esm2_t30_150M_ptm_qlora_2100K / trainer_state.json
AmelieSchreiber's picture
Upload 11 files
ccad8a7
{
"best_metric": 0.10229773443883387,
"best_model_checkpoint": "esm2_t30_150M_qlora_ptm_sites_2023-10-17_05-49-32/checkpoint-21690",
"epoch": 1.0,
"eval_steps": 500,
"global_step": 21690,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"learning_rate": 0.0003700791565537852,
"loss": 0.2788,
"step": 200
},
{
"epoch": 0.02,
"learning_rate": 0.0003698462746319928,
"loss": 0.0908,
"step": 400
},
{
"epoch": 0.03,
"learning_rate": 0.00036945835522359584,
"loss": 0.061,
"step": 600
},
{
"epoch": 0.04,
"learning_rate": 0.00036891572382917795,
"loss": 0.0467,
"step": 800
},
{
"epoch": 0.05,
"learning_rate": 0.0003682188357671161,
"loss": 0.0447,
"step": 1000
},
{
"epoch": 0.06,
"learning_rate": 0.00036736827579152647,
"loss": 0.0411,
"step": 1200
},
{
"epoch": 0.06,
"learning_rate": 0.00036637015433638333,
"loss": 0.039,
"step": 1400
},
{
"epoch": 0.07,
"learning_rate": 0.0003652152782523256,
"loss": 0.0389,
"step": 1600
},
{
"epoch": 0.08,
"learning_rate": 0.0003639092505197804,
"loss": 0.0307,
"step": 1800
},
{
"epoch": 0.09,
"learning_rate": 0.0003624531670178909,
"loss": 0.0333,
"step": 2000
},
{
"epoch": 0.1,
"learning_rate": 0.0003608482495365936,
"loss": 0.0289,
"step": 2200
},
{
"epoch": 0.11,
"learning_rate": 0.00035909584475142265,
"loss": 0.0312,
"step": 2400
},
{
"epoch": 0.12,
"learning_rate": 0.0003571974230935234,
"loss": 0.029,
"step": 2600
},
{
"epoch": 0.13,
"learning_rate": 0.00035515457751582335,
"loss": 0.0278,
"step": 2800
},
{
"epoch": 0.14,
"learning_rate": 0.000352969022156396,
"loss": 0.0276,
"step": 3000
},
{
"epoch": 0.15,
"learning_rate": 0.000350642590900139,
"loss": 0.0284,
"step": 3200
},
{
"epoch": 0.16,
"learning_rate": 0.00034817723583997393,
"loss": 0.0265,
"step": 3400
},
{
"epoch": 0.17,
"learning_rate": 0.0003455750256388576,
"loss": 0.0263,
"step": 3600
},
{
"epoch": 0.18,
"learning_rate": 0.0003428381437939813,
"loss": 0.0266,
"step": 3800
},
{
"epoch": 0.18,
"learning_rate": 0.00033996888680461266,
"loss": 0.0295,
"step": 4000
},
{
"epoch": 0.19,
"learning_rate": 0.0003369696622451184,
"loss": 0.0259,
"step": 4200
},
{
"epoch": 0.2,
"learning_rate": 0.00033384298674478505,
"loss": 0.0229,
"step": 4400
},
{
"epoch": 0.21,
"learning_rate": 0.0003305914838761317,
"loss": 0.0228,
"step": 4600
},
{
"epoch": 0.22,
"learning_rate": 0.00032721788195348825,
"loss": 0.0236,
"step": 4800
},
{
"epoch": 0.23,
"learning_rate": 0.0003237427679674208,
"loss": 0.0241,
"step": 5000
},
{
"epoch": 0.24,
"learning_rate": 0.00032013413454873025,
"loss": 0.0278,
"step": 5200
},
{
"epoch": 0.25,
"learning_rate": 0.000316412176767992,
"loss": 0.0231,
"step": 5400
},
{
"epoch": 0.26,
"learning_rate": 0.00031258001769510885,
"loss": 0.0247,
"step": 5600
},
{
"epoch": 0.27,
"learning_rate": 0.0003086408728691517,
"loss": 0.0243,
"step": 5800
},
{
"epoch": 0.28,
"learning_rate": 0.0003045980476002217,
"loss": 0.0236,
"step": 6000
},
{
"epoch": 0.29,
"learning_rate": 0.00030045493419598703,
"loss": 0.0213,
"step": 6200
},
{
"epoch": 0.3,
"learning_rate": 0.00029621500911522,
"loss": 0.0244,
"step": 6400
},
{
"epoch": 0.3,
"learning_rate": 0.00029188183005072376,
"loss": 0.019,
"step": 6600
},
{
"epoch": 0.31,
"learning_rate": 0.0002874590329440961,
"loss": 0.0223,
"step": 6800
},
{
"epoch": 0.32,
"learning_rate": 0.0002829503289348351,
"loss": 0.0221,
"step": 7000
},
{
"epoch": 0.33,
"learning_rate": 0.00027835950124634694,
"loss": 0.0215,
"step": 7200
},
{
"epoch": 0.34,
"learning_rate": 0.000273690402011468,
"loss": 0.0235,
"step": 7400
},
{
"epoch": 0.35,
"learning_rate": 0.0002689469490401664,
"loss": 0.021,
"step": 7600
},
{
"epoch": 0.36,
"learning_rate": 0.0002641331225321336,
"loss": 0.0198,
"step": 7800
},
{
"epoch": 0.37,
"learning_rate": 0.0002592529617370253,
"loss": 0.0184,
"step": 8000
},
{
"epoch": 0.38,
"learning_rate": 0.0002543105615651547,
"loss": 0.0208,
"step": 8200
},
{
"epoch": 0.39,
"learning_rate": 0.00024931006915147966,
"loss": 0.02,
"step": 8400
},
{
"epoch": 0.4,
"learning_rate": 0.0002442556803757701,
"loss": 0.0201,
"step": 8600
},
{
"epoch": 0.41,
"learning_rate": 0.00023915163634187332,
"loss": 0.0216,
"step": 8800
},
{
"epoch": 0.41,
"learning_rate": 0.00023400221981903174,
"loss": 0.0216,
"step": 9000
},
{
"epoch": 0.42,
"learning_rate": 0.00022881175164824033,
"loss": 0.0196,
"step": 9200
},
{
"epoch": 0.43,
"learning_rate": 0.00022358458711665722,
"loss": 0.0192,
"step": 9400
},
{
"epoch": 0.44,
"learning_rate": 0.00021832511230311189,
"loss": 0.0203,
"step": 9600
},
{
"epoch": 0.45,
"learning_rate": 0.00021303774039777544,
"loss": 0.0204,
"step": 9800
},
{
"epoch": 0.46,
"learning_rate": 0.0002077269079990823,
"loss": 0.0204,
"step": 10000
},
{
"epoch": 0.47,
"learning_rate": 0.0002023970713910101,
"loss": 0.0212,
"step": 10200
},
{
"epoch": 0.48,
"learning_rate": 0.00019705270280384151,
"loss": 0.0204,
"step": 10400
},
{
"epoch": 0.49,
"learning_rate": 0.00019169828666154592,
"loss": 0.0184,
"step": 10600
},
{
"epoch": 0.5,
"learning_rate": 0.00018633831581892887,
"loss": 0.0202,
"step": 10800
},
{
"epoch": 0.51,
"learning_rate": 0.00018097728779170797,
"loss": 0.0198,
"step": 11000
},
{
"epoch": 0.52,
"learning_rate": 0.0001756464729181275,
"loss": 0.0189,
"step": 11200
},
{
"epoch": 0.53,
"learning_rate": 0.00017029677198576599,
"loss": 0.0164,
"step": 11400
},
{
"epoch": 0.53,
"learning_rate": 0.00016495947422103804,
"loss": 0.0191,
"step": 11600
},
{
"epoch": 0.54,
"learning_rate": 0.00015963905811483306,
"loss": 0.0195,
"step": 11800
},
{
"epoch": 0.55,
"learning_rate": 0.0001543399879927535,
"loss": 0.0188,
"step": 12000
},
{
"epoch": 0.56,
"learning_rate": 0.00014906671026912937,
"loss": 0.0185,
"step": 12200
},
{
"epoch": 0.57,
"learning_rate": 0.0001438236497160614,
"loss": 0.0189,
"step": 12400
},
{
"epoch": 0.58,
"learning_rate": 0.0001386152057506243,
"loss": 0.0194,
"step": 12600
},
{
"epoch": 0.59,
"learning_rate": 0.00013344574874334572,
"loss": 0.0191,
"step": 12800
},
{
"epoch": 0.6,
"learning_rate": 0.00012831961635105695,
"loss": 0.0172,
"step": 13000
},
{
"epoch": 0.61,
"learning_rate": 0.00012326637687353276,
"loss": 0.0179,
"step": 13200
},
{
"epoch": 0.62,
"learning_rate": 0.00011826448612580445,
"loss": 0.0196,
"step": 13400
},
{
"epoch": 0.63,
"learning_rate": 0.00011329339019600353,
"loss": 0.0182,
"step": 13600
},
{
"epoch": 0.64,
"learning_rate": 0.00010838252859058713,
"loss": 0.0191,
"step": 13800
},
{
"epoch": 0.65,
"learning_rate": 0.00010353602198071279,
"loss": 0.0209,
"step": 14000
},
{
"epoch": 0.65,
"learning_rate": 9.875793703769172e-05,
"loss": 0.0196,
"step": 14200
},
{
"epoch": 0.66,
"learning_rate": 9.405228302067126e-05,
"loss": 0.0186,
"step": 14400
},
{
"epoch": 0.67,
"learning_rate": 8.942300841249284e-05,
"loss": 0.0186,
"step": 14600
},
{
"epoch": 0.68,
"learning_rate": 8.48739976065466e-05,
"loss": 0.0206,
"step": 14800
},
{
"epoch": 0.69,
"learning_rate": 8.040906764740216e-05,
"loss": 0.0191,
"step": 15000
},
{
"epoch": 0.7,
"learning_rate": 7.603196502795268e-05,
"loss": 0.0177,
"step": 15200
},
{
"epoch": 0.71,
"learning_rate": 7.17463625457574e-05,
"loss": 0.0167,
"step": 15400
},
{
"epoch": 0.72,
"learning_rate": 6.757656635075146e-05,
"loss": 0.0169,
"step": 15600
},
{
"epoch": 0.73,
"learning_rate": 6.348417073686787e-05,
"loss": 0.0187,
"step": 15800
},
{
"epoch": 0.74,
"learning_rate": 5.949380403058096e-05,
"loss": 0.0188,
"step": 16000
},
{
"epoch": 0.75,
"learning_rate": 5.560881452192967e-05,
"loss": 0.0185,
"step": 16200
},
{
"epoch": 0.76,
"learning_rate": 5.18324620796504e-05,
"loss": 0.0156,
"step": 16400
},
{
"epoch": 0.77,
"learning_rate": 4.816791541584292e-05,
"loss": 0.0157,
"step": 16600
},
{
"epoch": 0.77,
"learning_rate": 4.4618249427125566e-05,
"loss": 0.0159,
"step": 16800
},
{
"epoch": 0.78,
"learning_rate": 4.1186442614509987e-05,
"loss": 0.0163,
"step": 17000
},
{
"epoch": 0.79,
"learning_rate": 3.787537458416044e-05,
"loss": 0.0155,
"step": 17200
},
{
"epoch": 0.8,
"learning_rate": 3.468782363113597e-05,
"loss": 0.0175,
"step": 17400
},
{
"epoch": 0.81,
"learning_rate": 3.164145300855508e-05,
"loss": 0.0159,
"step": 17600
},
{
"epoch": 0.82,
"learning_rate": 2.8708204267220065e-05,
"loss": 0.0171,
"step": 17800
},
{
"epoch": 0.83,
"learning_rate": 2.590616471456609e-05,
"loss": 0.0203,
"step": 18000
},
{
"epoch": 0.84,
"learning_rate": 2.3237685523257935e-05,
"loss": 0.0177,
"step": 18200
},
{
"epoch": 0.85,
"learning_rate": 2.0705005796353695e-05,
"loss": 0.0189,
"step": 18400
},
{
"epoch": 0.86,
"learning_rate": 1.832187800404772e-05,
"loss": 0.018,
"step": 18600
},
{
"epoch": 0.87,
"learning_rate": 1.6066352463134496e-05,
"loss": 0.0173,
"step": 18800
},
{
"epoch": 0.88,
"learning_rate": 1.3952643804275625e-05,
"loss": 0.0173,
"step": 19000
},
{
"epoch": 0.89,
"learning_rate": 1.1982525626282391e-05,
"loss": 0.0168,
"step": 19200
},
{
"epoch": 0.89,
"learning_rate": 1.0157651042153225e-05,
"loss": 0.0168,
"step": 19400
},
{
"epoch": 0.9,
"learning_rate": 8.479551291957473e-06,
"loss": 0.0162,
"step": 19600
},
{
"epoch": 0.91,
"learning_rate": 6.949634457982115e-06,
"loss": 0.017,
"step": 19800
},
{
"epoch": 0.92,
"learning_rate": 5.569184283219286e-06,
"loss": 0.0169,
"step": 20000
},
{
"epoch": 0.93,
"learning_rate": 4.339359094186193e-06,
"loss": 0.0162,
"step": 20200
},
{
"epoch": 0.94,
"learning_rate": 3.2611908289809675e-06,
"loss": 0.0166,
"step": 20400
},
{
"epoch": 0.95,
"learning_rate": 2.3355841713904373e-06,
"loss": 0.0163,
"step": 20600
},
{
"epoch": 0.96,
"learning_rate": 1.5633157917758234e-06,
"loss": 0.0168,
"step": 20800
},
{
"epoch": 0.97,
"learning_rate": 9.450336953740333e-07,
"loss": 0.0168,
"step": 21000
},
{
"epoch": 0.98,
"learning_rate": 4.812566785606635e-07,
"loss": 0.0178,
"step": 21200
},
{
"epoch": 0.99,
"learning_rate": 1.7237389353147272e-07,
"loss": 0.0196,
"step": 21400
},
{
"epoch": 1.0,
"learning_rate": 1.864452176733613e-08,
"loss": 0.0154,
"step": 21600
},
{
"epoch": 1.0,
"eval_accuracy": 0.9653004077243341,
"eval_auc": 0.8861627985063656,
"eval_f1": 0.10229773443883387,
"eval_loss": 0.36874082684516907,
"eval_mcc": 0.20444634920778376,
"eval_precision": 0.05461181025780778,
"eval_recall": 0.8066353389360433,
"eval_runtime": 12018.8462,
"eval_samples_per_second": 43.311,
"eval_steps_per_second": 0.451,
"step": 21690
}
],
"logging_steps": 200,
"max_steps": 21690,
"num_train_epochs": 1,
"save_steps": 500,
"total_flos": 9.463065703778611e+17,
"trial_name": null,
"trial_params": null
}