esm2_t6_8M_ptm_lora_500K / trainer_state.json
AmelieSchreiber's picture
Upload 11 files
00429e4
{
"best_metric": 0.0837364188034316,
"best_model_checkpoint": "esm2_t6_8M_lora_ptm_sites_2023-10-09_03-08-23/checkpoint-23309",
"epoch": 1.0,
"eval_steps": 500,
"global_step": 23309,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"learning_rate": 0.00037014933433791045,
"loss": 0.4563,
"step": 200
},
{
"epoch": 0.02,
"learning_rate": 0.0003701270704479516,
"loss": 0.2291,
"step": 400
},
{
"epoch": 0.03,
"learning_rate": 0.00037008979195075377,
"loss": 0.1662,
"step": 600
},
{
"epoch": 0.03,
"learning_rate": 0.0003700375763811688,
"loss": 0.1448,
"step": 800
},
{
"epoch": 0.04,
"learning_rate": 0.00036997042795487093,
"loss": 0.1194,
"step": 1000
},
{
"epoch": 0.05,
"learning_rate": 0.0003698887995939965,
"loss": 0.1104,
"step": 1200
},
{
"epoch": 0.06,
"learning_rate": 0.00036979187750890046,
"loss": 0.1126,
"step": 1400
},
{
"epoch": 0.07,
"learning_rate": 0.00036968004240382616,
"loss": 0.0939,
"step": 1600
},
{
"epoch": 0.08,
"learning_rate": 0.0003695533033078884,
"loss": 0.0985,
"step": 1800
},
{
"epoch": 0.09,
"learning_rate": 0.00036941167045348997,
"loss": 0.088,
"step": 2000
},
{
"epoch": 0.09,
"learning_rate": 0.0003692551552754955,
"loss": 0.0839,
"step": 2200
},
{
"epoch": 0.1,
"learning_rate": 0.0003690837704103084,
"loss": 0.0898,
"step": 2400
},
{
"epoch": 0.11,
"learning_rate": 0.00036889752969485046,
"loss": 0.0825,
"step": 2600
},
{
"epoch": 0.12,
"learning_rate": 0.00036869644816544503,
"loss": 0.0825,
"step": 2800
},
{
"epoch": 0.13,
"learning_rate": 0.00036848054205660264,
"loss": 0.0798,
"step": 3000
},
{
"epoch": 0.14,
"learning_rate": 0.00036824982879971063,
"loss": 0.0692,
"step": 3200
},
{
"epoch": 0.15,
"learning_rate": 0.0003680055912846709,
"loss": 0.0622,
"step": 3400
},
{
"epoch": 0.15,
"learning_rate": 0.00036774539459843937,
"loss": 0.0686,
"step": 3600
},
{
"epoch": 0.16,
"learning_rate": 0.00036747045011699506,
"loss": 0.0755,
"step": 3800
},
{
"epoch": 0.17,
"learning_rate": 0.00036718078003824503,
"loss": 0.0675,
"step": 4000
},
{
"epoch": 0.18,
"learning_rate": 0.0003668764077489815,
"loss": 0.0695,
"step": 4200
},
{
"epoch": 0.19,
"learning_rate": 0.00036655735782299405,
"loss": 0.0738,
"step": 4400
},
{
"epoch": 0.2,
"learning_rate": 0.00036622365601908513,
"loss": 0.0601,
"step": 4600
},
{
"epoch": 0.21,
"learning_rate": 0.0003658753292789907,
"loss": 0.0707,
"step": 4800
},
{
"epoch": 0.21,
"learning_rate": 0.000365512405725205,
"loss": 0.0642,
"step": 5000
},
{
"epoch": 0.22,
"learning_rate": 0.0003651349146587103,
"loss": 0.0529,
"step": 5200
},
{
"epoch": 0.23,
"learning_rate": 0.00036474288655661055,
"loss": 0.0527,
"step": 5400
},
{
"epoch": 0.24,
"learning_rate": 0.0003643384217654436,
"loss": 0.0625,
"step": 5600
},
{
"epoch": 0.25,
"learning_rate": 0.00036391748799477383,
"loss": 0.0546,
"step": 5800
},
{
"epoch": 0.26,
"learning_rate": 0.00036348211547861136,
"loss": 0.069,
"step": 6000
},
{
"epoch": 0.27,
"learning_rate": 0.00036303233936717483,
"loss": 0.0597,
"step": 6200
},
{
"epoch": 0.27,
"learning_rate": 0.00036256819597357105,
"loss": 0.0598,
"step": 6400
},
{
"epoch": 0.28,
"learning_rate": 0.00036208972277086324,
"loss": 0.0499,
"step": 6600
},
{
"epoch": 0.29,
"learning_rate": 0.00036159695838904537,
"loss": 0.0526,
"step": 6800
},
{
"epoch": 0.3,
"learning_rate": 0.0003610899426119236,
"loss": 0.0708,
"step": 7000
},
{
"epoch": 0.31,
"learning_rate": 0.00036056871637390423,
"loss": 0.0506,
"step": 7200
},
{
"epoch": 0.32,
"learning_rate": 0.0003600333217566887,
"loss": 0.0654,
"step": 7400
},
{
"epoch": 0.33,
"learning_rate": 0.00035948380198587616,
"loss": 0.0529,
"step": 7600
},
{
"epoch": 0.33,
"learning_rate": 0.0003589230543814262,
"loss": 0.0698,
"step": 7800
},
{
"epoch": 0.34,
"learning_rate": 0.00035834548859962624,
"loss": 0.0499,
"step": 8000
},
{
"epoch": 0.35,
"learning_rate": 0.0003577539339330621,
"loss": 0.0475,
"step": 8200
},
{
"epoch": 0.36,
"learning_rate": 0.0003571484381414685,
"loss": 0.0573,
"step": 8400
},
{
"epoch": 0.37,
"learning_rate": 0.00035652905011013025,
"loss": 0.0445,
"step": 8600
},
{
"epoch": 0.38,
"learning_rate": 0.00035589581984593567,
"loss": 0.0572,
"step": 8800
},
{
"epoch": 0.39,
"learning_rate": 0.00035524879847333905,
"loss": 0.0552,
"step": 9000
},
{
"epoch": 0.39,
"learning_rate": 0.00035458803823023286,
"loss": 0.0368,
"step": 9200
},
{
"epoch": 0.4,
"learning_rate": 0.0003539135924637307,
"loss": 0.0464,
"step": 9400
},
{
"epoch": 0.41,
"learning_rate": 0.00035322551562586,
"loss": 0.0478,
"step": 9600
},
{
"epoch": 0.42,
"learning_rate": 0.0003525238632691658,
"loss": 0.064,
"step": 9800
},
{
"epoch": 0.43,
"learning_rate": 0.0003518086920422257,
"loss": 0.0501,
"step": 10000
},
{
"epoch": 0.44,
"learning_rate": 0.00035108373623456347,
"loss": 0.0554,
"step": 10200
},
{
"epoch": 0.45,
"learning_rate": 0.000350341768437432,
"loss": 0.0433,
"step": 10400
},
{
"epoch": 0.45,
"learning_rate": 0.00034958645794357904,
"loss": 0.0546,
"step": 10600
},
{
"epoch": 0.46,
"learning_rate": 0.0003488178657337234,
"loss": 0.0556,
"step": 10800
},
{
"epoch": 0.47,
"learning_rate": 0.0003480360538608961,
"loss": 0.0597,
"step": 11000
},
{
"epoch": 0.48,
"learning_rate": 0.0003472410854454305,
"loss": 0.0526,
"step": 11200
},
{
"epoch": 0.49,
"learning_rate": 0.00034643302466986575,
"loss": 0.0559,
"step": 11400
},
{
"epoch": 0.5,
"learning_rate": 0.0003456119367737657,
"loss": 0.0514,
"step": 11600
},
{
"epoch": 0.51,
"learning_rate": 0.00034477788804845107,
"loss": 0.056,
"step": 11800
},
{
"epoch": 0.51,
"learning_rate": 0.00034393094583164754,
"loss": 0.0491,
"step": 12000
},
{
"epoch": 0.52,
"learning_rate": 0.00034307117850204926,
"loss": 0.0386,
"step": 12200
},
{
"epoch": 0.53,
"learning_rate": 0.0003421986554737983,
"loss": 0.047,
"step": 12400
},
{
"epoch": 0.54,
"learning_rate": 0.0003413134471908803,
"loss": 0.0587,
"step": 12600
},
{
"epoch": 0.55,
"learning_rate": 0.00034041562512143693,
"loss": 0.0442,
"step": 12800
},
{
"epoch": 0.56,
"learning_rate": 0.00033950526175199636,
"loss": 0.0517,
"step": 13000
},
{
"epoch": 0.57,
"learning_rate": 0.0003385824305816204,
"loss": 0.0446,
"step": 13200
},
{
"epoch": 0.57,
"learning_rate": 0.0003376472061159709,
"loss": 0.0525,
"step": 13400
},
{
"epoch": 0.58,
"learning_rate": 0.00033670443208693313,
"loss": 0.0482,
"step": 13600
},
{
"epoch": 0.59,
"learning_rate": 0.00033574470955847287,
"loss": 0.0489,
"step": 13800
},
{
"epoch": 0.6,
"learning_rate": 0.0003347728228408742,
"loss": 0.0684,
"step": 14000
},
{
"epoch": 0.61,
"learning_rate": 0.0003337888504003458,
"loss": 0.0506,
"step": 14200
},
{
"epoch": 0.62,
"learning_rate": 0.00033279287167884867,
"loss": 0.0563,
"step": 14400
},
{
"epoch": 0.63,
"learning_rate": 0.00033178496708768257,
"loss": 0.0356,
"step": 14600
},
{
"epoch": 0.63,
"learning_rate": 0.0003307652180009936,
"loss": 0.0573,
"step": 14800
},
{
"epoch": 0.64,
"learning_rate": 0.0003297337067492049,
"loss": 0.0498,
"step": 15000
},
{
"epoch": 0.65,
"learning_rate": 0.00032869051661236907,
"loss": 0.0493,
"step": 15200
},
{
"epoch": 0.66,
"learning_rate": 0.00032763573181344475,
"loss": 0.0479,
"step": 15400
},
{
"epoch": 0.67,
"learning_rate": 0.0003265694375114969,
"loss": 0.0497,
"step": 15600
},
{
"epoch": 0.68,
"learning_rate": 0.0003254917197948212,
"loss": 0.0543,
"step": 15800
},
{
"epoch": 0.69,
"learning_rate": 0.0003244081389988359,
"loss": 0.0537,
"step": 16000
},
{
"epoch": 0.7,
"learning_rate": 0.00032330789242185896,
"loss": 0.0743,
"step": 16200
},
{
"epoch": 0.7,
"learning_rate": 0.00032219648575413284,
"loss": 0.0346,
"step": 16400
},
{
"epoch": 0.71,
"learning_rate": 0.0003210740087261434,
"loss": 0.0454,
"step": 16600
},
{
"epoch": 0.72,
"learning_rate": 0.00031994055196215285,
"loss": 0.0563,
"step": 16800
},
{
"epoch": 0.73,
"learning_rate": 0.0003187962069728827,
"loss": 0.0456,
"step": 17000
},
{
"epoch": 0.74,
"learning_rate": 0.000317641066148126,
"loss": 0.0445,
"step": 17200
},
{
"epoch": 0.75,
"learning_rate": 0.0003164752227492878,
"loss": 0.0416,
"step": 17400
},
{
"epoch": 0.76,
"learning_rate": 0.00031529877090185585,
"loss": 0.0547,
"step": 17600
},
{
"epoch": 0.76,
"learning_rate": 0.00031411180558780117,
"loss": 0.0504,
"step": 17800
},
{
"epoch": 0.77,
"learning_rate": 0.0003129144226379097,
"loss": 0.0437,
"step": 18000
},
{
"epoch": 0.78,
"learning_rate": 0.00031171278275606484,
"loss": 0.0479,
"step": 18200
},
{
"epoch": 0.79,
"learning_rate": 0.0003104949062567452,
"loss": 0.0512,
"step": 18400
},
{
"epoch": 0.8,
"learning_rate": 0.0003092669041354337,
"loss": 0.0477,
"step": 18600
},
{
"epoch": 0.81,
"learning_rate": 0.0003080288755360624,
"loss": 0.0431,
"step": 18800
},
{
"epoch": 0.82,
"learning_rate": 0.0003067809204120605,
"loss": 0.05,
"step": 19000
},
{
"epoch": 0.82,
"learning_rate": 0.00030552313951828493,
"loss": 0.0557,
"step": 19200
},
{
"epoch": 0.83,
"learning_rate": 0.0003042556344028853,
"loss": 0.0388,
"step": 19400
},
{
"epoch": 0.84,
"learning_rate": 0.00030297850739910594,
"loss": 0.0515,
"step": 19600
},
{
"epoch": 0.85,
"learning_rate": 0.00030169186161702326,
"loss": 0.0438,
"step": 19800
},
{
"epoch": 0.86,
"learning_rate": 0.0003003958009352217,
"loss": 0.0512,
"step": 20000
},
{
"epoch": 0.87,
"learning_rate": 0.0002990904299924064,
"loss": 0.0499,
"step": 20200
},
{
"epoch": 0.88,
"learning_rate": 0.00029778244977988057,
"loss": 0.0505,
"step": 20400
},
{
"epoch": 0.88,
"learning_rate": 0.000296458820457796,
"loss": 0.0458,
"step": 20600
},
{
"epoch": 0.89,
"learning_rate": 0.0002951261987306019,
"loss": 0.0368,
"step": 20800
},
{
"epoch": 0.9,
"learning_rate": 0.0002937846921887947,
"loss": 0.0575,
"step": 21000
},
{
"epoch": 0.91,
"learning_rate": 0.00029243440914019487,
"loss": 0.0444,
"step": 21200
},
{
"epoch": 0.92,
"learning_rate": 0.0002910754586012025,
"loss": 0.0467,
"step": 21400
},
{
"epoch": 0.93,
"learning_rate": 0.00028970795028799613,
"loss": 0.0383,
"step": 21600
},
{
"epoch": 0.94,
"learning_rate": 0.0002883319946076741,
"loss": 0.0486,
"step": 21800
},
{
"epoch": 0.94,
"learning_rate": 0.0002869477026493413,
"loss": 0.0576,
"step": 22000
},
{
"epoch": 0.95,
"learning_rate": 0.0002855551861751398,
"loss": 0.0551,
"step": 22200
},
{
"epoch": 0.96,
"learning_rate": 0.00028415455761122624,
"loss": 0.0451,
"step": 22400
},
{
"epoch": 0.97,
"learning_rate": 0.00028274593003869416,
"loss": 0.0479,
"step": 22600
},
{
"epoch": 0.98,
"learning_rate": 0.0002813365191744383,
"loss": 0.0519,
"step": 22800
},
{
"epoch": 0.99,
"learning_rate": 0.0002799122739710694,
"loss": 0.049,
"step": 23000
},
{
"epoch": 1.0,
"learning_rate": 0.00027848037226393073,
"loss": 0.0489,
"step": 23200
},
{
"epoch": 1.0,
"eval_accuracy": 0.9363692882415535,
"eval_auc": 0.853910045782194,
"eval_f1": 0.0837364188034316,
"eval_loss": 0.6555151343345642,
"eval_mcc": 0.17516389459528245,
"eval_precision": 0.04427294640937209,
"eval_recall": 0.7708239966353413,
"eval_runtime": 1080.6969,
"eval_samples_per_second": 102.587,
"eval_steps_per_second": 5.399,
"step": 23309
}
],
"logging_steps": 200,
"max_steps": 69927,
"num_train_epochs": 3,
"save_steps": 500,
"total_flos": 1.97635804552818e+16,
"trial_name": null,
"trial_params": null
}