tuanio's picture
End of training
fca6f63
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 100.0,
"eval_steps": 200,
"global_step": 10600,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.94,
"learning_rate": 1.4528301886792454e-06,
"loss": 129.1763,
"step": 100
},
{
"epoch": 1.89,
"learning_rate": 3.3207547169811323e-06,
"loss": 118.0881,
"step": 200
},
{
"epoch": 2.83,
"learning_rate": 5.188679245283019e-06,
"loss": 98.9424,
"step": 300
},
{
"epoch": 3.77,
"learning_rate": 7.0754716981132075e-06,
"loss": 43.3696,
"step": 400
},
{
"epoch": 4.72,
"learning_rate": 8.962264150943398e-06,
"loss": 23.1622,
"step": 500
},
{
"epoch": 5.66,
"learning_rate": 1.0849056603773586e-05,
"loss": 16.4914,
"step": 600
},
{
"epoch": 6.6,
"learning_rate": 1.2735849056603775e-05,
"loss": 11.8071,
"step": 700
},
{
"epoch": 7.55,
"learning_rate": 1.4622641509433963e-05,
"loss": 8.6248,
"step": 800
},
{
"epoch": 8.49,
"learning_rate": 1.650943396226415e-05,
"loss": 6.542,
"step": 900
},
{
"epoch": 9.43,
"learning_rate": 1.839622641509434e-05,
"loss": 5.467,
"step": 1000
},
{
"epoch": 10.38,
"learning_rate": 1.9999878001459436e-05,
"loss": 4.8661,
"step": 1100
},
{
"epoch": 11.32,
"learning_rate": 1.9992830039282252e-05,
"loss": 4.5895,
"step": 1200
},
{
"epoch": 12.26,
"learning_rate": 1.997494649494382e-05,
"loss": 4.4634,
"step": 1300
},
{
"epoch": 13.21,
"learning_rate": 1.994624676020936e-05,
"loss": 4.4113,
"step": 1400
},
{
"epoch": 14.15,
"learning_rate": 1.990676195522527e-05,
"loss": 4.3831,
"step": 1500
},
{
"epoch": 15.09,
"learning_rate": 1.9856534894774423e-05,
"loss": 4.3747,
"step": 1600
},
{
"epoch": 16.04,
"learning_rate": 1.9795620041850602e-05,
"loss": 4.3661,
"step": 1700
},
{
"epoch": 16.98,
"learning_rate": 1.972408344860231e-05,
"loss": 4.3533,
"step": 1800
},
{
"epoch": 17.92,
"learning_rate": 1.9642002684710065e-05,
"loss": 4.3362,
"step": 1900
},
{
"epoch": 18.87,
"learning_rate": 1.9549466753274843e-05,
"loss": 4.3183,
"step": 2000
},
{
"epoch": 19.81,
"learning_rate": 1.9446575994308842e-05,
"loss": 4.3012,
"step": 2100
},
{
"epoch": 20.75,
"learning_rate": 1.933344197593319e-05,
"loss": 4.2824,
"step": 2200
},
{
"epoch": 21.7,
"learning_rate": 1.921018737340071e-05,
"loss": 4.2574,
"step": 2300
},
{
"epoch": 22.64,
"learning_rate": 1.9076945836074716e-05,
"loss": 4.2355,
"step": 2400
},
{
"epoch": 23.58,
"learning_rate": 1.8933861842508258e-05,
"loss": 4.2128,
"step": 2500
},
{
"epoch": 24.53,
"learning_rate": 1.8781090543780876e-05,
"loss": 4.1852,
"step": 2600
},
{
"epoch": 25.47,
"learning_rate": 1.8618797595262748e-05,
"loss": 4.153,
"step": 2700
},
{
"epoch": 26.42,
"learning_rate": 1.8447158976988656e-05,
"loss": 4.0818,
"step": 2800
},
{
"epoch": 27.36,
"learning_rate": 1.8266360802836542e-05,
"loss": 3.9215,
"step": 2900
},
{
"epoch": 28.3,
"learning_rate": 1.8076599118717634e-05,
"loss": 3.7038,
"step": 3000
},
{
"epoch": 29.25,
"learning_rate": 1.7878079689996806e-05,
"loss": 3.3915,
"step": 3100
},
{
"epoch": 30.19,
"learning_rate": 1.7671017778373914e-05,
"loss": 2.9503,
"step": 3200
},
{
"epoch": 31.13,
"learning_rate": 1.745563790846781e-05,
"loss": 2.5289,
"step": 3300
},
{
"epoch": 32.08,
"learning_rate": 1.7232173624356307e-05,
"loss": 2.1726,
"step": 3400
},
{
"epoch": 33.02,
"learning_rate": 1.7000867236335987e-05,
"loss": 1.8909,
"step": 3500
},
{
"epoch": 33.96,
"learning_rate": 1.676196955817653e-05,
"loss": 1.6745,
"step": 3600
},
{
"epoch": 34.91,
"learning_rate": 1.651573963515436e-05,
"loss": 1.5141,
"step": 3700
},
{
"epoch": 35.85,
"learning_rate": 1.6262444463160627e-05,
"loss": 1.3735,
"step": 3800
},
{
"epoch": 36.79,
"learning_rate": 1.6002358699188035e-05,
"loss": 1.2418,
"step": 3900
},
{
"epoch": 37.74,
"learning_rate": 1.573576436351046e-05,
"loss": 1.1395,
"step": 4000
},
{
"epoch": 38.68,
"learning_rate": 1.5462950533878318e-05,
"loss": 1.0523,
"step": 4100
},
{
"epoch": 39.62,
"learning_rate": 1.5184213032061217e-05,
"loss": 0.9775,
"step": 4200
},
{
"epoch": 40.57,
"learning_rate": 1.490272451088505e-05,
"loss": 0.9198,
"step": 4300
},
{
"epoch": 41.51,
"learning_rate": 1.4613104080651812e-05,
"loss": 0.8618,
"step": 4400
},
{
"epoch": 42.45,
"learning_rate": 1.4321451438907927e-05,
"loss": 0.8094,
"step": 4500
},
{
"epoch": 43.4,
"learning_rate": 1.4022191400246923e-05,
"loss": 0.7578,
"step": 4600
},
{
"epoch": 44.34,
"learning_rate": 1.3718569955936632e-05,
"loss": 0.7338,
"step": 4700
},
{
"epoch": 45.28,
"learning_rate": 1.3410916333542502e-05,
"loss": 0.7023,
"step": 4800
},
{
"epoch": 46.23,
"learning_rate": 1.309956413286455e-05,
"loss": 0.6598,
"step": 4900
},
{
"epoch": 47.17,
"learning_rate": 1.2784850964203184e-05,
"loss": 0.6316,
"step": 5000
},
{
"epoch": 48.11,
"learning_rate": 1.2467118082276324e-05,
"loss": 0.6309,
"step": 5100
},
{
"epoch": 49.06,
"learning_rate": 1.2146710016184722e-05,
"loss": 0.5876,
"step": 5200
},
{
"epoch": 50.0,
"learning_rate": 1.1823974195826782e-05,
"loss": 0.5738,
"step": 5300
},
{
"epoch": 50.94,
"learning_rate": 1.1499260575167908e-05,
"loss": 0.5557,
"step": 5400
},
{
"epoch": 51.89,
"learning_rate": 1.1172921252772938e-05,
"loss": 0.5642,
"step": 5500
},
{
"epoch": 52.83,
"learning_rate": 1.0845310090013116e-05,
"loss": 0.5474,
"step": 5600
},
{
"epoch": 53.77,
"learning_rate": 1.0516782327361563e-05,
"loss": 0.5307,
"step": 5700
},
{
"epoch": 54.72,
"learning_rate": 1.0187694199193371e-05,
"loss": 0.5044,
"step": 5800
},
{
"epoch": 55.66,
"learning_rate": 9.858402547507968e-06,
"loss": 0.4961,
"step": 5900
},
{
"epoch": 56.6,
"learning_rate": 9.529264434992576e-06,
"loss": 0.5012,
"step": 6000
},
{
"epoch": 57.55,
"learning_rate": 9.203919337260903e-06,
"loss": 0.4874,
"step": 6100
},
{
"epoch": 58.49,
"learning_rate": 8.876148009288813e-06,
"loss": 0.4643,
"step": 6200
},
{
"epoch": 59.43,
"learning_rate": 8.549595314129907e-06,
"loss": 0.4569,
"step": 6300
},
{
"epoch": 60.38,
"learning_rate": 8.224615344523123e-06,
"loss": 0.4487,
"step": 6400
},
{
"epoch": 61.32,
"learning_rate": 7.901560487844973e-06,
"loss": 0.4353,
"step": 6500
},
{
"epoch": 62.26,
"learning_rate": 7.580781044003324e-06,
"loss": 0.4271,
"step": 6600
},
{
"epoch": 63.21,
"learning_rate": 7.262624845594721e-06,
"loss": 0.4138,
"step": 6700
},
{
"epoch": 64.15,
"learning_rate": 6.947436880737089e-06,
"loss": 0.4108,
"step": 6800
},
{
"epoch": 65.09,
"learning_rate": 6.635558918986797e-06,
"loss": 0.4056,
"step": 6900
},
{
"epoch": 66.04,
"learning_rate": 6.327329140745751e-06,
"loss": 0.3992,
"step": 7000
},
{
"epoch": 66.98,
"learning_rate": 6.023081770560307e-06,
"loss": 0.3955,
"step": 7100
},
{
"epoch": 67.92,
"learning_rate": 5.723146714709664e-06,
"loss": 0.3944,
"step": 7200
},
{
"epoch": 68.87,
"learning_rate": 5.427849203476738e-06,
"loss": 0.3883,
"step": 7300
},
{
"epoch": 69.81,
"learning_rate": 5.140387256937587e-06,
"loss": 0.3825,
"step": 7400
},
{
"epoch": 70.75,
"learning_rate": 4.855265798347217e-06,
"loss": 0.3769,
"step": 7500
},
{
"epoch": 71.7,
"learning_rate": 4.575722958618642e-06,
"loss": 0.3704,
"step": 7600
},
{
"epoch": 72.64,
"learning_rate": 4.302061856027213e-06,
"loss": 0.3676,
"step": 7700
},
{
"epoch": 73.58,
"learning_rate": 4.034579231070851e-06,
"loss": 0.3682,
"step": 7800
},
{
"epoch": 74.53,
"learning_rate": 3.7735651247037275e-06,
"loss": 0.3609,
"step": 7900
},
{
"epoch": 75.47,
"learning_rate": 3.5193025638344956e-06,
"loss": 0.3605,
"step": 8000
},
{
"epoch": 76.42,
"learning_rate": 3.272067254430137e-06,
"loss": 0.3554,
"step": 8100
},
{
"epoch": 77.36,
"learning_rate": 3.0321272825581327e-06,
"loss": 0.3493,
"step": 8200
},
{
"epoch": 78.3,
"learning_rate": 2.7997428236911684e-06,
"loss": 0.347,
"step": 8300
},
{
"epoch": 79.25,
"learning_rate": 2.5751658605896003e-06,
"loss": 0.3516,
"step": 8400
},
{
"epoch": 80.19,
"learning_rate": 2.3586399100675516e-06,
"loss": 0.3489,
"step": 8500
},
{
"epoch": 81.13,
"learning_rate": 2.1503997589389334e-06,
"loss": 0.3416,
"step": 8600
},
{
"epoch": 82.08,
"learning_rate": 1.9506712094297386e-06,
"loss": 0.3398,
"step": 8700
},
{
"epoch": 83.02,
"learning_rate": 1.7596708343326285e-06,
"loss": 0.3412,
"step": 8800
},
{
"epoch": 83.96,
"learning_rate": 1.5776057421693126e-06,
"loss": 0.3394,
"step": 8900
},
{
"epoch": 84.91,
"learning_rate": 1.4046733526154088e-06,
"loss": 0.3409,
"step": 9000
},
{
"epoch": 85.85,
"learning_rate": 1.2410611824312334e-06,
"loss": 0.3364,
"step": 9100
},
{
"epoch": 86.79,
"learning_rate": 1.0884402169237329e-06,
"loss": 0.338,
"step": 9200
},
{
"epoch": 87.74,
"learning_rate": 9.438929780459661e-07,
"loss": 0.3391,
"step": 9300
},
{
"epoch": 88.68,
"learning_rate": 8.091655991389668e-07,
"loss": 0.3327,
"step": 9400
},
{
"epoch": 89.62,
"learning_rate": 6.844041699068349e-07,
"loss": 0.3319,
"step": 9500
},
{
"epoch": 90.57,
"learning_rate": 5.708401570907373e-07,
"loss": 0.3362,
"step": 9600
},
{
"epoch": 91.51,
"learning_rate": 4.6630268912521757e-07,
"loss": 0.3307,
"step": 9700
},
{
"epoch": 92.45,
"learning_rate": 3.7210294932142566e-07,
"loss": 0.3338,
"step": 9800
},
{
"epoch": 93.4,
"learning_rate": 2.8834308181833304e-07,
"loss": 0.3329,
"step": 9900
},
{
"epoch": 94.34,
"learning_rate": 2.151139104288702e-07,
"loss": 0.3297,
"step": 10000
},
{
"epoch": 95.28,
"learning_rate": 1.5249484015643078e-07,
"loss": 0.332,
"step": 10100
},
{
"epoch": 96.23,
"learning_rate": 1.0055377109319475e-07,
"loss": 0.3324,
"step": 10200
},
{
"epoch": 97.17,
"learning_rate": 5.934702479363341e-08,
"loss": 0.3319,
"step": 10300
},
{
"epoch": 98.11,
"learning_rate": 2.8919283203038274e-08,
"loss": 0.3301,
"step": 10400
},
{
"epoch": 99.06,
"learning_rate": 9.30354020729496e-09,
"loss": 0.333,
"step": 10500
},
{
"epoch": 100.0,
"learning_rate": 5.210658564336335e-10,
"loss": 0.3296,
"step": 10600
},
{
"epoch": 100.0,
"step": 10600,
"total_flos": 6.483430015500288e+16,
"train_loss": 5.64003086198051,
"train_runtime": 5748.6483,
"train_samples_per_second": 58.866,
"train_steps_per_second": 1.844
}
],
"logging_steps": 100,
"max_steps": 10600,
"num_train_epochs": 100,
"save_steps": 100,
"total_flos": 6.483430015500288e+16,
"trial_name": null,
"trial_params": null
}