"grad_norm": 6.310654640197754,
"learning_rate": 7.698461538461539e-06,
"loss": 0.0518,
"step": 2000
},
{
"epoch": 48.78048780487805,
"eval_loss": 4.87410306930542,
"eval_runtime": 62.3697,
"eval_samples_per_second": 2.581,
"eval_steps_per_second": 0.337,
"eval_wer": 94.92795389048992,
"step": 2000
},
{
"epoch": 49.390243902439025,
"grad_norm": 5.030487060546875,
"learning_rate": 7.660000000000001e-06,
"loss": 0.0447,
"step": 2025
},
{
"epoch": 50.0,
"grad_norm": 9.628267288208008,
"learning_rate": 7.6215384615384625e-06,
"loss": 0.048,
"step": 2050
},
{
"epoch": 50.609756097560975,
"grad_norm": 9.77782917022705,
"learning_rate": 7.583076923076924e-06,
"loss": 0.0428,
"step": 2075
},
{
"epoch": 51.21951219512195,
"grad_norm": 5.163978099822998,
"learning_rate": 7.544615384615386e-06,
"loss": 0.0444,
"step": 2100
},
{
"epoch": 51.829268292682926,
"grad_norm": 3.368098497390747,
"learning_rate": 7.5061538461538465e-06,
"loss": 0.0421,
"step": 2125
},
{
"epoch": 52.4390243902439,
"grad_norm": 2.6525158882141113,
"learning_rate": 7.467692307692308e-06,
"loss": 0.0381,
"step": 2150
},
{
"epoch": 53.048780487804876,
"grad_norm": 5.2876410484313965,
"learning_rate": 7.42923076923077e-06,
"loss": 0.0421,
"step": 2175
},
{
"epoch": 53.65853658536585,
"grad_norm": 2.5063223838806152,
"learning_rate": 7.390769230769231e-06,
"loss": 0.0379,
"step": 2200
},
{
"epoch": 54.26829268292683,
"grad_norm": 7.786949157714844,
"learning_rate": 7.3523076923076935e-06,
"loss": 0.0365,
"step": 2225
},
{
"epoch": 54.8780487804878,
"grad_norm": 4.044300556182861,
"learning_rate": 7.313846153846155e-06,
"loss": 0.0359,
"step": 2250
},
{
"epoch": 55.48780487804878,
"grad_norm": 4.029970169067383,
"learning_rate": 7.275384615384616e-06,
"loss": 0.0348,
"step": 2275
},
{
"epoch": 56.09756097560975,
"grad_norm": 2.0436642169952393,
"learning_rate": 7.236923076923078e-06,
"loss": 0.0366,
"step": 2300
},
{
"epoch": 56.707317073170735,
"grad_norm": 5.075358867645264,
"learning_rate": 7.198461538461539e-06,
"loss": 0.0345,
"step": 2325
},
{
"epoch": 57.31707317073171,
"grad_norm": 1.7547271251678467,
"learning_rate": 7.16e-06,
"loss": 0.0304,
"step": 2350
},
{
"epoch": 57.926829268292686,
"grad_norm": 3.1869213581085205,
"learning_rate": 7.121538461538462e-06,
"loss": 0.0354,
"step": 2375
},
{
"epoch": 58.53658536585366,
"grad_norm": 2.8294670581817627,
"learning_rate": 7.0830769230769235e-06,
"loss": 0.0281,
"step": 2400
},
{
"epoch": 59.146341463414636,
"grad_norm": 2.4291458129882812,
"learning_rate": 7.044615384615386e-06,
"loss": 0.0334,
"step": 2425
},
{
"epoch": 59.75609756097561,
"grad_norm": 3.106226921081543,
"learning_rate": 7.006153846153847e-06,
"loss": 0.0306,
"step": 2450
},
{
"epoch": 60.36585365853659,
"grad_norm": 2.7606558799743652,
"learning_rate": 6.967692307692308e-06,
"loss": 0.0311,
"step": 2475
},
{
"epoch": 60.97560975609756,
"grad_norm": 3.0400702953338623,
"learning_rate": 6.9292307692307705e-06,
"loss": 0.0305,
"step": 2500
},
{
"epoch": 61.58536585365854,
"grad_norm": 3.5321948528289795,
"learning_rate": 6.890769230769231e-06,
"loss": 0.0275,
"step": 2525
},
{
"epoch": 62.19512195121951,
"grad_norm": 1.7394859790802002,
"learning_rate": 6.852307692307692e-06,
"loss": 0.0325,
"step": 2550
},
{
"epoch": 62.80487804878049,
"grad_norm": 3.530212163925171,
"learning_rate": 6.8138461538461545e-06,
"loss": 0.0285,
"step": 2575
},
{
"epoch": 63.41463414634146,
"grad_norm": 1.9453763961791992,
"learning_rate": 6.775384615384616e-06,
"loss": 0.0255,
"step": 2600
},
{
"epoch": 64.02439024390245,
"grad_norm": 2.1866273880004883,
"learning_rate": 6.736923076923078e-06,
"loss": 0.0311,
"step": 2625
},
{
"epoch": 64.63414634146342,
"grad_norm": 2.9744958877563477,
"learning_rate": 6.698461538461539e-06,
"loss": 0.0234,
"step": 2650
},
{
"epoch": 65.2439024390244,
"grad_norm": 1.6761668920516968,
"learning_rate": 6.660000000000001e-06,
"loss": 0.027,
"step": 2675
},
{
"epoch": 65.85365853658537,
"grad_norm": 1.5864598751068115,
"learning_rate": 6.621538461538463e-06,
"loss": 0.0238,
"step": 2700
},
{
"epoch": 66.46341463414635,
"grad_norm": 2.605698347091675,
"learning_rate": 6.583076923076923e-06,
"loss": 0.0236,
"step": 2725
},
{
"epoch": 67.07317073170732,
"grad_norm": 1.8222204446792603,
"learning_rate": 6.5446153846153846e-06,
"loss": 0.0299,
"step": 2750
},
{
"epoch": 67.6829268292683,
"grad_norm": 2.6547422409057617,
"learning_rate": 6.506153846153847e-06,
"loss": 0.0203,
"step": 2775
},
{
"epoch": 68.29268292682927,
"grad_norm": 1.494598627090454,
"learning_rate": 6.467692307692308e-06,
"loss": 0.0268,
"step": 2800
},
{
"epoch": 68.90243902439025,
"grad_norm": 1.973913311958313,
"learning_rate": 6.42923076923077e-06,
"loss": 0.0235,
"step": 2825
},
{
"epoch": 69.51219512195122,
"grad_norm": 1.9251242876052856,
"learning_rate": 6.3907692307692315e-06,
"loss": 0.0232,
"step": 2850
},
{
"epoch": 70.1219512195122,
"grad_norm": 2.214128017425537,
"learning_rate": 6.352307692307693e-06,
"loss": 0.0239,
"step": 2875
},
{
"epoch": 70.73170731707317,
"grad_norm": 2.427274703979492,
"learning_rate": 6.313846153846155e-06,
"loss": 0.0223,
"step": 2900
},
{
"epoch": 71.34146341463415,
"grad_norm": 2.7526907920837402,
"learning_rate": 6.2753846153846155e-06,
"loss": 0.0215,
"step": 2925
},
{
"epoch": 71.95121951219512,
"grad_norm": 2.739189386367798,
"learning_rate": 6.236923076923077e-06,
"loss": 0.0243,
"step": 2950
},
{
"epoch": 72.5609756097561,
"grad_norm": 1.61655592918396,
"learning_rate": 6.198461538461539e-06,
"loss": 0.0188,
"step": 2975
},
{
"epoch": 73.17073170731707,
"grad_norm": 2.1026246547698975,
"learning_rate": 6.16e-06,
"loss": 0.0234,
"step": 3000
},
{
"epoch": 73.17073170731707,
"eval_loss": 5.154399871826172,
"eval_runtime": 55.0306,
"eval_samples_per_second": 2.926,
"eval_steps_per_second": 0.382,
"eval_wer": 93.11239193083574,
"step": 3000
},
{
"epoch": 73.78048780487805,
"grad_norm": 2.770653247833252,
"learning_rate": 6.1215384615384625e-06,
"loss": 0.0206,
"step": 3025
},
{
"epoch": 74.39024390243902,
"grad_norm": 1.6438905000686646,
"learning_rate": 6.083076923076924e-06,
"loss": 0.019,
"step": 3050
},
{
"epoch": 75.0,
"grad_norm": 1.086746096611023,
"learning_rate": 6.044615384615385e-06,
"loss": 0.0229,
"step": 3075
},
{
"epoch": 75.60975609756098,
"grad_norm": 2.3001186847686768,
"learning_rate": 6.006153846153847e-06,
"loss": 0.0188,
"step": 3100
},
{
"epoch": 76.21951219512195,
"grad_norm": 1.9896184206008911,
"learning_rate": 5.967692307692308e-06,
"loss": 0.0208,
"step": 3125
},
{
"epoch": 76.82926829268293,
"grad_norm": 1.5976531505584717,
"learning_rate": 5.929230769230769e-06,
"loss": 0.0206,
"step": 3150
},
{
"epoch": 77.4390243902439,
"grad_norm": 2.1733193397521973,
"learning_rate": 5.890769230769231e-06,
"loss": 0.0194,
"step": 3175
},
{
"epoch": 78.04878048780488,
"grad_norm": 1.85785710811615,
"learning_rate": 5.8523076923076926e-06,
"loss": 0.0206,
"step": 3200
},
{
"epoch": 78.65853658536585,
"grad_norm": 3.4088730812072754,
"learning_rate": 5.813846153846155e-06,
"loss": 0.0174,
"step": 3225
},
{
"epoch": 79.26829268292683,
"grad_norm": 2.442415237426758,
"learning_rate": 5.775384615384616e-06,
"loss": 0.0203,
"step": 3250
},
{
"epoch": 79.8780487804878,
"grad_norm": 2.473065137863159,
"learning_rate": 5.736923076923077e-06,
"loss": 0.0196,
"step": 3275
},
{
"epoch": 80.48780487804878,
"grad_norm": 1.7116321325302124,
"learning_rate": 5.6984615384615395e-06,
"loss": 0.0184,
"step": 3300
},
{
"epoch": 81.09756097560975,
"grad_norm": 0.6662659645080566,
"learning_rate": 5.66e-06,
"loss": 0.0191,
"step": 3325
},
{
"epoch": 81.70731707317073,
"grad_norm": 1.450646162033081,
"learning_rate": 5.621538461538461e-06,
"loss": 0.0173,
"step": 3350
},
{
"epoch": 82.3170731707317,
"grad_norm": 1.3656872510910034,
"learning_rate": 5.5830769230769235e-06,
"loss": 0.0161,
"step": 3375
},
{
"epoch": 82.92682926829268,
"grad_norm": 2.8566460609436035,
"learning_rate": 5.544615384615385e-06,
"loss": 0.0217,
"step": 3400
},
{
"epoch": 83.53658536585365,
"grad_norm": 2.3349201679229736,
"learning_rate": 5.506153846153847e-06,
"loss": 0.0153,
"step": 3425
},
{
"epoch": 84.14634146341463,
"grad_norm": 1.6854404211044312,
"learning_rate": 5.467692307692308e-06,
"loss": 0.0189,
"step": 3450
},
{
"epoch": 84.7560975609756,
"grad_norm": 2.3685340881347656,
"learning_rate": 5.42923076923077e-06,
"loss": 0.0188,
"step": 3475
},
{
"epoch": 85.36585365853658,
"grad_norm": 1.909439206123352,
"learning_rate": 5.390769230769232e-06,
"loss": 0.0163,
"step": 3500
},
{
"epoch": 85.97560975609755,
"grad_norm": 2.5704216957092285,
"learning_rate": 5.352307692307692e-06,
"loss": 0.0186,
"step": 3525
},
{
"epoch": 86.58536585365853,
"grad_norm": 2.1618387699127197,
"learning_rate": 5.313846153846154e-06,
"loss": 0.015,
"step": 3550
},
{
"epoch": 87.1951219512195,
"grad_norm": 1.4021707773208618,
"learning_rate": 5.275384615384616e-06,
"loss": 0.0191,
"step": 3575
},
{
"epoch": 87.8048780487805,
"grad_norm": 1.4248768091201782,
"learning_rate": 5.236923076923077e-06,
"loss": 0.0181,
"step": 3600
},
{
"epoch": 88.41463414634147,
"grad_norm": 1.5336748361587524,
"learning_rate": 5.198461538461539e-06,
"loss": 0.0152,
"step": 3625
},
{
"epoch": 89.02439024390245,
"grad_norm": 1.369664192199707,
"learning_rate": 5.1600000000000006e-06,
"loss": 0.0193,
"step": 3650
},
{
"epoch": 89.63414634146342,
"grad_norm": 2.177635669708252,
"learning_rate": 5.121538461538462e-06,
"loss": 0.0155,
"step": 3675
},
{
"epoch": 90.2439024390244,
"grad_norm": 1.5026379823684692,
"learning_rate": 5.083076923076924e-06,
"loss": 0.017,
"step": 3700
},
{
"epoch": 90.85365853658537,
"grad_norm": 1.1004050970077515,
"learning_rate": 5.044615384615385e-06,
"loss": 0.0157,
"step": 3725
},
{
"epoch": 91.46341463414635,
"grad_norm": 1.5270745754241943,
"learning_rate": 5.006153846153846e-06,
"loss": 0.0148,
"step": 3750
},
{
"epoch": 92.07317073170732,
"grad_norm": 2.0228114128112793,
"learning_rate": 4.967692307692308e-06,
"loss": 0.0198,
"step": 3775
},
{
"epoch": 92.6829268292683,
"grad_norm": 2.0614633560180664,
"learning_rate": 4.929230769230769e-06,
"loss": 0.0136,
"step": 3800
},
{
"epoch": 93.29268292682927,
"grad_norm": 0.9733468294143677,
"learning_rate": 4.8907692307692315e-06,
"loss": 0.0161,
"step": 3825
},
{
"epoch": 93.90243902439025,
"grad_norm": 2.1101596355438232,
"learning_rate": 4.852307692307693e-06,
"loss": 0.0165,
"step": 3850
},
{
"epoch": 94.51219512195122,
"grad_norm": 1.4769028425216675,
"learning_rate": 4.813846153846154e-06,
"loss": 0.015,
"step": 3875
},
{
"epoch": 95.1219512195122,
"grad_norm": 2.2220544815063477,
"learning_rate": 4.7753846153846155e-06,
"loss": 0.019,
"step": 3900
},
{
"epoch": 95.73170731707317,
"grad_norm": 1.2263191938400269,
"learning_rate": 4.736923076923078e-06,
"loss": 0.0145,
"step": 3925
},
{
"epoch": 96.34146341463415,
"grad_norm": 1.1551388502120972,
"learning_rate": 4.698461538461539e-06,
"loss": 0.0146,
"step": 3950
},
{
"epoch": 96.95121951219512,
"grad_norm": 1.4800796508789062,
"learning_rate": 4.66e-06,
"loss": 0.0167,
"step": 3975
},
{
"epoch": 97.5609756097561,
"grad_norm": 1.9003641605377197,
"learning_rate": 4.621538461538462e-06,
"loss": 0.0148,
"step": 4000
},
{
"epoch": 97.5609756097561,
"eval_loss": 5.350304126739502,
"eval_runtime": 47.6459,
"eval_samples_per_second": 3.379,
"eval_steps_per_second": 0.441,
"eval_wer": 93.42939481268012,
"step": 4000
}
],
"logging_steps": 25,
"max_steps": 7000,
"num_input_tokens_seen": 0,
"num_train_epochs": 171,
"save_steps": 1000,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 1.54456423354368e+18,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}