kcbert-sentiment / trainer_state.json
lkw2024's picture
Upload 11 files
cdab7e5 verified
{
"best_metric": 0.8851567498400512,
"best_model_checkpoint": "./checkpoints/beomi_kcbert-base\\checkpoint-4689",
"epoch": 3.0,
"eval_steps": 500,
"global_step": 4689,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.03198976327575176,
"grad_norm": 8.217513084411621,
"learning_rate": 1.9786734911494988e-05,
"loss": 0.89,
"step": 50
},
{
"epoch": 0.06397952655150352,
"grad_norm": 14.240294456481934,
"learning_rate": 1.9573469822989978e-05,
"loss": 0.6064,
"step": 100
},
{
"epoch": 0.09596928982725528,
"grad_norm": 13.745306015014648,
"learning_rate": 1.9360204734484968e-05,
"loss": 0.5601,
"step": 150
},
{
"epoch": 0.12795905310300704,
"grad_norm": 15.16039752960205,
"learning_rate": 1.9146939645979955e-05,
"loss": 0.4447,
"step": 200
},
{
"epoch": 0.1599488163787588,
"grad_norm": 13.24716567993164,
"learning_rate": 1.8933674557474945e-05,
"loss": 0.5015,
"step": 250
},
{
"epoch": 0.19193857965451055,
"grad_norm": 5.5529327392578125,
"learning_rate": 1.872040946896993e-05,
"loss": 0.4449,
"step": 300
},
{
"epoch": 0.22392834293026231,
"grad_norm": 13.999655723571777,
"learning_rate": 1.8507144380464918e-05,
"loss": 0.4436,
"step": 350
},
{
"epoch": 0.2559181062060141,
"grad_norm": 7.681272029876709,
"learning_rate": 1.8293879291959908e-05,
"loss": 0.4384,
"step": 400
},
{
"epoch": 0.28790786948176583,
"grad_norm": 10.44015121459961,
"learning_rate": 1.8080614203454897e-05,
"loss": 0.4037,
"step": 450
},
{
"epoch": 0.3198976327575176,
"grad_norm": 10.103533744812012,
"learning_rate": 1.7867349114949884e-05,
"loss": 0.4119,
"step": 500
},
{
"epoch": 0.35188739603326935,
"grad_norm": 10.231204986572266,
"learning_rate": 1.7654084026444874e-05,
"loss": 0.4122,
"step": 550
},
{
"epoch": 0.3838771593090211,
"grad_norm": 5.318388938903809,
"learning_rate": 1.744081893793986e-05,
"loss": 0.3774,
"step": 600
},
{
"epoch": 0.41586692258477287,
"grad_norm": 4.911611557006836,
"learning_rate": 1.7227553849434847e-05,
"loss": 0.4413,
"step": 650
},
{
"epoch": 0.44785668586052463,
"grad_norm": 9.968073844909668,
"learning_rate": 1.7014288760929837e-05,
"loss": 0.3951,
"step": 700
},
{
"epoch": 0.4798464491362764,
"grad_norm": 9.88436222076416,
"learning_rate": 1.6801023672424827e-05,
"loss": 0.3998,
"step": 750
},
{
"epoch": 0.5118362124120281,
"grad_norm": 9.33141803741455,
"learning_rate": 1.6587758583919813e-05,
"loss": 0.3632,
"step": 800
},
{
"epoch": 0.5438259756877799,
"grad_norm": 9.803372383117676,
"learning_rate": 1.6374493495414803e-05,
"loss": 0.4311,
"step": 850
},
{
"epoch": 0.5758157389635317,
"grad_norm": 6.760705947875977,
"learning_rate": 1.616122840690979e-05,
"loss": 0.4055,
"step": 900
},
{
"epoch": 0.6078055022392834,
"grad_norm": 6.0470290184021,
"learning_rate": 1.5947963318404776e-05,
"loss": 0.384,
"step": 950
},
{
"epoch": 0.6397952655150352,
"grad_norm": 11.182646751403809,
"learning_rate": 1.5734698229899766e-05,
"loss": 0.4165,
"step": 1000
},
{
"epoch": 0.6717850287907869,
"grad_norm": 15.545323371887207,
"learning_rate": 1.5521433141394756e-05,
"loss": 0.3785,
"step": 1050
},
{
"epoch": 0.7037747920665387,
"grad_norm": 9.248072624206543,
"learning_rate": 1.5308168052889743e-05,
"loss": 0.3709,
"step": 1100
},
{
"epoch": 0.7357645553422905,
"grad_norm": 1.7462440729141235,
"learning_rate": 1.5094902964384733e-05,
"loss": 0.4148,
"step": 1150
},
{
"epoch": 0.7677543186180422,
"grad_norm": 7.342015266418457,
"learning_rate": 1.488163787587972e-05,
"loss": 0.402,
"step": 1200
},
{
"epoch": 0.799744081893794,
"grad_norm": 8.367056846618652,
"learning_rate": 1.4668372787374708e-05,
"loss": 0.3647,
"step": 1250
},
{
"epoch": 0.8317338451695457,
"grad_norm": 20.332612991333008,
"learning_rate": 1.4455107698869698e-05,
"loss": 0.3474,
"step": 1300
},
{
"epoch": 0.8637236084452975,
"grad_norm": 5.561351776123047,
"learning_rate": 1.4241842610364684e-05,
"loss": 0.3877,
"step": 1350
},
{
"epoch": 0.8957133717210493,
"grad_norm": 6.848100185394287,
"learning_rate": 1.4028577521859672e-05,
"loss": 0.3215,
"step": 1400
},
{
"epoch": 0.927703134996801,
"grad_norm": 3.406212568283081,
"learning_rate": 1.3815312433354662e-05,
"loss": 0.3549,
"step": 1450
},
{
"epoch": 0.9596928982725528,
"grad_norm": 12.684820175170898,
"learning_rate": 1.3602047344849649e-05,
"loss": 0.3429,
"step": 1500
},
{
"epoch": 0.9916826615483045,
"grad_norm": 7.167674541473389,
"learning_rate": 1.3388782256344637e-05,
"loss": 0.2946,
"step": 1550
},
{
"epoch": 1.0,
"eval_accuracy": 0.8694817658349329,
"eval_f1": 0.8560898036594571,
"eval_loss": 0.3504190444946289,
"eval_runtime": 65.935,
"eval_samples_per_second": 47.41,
"eval_steps_per_second": 2.973,
"step": 1563
},
{
"epoch": 1.0236724248240563,
"grad_norm": 6.036362171173096,
"learning_rate": 1.3175517167839627e-05,
"loss": 0.2343,
"step": 1600
},
{
"epoch": 1.055662188099808,
"grad_norm": 12.397319793701172,
"learning_rate": 1.2962252079334613e-05,
"loss": 0.2676,
"step": 1650
},
{
"epoch": 1.0876519513755598,
"grad_norm": 14.472246170043945,
"learning_rate": 1.2748986990829602e-05,
"loss": 0.2351,
"step": 1700
},
{
"epoch": 1.1196417146513116,
"grad_norm": 4.455846786499023,
"learning_rate": 1.2535721902324592e-05,
"loss": 0.2011,
"step": 1750
},
{
"epoch": 1.1516314779270633,
"grad_norm": 2.5196011066436768,
"learning_rate": 1.2322456813819578e-05,
"loss": 0.2774,
"step": 1800
},
{
"epoch": 1.183621241202815,
"grad_norm": 4.304745197296143,
"learning_rate": 1.2109191725314566e-05,
"loss": 0.2192,
"step": 1850
},
{
"epoch": 1.2156110044785668,
"grad_norm": 3.972710609436035,
"learning_rate": 1.1895926636809556e-05,
"loss": 0.2787,
"step": 1900
},
{
"epoch": 1.2476007677543186,
"grad_norm": 10.098494529724121,
"learning_rate": 1.1682661548304543e-05,
"loss": 0.2402,
"step": 1950
},
{
"epoch": 1.2795905310300704,
"grad_norm": 3.2702131271362305,
"learning_rate": 1.1469396459799531e-05,
"loss": 0.2567,
"step": 2000
},
{
"epoch": 1.3115802943058221,
"grad_norm": 3.6833343505859375,
"learning_rate": 1.1256131371294521e-05,
"loss": 0.2429,
"step": 2050
},
{
"epoch": 1.3435700575815739,
"grad_norm": 8.543336868286133,
"learning_rate": 1.1042866282789508e-05,
"loss": 0.2178,
"step": 2100
},
{
"epoch": 1.3755598208573256,
"grad_norm": 8.93952751159668,
"learning_rate": 1.0829601194284496e-05,
"loss": 0.223,
"step": 2150
},
{
"epoch": 1.4075495841330774,
"grad_norm": 0.37753650546073914,
"learning_rate": 1.0616336105779486e-05,
"loss": 0.202,
"step": 2200
},
{
"epoch": 1.4395393474088292,
"grad_norm": 11.612641334533691,
"learning_rate": 1.0403071017274472e-05,
"loss": 0.242,
"step": 2250
},
{
"epoch": 1.471529110684581,
"grad_norm": 28.171804428100586,
"learning_rate": 1.018980592876946e-05,
"loss": 0.2103,
"step": 2300
},
{
"epoch": 1.5035188739603327,
"grad_norm": 4.860315322875977,
"learning_rate": 9.976540840264449e-06,
"loss": 0.2168,
"step": 2350
},
{
"epoch": 1.5355086372360844,
"grad_norm": 6.9845499992370605,
"learning_rate": 9.763275751759437e-06,
"loss": 0.2225,
"step": 2400
},
{
"epoch": 1.5674984005118362,
"grad_norm": 9.760077476501465,
"learning_rate": 9.550010663254427e-06,
"loss": 0.2069,
"step": 2450
},
{
"epoch": 1.599488163787588,
"grad_norm": 19.529109954833984,
"learning_rate": 9.336745574749414e-06,
"loss": 0.2546,
"step": 2500
},
{
"epoch": 1.6314779270633397,
"grad_norm": 13.040373802185059,
"learning_rate": 9.123480486244403e-06,
"loss": 0.2327,
"step": 2550
},
{
"epoch": 1.6634676903390915,
"grad_norm": 30.015762329101562,
"learning_rate": 8.910215397739392e-06,
"loss": 0.2421,
"step": 2600
},
{
"epoch": 1.6954574536148432,
"grad_norm": 7.028976917266846,
"learning_rate": 8.696950309234378e-06,
"loss": 0.227,
"step": 2650
},
{
"epoch": 1.727447216890595,
"grad_norm": 8.363308906555176,
"learning_rate": 8.483685220729368e-06,
"loss": 0.226,
"step": 2700
},
{
"epoch": 1.7594369801663468,
"grad_norm": 6.349699020385742,
"learning_rate": 8.270420132224356e-06,
"loss": 0.2472,
"step": 2750
},
{
"epoch": 1.7914267434420985,
"grad_norm": 6.191136837005615,
"learning_rate": 8.057155043719343e-06,
"loss": 0.2457,
"step": 2800
},
{
"epoch": 1.8234165067178503,
"grad_norm": 0.29677829146385193,
"learning_rate": 7.843889955214333e-06,
"loss": 0.2324,
"step": 2850
},
{
"epoch": 1.855406269993602,
"grad_norm": 10.708757400512695,
"learning_rate": 7.630624866709321e-06,
"loss": 0.2525,
"step": 2900
},
{
"epoch": 1.8873960332693538,
"grad_norm": 9.20060920715332,
"learning_rate": 7.4173597782043085e-06,
"loss": 0.2107,
"step": 2950
},
{
"epoch": 1.9193857965451055,
"grad_norm": 17.06194305419922,
"learning_rate": 7.204094689699297e-06,
"loss": 0.2025,
"step": 3000
},
{
"epoch": 1.9513755598208573,
"grad_norm": 8.665574073791504,
"learning_rate": 6.990829601194286e-06,
"loss": 0.2297,
"step": 3050
},
{
"epoch": 1.983365323096609,
"grad_norm": 6.722887992858887,
"learning_rate": 6.777564512689273e-06,
"loss": 0.195,
"step": 3100
},
{
"epoch": 2.0,
"eval_accuracy": 0.8829174664107485,
"eval_f1": 0.8726527951413109,
"eval_loss": 0.3595542907714844,
"eval_runtime": 66.0769,
"eval_samples_per_second": 47.309,
"eval_steps_per_second": 2.966,
"step": 3126
},
{
"epoch": 2.015355086372361,
"grad_norm": 7.179947853088379,
"learning_rate": 6.5642994241842614e-06,
"loss": 0.1669,
"step": 3150
},
{
"epoch": 2.0473448496481126,
"grad_norm": 12.716062545776367,
"learning_rate": 6.3510343356792505e-06,
"loss": 0.1207,
"step": 3200
},
{
"epoch": 2.0793346129238643,
"grad_norm": 10.967968940734863,
"learning_rate": 6.137769247174238e-06,
"loss": 0.1268,
"step": 3250
},
{
"epoch": 2.111324376199616,
"grad_norm": 0.44536763429641724,
"learning_rate": 5.924504158669226e-06,
"loss": 0.1287,
"step": 3300
},
{
"epoch": 2.143314139475368,
"grad_norm": 12.086326599121094,
"learning_rate": 5.711239070164215e-06,
"loss": 0.1008,
"step": 3350
},
{
"epoch": 2.1753039027511196,
"grad_norm": 21.13951301574707,
"learning_rate": 5.497973981659203e-06,
"loss": 0.1798,
"step": 3400
},
{
"epoch": 2.2072936660268714,
"grad_norm": 24.906211853027344,
"learning_rate": 5.284708893154191e-06,
"loss": 0.1484,
"step": 3450
},
{
"epoch": 2.239283429302623,
"grad_norm": 20.485675811767578,
"learning_rate": 5.07144380464918e-06,
"loss": 0.1568,
"step": 3500
},
{
"epoch": 2.271273192578375,
"grad_norm": 12.844356536865234,
"learning_rate": 4.858178716144167e-06,
"loss": 0.1197,
"step": 3550
},
{
"epoch": 2.3032629558541267,
"grad_norm": 16.984413146972656,
"learning_rate": 4.644913627639156e-06,
"loss": 0.1296,
"step": 3600
},
{
"epoch": 2.3352527191298784,
"grad_norm": 0.032408133149147034,
"learning_rate": 4.431648539134144e-06,
"loss": 0.1151,
"step": 3650
},
{
"epoch": 2.36724248240563,
"grad_norm": 0.1846800148487091,
"learning_rate": 4.218383450629132e-06,
"loss": 0.1797,
"step": 3700
},
{
"epoch": 2.399232245681382,
"grad_norm": 7.689809799194336,
"learning_rate": 4.005118362124121e-06,
"loss": 0.1725,
"step": 3750
},
{
"epoch": 2.4312220089571337,
"grad_norm": 10.447749137878418,
"learning_rate": 3.791853273619109e-06,
"loss": 0.1244,
"step": 3800
},
{
"epoch": 2.4632117722328855,
"grad_norm": 11.911157608032227,
"learning_rate": 3.5785881851140968e-06,
"loss": 0.1316,
"step": 3850
},
{
"epoch": 2.495201535508637,
"grad_norm": 8.964936256408691,
"learning_rate": 3.3653230966090854e-06,
"loss": 0.1384,
"step": 3900
},
{
"epoch": 2.527191298784389,
"grad_norm": 1.8026219606399536,
"learning_rate": 3.1520580081040737e-06,
"loss": 0.1148,
"step": 3950
},
{
"epoch": 2.5591810620601407,
"grad_norm": 13.85440444946289,
"learning_rate": 2.9387929195990615e-06,
"loss": 0.1436,
"step": 4000
},
{
"epoch": 2.5911708253358925,
"grad_norm": 0.33746129274368286,
"learning_rate": 2.72552783109405e-06,
"loss": 0.1175,
"step": 4050
},
{
"epoch": 2.6231605886116443,
"grad_norm": 23.944438934326172,
"learning_rate": 2.5122627425890384e-06,
"loss": 0.1361,
"step": 4100
},
{
"epoch": 2.655150351887396,
"grad_norm": 46.74479293823242,
"learning_rate": 2.2989976540840266e-06,
"loss": 0.1624,
"step": 4150
},
{
"epoch": 2.6871401151631478,
"grad_norm": 1.4517056941986084,
"learning_rate": 2.085732565579015e-06,
"loss": 0.1308,
"step": 4200
},
{
"epoch": 2.7191298784388995,
"grad_norm": 16.07023048400879,
"learning_rate": 1.872467477074003e-06,
"loss": 0.1477,
"step": 4250
},
{
"epoch": 2.7511196417146513,
"grad_norm": 0.09992417693138123,
"learning_rate": 1.6592023885689915e-06,
"loss": 0.1426,
"step": 4300
},
{
"epoch": 2.783109404990403,
"grad_norm": 24.815753936767578,
"learning_rate": 1.4459373000639796e-06,
"loss": 0.1494,
"step": 4350
},
{
"epoch": 2.815099168266155,
"grad_norm": 0.5908675789833069,
"learning_rate": 1.2326722115589678e-06,
"loss": 0.1505,
"step": 4400
},
{
"epoch": 2.8470889315419066,
"grad_norm": 19.279033660888672,
"learning_rate": 1.019407123053956e-06,
"loss": 0.1432,
"step": 4450
},
{
"epoch": 2.8790786948176583,
"grad_norm": 20.35901641845703,
"learning_rate": 8.061420345489445e-07,
"loss": 0.1479,
"step": 4500
},
{
"epoch": 2.91106845809341,
"grad_norm": 38.45198059082031,
"learning_rate": 5.928769460439326e-07,
"loss": 0.1184,
"step": 4550
},
{
"epoch": 2.943058221369162,
"grad_norm": 15.500663757324219,
"learning_rate": 3.796118575389209e-07,
"loss": 0.1638,
"step": 4600
},
{
"epoch": 2.9750479846449136,
"grad_norm": 7.924103260040283,
"learning_rate": 1.6634676903390917e-07,
"loss": 0.1474,
"step": 4650
},
{
"epoch": 3.0,
"eval_accuracy": 0.8851567498400512,
"eval_f1": 0.8744039386565773,
"eval_loss": 0.48145365715026855,
"eval_runtime": 67.22,
"eval_samples_per_second": 46.504,
"eval_steps_per_second": 2.916,
"step": 4689
}
],
"logging_steps": 50,
"max_steps": 4689,
"num_input_tokens_seen": 0,
"num_train_epochs": 3,
"save_steps": 500,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 766896986987838.0,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}