AndrewMcDowell's picture
End of training
474c43a
raw
history blame
24.1 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 50.0,
"global_step": 15900,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.31,
"learning_rate": 4.75e-06,
"loss": 19.9321,
"step": 100
},
{
"epoch": 0.63,
"learning_rate": 9.750000000000002e-06,
"loss": 4.3153,
"step": 200
},
{
"epoch": 0.94,
"learning_rate": 1.475e-05,
"loss": 4.0475,
"step": 300
},
{
"epoch": 1.26,
"learning_rate": 1.9750000000000002e-05,
"loss": 3.3541,
"step": 400
},
{
"epoch": 1.57,
"learning_rate": 2.4750000000000002e-05,
"loss": 2.4695,
"step": 500
},
{
"epoch": 1.89,
"learning_rate": 2.975e-05,
"loss": 2.176,
"step": 600
},
{
"epoch": 2.2,
"learning_rate": 3.475e-05,
"loss": 2.0514,
"step": 700
},
{
"epoch": 2.52,
"learning_rate": 3.9750000000000004e-05,
"loss": 1.9917,
"step": 800
},
{
"epoch": 2.83,
"learning_rate": 4.4750000000000004e-05,
"loss": 1.9838,
"step": 900
},
{
"epoch": 3.14,
"learning_rate": 4.975e-05,
"loss": 1.9321,
"step": 1000
},
{
"epoch": 3.14,
"eval_cer": 0.2635160121292804,
"eval_loss": 1.0115777254104614,
"eval_runtime": 309.9157,
"eval_samples_per_second": 14.41,
"eval_steps_per_second": 1.804,
"eval_wer": 0.9823226672633699,
"step": 1000
},
{
"epoch": 3.46,
"learning_rate": 5.475e-05,
"loss": 1.9148,
"step": 1100
},
{
"epoch": 3.77,
"learning_rate": 5.975000000000001e-05,
"loss": 1.89,
"step": 1200
},
{
"epoch": 4.09,
"learning_rate": 6.475e-05,
"loss": 1.8953,
"step": 1300
},
{
"epoch": 4.4,
"learning_rate": 6.975e-05,
"loss": 1.9038,
"step": 1400
},
{
"epoch": 4.72,
"learning_rate": 7.475000000000001e-05,
"loss": 1.9498,
"step": 1500
},
{
"epoch": 5.03,
"learning_rate": 7.975e-05,
"loss": 2.0113,
"step": 1600
},
{
"epoch": 5.35,
"learning_rate": 8.475000000000001e-05,
"loss": 2.0294,
"step": 1700
},
{
"epoch": 5.66,
"learning_rate": 8.975e-05,
"loss": 2.0839,
"step": 1800
},
{
"epoch": 5.97,
"learning_rate": 9.475e-05,
"loss": 2.056,
"step": 1900
},
{
"epoch": 6.29,
"learning_rate": 9.975000000000001e-05,
"loss": 2.0934,
"step": 2000
},
{
"epoch": 6.29,
"eval_cer": 0.29319823484456276,
"eval_loss": 1.1241235733032227,
"eval_runtime": 306.3033,
"eval_samples_per_second": 14.58,
"eval_steps_per_second": 1.825,
"eval_wer": 1.0221526068471694,
"step": 2000
},
{
"epoch": 6.6,
"learning_rate": 9.931654676258994e-05,
"loss": 2.0845,
"step": 2100
},
{
"epoch": 6.92,
"learning_rate": 9.859712230215827e-05,
"loss": 2.0873,
"step": 2200
},
{
"epoch": 7.23,
"learning_rate": 9.787769784172662e-05,
"loss": 2.0782,
"step": 2300
},
{
"epoch": 7.55,
"learning_rate": 9.715827338129497e-05,
"loss": 2.1767,
"step": 2400
},
{
"epoch": 7.86,
"learning_rate": 9.643884892086331e-05,
"loss": 2.137,
"step": 2500
},
{
"epoch": 8.18,
"learning_rate": 9.571942446043166e-05,
"loss": 2.113,
"step": 2600
},
{
"epoch": 8.49,
"learning_rate": 9.5e-05,
"loss": 2.0582,
"step": 2700
},
{
"epoch": 8.81,
"learning_rate": 9.428057553956835e-05,
"loss": 2.0384,
"step": 2800
},
{
"epoch": 9.12,
"learning_rate": 9.35611510791367e-05,
"loss": 2.0837,
"step": 2900
},
{
"epoch": 9.43,
"learning_rate": 9.284172661870504e-05,
"loss": 2.0389,
"step": 3000
},
{
"epoch": 9.43,
"eval_cer": 0.3344838070819877,
"eval_loss": 1.2066822052001953,
"eval_runtime": 306.7837,
"eval_samples_per_second": 14.557,
"eval_steps_per_second": 1.822,
"eval_wer": 1.1324681136719623,
"step": 3000
},
{
"epoch": 9.75,
"learning_rate": 9.212230215827339e-05,
"loss": 2.0286,
"step": 3100
},
{
"epoch": 10.06,
"learning_rate": 9.140287769784172e-05,
"loss": 2.0073,
"step": 3200
},
{
"epoch": 10.38,
"learning_rate": 9.068345323741007e-05,
"loss": 1.9767,
"step": 3300
},
{
"epoch": 10.69,
"learning_rate": 8.996402877697842e-05,
"loss": 1.9924,
"step": 3400
},
{
"epoch": 11.01,
"learning_rate": 8.924460431654677e-05,
"loss": 2.003,
"step": 3500
},
{
"epoch": 11.32,
"learning_rate": 8.852517985611512e-05,
"loss": 1.9925,
"step": 3600
},
{
"epoch": 11.64,
"learning_rate": 8.780575539568345e-05,
"loss": 1.9533,
"step": 3700
},
{
"epoch": 11.95,
"learning_rate": 8.70863309352518e-05,
"loss": 1.9474,
"step": 3800
},
{
"epoch": 12.26,
"learning_rate": 8.636690647482015e-05,
"loss": 1.949,
"step": 3900
},
{
"epoch": 12.58,
"learning_rate": 8.564748201438849e-05,
"loss": 1.9569,
"step": 4000
},
{
"epoch": 12.58,
"eval_cer": 0.26565260623392417,
"eval_loss": 0.9817910194396973,
"eval_runtime": 306.1977,
"eval_samples_per_second": 14.585,
"eval_steps_per_second": 1.826,
"eval_wer": 1.0089505482210785,
"step": 4000
},
{
"epoch": 12.89,
"learning_rate": 8.492805755395684e-05,
"loss": 1.9532,
"step": 4100
},
{
"epoch": 13.21,
"learning_rate": 8.420863309352519e-05,
"loss": 1.9614,
"step": 4200
},
{
"epoch": 13.52,
"learning_rate": 8.348920863309353e-05,
"loss": 1.9632,
"step": 4300
},
{
"epoch": 13.84,
"learning_rate": 8.276978417266188e-05,
"loss": 1.9463,
"step": 4400
},
{
"epoch": 14.15,
"learning_rate": 8.205035971223022e-05,
"loss": 1.9153,
"step": 4500
},
{
"epoch": 14.47,
"learning_rate": 8.133093525179857e-05,
"loss": 1.8858,
"step": 4600
},
{
"epoch": 14.78,
"learning_rate": 8.06115107913669e-05,
"loss": 1.873,
"step": 4700
},
{
"epoch": 15.09,
"learning_rate": 7.989208633093525e-05,
"loss": 1.8554,
"step": 4800
},
{
"epoch": 15.41,
"learning_rate": 7.91726618705036e-05,
"loss": 1.8789,
"step": 4900
},
{
"epoch": 15.72,
"learning_rate": 7.845323741007195e-05,
"loss": 1.8409,
"step": 5000
},
{
"epoch": 15.72,
"eval_cer": 0.3740518863660643,
"eval_loss": 1.0381526947021484,
"eval_runtime": 306.1278,
"eval_samples_per_second": 14.589,
"eval_steps_per_second": 1.826,
"eval_wer": 1.6480196912060863,
"step": 5000
},
{
"epoch": 16.04,
"learning_rate": 7.77338129496403e-05,
"loss": 1.8599,
"step": 5100
},
{
"epoch": 16.35,
"learning_rate": 7.701438848920864e-05,
"loss": 1.8393,
"step": 5200
},
{
"epoch": 16.67,
"learning_rate": 7.629496402877698e-05,
"loss": 1.8435,
"step": 5300
},
{
"epoch": 16.98,
"learning_rate": 7.557553956834533e-05,
"loss": 1.8443,
"step": 5400
},
{
"epoch": 17.3,
"learning_rate": 7.485611510791367e-05,
"loss": 1.7856,
"step": 5500
},
{
"epoch": 17.61,
"learning_rate": 7.413669064748202e-05,
"loss": 1.7948,
"step": 5600
},
{
"epoch": 17.92,
"learning_rate": 7.341726618705035e-05,
"loss": 1.8169,
"step": 5700
},
{
"epoch": 18.24,
"learning_rate": 7.269784172661872e-05,
"loss": 1.7683,
"step": 5800
},
{
"epoch": 18.55,
"learning_rate": 7.197841726618706e-05,
"loss": 1.7461,
"step": 5900
},
{
"epoch": 18.87,
"learning_rate": 7.12589928057554e-05,
"loss": 1.7449,
"step": 6000
},
{
"epoch": 18.87,
"eval_cer": 0.34543796070310384,
"eval_loss": 0.9962468147277832,
"eval_runtime": 306.247,
"eval_samples_per_second": 14.583,
"eval_steps_per_second": 1.825,
"eval_wer": 1.6267621391810247,
"step": 6000
},
{
"epoch": 19.18,
"learning_rate": 7.053956834532375e-05,
"loss": 1.7904,
"step": 6100
},
{
"epoch": 19.5,
"learning_rate": 6.982014388489208e-05,
"loss": 1.7315,
"step": 6200
},
{
"epoch": 19.81,
"learning_rate": 6.910071942446043e-05,
"loss": 1.7591,
"step": 6300
},
{
"epoch": 20.13,
"learning_rate": 6.83884892086331e-05,
"loss": 1.7338,
"step": 6400
},
{
"epoch": 20.44,
"learning_rate": 6.766906474820143e-05,
"loss": 1.7211,
"step": 6500
},
{
"epoch": 20.75,
"learning_rate": 6.694964028776978e-05,
"loss": 1.7398,
"step": 6600
},
{
"epoch": 21.07,
"learning_rate": 6.623021582733813e-05,
"loss": 1.7244,
"step": 6700
},
{
"epoch": 21.38,
"learning_rate": 6.551079136690648e-05,
"loss": 1.7823,
"step": 6800
},
{
"epoch": 21.7,
"learning_rate": 6.479136690647483e-05,
"loss": 1.7602,
"step": 6900
},
{
"epoch": 22.01,
"learning_rate": 6.407194244604317e-05,
"loss": 1.7349,
"step": 7000
},
{
"epoch": 22.01,
"eval_cer": 0.25974410176762075,
"eval_loss": 0.9559761881828308,
"eval_runtime": 307.762,
"eval_samples_per_second": 14.511,
"eval_steps_per_second": 1.816,
"eval_wer": 0.9850078317296934,
"step": 7000
},
{
"epoch": 22.33,
"learning_rate": 6.335251798561152e-05,
"loss": 1.7176,
"step": 7100
},
{
"epoch": 22.64,
"learning_rate": 6.263309352517986e-05,
"loss": 1.7154,
"step": 7200
},
{
"epoch": 22.96,
"learning_rate": 6.19136690647482e-05,
"loss": 1.7136,
"step": 7300
},
{
"epoch": 23.27,
"learning_rate": 6.119424460431655e-05,
"loss": 1.6846,
"step": 7400
},
{
"epoch": 23.58,
"learning_rate": 6.047482014388489e-05,
"loss": 1.6878,
"step": 7500
},
{
"epoch": 23.9,
"learning_rate": 5.975539568345324e-05,
"loss": 1.7066,
"step": 7600
},
{
"epoch": 24.21,
"learning_rate": 5.903597122302159e-05,
"loss": 1.6664,
"step": 7700
},
{
"epoch": 24.53,
"learning_rate": 5.831654676258993e-05,
"loss": 1.6676,
"step": 7800
},
{
"epoch": 24.84,
"learning_rate": 5.759712230215828e-05,
"loss": 1.6377,
"step": 7900
},
{
"epoch": 25.16,
"learning_rate": 5.6877697841726616e-05,
"loss": 1.6857,
"step": 8000
},
{
"epoch": 25.16,
"eval_cer": 0.23246143858524598,
"eval_loss": 0.8722429275512695,
"eval_runtime": 307.5785,
"eval_samples_per_second": 14.52,
"eval_steps_per_second": 1.817,
"eval_wer": 0.9668829715820094,
"step": 8000
},
{
"epoch": 25.47,
"learning_rate": 5.615827338129497e-05,
"loss": 1.6433,
"step": 8100
},
{
"epoch": 25.79,
"learning_rate": 5.543884892086331e-05,
"loss": 1.6355,
"step": 8200
},
{
"epoch": 26.1,
"learning_rate": 5.4719424460431656e-05,
"loss": 1.6345,
"step": 8300
},
{
"epoch": 26.42,
"learning_rate": 5.4000000000000005e-05,
"loss": 1.6166,
"step": 8400
},
{
"epoch": 26.73,
"learning_rate": 5.328057553956835e-05,
"loss": 1.6357,
"step": 8500
},
{
"epoch": 27.04,
"learning_rate": 5.2561151079136696e-05,
"loss": 1.6294,
"step": 8600
},
{
"epoch": 27.36,
"learning_rate": 5.1848920863309355e-05,
"loss": 1.5973,
"step": 8700
},
{
"epoch": 27.67,
"learning_rate": 5.11294964028777e-05,
"loss": 1.5881,
"step": 8800
},
{
"epoch": 27.99,
"learning_rate": 5.0410071942446046e-05,
"loss": 1.6079,
"step": 8900
},
{
"epoch": 28.3,
"learning_rate": 4.969064748201439e-05,
"loss": 1.5637,
"step": 9000
},
{
"epoch": 28.3,
"eval_cer": 0.3422495048854046,
"eval_loss": 0.7635577321052551,
"eval_runtime": 305.1268,
"eval_samples_per_second": 14.637,
"eval_steps_per_second": 1.832,
"eval_wer": 1.8071156858357575,
"step": 9000
},
{
"epoch": 28.62,
"learning_rate": 4.897122302158274e-05,
"loss": 1.5611,
"step": 9100
},
{
"epoch": 28.93,
"learning_rate": 4.825179856115108e-05,
"loss": 1.5648,
"step": 9200
},
{
"epoch": 29.25,
"learning_rate": 4.753237410071943e-05,
"loss": 1.5724,
"step": 9300
},
{
"epoch": 29.56,
"learning_rate": 4.681294964028777e-05,
"loss": 1.5375,
"step": 9400
},
{
"epoch": 29.87,
"learning_rate": 4.609352517985612e-05,
"loss": 1.5468,
"step": 9500
},
{
"epoch": 30.19,
"learning_rate": 4.537410071942446e-05,
"loss": 1.5512,
"step": 9600
},
{
"epoch": 30.5,
"learning_rate": 4.4654676258992804e-05,
"loss": 1.5394,
"step": 9700
},
{
"epoch": 30.82,
"learning_rate": 4.393525179856115e-05,
"loss": 1.5147,
"step": 9800
},
{
"epoch": 31.13,
"learning_rate": 4.32158273381295e-05,
"loss": 1.5102,
"step": 9900
},
{
"epoch": 31.45,
"learning_rate": 4.2496402877697845e-05,
"loss": 1.5088,
"step": 10000
},
{
"epoch": 31.45,
"eval_cer": 0.20850693160433564,
"eval_loss": 0.7290384769439697,
"eval_runtime": 306.7599,
"eval_samples_per_second": 14.559,
"eval_steps_per_second": 1.822,
"eval_wer": 1.0398299395837995,
"step": 10000
},
{
"epoch": 31.76,
"learning_rate": 4.177697841726619e-05,
"loss": 1.503,
"step": 10100
},
{
"epoch": 32.08,
"learning_rate": 4.1057553956834536e-05,
"loss": 1.4922,
"step": 10200
},
{
"epoch": 32.39,
"learning_rate": 4.0338129496402885e-05,
"loss": 1.5002,
"step": 10300
},
{
"epoch": 32.7,
"learning_rate": 3.961870503597123e-05,
"loss": 1.4877,
"step": 10400
},
{
"epoch": 33.02,
"learning_rate": 3.889928057553957e-05,
"loss": 1.5061,
"step": 10500
},
{
"epoch": 33.33,
"learning_rate": 3.817985611510791e-05,
"loss": 1.4751,
"step": 10600
},
{
"epoch": 33.65,
"learning_rate": 3.746043165467626e-05,
"loss": 1.4671,
"step": 10700
},
{
"epoch": 33.96,
"learning_rate": 3.674820143884892e-05,
"loss": 1.4496,
"step": 10800
},
{
"epoch": 34.28,
"learning_rate": 3.602877697841727e-05,
"loss": 1.4533,
"step": 10900
},
{
"epoch": 34.59,
"learning_rate": 3.530935251798561e-05,
"loss": 1.4298,
"step": 11000
},
{
"epoch": 34.59,
"eval_cer": 0.21042164862888182,
"eval_loss": 0.7575956583023071,
"eval_runtime": 307.6821,
"eval_samples_per_second": 14.515,
"eval_steps_per_second": 1.817,
"eval_wer": 1.0165585142089952,
"step": 11000
},
{
"epoch": 34.91,
"learning_rate": 3.458992805755396e-05,
"loss": 1.4401,
"step": 11100
},
{
"epoch": 35.22,
"learning_rate": 3.38705035971223e-05,
"loss": 1.4155,
"step": 11200
},
{
"epoch": 35.53,
"learning_rate": 3.315107913669065e-05,
"loss": 1.4119,
"step": 11300
},
{
"epoch": 35.85,
"learning_rate": 3.243165467625899e-05,
"loss": 1.4156,
"step": 11400
},
{
"epoch": 36.16,
"learning_rate": 3.1712230215827335e-05,
"loss": 1.4316,
"step": 11500
},
{
"epoch": 36.48,
"learning_rate": 3.0992805755395684e-05,
"loss": 1.3815,
"step": 11600
},
{
"epoch": 36.79,
"learning_rate": 3.027338129496403e-05,
"loss": 1.4087,
"step": 11700
},
{
"epoch": 37.11,
"learning_rate": 2.9553956834532376e-05,
"loss": 1.3889,
"step": 11800
},
{
"epoch": 37.42,
"learning_rate": 2.883453237410072e-05,
"loss": 1.3722,
"step": 11900
},
{
"epoch": 37.74,
"learning_rate": 2.8115107913669063e-05,
"loss": 1.3716,
"step": 12000
},
{
"epoch": 37.74,
"eval_cer": 0.21379089317851244,
"eval_loss": 0.7045890092849731,
"eval_runtime": 307.0323,
"eval_samples_per_second": 14.546,
"eval_steps_per_second": 1.821,
"eval_wer": 1.1275453121503691,
"step": 12000
},
{
"epoch": 38.05,
"learning_rate": 2.7395683453237412e-05,
"loss": 1.3782,
"step": 12100
},
{
"epoch": 38.36,
"learning_rate": 2.6676258992805758e-05,
"loss": 1.3496,
"step": 12200
},
{
"epoch": 38.68,
"learning_rate": 2.5956834532374104e-05,
"loss": 1.3645,
"step": 12300
},
{
"epoch": 38.99,
"learning_rate": 2.5237410071942446e-05,
"loss": 1.4,
"step": 12400
},
{
"epoch": 39.31,
"learning_rate": 2.4517985611510795e-05,
"loss": 1.3441,
"step": 12500
},
{
"epoch": 39.62,
"learning_rate": 2.3798561151079137e-05,
"loss": 1.3512,
"step": 12600
},
{
"epoch": 39.94,
"learning_rate": 2.3079136690647483e-05,
"loss": 1.3429,
"step": 12700
},
{
"epoch": 40.25,
"learning_rate": 2.2366906474820145e-05,
"loss": 1.3285,
"step": 12800
},
{
"epoch": 40.57,
"learning_rate": 2.164748201438849e-05,
"loss": 1.3222,
"step": 12900
},
{
"epoch": 40.88,
"learning_rate": 2.0928057553956836e-05,
"loss": 1.3185,
"step": 13000
},
{
"epoch": 40.88,
"eval_cer": 0.21792438100403488,
"eval_loss": 0.7010895013809204,
"eval_runtime": 308.0387,
"eval_samples_per_second": 14.498,
"eval_steps_per_second": 1.815,
"eval_wer": 1.1696128887894384,
"step": 13000
},
{
"epoch": 41.19,
"learning_rate": 2.02158273381295e-05,
"loss": 1.3151,
"step": 13100
},
{
"epoch": 41.51,
"learning_rate": 1.949640287769784e-05,
"loss": 1.2889,
"step": 13200
},
{
"epoch": 41.82,
"learning_rate": 1.877697841726619e-05,
"loss": 1.3339,
"step": 13300
},
{
"epoch": 42.14,
"learning_rate": 1.8057553956834532e-05,
"loss": 1.3292,
"step": 13400
},
{
"epoch": 42.45,
"learning_rate": 1.7338129496402878e-05,
"loss": 1.2849,
"step": 13500
},
{
"epoch": 42.77,
"learning_rate": 1.6618705035971224e-05,
"loss": 1.3046,
"step": 13600
},
{
"epoch": 43.08,
"learning_rate": 1.589928057553957e-05,
"loss": 1.2722,
"step": 13700
},
{
"epoch": 43.4,
"learning_rate": 1.5179856115107915e-05,
"loss": 1.2704,
"step": 13800
},
{
"epoch": 43.71,
"learning_rate": 1.446043165467626e-05,
"loss": 1.279,
"step": 13900
},
{
"epoch": 44.03,
"learning_rate": 1.3741007194244604e-05,
"loss": 1.28,
"step": 14000
},
{
"epoch": 44.03,
"eval_cer": 0.2023847677275678,
"eval_loss": 0.675375759601593,
"eval_runtime": 307.5046,
"eval_samples_per_second": 14.523,
"eval_steps_per_second": 1.818,
"eval_wer": 1.1315730588498545,
"step": 14000
},
{
"epoch": 44.34,
"learning_rate": 1.3021582733812952e-05,
"loss": 1.2604,
"step": 14100
},
{
"epoch": 44.65,
"learning_rate": 1.2302158273381296e-05,
"loss": 1.2631,
"step": 14200
},
{
"epoch": 44.97,
"learning_rate": 1.1582733812949641e-05,
"loss": 1.2742,
"step": 14300
},
{
"epoch": 45.28,
"learning_rate": 1.0863309352517987e-05,
"loss": 1.2552,
"step": 14400
},
{
"epoch": 45.6,
"learning_rate": 1.0143884892086332e-05,
"loss": 1.2328,
"step": 14500
},
{
"epoch": 45.91,
"learning_rate": 9.424460431654676e-06,
"loss": 1.2562,
"step": 14600
},
{
"epoch": 46.23,
"learning_rate": 8.705035971223022e-06,
"loss": 1.2239,
"step": 14700
},
{
"epoch": 46.54,
"learning_rate": 7.985611510791367e-06,
"loss": 1.2502,
"step": 14800
},
{
"epoch": 46.86,
"learning_rate": 7.266187050359713e-06,
"loss": 1.2265,
"step": 14900
},
{
"epoch": 47.17,
"learning_rate": 6.546762589928059e-06,
"loss": 1.2368,
"step": 15000
},
{
"epoch": 47.17,
"eval_cer": 0.19226881640904273,
"eval_loss": 0.6925497055053711,
"eval_runtime": 307.2836,
"eval_samples_per_second": 14.534,
"eval_steps_per_second": 1.819,
"eval_wer": 1.0516894159767285,
"step": 15000
},
{
"epoch": 47.48,
"learning_rate": 5.8273381294964035e-06,
"loss": 1.2265,
"step": 15100
},
{
"epoch": 47.8,
"learning_rate": 5.107913669064748e-06,
"loss": 1.2267,
"step": 15200
},
{
"epoch": 48.11,
"learning_rate": 4.388489208633094e-06,
"loss": 1.2267,
"step": 15300
},
{
"epoch": 48.43,
"learning_rate": 3.669064748201439e-06,
"loss": 1.2095,
"step": 15400
},
{
"epoch": 48.74,
"learning_rate": 2.9496402877697846e-06,
"loss": 1.1985,
"step": 15500
},
{
"epoch": 49.06,
"learning_rate": 2.23021582733813e-06,
"loss": 1.2141,
"step": 15600
},
{
"epoch": 49.37,
"learning_rate": 1.510791366906475e-06,
"loss": 1.2087,
"step": 15700
},
{
"epoch": 49.69,
"learning_rate": 7.913669064748202e-07,
"loss": 1.2109,
"step": 15800
},
{
"epoch": 50.0,
"learning_rate": 7.913669064748202e-08,
"loss": 1.2123,
"step": 15900
},
{
"epoch": 50.0,
"step": 15900,
"total_flos": 2.3890518741171333e+20,
"train_loss": 1.801840565999349,
"train_runtime": 49502.0749,
"train_samples_per_second": 10.254,
"train_steps_per_second": 0.321
}
],
"max_steps": 15900,
"num_train_epochs": 50,
"total_flos": 2.3890518741171333e+20,
"trial_name": null,
"trial_params": null
}