tuanio's picture
End of training
4fd790e
{
"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": 1.3773584905660377e-06,
"loss": 126.8518,
"step": 100
},
{
"epoch": 1.89,
"learning_rate": 3.2641509433962264e-06,
"loss": 93.0988,
"step": 200
},
{
"epoch": 2.83,
"learning_rate": 5.150943396226416e-06,
"loss": 47.0116,
"step": 300
},
{
"epoch": 3.77,
"learning_rate": 7.037735849056604e-06,
"loss": 28.9934,
"step": 400
},
{
"epoch": 4.72,
"learning_rate": 8.924528301886793e-06,
"loss": 21.351,
"step": 500
},
{
"epoch": 5.66,
"learning_rate": 1.0811320754716984e-05,
"loss": 16.0057,
"step": 600
},
{
"epoch": 6.6,
"learning_rate": 1.2698113207547172e-05,
"loss": 11.6529,
"step": 700
},
{
"epoch": 7.55,
"learning_rate": 1.458490566037736e-05,
"loss": 8.5554,
"step": 800
},
{
"epoch": 8.49,
"learning_rate": 1.6471698113207547e-05,
"loss": 6.564,
"step": 900
},
{
"epoch": 9.43,
"learning_rate": 1.8358490566037736e-05,
"loss": 5.4798,
"step": 1000
},
{
"epoch": 10.38,
"learning_rate": 1.9999908365494266e-05,
"loss": 4.8443,
"step": 1100
},
{
"epoch": 11.32,
"learning_rate": 1.999307723208382e-05,
"loss": 4.5809,
"step": 1200
},
{
"epoch": 12.26,
"learning_rate": 1.9975410248472147e-05,
"loss": 4.4584,
"step": 1300
},
{
"epoch": 13.21,
"learning_rate": 1.9946926571599945e-05,
"loss": 4.4059,
"step": 1400
},
{
"epoch": 14.15,
"learning_rate": 1.9907657087334364e-05,
"loss": 4.3875,
"step": 1500
},
{
"epoch": 15.09,
"learning_rate": 1.9857644376978347e-05,
"loss": 4.3756,
"step": 1600
},
{
"epoch": 16.04,
"learning_rate": 1.9796942671098212e-05,
"loss": 4.3675,
"step": 1700
},
{
"epoch": 16.98,
"learning_rate": 1.9725617790719523e-05,
"loss": 4.3606,
"step": 1800
},
{
"epoch": 17.92,
"learning_rate": 1.9643747075954974e-05,
"loss": 4.3517,
"step": 1900
},
{
"epoch": 18.87,
"learning_rate": 1.9551419302141784e-05,
"loss": 4.3407,
"step": 2000
},
{
"epoch": 19.81,
"learning_rate": 1.9448734583579404e-05,
"loss": 4.3216,
"step": 2100
},
{
"epoch": 20.75,
"learning_rate": 1.9335804264972018e-05,
"loss": 4.3099,
"step": 2200
},
{
"epoch": 21.7,
"learning_rate": 1.9212750800693506e-05,
"loss": 4.2844,
"step": 2300
},
{
"epoch": 22.64,
"learning_rate": 1.90797076220058e-05,
"loss": 4.2608,
"step": 2400
},
{
"epoch": 23.58,
"learning_rate": 1.8936818992374606e-05,
"loss": 4.2275,
"step": 2500
},
{
"epoch": 24.53,
"learning_rate": 1.8784239851039375e-05,
"loss": 4.1848,
"step": 2600
},
{
"epoch": 25.47,
"learning_rate": 1.8622135645007168e-05,
"loss": 4.0996,
"step": 2700
},
{
"epoch": 26.42,
"learning_rate": 1.845068214965255e-05,
"loss": 3.8911,
"step": 2800
},
{
"epoch": 27.36,
"learning_rate": 1.8270065278118092e-05,
"loss": 3.6114,
"step": 2900
},
{
"epoch": 28.3,
"learning_rate": 1.8080480879722103e-05,
"loss": 3.2303,
"step": 3000
},
{
"epoch": 29.25,
"learning_rate": 1.7882134527592244e-05,
"loss": 2.757,
"step": 3100
},
{
"epoch": 30.19,
"learning_rate": 1.7677351806066863e-05,
"loss": 2.3513,
"step": 3200
},
{
"epoch": 31.13,
"learning_rate": 1.7462218121286224e-05,
"loss": 2.0624,
"step": 3300
},
{
"epoch": 32.08,
"learning_rate": 1.7241264405022477e-05,
"loss": 1.8598,
"step": 3400
},
{
"epoch": 33.02,
"learning_rate": 1.701026694125833e-05,
"loss": 1.6587,
"step": 3500
},
{
"epoch": 33.96,
"learning_rate": 1.6771667994919608e-05,
"loss": 1.5081,
"step": 3600
},
{
"epoch": 34.91,
"learning_rate": 1.6525726287357162e-05,
"loss": 1.368,
"step": 3700
},
{
"epoch": 35.85,
"learning_rate": 1.6275272811746063e-05,
"loss": 1.2771,
"step": 3800
},
{
"epoch": 36.79,
"learning_rate": 1.6015519939371957e-05,
"loss": 1.1754,
"step": 3900
},
{
"epoch": 37.74,
"learning_rate": 1.5749244224090537e-05,
"loss": 1.079,
"step": 4000
},
{
"epoch": 38.68,
"learning_rate": 1.5479489390466186e-05,
"loss": 1.026,
"step": 4100
},
{
"epoch": 39.62,
"learning_rate": 1.520109884810164e-05,
"loss": 0.9535,
"step": 4200
},
{
"epoch": 40.57,
"learning_rate": 1.4917068568678037e-05,
"loss": 0.9007,
"step": 4300
},
{
"epoch": 41.51,
"learning_rate": 1.4627706536361717e-05,
"loss": 0.8523,
"step": 4400
},
{
"epoch": 42.45,
"learning_rate": 1.4333326516728909e-05,
"loss": 0.8297,
"step": 4500
},
{
"epoch": 43.4,
"learning_rate": 1.4034247716538545e-05,
"loss": 0.7856,
"step": 4600
},
{
"epoch": 44.34,
"learning_rate": 1.373079443760497e-05,
"loss": 0.7526,
"step": 4700
},
{
"epoch": 45.28,
"learning_rate": 1.3423295725145908e-05,
"loss": 0.7318,
"step": 4800
},
{
"epoch": 46.23,
"learning_rate": 1.3112085010987002e-05,
"loss": 0.7039,
"step": 4900
},
{
"epoch": 47.17,
"learning_rate": 1.279749975200978e-05,
"loss": 0.6837,
"step": 5000
},
{
"epoch": 48.11,
"learning_rate": 1.247988106423513e-05,
"loss": 0.6541,
"step": 5100
},
{
"epoch": 49.06,
"learning_rate": 1.2159573352939014e-05,
"loss": 0.6463,
"step": 5200
},
{
"epoch": 50.0,
"learning_rate": 1.1840160877917335e-05,
"loss": 0.6415,
"step": 5300
},
{
"epoch": 50.94,
"learning_rate": 1.1515537801215944e-05,
"loss": 0.6578,
"step": 5400
},
{
"epoch": 51.89,
"learning_rate": 1.1189271372801474e-05,
"loss": 0.6162,
"step": 5500
},
{
"epoch": 52.83,
"learning_rate": 1.0861715375003623e-05,
"loss": 0.5861,
"step": 5600
},
{
"epoch": 53.77,
"learning_rate": 1.0533224988478176e-05,
"loss": 0.5669,
"step": 5700
},
{
"epoch": 54.72,
"learning_rate": 1.0204156407071964e-05,
"loss": 0.573,
"step": 5800
},
{
"epoch": 55.66,
"learning_rate": 9.874866451589151e-06,
"loss": 0.5379,
"step": 5900
},
{
"epoch": 56.6,
"learning_rate": 9.545712182877714e-06,
"loss": 0.5475,
"step": 6000
},
{
"epoch": 57.55,
"learning_rate": 9.21705051465562e-06,
"loss": 0.5475,
"step": 6100
},
{
"epoch": 58.49,
"learning_rate": 8.889237826496551e-06,
"loss": 0.544,
"step": 6200
},
{
"epoch": 59.43,
"learning_rate": 8.562629577394817e-06,
"loss": 0.5581,
"step": 6300
},
{
"epoch": 60.38,
"learning_rate": 8.237579920328478e-06,
"loss": 0.5399,
"step": 6400
},
{
"epoch": 61.32,
"learning_rate": 7.91444131823864e-06,
"loss": 0.5034,
"step": 6500
},
{
"epoch": 62.26,
"learning_rate": 7.593564161841318e-06,
"loss": 0.5139,
"step": 6600
},
{
"epoch": 63.21,
"learning_rate": 7.275296389686258e-06,
"loss": 0.4803,
"step": 6700
},
{
"epoch": 64.15,
"learning_rate": 6.959983110874782e-06,
"loss": 0.4803,
"step": 6800
},
{
"epoch": 65.09,
"learning_rate": 6.647966230845655e-06,
"loss": 0.468,
"step": 6900
},
{
"epoch": 66.04,
"learning_rate": 6.339584080634824e-06,
"loss": 0.4635,
"step": 7000
},
{
"epoch": 66.98,
"learning_rate": 6.03517105001101e-06,
"loss": 0.4638,
"step": 7100
},
{
"epoch": 67.92,
"learning_rate": 5.735057224884939e-06,
"loss": 0.4618,
"step": 7200
},
{
"epoch": 68.87,
"learning_rate": 5.439568029385422e-06,
"loss": 0.4479,
"step": 7300
},
{
"epoch": 69.81,
"learning_rate": 5.14902387299036e-06,
"loss": 0.4394,
"step": 7400
},
{
"epoch": 70.75,
"learning_rate": 4.866565585739835e-06,
"loss": 0.4445,
"step": 7500
},
{
"epoch": 71.7,
"learning_rate": 4.586793740042181e-06,
"loss": 0.4502,
"step": 7600
},
{
"epoch": 72.64,
"learning_rate": 4.312891627038364e-06,
"loss": 0.436,
"step": 7700
},
{
"epoch": 73.58,
"learning_rate": 4.0451562485624955e-06,
"loss": 0.4338,
"step": 7800
},
{
"epoch": 74.53,
"learning_rate": 3.7838779196384036e-06,
"loss": 0.4308,
"step": 7900
},
{
"epoch": 75.47,
"learning_rate": 3.5293399536806903e-06,
"loss": 0.4305,
"step": 8000
},
{
"epoch": 76.42,
"learning_rate": 3.2818183552879e-06,
"loss": 0.4183,
"step": 8100
},
{
"epoch": 77.36,
"learning_rate": 3.041581520960881e-06,
"loss": 0.4282,
"step": 8200
},
{
"epoch": 78.3,
"learning_rate": 2.8088899480708677e-06,
"loss": 0.4261,
"step": 8300
},
{
"epoch": 79.25,
"learning_rate": 2.5839959523929126e-06,
"loss": 0.4111,
"step": 8400
},
{
"epoch": 80.19,
"learning_rate": 2.3671433945108712e-06,
"loss": 0.4201,
"step": 8500
},
{
"epoch": 81.13,
"learning_rate": 2.15856741539067e-06,
"loss": 0.4323,
"step": 8600
},
{
"epoch": 82.08,
"learning_rate": 1.9584941814085756e-06,
"loss": 0.4071,
"step": 8700
},
{
"epoch": 83.02,
"learning_rate": 1.7671406391109204e-06,
"loss": 0.4129,
"step": 8800
},
{
"epoch": 83.96,
"learning_rate": 1.5847142799712112e-06,
"loss": 0.4128,
"step": 8900
},
{
"epoch": 84.91,
"learning_rate": 1.4114129153997401e-06,
"loss": 0.4128,
"step": 9000
},
{
"epoch": 85.85,
"learning_rate": 1.2474244622496002e-06,
"loss": 0.407,
"step": 9100
},
{
"epoch": 86.79,
"learning_rate": 1.0929267390517373e-06,
"loss": 0.4093,
"step": 9200
},
{
"epoch": 87.74,
"learning_rate": 9.48087273199979e-07,
"loss": 0.4038,
"step": 9300
},
{
"epoch": 88.68,
"learning_rate": 8.143642853931688e-07,
"loss": 0.4023,
"step": 9400
},
{
"epoch": 89.62,
"learning_rate": 6.892015483135095e-07,
"loss": 0.4031,
"step": 9500
},
{
"epoch": 90.57,
"learning_rate": 5.741348422094128e-07,
"loss": 0.3969,
"step": 9600
},
{
"epoch": 91.51,
"learning_rate": 4.69288938015946e-07,
"loss": 0.4251,
"step": 9700
},
{
"epoch": 92.45,
"learning_rate": 3.747775238879847e-07,
"loss": 0.3997,
"step": 9800
},
{
"epoch": 93.4,
"learning_rate": 2.907030819240908e-07,
"loss": 0.4096,
"step": 9900
},
{
"epoch": 94.34,
"learning_rate": 2.1783985444653433e-07,
"loss": 0.3956,
"step": 10000
},
{
"epoch": 95.28,
"learning_rate": 1.5479499961417067e-07,
"loss": 0.3984,
"step": 10100
},
{
"epoch": 96.23,
"learning_rate": 1.024256518460387e-07,
"loss": 0.4015,
"step": 10200
},
{
"epoch": 97.17,
"learning_rate": 6.115168487157097e-08,
"loss": 0.4019,
"step": 10300
},
{
"epoch": 98.11,
"learning_rate": 3.018411211767158e-08,
"loss": 0.4115,
"step": 10400
},
{
"epoch": 99.06,
"learning_rate": 1.002716645950197e-08,
"loss": 0.403,
"step": 10500
},
{
"epoch": 100.0,
"learning_rate": 7.027047924512698e-10,
"loss": 0.408,
"step": 10600
},
{
"epoch": 100.0,
"step": 10600,
"total_flos": 6.511706405810995e+16,
"train_loss": 4.735838983643729,
"train_runtime": 6086.8056,
"train_samples_per_second": 55.596,
"train_steps_per_second": 1.741
}
],
"logging_steps": 100,
"max_steps": 10600,
"num_train_epochs": 100,
"save_steps": 100,
"total_flos": 6.511706405810995e+16,
"trial_name": null,
"trial_params": null
}