polaris_qa_qg_model_stg_11 / trainer_state.json
logoyazilim's picture
commit files to HF hub
3433b13
raw
history blame
11.2 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 11.999931153184166,
"global_step": 43572,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.14,
"learning_rate": 0.0009885247406591388,
"loss": 0.4139,
"step": 500
},
{
"epoch": 0.28,
"learning_rate": 0.0009770494813182778,
"loss": 0.315,
"step": 1000
},
{
"epoch": 0.41,
"learning_rate": 0.0009655742219774168,
"loss": 0.2807,
"step": 1500
},
{
"epoch": 0.55,
"learning_rate": 0.0009540989626365555,
"loss": 0.2624,
"step": 2000
},
{
"epoch": 0.69,
"learning_rate": 0.0009426237032956944,
"loss": 0.257,
"step": 2500
},
{
"epoch": 0.83,
"learning_rate": 0.0009311484439548334,
"loss": 0.2378,
"step": 3000
},
{
"epoch": 0.96,
"learning_rate": 0.0009196731846139723,
"loss": 0.2245,
"step": 3500
},
{
"epoch": 1.1,
"learning_rate": 0.0009081979252731112,
"loss": 0.1983,
"step": 4000
},
{
"epoch": 1.24,
"learning_rate": 0.0008967226659322501,
"loss": 0.1987,
"step": 4500
},
{
"epoch": 1.38,
"learning_rate": 0.0008852474065913891,
"loss": 0.1904,
"step": 5000
},
{
"epoch": 1.51,
"learning_rate": 0.0008737721472505279,
"loss": 0.1829,
"step": 5500
},
{
"epoch": 1.65,
"learning_rate": 0.0008622968879096667,
"loss": 0.1773,
"step": 6000
},
{
"epoch": 1.79,
"learning_rate": 0.0008508216285688056,
"loss": 0.179,
"step": 6500
},
{
"epoch": 1.93,
"learning_rate": 0.0008393463692279446,
"loss": 0.177,
"step": 7000
},
{
"epoch": 2.07,
"learning_rate": 0.0008278711098870835,
"loss": 0.1577,
"step": 7500
},
{
"epoch": 2.2,
"learning_rate": 0.0008163958505462224,
"loss": 0.1529,
"step": 8000
},
{
"epoch": 2.34,
"learning_rate": 0.0008049205912053612,
"loss": 0.1468,
"step": 8500
},
{
"epoch": 2.48,
"learning_rate": 0.0007934453318645002,
"loss": 0.1492,
"step": 9000
},
{
"epoch": 2.62,
"learning_rate": 0.000781970072523639,
"loss": 0.1426,
"step": 9500
},
{
"epoch": 2.75,
"learning_rate": 0.0007704948131827779,
"loss": 0.1368,
"step": 10000
},
{
"epoch": 2.89,
"learning_rate": 0.0007590195538419168,
"loss": 0.1443,
"step": 10500
},
{
"epoch": 3.03,
"learning_rate": 0.0007475442945010558,
"loss": 0.1364,
"step": 11000
},
{
"epoch": 3.17,
"learning_rate": 0.0007360690351601946,
"loss": 0.1221,
"step": 11500
},
{
"epoch": 3.3,
"learning_rate": 0.0007245937758193335,
"loss": 0.1188,
"step": 12000
},
{
"epoch": 3.44,
"learning_rate": 0.0007131185164784725,
"loss": 0.1218,
"step": 12500
},
{
"epoch": 3.58,
"learning_rate": 0.0007016432571376114,
"loss": 0.1175,
"step": 13000
},
{
"epoch": 3.72,
"learning_rate": 0.0006901679977967502,
"loss": 0.117,
"step": 13500
},
{
"epoch": 3.86,
"learning_rate": 0.000678692738455889,
"loss": 0.1154,
"step": 14000
},
{
"epoch": 3.99,
"learning_rate": 0.000667217479115028,
"loss": 0.1124,
"step": 14500
},
{
"epoch": 4.13,
"learning_rate": 0.0006557422197741669,
"loss": 0.0995,
"step": 15000
},
{
"epoch": 4.27,
"learning_rate": 0.0006442669604333058,
"loss": 0.0948,
"step": 15500
},
{
"epoch": 4.41,
"learning_rate": 0.0006327917010924447,
"loss": 0.0972,
"step": 16000
},
{
"epoch": 4.54,
"learning_rate": 0.0006213164417515837,
"loss": 0.0967,
"step": 16500
},
{
"epoch": 4.68,
"learning_rate": 0.0006098411824107226,
"loss": 0.1003,
"step": 17000
},
{
"epoch": 4.82,
"learning_rate": 0.0005983659230698613,
"loss": 0.0961,
"step": 17500
},
{
"epoch": 4.96,
"learning_rate": 0.0005868906637290002,
"loss": 0.0967,
"step": 18000
},
{
"epoch": 5.1,
"learning_rate": 0.0005754154043881392,
"loss": 0.0825,
"step": 18500
},
{
"epoch": 5.23,
"learning_rate": 0.0005639401450472781,
"loss": 0.081,
"step": 19000
},
{
"epoch": 5.37,
"learning_rate": 0.000552464885706417,
"loss": 0.0827,
"step": 19500
},
{
"epoch": 5.51,
"learning_rate": 0.0005409896263655559,
"loss": 0.08,
"step": 20000
},
{
"epoch": 5.65,
"learning_rate": 0.0005295143670246948,
"loss": 0.0842,
"step": 20500
},
{
"epoch": 5.78,
"learning_rate": 0.0005180391076838336,
"loss": 0.0798,
"step": 21000
},
{
"epoch": 5.92,
"learning_rate": 0.0005065638483429725,
"loss": 0.083,
"step": 21500
},
{
"epoch": 6.06,
"learning_rate": 0.0004950885890021115,
"loss": 0.0749,
"step": 22000
},
{
"epoch": 6.2,
"learning_rate": 0.00048361332966125037,
"loss": 0.0667,
"step": 22500
},
{
"epoch": 6.33,
"learning_rate": 0.00047213807032038925,
"loss": 0.0677,
"step": 23000
},
{
"epoch": 6.47,
"learning_rate": 0.0004606628109795282,
"loss": 0.0699,
"step": 23500
},
{
"epoch": 6.61,
"learning_rate": 0.000449187551638667,
"loss": 0.0694,
"step": 24000
},
{
"epoch": 6.75,
"learning_rate": 0.00043771229229780595,
"loss": 0.0674,
"step": 24500
},
{
"epoch": 6.89,
"learning_rate": 0.00042623703295694484,
"loss": 0.0652,
"step": 25000
},
{
"epoch": 7.02,
"learning_rate": 0.00041476177361608377,
"loss": 0.0668,
"step": 25500
},
{
"epoch": 7.16,
"learning_rate": 0.0004032865142752226,
"loss": 0.0573,
"step": 26000
},
{
"epoch": 7.3,
"learning_rate": 0.00039181125493436154,
"loss": 0.0564,
"step": 26500
},
{
"epoch": 7.44,
"learning_rate": 0.0003803359955935004,
"loss": 0.0573,
"step": 27000
},
{
"epoch": 7.57,
"learning_rate": 0.0003688607362526393,
"loss": 0.0572,
"step": 27500
},
{
"epoch": 7.71,
"learning_rate": 0.0003573854769117782,
"loss": 0.058,
"step": 28000
},
{
"epoch": 7.85,
"learning_rate": 0.0003459102175709171,
"loss": 0.0583,
"step": 28500
},
{
"epoch": 7.99,
"learning_rate": 0.000334434958230056,
"loss": 0.0581,
"step": 29000
},
{
"epoch": 8.12,
"learning_rate": 0.0003229596988891949,
"loss": 0.0483,
"step": 29500
},
{
"epoch": 8.26,
"learning_rate": 0.00031148443954833377,
"loss": 0.0474,
"step": 30000
},
{
"epoch": 8.4,
"learning_rate": 0.0003000091802074727,
"loss": 0.0497,
"step": 30500
},
{
"epoch": 8.54,
"learning_rate": 0.0002885339208666116,
"loss": 0.0487,
"step": 31000
},
{
"epoch": 8.68,
"learning_rate": 0.00027705866152575047,
"loss": 0.0469,
"step": 31500
},
{
"epoch": 8.81,
"learning_rate": 0.0002655834021848894,
"loss": 0.0492,
"step": 32000
},
{
"epoch": 8.95,
"learning_rate": 0.0002541081428440283,
"loss": 0.0494,
"step": 32500
},
{
"epoch": 9.09,
"learning_rate": 0.00024263288350316717,
"loss": 0.0426,
"step": 33000
},
{
"epoch": 9.23,
"learning_rate": 0.00023115762416230608,
"loss": 0.0406,
"step": 33500
},
{
"epoch": 9.36,
"learning_rate": 0.00021968236482144496,
"loss": 0.0405,
"step": 34000
},
{
"epoch": 9.5,
"learning_rate": 0.00020820710548058387,
"loss": 0.043,
"step": 34500
},
{
"epoch": 9.64,
"learning_rate": 0.00019673184613972275,
"loss": 0.0411,
"step": 35000
},
{
"epoch": 9.78,
"learning_rate": 0.00018525658679886166,
"loss": 0.0419,
"step": 35500
},
{
"epoch": 9.91,
"learning_rate": 0.00017378132745800054,
"loss": 0.0409,
"step": 36000
},
{
"epoch": 10.05,
"learning_rate": 0.00016230606811713943,
"loss": 0.0388,
"step": 36500
},
{
"epoch": 10.19,
"learning_rate": 0.00015083080877627833,
"loss": 0.0367,
"step": 37000
},
{
"epoch": 10.33,
"learning_rate": 0.00013935554943541724,
"loss": 0.0355,
"step": 37500
},
{
"epoch": 10.47,
"learning_rate": 0.00012788029009455615,
"loss": 0.0358,
"step": 38000
},
{
"epoch": 10.6,
"learning_rate": 0.00011640503075369504,
"loss": 0.0369,
"step": 38500
},
{
"epoch": 10.74,
"learning_rate": 0.00010492977141283393,
"loss": 0.0339,
"step": 39000
},
{
"epoch": 10.88,
"learning_rate": 9.345451207197283e-05,
"loss": 0.0359,
"step": 39500
},
{
"epoch": 11.02,
"learning_rate": 8.197925273111174e-05,
"loss": 0.0364,
"step": 40000
},
{
"epoch": 11.15,
"learning_rate": 7.050399339025063e-05,
"loss": 0.0325,
"step": 40500
},
{
"epoch": 11.29,
"learning_rate": 5.902873404938952e-05,
"loss": 0.0321,
"step": 41000
},
{
"epoch": 11.43,
"learning_rate": 4.755347470852842e-05,
"loss": 0.0328,
"step": 41500
},
{
"epoch": 11.57,
"learning_rate": 3.607821536766731e-05,
"loss": 0.032,
"step": 42000
},
{
"epoch": 11.7,
"learning_rate": 2.4602956026806205e-05,
"loss": 0.0315,
"step": 42500
},
{
"epoch": 11.84,
"learning_rate": 1.3127696685945103e-05,
"loss": 0.0318,
"step": 43000
},
{
"epoch": 11.98,
"learning_rate": 1.6524373450839989e-06,
"loss": 0.0314,
"step": 43500
},
{
"epoch": 12.0,
"step": 43572,
"total_flos": 2.93279124897792e+17,
"train_loss": 0.09914802481164445,
"train_runtime": 67574.8841,
"train_samples_per_second": 20.634,
"train_steps_per_second": 0.645
}
],
"max_steps": 43572,
"num_train_epochs": 12,
"total_flos": 2.93279124897792e+17,
"trial_name": null,
"trial_params": null
}