tuanio's picture
End of training
edf185e
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 100.0,
"eval_steps": 100,
"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": 1.358490566037736e-06,
"loss": 127.0899,
"step": 100
},
{
"epoch": 1.89,
"learning_rate": 3.2452830188679247e-06,
"loss": 93.6454,
"step": 200
},
{
"epoch": 2.83,
"learning_rate": 5.1320754716981136e-06,
"loss": 47.1864,
"step": 300
},
{
"epoch": 3.77,
"learning_rate": 7.018867924528302e-06,
"loss": 29.1895,
"step": 400
},
{
"epoch": 4.72,
"learning_rate": 8.905660377358491e-06,
"loss": 21.4491,
"step": 500
},
{
"epoch": 5.66,
"learning_rate": 1.079245283018868e-05,
"loss": 16.0656,
"step": 600
},
{
"epoch": 6.6,
"learning_rate": 1.2679245283018868e-05,
"loss": 11.7189,
"step": 700
},
{
"epoch": 7.55,
"learning_rate": 1.4566037735849057e-05,
"loss": 8.5375,
"step": 800
},
{
"epoch": 8.49,
"learning_rate": 1.6452830188679247e-05,
"loss": 6.5608,
"step": 900
},
{
"epoch": 9.43,
"learning_rate": 1.8339622641509435e-05,
"loss": 5.4918,
"step": 1000
},
{
"epoch": 10.38,
"learning_rate": 1.9999921920876882e-05,
"loss": 4.8475,
"step": 1100
},
{
"epoch": 11.32,
"learning_rate": 1.999319920296687e-05,
"loss": 4.582,
"step": 1200
},
{
"epoch": 12.26,
"learning_rate": 1.9975640502598243e-05,
"loss": 4.4601,
"step": 1300
},
{
"epoch": 13.21,
"learning_rate": 1.994726485929633e-05,
"loss": 4.4055,
"step": 1400
},
{
"epoch": 14.15,
"learning_rate": 1.9908103041783615e-05,
"loss": 4.386,
"step": 1500
},
{
"epoch": 15.09,
"learning_rate": 1.9858197514616142e-05,
"loss": 4.375,
"step": 1600
},
{
"epoch": 16.04,
"learning_rate": 1.9797602392137678e-05,
"loss": 4.3661,
"step": 1700
},
{
"epoch": 16.98,
"learning_rate": 1.9726383379801593e-05,
"loss": 4.3576,
"step": 1800
},
{
"epoch": 17.92,
"learning_rate": 1.9644617702924093e-05,
"loss": 4.348,
"step": 1900
},
{
"epoch": 18.87,
"learning_rate": 1.9552394022946068e-05,
"loss": 4.3346,
"step": 2000
},
{
"epoch": 19.81,
"learning_rate": 1.9449812341294302e-05,
"loss": 4.3124,
"step": 2100
},
{
"epoch": 20.75,
"learning_rate": 1.9336983890946383e-05,
"loss": 4.302,
"step": 2200
},
{
"epoch": 21.7,
"learning_rate": 1.9214031015816803e-05,
"loss": 4.2777,
"step": 2300
},
{
"epoch": 22.64,
"learning_rate": 1.9081087038095094e-05,
"loss": 4.2489,
"step": 2400
},
{
"epoch": 23.58,
"learning_rate": 1.8938296113679814e-05,
"loss": 4.2226,
"step": 2500
},
{
"epoch": 24.53,
"learning_rate": 1.8785813075865164e-05,
"loss": 4.1862,
"step": 2600
},
{
"epoch": 25.47,
"learning_rate": 1.8623803267449722e-05,
"loss": 4.1183,
"step": 2700
},
{
"epoch": 26.42,
"learning_rate": 1.8452442361449353e-05,
"loss": 3.9384,
"step": 2800
},
{
"epoch": 27.36,
"learning_rate": 1.827191617060869e-05,
"loss": 3.6958,
"step": 2900
},
{
"epoch": 28.3,
"learning_rate": 1.8082420445917727e-05,
"loss": 3.3496,
"step": 3000
},
{
"epoch": 29.25,
"learning_rate": 1.7884160664352062e-05,
"loss": 2.9018,
"step": 3100
},
{
"epoch": 30.19,
"learning_rate": 1.7677351806066863e-05,
"loss": 2.4478,
"step": 3200
},
{
"epoch": 31.13,
"learning_rate": 1.7462218121286224e-05,
"loss": 2.1227,
"step": 3300
},
{
"epoch": 32.08,
"learning_rate": 1.723899288714064e-05,
"loss": 1.9402,
"step": 3400
},
{
"epoch": 33.02,
"learning_rate": 1.701026694125833e-05,
"loss": 1.7127,
"step": 3500
},
{
"epoch": 33.96,
"learning_rate": 1.6771667994919608e-05,
"loss": 1.5356,
"step": 3600
},
{
"epoch": 34.91,
"learning_rate": 1.6525726287357162e-05,
"loss": 1.3928,
"step": 3700
},
{
"epoch": 35.85,
"learning_rate": 1.6272708501939845e-05,
"loss": 1.2609,
"step": 3800
},
{
"epoch": 36.79,
"learning_rate": 1.6012888994880194e-05,
"loss": 1.1638,
"step": 3900
},
{
"epoch": 37.74,
"learning_rate": 1.574654949774022e-05,
"loss": 1.0668,
"step": 4000
},
{
"epoch": 38.68,
"learning_rate": 1.5473978811939808e-05,
"loss": 1.016,
"step": 4100
},
{
"epoch": 39.62,
"learning_rate": 1.5195472495599042e-05,
"loss": 0.9288,
"step": 4200
},
{
"epoch": 40.57,
"learning_rate": 1.4911332543053986e-05,
"loss": 0.8772,
"step": 4300
},
{
"epoch": 41.51,
"learning_rate": 1.4621867057393465e-05,
"loss": 0.8202,
"step": 4400
},
{
"epoch": 42.45,
"learning_rate": 1.4327389916371856e-05,
"loss": 0.7753,
"step": 4500
},
{
"epoch": 43.4,
"learning_rate": 1.4031234292879726e-05,
"loss": 0.7272,
"step": 4600
},
{
"epoch": 44.34,
"learning_rate": 1.373079443760497e-05,
"loss": 0.7156,
"step": 4700
},
{
"epoch": 45.28,
"learning_rate": 1.3423295725145908e-05,
"loss": 0.6848,
"step": 4800
},
{
"epoch": 46.23,
"learning_rate": 1.3115214387653711e-05,
"loss": 0.6478,
"step": 4900
},
{
"epoch": 47.17,
"learning_rate": 1.2800661191392916e-05,
"loss": 0.6204,
"step": 5000
},
{
"epoch": 48.11,
"learning_rate": 1.2483071138273168e-05,
"loss": 0.5965,
"step": 5100
},
{
"epoch": 49.06,
"learning_rate": 1.2162788602520851e-05,
"loss": 0.5816,
"step": 5200
},
{
"epoch": 50.0,
"learning_rate": 1.1840160877917335e-05,
"loss": 0.5519,
"step": 5300
},
{
"epoch": 50.94,
"learning_rate": 1.1515537801215944e-05,
"loss": 0.5382,
"step": 5400
},
{
"epoch": 51.89,
"learning_rate": 1.1189271372801474e-05,
"loss": 0.5363,
"step": 5500
},
{
"epoch": 52.83,
"learning_rate": 1.0861715375003623e-05,
"loss": 0.5102,
"step": 5600
},
{
"epoch": 53.77,
"learning_rate": 1.0533224988478176e-05,
"loss": 0.4949,
"step": 5700
},
{
"epoch": 54.72,
"learning_rate": 1.0204156407071964e-05,
"loss": 0.4994,
"step": 5800
},
{
"epoch": 55.66,
"learning_rate": 9.874866451589151e-06,
"loss": 0.4642,
"step": 5900
},
{
"epoch": 56.6,
"learning_rate": 9.545712182877714e-06,
"loss": 0.4648,
"step": 6000
},
{
"epoch": 57.55,
"learning_rate": 9.22033352215851e-06,
"loss": 0.4518,
"step": 6100
},
{
"epoch": 58.49,
"learning_rate": 8.892510582822682e-06,
"loss": 0.4469,
"step": 6200
},
{
"epoch": 59.43,
"learning_rate": 8.565888533777731e-06,
"loss": 0.4384,
"step": 6300
},
{
"epoch": 60.38,
"learning_rate": 8.240821542965487e-06,
"loss": 0.4314,
"step": 6400
},
{
"epoch": 61.32,
"learning_rate": 7.917662092122658e-06,
"loss": 0.4244,
"step": 6500
},
{
"epoch": 62.26,
"learning_rate": 7.5967605945723e-06,
"loss": 0.4226,
"step": 6600
},
{
"epoch": 63.21,
"learning_rate": 7.278465015258145e-06,
"loss": 0.404,
"step": 6700
},
{
"epoch": 64.15,
"learning_rate": 6.963120493433808e-06,
"loss": 0.4041,
"step": 6800
},
{
"epoch": 65.09,
"learning_rate": 6.65106896841597e-06,
"loss": 0.3987,
"step": 6900
},
{
"epoch": 66.04,
"learning_rate": 6.342648808807368e-06,
"loss": 0.3954,
"step": 7000
},
{
"epoch": 66.98,
"learning_rate": 6.038194445591676e-06,
"loss": 0.3887,
"step": 7100
},
{
"epoch": 67.92,
"learning_rate": 5.738036009498027e-06,
"loss": 0.3787,
"step": 7200
},
{
"epoch": 68.87,
"learning_rate": 5.442498973028489e-06,
"loss": 0.3791,
"step": 7300
},
{
"epoch": 69.81,
"learning_rate": 5.151903797536631e-06,
"loss": 0.3688,
"step": 7400
},
{
"epoch": 70.75,
"learning_rate": 4.866565585739835e-06,
"loss": 0.3722,
"step": 7500
},
{
"epoch": 71.7,
"learning_rate": 4.586793740042181e-06,
"loss": 0.3678,
"step": 7600
},
{
"epoch": 72.64,
"learning_rate": 4.312891627038364e-06,
"loss": 0.3582,
"step": 7700
},
{
"epoch": 73.58,
"learning_rate": 4.047802118060723e-06,
"loss": 0.3538,
"step": 7800
},
{
"epoch": 74.53,
"learning_rate": 3.7864578043142407e-06,
"loss": 0.3532,
"step": 7900
},
{
"epoch": 75.47,
"learning_rate": 3.5318510560731366e-06,
"loss": 0.3502,
"step": 8000
},
{
"epoch": 76.42,
"learning_rate": 3.284257952519031e-06,
"loss": 0.3423,
"step": 8100
},
{
"epoch": 77.36,
"learning_rate": 3.0439469676883693e-06,
"loss": 0.345,
"step": 8200
},
{
"epoch": 78.3,
"learning_rate": 2.811178679356429e-06,
"loss": 0.3505,
"step": 8300
},
{
"epoch": 79.25,
"learning_rate": 2.586205486483554e-06,
"loss": 0.3284,
"step": 8400
},
{
"epoch": 80.19,
"learning_rate": 2.369271335529936e-06,
"loss": 0.3352,
"step": 8500
},
{
"epoch": 81.13,
"learning_rate": 2.160611455935786e-06,
"loss": 0.3485,
"step": 8600
},
{
"epoch": 82.08,
"learning_rate": 1.9604521050536506e-06,
"loss": 0.3355,
"step": 8700
},
{
"epoch": 83.02,
"learning_rate": 1.7690103228094845e-06,
"loss": 0.3363,
"step": 8800
},
{
"epoch": 83.96,
"learning_rate": 1.5864936963585164e-06,
"loss": 0.331,
"step": 8900
},
{
"epoch": 84.91,
"learning_rate": 1.413100134991111e-06,
"loss": 0.3349,
"step": 9000
},
{
"epoch": 85.85,
"learning_rate": 1.2490176555326306e-06,
"loss": 0.3332,
"step": 9100
},
{
"epoch": 86.79,
"learning_rate": 1.0944241784701025e-06,
"loss": 0.3303,
"step": 9200
},
{
"epoch": 87.74,
"learning_rate": 9.49487335026692e-07,
"loss": 0.3241,
"step": 9300
},
{
"epoch": 88.68,
"learning_rate": 8.143642853931688e-07,
"loss": 0.3315,
"step": 9400
},
{
"epoch": 89.62,
"learning_rate": 6.892015483135095e-07,
"loss": 0.3249,
"step": 9500
},
{
"epoch": 90.57,
"learning_rate": 5.741348422094128e-07,
"loss": 0.3211,
"step": 9600
},
{
"epoch": 91.51,
"learning_rate": 4.702864215416847e-07,
"loss": 0.3371,
"step": 9700
},
{
"epoch": 92.45,
"learning_rate": 3.7567113645558537e-07,
"loss": 0.3143,
"step": 9800
},
{
"epoch": 93.4,
"learning_rate": 2.914918545575618e-07,
"loss": 0.3307,
"step": 9900
},
{
"epoch": 94.34,
"learning_rate": 2.1783985444653433e-07,
"loss": 0.3214,
"step": 10000
},
{
"epoch": 95.28,
"learning_rate": 1.5479499961417067e-07,
"loss": 0.3192,
"step": 10100
},
{
"epoch": 96.23,
"learning_rate": 1.024256518460387e-07,
"loss": 0.3302,
"step": 10200
},
{
"epoch": 97.17,
"learning_rate": 6.078859709449215e-08,
"loss": 0.3241,
"step": 10300
},
{
"epoch": 98.11,
"learning_rate": 2.992898390362875e-08,
"loss": 0.3259,
"step": 10400
},
{
"epoch": 99.06,
"learning_rate": 9.88027445312123e-09,
"loss": 0.3279,
"step": 10500
},
{
"epoch": 100.0,
"learning_rate": 7.027047924512698e-10,
"loss": 0.3253,
"step": 10600
},
{
"epoch": 100.0,
"step": 10600,
"total_flos": 6.510810385036083e+16,
"train_loss": 4.708893706123784,
"train_runtime": 6081.8512,
"train_samples_per_second": 55.641,
"train_steps_per_second": 1.743
}
],
"logging_steps": 100,
"max_steps": 10600,
"num_train_epochs": 100,
"save_steps": 100,
"total_flos": 6.510810385036083e+16,
"trial_name": null,
"trial_params": null
}