| { | |
| "best_metric": 0.257554292678833, | |
| "best_model_checkpoint": "./checkpoint/checkpoint-45000", | |
| "epoch": 1.9922082521692934, | |
| "eval_steps": 1000, | |
| "global_step": 45000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 16.78230857849121, | |
| "learning_rate": 1.987315680989377e-05, | |
| "loss": 0.7008, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "eval_LOC_f1": 0.7083043365359647, | |
| "eval_ORG_f1": 0.5623498694516972, | |
| "eval_PER_f1": 0.7100647964924535, | |
| "eval_loss": 0.525841236114502, | |
| "eval_overall_accuracy": 0.839431747364137, | |
| "eval_overall_f1": 0.6684494885258021, | |
| "eval_overall_precision": 0.6813296398891967, | |
| "eval_overall_recall": 0.656047285763059, | |
| "eval_runtime": 178.5538, | |
| "eval_samples_per_second": 365.156, | |
| "eval_steps_per_second": 22.822, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 9.70958137512207, | |
| "learning_rate": 1.974631361978754e-05, | |
| "loss": 0.4865, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "eval_LOC_f1": 0.7016948849995092, | |
| "eval_ORG_f1": 0.590386287417712, | |
| "eval_PER_f1": 0.744081694250214, | |
| "eval_loss": 0.4681616723537445, | |
| "eval_overall_accuracy": 0.8518320678647284, | |
| "eval_overall_f1": 0.6854302821050324, | |
| "eval_overall_precision": 0.6610573746663495, | |
| "eval_overall_recall": 0.7116692272675543, | |
| "eval_runtime": 178.958, | |
| "eval_samples_per_second": 364.331, | |
| "eval_steps_per_second": 22.771, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 15.70914077758789, | |
| "learning_rate": 1.961947042968131e-05, | |
| "loss": 0.4382, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "eval_LOC_f1": 0.7578888199720886, | |
| "eval_ORG_f1": 0.6413910093299406, | |
| "eval_PER_f1": 0.7750323494978126, | |
| "eval_loss": 0.4217945635318756, | |
| "eval_overall_accuracy": 0.8756332018861953, | |
| "eval_overall_f1": 0.7298225286836603, | |
| "eval_overall_precision": 0.7249466651695486, | |
| "eval_overall_recall": 0.7347644247183339, | |
| "eval_runtime": 177.7042, | |
| "eval_samples_per_second": 366.902, | |
| "eval_steps_per_second": 22.931, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.9930562376976013, | |
| "learning_rate": 1.9492627239575078e-05, | |
| "loss": 0.3973, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "eval_LOC_f1": 0.7571125548125672, | |
| "eval_ORG_f1": 0.648735892760337, | |
| "eval_PER_f1": 0.7869654790165547, | |
| "eval_loss": 0.37678804993629456, | |
| "eval_overall_accuracy": 0.8835828734179773, | |
| "eval_overall_f1": 0.7360367198972751, | |
| "eval_overall_precision": 0.7341253555609017, | |
| "eval_overall_recall": 0.7379580630476841, | |
| "eval_runtime": 176.7693, | |
| "eval_samples_per_second": 368.842, | |
| "eval_steps_per_second": 23.053, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 10.682718276977539, | |
| "learning_rate": 1.9365784049468845e-05, | |
| "loss": 0.3937, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_LOC_f1": 0.7778496267121274, | |
| "eval_ORG_f1": 0.6430323934146428, | |
| "eval_PER_f1": 0.7986373262130558, | |
| "eval_loss": 0.4208361804485321, | |
| "eval_overall_accuracy": 0.8810407141986101, | |
| "eval_overall_f1": 0.7449750865636349, | |
| "eval_overall_precision": 0.7371928798261742, | |
| "eval_overall_recall": 0.7529233526800956, | |
| "eval_runtime": 175.2138, | |
| "eval_samples_per_second": 372.117, | |
| "eval_steps_per_second": 23.257, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 9.084446907043457, | |
| "learning_rate": 1.924106352298313e-05, | |
| "loss": 0.3536, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "eval_LOC_f1": 0.7828708236001125, | |
| "eval_ORG_f1": 0.6829649863421904, | |
| "eval_PER_f1": 0.799357419856545, | |
| "eval_loss": 0.36626189947128296, | |
| "eval_overall_accuracy": 0.8907383203725096, | |
| "eval_overall_f1": 0.760005685048323, | |
| "eval_overall_precision": 0.7663556759390631, | |
| "eval_overall_recall": 0.7537600608939571, | |
| "eval_runtime": 1096.1574, | |
| "eval_samples_per_second": 59.937, | |
| "eval_steps_per_second": 0.118, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 17.549041748046875, | |
| "learning_rate": 1.9114574110146982e-05, | |
| "loss": 0.3432, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eval_LOC_f1": 0.7952019006963426, | |
| "eval_ORG_f1": 0.6895359096523734, | |
| "eval_PER_f1": 0.8176749430952034, | |
| "eval_loss": 0.3385719060897827, | |
| "eval_overall_accuracy": 0.8962171463957795, | |
| "eval_overall_f1": 0.771180360157569, | |
| "eval_overall_precision": 0.7696968420461406, | |
| "eval_overall_recall": 0.772669607995151, | |
| "eval_runtime": 858.9522, | |
| "eval_samples_per_second": 76.489, | |
| "eval_steps_per_second": 0.15, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 24.70513916015625, | |
| "learning_rate": 1.8988084697310836e-05, | |
| "loss": 0.3278, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "eval_LOC_f1": 0.8013627254509018, | |
| "eval_ORG_f1": 0.6951269959460578, | |
| "eval_PER_f1": 0.8197644808913556, | |
| "eval_loss": 0.34438127279281616, | |
| "eval_overall_accuracy": 0.898079396458301, | |
| "eval_overall_f1": 0.7756412540713129, | |
| "eval_overall_precision": 0.7844467189482182, | |
| "eval_overall_recall": 0.7670312786321413, | |
| "eval_runtime": 858.3259, | |
| "eval_samples_per_second": 76.544, | |
| "eval_steps_per_second": 0.15, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 9.711076736450195, | |
| "learning_rate": 1.8861595284474693e-05, | |
| "loss": 0.3262, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_LOC_f1": 0.813213095471476, | |
| "eval_ORG_f1": 0.7107696972759894, | |
| "eval_PER_f1": 0.8219371637919723, | |
| "eval_loss": 0.3584470748901367, | |
| "eval_overall_accuracy": 0.898605619797311, | |
| "eval_overall_f1": 0.7873656115410829, | |
| "eval_overall_precision": 0.7973837314349728, | |
| "eval_overall_recall": 0.7775960982760808, | |
| "eval_runtime": 870.9363, | |
| "eval_samples_per_second": 75.436, | |
| "eval_steps_per_second": 0.148, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 1.0706424713134766, | |
| "learning_rate": 1.8735105871638546e-05, | |
| "loss": 0.3224, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_LOC_f1": 0.8127321363277717, | |
| "eval_ORG_f1": 0.7053217033370902, | |
| "eval_PER_f1": 0.8284614520690605, | |
| "eval_loss": 0.3149181008338928, | |
| "eval_overall_accuracy": 0.9003740874825833, | |
| "eval_overall_f1": 0.78723937113973, | |
| "eval_overall_precision": 0.7880037567703074, | |
| "eval_overall_recall": 0.7864764670228211, | |
| "eval_runtime": 905.7017, | |
| "eval_samples_per_second": 72.54, | |
| "eval_steps_per_second": 0.142, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 5.779551982879639, | |
| "learning_rate": 1.86086164588024e-05, | |
| "loss": 0.323, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "eval_LOC_f1": 0.8184425456746753, | |
| "eval_ORG_f1": 0.7099403917116094, | |
| "eval_PER_f1": 0.8274642760571583, | |
| "eval_loss": 0.32969576120376587, | |
| "eval_overall_accuracy": 0.8986934477379095, | |
| "eval_overall_f1": 0.787954918665525, | |
| "eval_overall_precision": 0.7845852840472364, | |
| "eval_overall_recall": 0.7913536219218246, | |
| "eval_runtime": 904.0471, | |
| "eval_samples_per_second": 72.673, | |
| "eval_steps_per_second": 0.143, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 0.5029723048210144, | |
| "learning_rate": 1.8482127045966253e-05, | |
| "loss": 0.307, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "eval_LOC_f1": 0.8216145583120352, | |
| "eval_ORG_f1": 0.7225526656894693, | |
| "eval_PER_f1": 0.8273035359038698, | |
| "eval_loss": 0.3300624191761017, | |
| "eval_overall_accuracy": 0.9015344583248979, | |
| "eval_overall_f1": 0.7950728742961997, | |
| "eval_overall_precision": 0.7871480673058326, | |
| "eval_overall_recall": 0.8031588740256262, | |
| "eval_runtime": 864.0266, | |
| "eval_samples_per_second": 76.039, | |
| "eval_steps_per_second": 0.149, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 3.1533098220825195, | |
| "learning_rate": 1.8355637633130106e-05, | |
| "loss": 0.3176, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_LOC_f1": 0.8250715828125148, | |
| "eval_ORG_f1": 0.7192632496804788, | |
| "eval_PER_f1": 0.8411864068909802, | |
| "eval_loss": 0.32228201627731323, | |
| "eval_overall_accuracy": 0.9048257731835991, | |
| "eval_overall_f1": 0.7993031971987119, | |
| "eval_overall_precision": 0.7982273018015168, | |
| "eval_overall_recall": 0.8003819968143439, | |
| "eval_runtime": 861.5394, | |
| "eval_samples_per_second": 76.259, | |
| "eval_steps_per_second": 0.15, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 27.121158599853516, | |
| "learning_rate": 1.8229148220293963e-05, | |
| "loss": 0.3026, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "eval_LOC_f1": 0.813366057700572, | |
| "eval_ORG_f1": 0.7255497858288067, | |
| "eval_PER_f1": 0.8334248928753359, | |
| "eval_loss": 0.31955230236053467, | |
| "eval_overall_accuracy": 0.9049277429112431, | |
| "eval_overall_f1": 0.7948808632215472, | |
| "eval_overall_precision": 0.7823057288325302, | |
| "eval_overall_recall": 0.8078668790437393, | |
| "eval_runtime": 870.9305, | |
| "eval_samples_per_second": 75.437, | |
| "eval_steps_per_second": 0.148, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 2.3121883869171143, | |
| "learning_rate": 1.8102658807457817e-05, | |
| "loss": 0.3148, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_LOC_f1": 0.8302503673286692, | |
| "eval_ORG_f1": 0.7304969576437721, | |
| "eval_PER_f1": 0.8389894805913836, | |
| "eval_loss": 0.3013811409473419, | |
| "eval_overall_accuracy": 0.9075328089459457, | |
| "eval_overall_f1": 0.8036533849292605, | |
| "eval_overall_precision": 0.8008272039526633, | |
| "eval_overall_recall": 0.8064995841732094, | |
| "eval_runtime": 915.5251, | |
| "eval_samples_per_second": 71.762, | |
| "eval_steps_per_second": 0.141, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 9.053048133850098, | |
| "learning_rate": 1.7976169394621674e-05, | |
| "loss": 0.2919, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "eval_LOC_f1": 0.8302478759061501, | |
| "eval_ORG_f1": 0.731782075129564, | |
| "eval_PER_f1": 0.8492362049256988, | |
| "eval_loss": 0.29880577325820923, | |
| "eval_overall_accuracy": 0.9072484845958724, | |
| "eval_overall_f1": 0.8070731973229327, | |
| "eval_overall_precision": 0.8054613737671545, | |
| "eval_overall_recall": 0.8086914847130795, | |
| "eval_runtime": 868.8985, | |
| "eval_samples_per_second": 75.613, | |
| "eval_steps_per_second": 0.148, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 10.181853294372559, | |
| "learning_rate": 1.7849679981785527e-05, | |
| "loss": 0.284, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eval_LOC_f1": 0.8266632184651193, | |
| "eval_ORG_f1": 0.7243731835870154, | |
| "eval_PER_f1": 0.8473984474857374, | |
| "eval_loss": 0.32140275835990906, | |
| "eval_overall_accuracy": 0.9021306462945541, | |
| "eval_overall_f1": 0.8045739646627278, | |
| "eval_overall_precision": 0.7991697840325968, | |
| "eval_overall_recall": 0.8100517316719056, | |
| "eval_runtime": 858.2577, | |
| "eval_samples_per_second": 76.55, | |
| "eval_steps_per_second": 0.15, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 5.25670862197876, | |
| "learning_rate": 1.772319056894938e-05, | |
| "loss": 0.3011, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_LOC_f1": 0.8265056399966245, | |
| "eval_ORG_f1": 0.7394759212033855, | |
| "eval_PER_f1": 0.8559909523111401, | |
| "eval_loss": 0.30769339203834534, | |
| "eval_overall_accuracy": 0.9103075764251944, | |
| "eval_overall_f1": 0.8115094912358725, | |
| "eval_overall_precision": 0.8128234861554996, | |
| "eval_overall_recall": 0.8101997378176846, | |
| "eval_runtime": 857.6527, | |
| "eval_samples_per_second": 76.604, | |
| "eval_steps_per_second": 0.15, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 3.439514398574829, | |
| "learning_rate": 1.7596701156113234e-05, | |
| "loss": 0.2913, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "eval_LOC_f1": 0.827463843173361, | |
| "eval_ORG_f1": 0.7474629390110435, | |
| "eval_PER_f1": 0.8510804051006602, | |
| "eval_loss": 0.3055616617202759, | |
| "eval_overall_accuracy": 0.9091010586988365, | |
| "eval_overall_f1": 0.8128348309192438, | |
| "eval_overall_precision": 0.817294400980434, | |
| "eval_overall_recall": 0.8084236640683365, | |
| "eval_runtime": 881.3321, | |
| "eval_samples_per_second": 74.546, | |
| "eval_steps_per_second": 0.146, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 4.887170791625977, | |
| "learning_rate": 1.747021174327709e-05, | |
| "loss": 0.2927, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "eval_LOC_f1": 0.8435358479700547, | |
| "eval_ORG_f1": 0.7490680162942847, | |
| "eval_PER_f1": 0.859867877786953, | |
| "eval_loss": 0.28957998752593994, | |
| "eval_overall_accuracy": 0.9142888616308011, | |
| "eval_overall_f1": 0.8220150311968236, | |
| "eval_overall_precision": 0.8269754768392371, | |
| "eval_overall_recall": 0.8171137391990753, | |
| "eval_runtime": 908.0817, | |
| "eval_samples_per_second": 72.35, | |
| "eval_steps_per_second": 0.142, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 24.44190216064453, | |
| "learning_rate": 1.7343722330440944e-05, | |
| "loss": 0.2833, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "eval_LOC_f1": 0.8368153269843636, | |
| "eval_ORG_f1": 0.7463167531278608, | |
| "eval_PER_f1": 0.8518246438142143, | |
| "eval_loss": 0.3060026466846466, | |
| "eval_overall_accuracy": 0.9097895404365793, | |
| "eval_overall_f1": 0.8159558988135214, | |
| "eval_overall_precision": 0.8099609727642672, | |
| "eval_overall_recall": 0.8220402294800051, | |
| "eval_runtime": 860.9742, | |
| "eval_samples_per_second": 76.309, | |
| "eval_steps_per_second": 0.15, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 9.784720420837402, | |
| "learning_rate": 1.7217232917604798e-05, | |
| "loss": 0.2862, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "eval_LOC_f1": 0.8464706167336854, | |
| "eval_ORG_f1": 0.7516536234342064, | |
| "eval_PER_f1": 0.8579948502396776, | |
| "eval_loss": 0.2908647060394287, | |
| "eval_overall_accuracy": 0.9129029665003394, | |
| "eval_overall_f1": 0.822899354947748, | |
| "eval_overall_precision": 0.8222161241750046, | |
| "eval_overall_recall": 0.823583722143129, | |
| "eval_runtime": 998.6892, | |
| "eval_samples_per_second": 65.786, | |
| "eval_steps_per_second": 0.515, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "grad_norm": 10.310678482055664, | |
| "learning_rate": 1.709074350476865e-05, | |
| "loss": 0.2578, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_LOC_f1": 0.841074796336982, | |
| "eval_ORG_f1": 0.7516604978958576, | |
| "eval_PER_f1": 0.8534542846463892, | |
| "eval_loss": 0.31625884771347046, | |
| "eval_overall_accuracy": 0.9131046730418835, | |
| "eval_overall_f1": 0.8203140861475114, | |
| "eval_overall_precision": 0.8292370287689482, | |
| "eval_overall_recall": 0.811581128511622, | |
| "eval_runtime": 992.6607, | |
| "eval_samples_per_second": 66.186, | |
| "eval_steps_per_second": 0.518, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "grad_norm": 12.428755760192871, | |
| "learning_rate": 1.6964254091932504e-05, | |
| "loss": 0.241, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "eval_LOC_f1": 0.8458245540438758, | |
| "eval_ORG_f1": 0.7417449698379979, | |
| "eval_PER_f1": 0.8449764637383674, | |
| "eval_loss": 0.31306418776512146, | |
| "eval_overall_accuracy": 0.910274082719034, | |
| "eval_overall_f1": 0.8166761890407407, | |
| "eval_overall_precision": 0.8123452830057111, | |
| "eval_overall_recall": 0.8210535218414784, | |
| "eval_runtime": 1000.0329, | |
| "eval_samples_per_second": 65.698, | |
| "eval_steps_per_second": 0.514, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "grad_norm": 7.654708385467529, | |
| "learning_rate": 1.683776467909636e-05, | |
| "loss": 0.2273, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "eval_LOC_f1": 0.8447393652873104, | |
| "eval_ORG_f1": 0.755644780920822, | |
| "eval_PER_f1": 0.8620841122450023, | |
| "eval_loss": 0.29408279061317444, | |
| "eval_overall_accuracy": 0.9140618487334913, | |
| "eval_overall_f1": 0.8249119505874042, | |
| "eval_overall_precision": 0.818763755528247, | |
| "eval_overall_recall": 0.8311531793129696, | |
| "eval_runtime": 941.7518, | |
| "eval_samples_per_second": 69.764, | |
| "eval_steps_per_second": 0.546, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "grad_norm": 0.1430547833442688, | |
| "learning_rate": 1.6711275266260215e-05, | |
| "loss": 0.2512, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "eval_LOC_f1": 0.8500541479262808, | |
| "eval_ORG_f1": 0.7624169717495258, | |
| "eval_PER_f1": 0.865798904312724, | |
| "eval_loss": 0.2906274199485779, | |
| "eval_overall_accuracy": 0.9169281656762454, | |
| "eval_overall_f1": 0.8295221228947797, | |
| "eval_overall_precision": 0.8304302989207211, | |
| "eval_overall_recall": 0.8286159310996152, | |
| "eval_runtime": 948.7716, | |
| "eval_samples_per_second": 69.247, | |
| "eval_steps_per_second": 0.542, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 2.0707778930664062, | |
| "learning_rate": 1.658478585342407e-05, | |
| "loss": 0.2355, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "eval_LOC_f1": 0.8528026151331602, | |
| "eval_ORG_f1": 0.7654581692212784, | |
| "eval_PER_f1": 0.8630401626842907, | |
| "eval_loss": 0.30151626467704773, | |
| "eval_overall_accuracy": 0.9169624036869872, | |
| "eval_overall_f1": 0.8306308694863459, | |
| "eval_overall_precision": 0.8223685119434121, | |
| "eval_overall_recall": 0.8390609362445908, | |
| "eval_runtime": 985.7403, | |
| "eval_samples_per_second": 66.65, | |
| "eval_steps_per_second": 0.521, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "grad_norm": 7.580867290496826, | |
| "learning_rate": 1.6458296440587925e-05, | |
| "loss": 0.2395, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "eval_LOC_f1": 0.8430476991285736, | |
| "eval_ORG_f1": 0.7574625494543183, | |
| "eval_PER_f1": 0.8637680570098303, | |
| "eval_loss": 0.3095172941684723, | |
| "eval_overall_accuracy": 0.914303747722428, | |
| "eval_overall_f1": 0.8256623551811597, | |
| "eval_overall_precision": 0.8230774618293879, | |
| "eval_overall_recall": 0.8282635355144271, | |
| "eval_runtime": 999.1758, | |
| "eval_samples_per_second": 65.754, | |
| "eval_steps_per_second": 0.514, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 0.15234589576721191, | |
| "learning_rate": 1.633180702775178e-05, | |
| "loss": 0.2322, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "eval_LOC_f1": 0.8442432749878247, | |
| "eval_ORG_f1": 0.7601534050352662, | |
| "eval_PER_f1": 0.8612064440586263, | |
| "eval_loss": 0.2927219569683075, | |
| "eval_overall_accuracy": 0.9170986114253731, | |
| "eval_overall_f1": 0.8258450109270158, | |
| "eval_overall_precision": 0.8182069728832319, | |
| "eval_overall_recall": 0.8336269963209901, | |
| "eval_runtime": 972.0367, | |
| "eval_samples_per_second": 67.59, | |
| "eval_steps_per_second": 0.529, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "grad_norm": 12.916611671447754, | |
| "learning_rate": 1.6205317614915632e-05, | |
| "loss": 0.2433, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "eval_LOC_f1": 0.8444784033582501, | |
| "eval_ORG_f1": 0.7596891611525098, | |
| "eval_PER_f1": 0.8642486197465865, | |
| "eval_loss": 0.2888515293598175, | |
| "eval_overall_accuracy": 0.91348501268295, | |
| "eval_overall_f1": 0.8271813480146815, | |
| "eval_overall_precision": 0.8160067477678877, | |
| "eval_overall_recall": 0.8386662531891801, | |
| "eval_runtime": 961.788, | |
| "eval_samples_per_second": 68.31, | |
| "eval_steps_per_second": 0.534, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "grad_norm": 0.4655471444129944, | |
| "learning_rate": 1.607882820207949e-05, | |
| "loss": 0.2358, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "eval_LOC_f1": 0.8478889967668045, | |
| "eval_ORG_f1": 0.7620526592328799, | |
| "eval_PER_f1": 0.8645434620748959, | |
| "eval_loss": 0.29194697737693787, | |
| "eval_overall_accuracy": 0.9150346548213073, | |
| "eval_overall_f1": 0.8278141773862643, | |
| "eval_overall_precision": 0.8183533624875177, | |
| "eval_overall_recall": 0.8374962998463555, | |
| "eval_runtime": 937.9473, | |
| "eval_samples_per_second": 70.047, | |
| "eval_steps_per_second": 0.548, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "grad_norm": 4.9739909172058105, | |
| "learning_rate": 1.5952338789243342e-05, | |
| "loss": 0.2258, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "eval_LOC_f1": 0.8552193079338859, | |
| "eval_ORG_f1": 0.763369485049633, | |
| "eval_PER_f1": 0.8649409287399143, | |
| "eval_loss": 0.2976300120353699, | |
| "eval_overall_accuracy": 0.9165649450405498, | |
| "eval_overall_f1": 0.8319835859514232, | |
| "eval_overall_precision": 0.8266665738956187, | |
| "eval_overall_recall": 0.8373694374356878, | |
| "eval_runtime": 961.624, | |
| "eval_samples_per_second": 68.322, | |
| "eval_steps_per_second": 0.535, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "grad_norm": 13.10745620727539, | |
| "learning_rate": 1.5825849376407196e-05, | |
| "loss": 0.2372, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "eval_LOC_f1": 0.8430370470368156, | |
| "eval_ORG_f1": 0.7732836874254703, | |
| "eval_PER_f1": 0.8708081138557172, | |
| "eval_loss": 0.2847268581390381, | |
| "eval_overall_accuracy": 0.91798879970466, | |
| "eval_overall_f1": 0.8324051096948847, | |
| "eval_overall_precision": 0.8326281644453847, | |
| "eval_overall_recall": 0.8321821744217188, | |
| "eval_runtime": 1011.399, | |
| "eval_samples_per_second": 64.96, | |
| "eval_steps_per_second": 0.508, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "grad_norm": 2.8712925910949707, | |
| "learning_rate": 1.569935996357105e-05, | |
| "loss": 0.2358, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "eval_LOC_f1": 0.8512200643745139, | |
| "eval_ORG_f1": 0.7646914589293013, | |
| "eval_PER_f1": 0.8725994049229104, | |
| "eval_loss": 0.28364238142967224, | |
| "eval_overall_accuracy": 0.9192392314013171, | |
| "eval_overall_f1": 0.8323193111236589, | |
| "eval_overall_precision": 0.8335312495582227, | |
| "eval_overall_recall": 0.831110891842747, | |
| "eval_runtime": 994.3994, | |
| "eval_samples_per_second": 66.07, | |
| "eval_steps_per_second": 0.517, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "grad_norm": 1.0189175605773926, | |
| "learning_rate": 1.5572870550734906e-05, | |
| "loss": 0.2314, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "eval_LOC_f1": 0.8507849855779469, | |
| "eval_ORG_f1": 0.7650251812241418, | |
| "eval_PER_f1": 0.8689379418372184, | |
| "eval_loss": 0.285095751285553, | |
| "eval_overall_accuracy": 0.9186356003858475, | |
| "eval_overall_f1": 0.8310027409855706, | |
| "eval_overall_precision": 0.8214160303503928, | |
| "eval_overall_recall": 0.8408158662588275, | |
| "eval_runtime": 935.5326, | |
| "eval_samples_per_second": 70.227, | |
| "eval_steps_per_second": 0.549, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "grad_norm": 3.663295269012451, | |
| "learning_rate": 1.544638113789876e-05, | |
| "loss": 0.234, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "eval_LOC_f1": 0.8537102746066029, | |
| "eval_ORG_f1": 0.7690944951636334, | |
| "eval_PER_f1": 0.8700584783258904, | |
| "eval_loss": 0.28537243604660034, | |
| "eval_overall_accuracy": 0.9194431708566052, | |
| "eval_overall_f1": 0.8341569064491033, | |
| "eval_overall_precision": 0.8295923488735222, | |
| "eval_overall_recall": 0.8387719718647365, | |
| "eval_runtime": 945.3448, | |
| "eval_samples_per_second": 69.498, | |
| "eval_steps_per_second": 0.544, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "grad_norm": 2.8524346351623535, | |
| "learning_rate": 1.5319891725062616e-05, | |
| "loss": 0.218, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "eval_LOC_f1": 0.8480498985496354, | |
| "eval_ORG_f1": 0.7720144752714113, | |
| "eval_PER_f1": 0.8681549995333838, | |
| "eval_loss": 0.28636589646339417, | |
| "eval_overall_accuracy": 0.9189809577115909, | |
| "eval_overall_f1": 0.8333121549691502, | |
| "eval_overall_precision": 0.8346791871137447, | |
| "eval_overall_recall": 0.8319495933354947, | |
| "eval_runtime": 971.8405, | |
| "eval_samples_per_second": 67.604, | |
| "eval_steps_per_second": 0.529, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "grad_norm": 4.014327049255371, | |
| "learning_rate": 1.5193402312226468e-05, | |
| "loss": 0.239, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "eval_LOC_f1": 0.856354181853526, | |
| "eval_ORG_f1": 0.7607628933739787, | |
| "eval_PER_f1": 0.8713080831892713, | |
| "eval_loss": 0.2887238562107086, | |
| "eval_overall_accuracy": 0.9170405556680282, | |
| "eval_overall_f1": 0.8318997253753843, | |
| "eval_overall_precision": 0.8238473394292507, | |
| "eval_overall_recall": 0.8401110750884513, | |
| "eval_runtime": 984.8035, | |
| "eval_samples_per_second": 66.714, | |
| "eval_steps_per_second": 0.522, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "grad_norm": 48.57558059692383, | |
| "learning_rate": 1.5066912899390323e-05, | |
| "loss": 0.2374, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "eval_LOC_f1": 0.8629891838741397, | |
| "eval_ORG_f1": 0.7734306355559895, | |
| "eval_PER_f1": 0.8760364502093424, | |
| "eval_loss": 0.28808096051216125, | |
| "eval_overall_accuracy": 0.921347101975682, | |
| "eval_overall_f1": 0.8414042377584563, | |
| "eval_overall_precision": 0.8495036133499992, | |
| "eval_overall_recall": 0.8334578464400998, | |
| "eval_runtime": 1002.3252, | |
| "eval_samples_per_second": 65.548, | |
| "eval_steps_per_second": 0.513, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "grad_norm": 18.295597076416016, | |
| "learning_rate": 1.4940423486554176e-05, | |
| "loss": 0.2327, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "eval_LOC_f1": 0.8568252173162995, | |
| "eval_ORG_f1": 0.765257865392274, | |
| "eval_PER_f1": 0.8650392170611746, | |
| "eval_loss": 0.27475544810295105, | |
| "eval_overall_accuracy": 0.918402633051887, | |
| "eval_overall_f1": 0.8338476462277082, | |
| "eval_overall_precision": 0.8316554025631678, | |
| "eval_overall_recall": 0.8360514779470842, | |
| "eval_runtime": 966.7516, | |
| "eval_samples_per_second": 67.96, | |
| "eval_steps_per_second": 0.532, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "grad_norm": 8.551658630371094, | |
| "learning_rate": 1.481393407371803e-05, | |
| "loss": 0.2266, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "eval_LOC_f1": 0.8626095376422174, | |
| "eval_ORG_f1": 0.7735328126106142, | |
| "eval_PER_f1": 0.8753750051370567, | |
| "eval_loss": 0.27126726508140564, | |
| "eval_overall_accuracy": 0.9233812863964941, | |
| "eval_overall_f1": 0.8405072872606979, | |
| "eval_overall_precision": 0.835852936051577, | |
| "eval_overall_recall": 0.845213763161975, | |
| "eval_runtime": 949.0619, | |
| "eval_samples_per_second": 69.226, | |
| "eval_steps_per_second": 0.542, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "grad_norm": 15.628961563110352, | |
| "learning_rate": 1.4687444660881885e-05, | |
| "loss": 0.2321, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "eval_LOC_f1": 0.8591221352011853, | |
| "eval_ORG_f1": 0.7785703957198631, | |
| "eval_PER_f1": 0.8756674763500703, | |
| "eval_loss": 0.2676946520805359, | |
| "eval_overall_accuracy": 0.9217490264496077, | |
| "eval_overall_f1": 0.8412990217616726, | |
| "eval_overall_precision": 0.8391781494337506, | |
| "eval_overall_recall": 0.8434306415009233, | |
| "eval_runtime": 933.8514, | |
| "eval_samples_per_second": 70.354, | |
| "eval_steps_per_second": 0.55, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 3.160637140274048, | |
| "learning_rate": 1.456095524804574e-05, | |
| "loss": 0.2103, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "eval_LOC_f1": 0.8572030651340997, | |
| "eval_ORG_f1": 0.7727868545480081, | |
| "eval_PER_f1": 0.8757355631042593, | |
| "eval_loss": 0.2812298536300659, | |
| "eval_overall_accuracy": 0.9198480725488561, | |
| "eval_overall_f1": 0.8381508761238307, | |
| "eval_overall_precision": 0.8330536854344057, | |
| "eval_overall_recall": 0.8433108270019594, | |
| "eval_runtime": 959.7072, | |
| "eval_samples_per_second": 68.458, | |
| "eval_steps_per_second": 0.536, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "grad_norm": 19.63943862915039, | |
| "learning_rate": 1.4434465835209595e-05, | |
| "loss": 0.2299, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "eval_LOC_f1": 0.8625562915195548, | |
| "eval_ORG_f1": 0.7738310591448732, | |
| "eval_PER_f1": 0.8727216526711657, | |
| "eval_loss": 0.2698276937007904, | |
| "eval_overall_accuracy": 0.9231393874075574, | |
| "eval_overall_f1": 0.8398996262454074, | |
| "eval_overall_precision": 0.8411791735887738, | |
| "eval_overall_recall": 0.8386239657189575, | |
| "eval_runtime": 959.292, | |
| "eval_samples_per_second": 68.488, | |
| "eval_steps_per_second": 0.536, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "grad_norm": 5.957523345947266, | |
| "learning_rate": 1.4307976422373449e-05, | |
| "loss": 0.217, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "eval_LOC_f1": 0.8548825805462975, | |
| "eval_ORG_f1": 0.7795103416137936, | |
| "eval_PER_f1": 0.8763479817569583, | |
| "eval_loss": 0.257554292678833, | |
| "eval_overall_accuracy": 0.9223258625001488, | |
| "eval_overall_f1": 0.8396542612618684, | |
| "eval_overall_precision": 0.8451246242708534, | |
| "eval_overall_recall": 0.834254260462625, | |
| "eval_runtime": 986.5064, | |
| "eval_samples_per_second": 66.599, | |
| "eval_steps_per_second": 0.521, | |
| "step": 45000 | |
| } | |
| ], | |
| "logging_steps": 1000, | |
| "max_steps": 158116, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "save_steps": 500, | |
| "total_flos": 1.2387920637624576e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |