| { | |
| "best_metric": 0.06995197385549545, | |
| "best_model_checkpoint": "output_pipe/prom_300_notata/origin/checkpoint-800", | |
| "epoch": 4.0, | |
| "eval_steps": 200, | |
| "global_step": 2656, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.15060240963855423, | |
| "grad_norm": 1.585874080657959, | |
| "learning_rate": 2.9424405218726017e-05, | |
| "loss": 0.3415, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.30120481927710846, | |
| "grad_norm": 3.314279317855835, | |
| "learning_rate": 2.827321565617805e-05, | |
| "loss": 0.1374, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.30120481927710846, | |
| "eval_accuracy": 0.9636329376295459, | |
| "eval_f1": 0.9635801730549014, | |
| "eval_loss": 0.11874407529830933, | |
| "eval_matthews_correlation": 0.9286443161593607, | |
| "eval_precision": 0.9653044555881855, | |
| "eval_recall": 0.9633419342852427, | |
| "eval_runtime": 2.0547, | |
| "eval_samples_per_second": 2582.881, | |
| "eval_steps_per_second": 40.396, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.45180722891566266, | |
| "grad_norm": 1.1476503610610962, | |
| "learning_rate": 2.7122026093630083e-05, | |
| "loss": 0.1131, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.6024096385542169, | |
| "grad_norm": 1.4456894397735596, | |
| "learning_rate": 2.597083653108212e-05, | |
| "loss": 0.0844, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.6024096385542169, | |
| "eval_accuracy": 0.9715470133785566, | |
| "eval_f1": 0.971545978842977, | |
| "eval_loss": 0.07299650460481644, | |
| "eval_matthews_correlation": 0.9431284574559641, | |
| "eval_precision": 0.9715395817834918, | |
| "eval_recall": 0.9715888769607457, | |
| "eval_runtime": 2.0485, | |
| "eval_samples_per_second": 2590.731, | |
| "eval_steps_per_second": 40.518, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.7530120481927711, | |
| "grad_norm": 0.18486830592155457, | |
| "learning_rate": 2.481964696853415e-05, | |
| "loss": 0.0896, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.9036144578313253, | |
| "grad_norm": 0.22208698093891144, | |
| "learning_rate": 2.3668457405986186e-05, | |
| "loss": 0.0785, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.9036144578313253, | |
| "eval_accuracy": 0.9739966082532504, | |
| "eval_f1": 0.9739843921840996, | |
| "eval_loss": 0.08863991498947144, | |
| "eval_matthews_correlation": 0.948218039694491, | |
| "eval_precision": 0.974341521021252, | |
| "eval_recall": 0.9738766326350019, | |
| "eval_runtime": 2.0499, | |
| "eval_samples_per_second": 2588.955, | |
| "eval_steps_per_second": 40.491, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.0542168674698795, | |
| "grad_norm": 2.187502384185791, | |
| "learning_rate": 2.251726784343822e-05, | |
| "loss": 0.0722, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.2048192771084336, | |
| "grad_norm": 2.979804515838623, | |
| "learning_rate": 2.1366078280890252e-05, | |
| "loss": 0.0416, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.2048192771084336, | |
| "eval_accuracy": 0.9781420765027322, | |
| "eval_f1": 0.9781331874627035, | |
| "eval_loss": 0.06995197385549545, | |
| "eval_matthews_correlation": 0.9564543617799154, | |
| "eval_precision": 0.9784162660123144, | |
| "eval_recall": 0.9780381704999754, | |
| "eval_runtime": 2.0486, | |
| "eval_samples_per_second": 2590.596, | |
| "eval_steps_per_second": 40.516, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.355421686746988, | |
| "grad_norm": 3.7425546646118164, | |
| "learning_rate": 2.021488871834229e-05, | |
| "loss": 0.0336, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.5060240963855422, | |
| "grad_norm": 0.8297073245048523, | |
| "learning_rate": 1.9063699155794322e-05, | |
| "loss": 0.035, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.5060240963855422, | |
| "eval_accuracy": 0.97719992462785, | |
| "eval_f1": 0.9771963977277897, | |
| "eval_loss": 0.08680247515439987, | |
| "eval_matthews_correlation": 0.954401194573866, | |
| "eval_precision": 0.9772252037763616, | |
| "eval_recall": 0.9771759920662539, | |
| "eval_runtime": 2.0487, | |
| "eval_samples_per_second": 2590.457, | |
| "eval_steps_per_second": 40.514, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.6566265060240963, | |
| "grad_norm": 0.46684613823890686, | |
| "learning_rate": 1.7912509593246355e-05, | |
| "loss": 0.0367, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.8072289156626506, | |
| "grad_norm": 1.1904711723327637, | |
| "learning_rate": 1.6761320030698388e-05, | |
| "loss": 0.0452, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.8072289156626506, | |
| "eval_accuracy": 0.979272658752591, | |
| "eval_f1": 0.9792632572070812, | |
| "eval_loss": 0.07334966957569122, | |
| "eval_matthews_correlation": 0.9587606831731992, | |
| "eval_precision": 0.9796036311772153, | |
| "eval_recall": 0.9791571559531644, | |
| "eval_runtime": 2.048, | |
| "eval_samples_per_second": 2591.3, | |
| "eval_steps_per_second": 40.527, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.9578313253012047, | |
| "grad_norm": 2.6981256008148193, | |
| "learning_rate": 1.5610130468150424e-05, | |
| "loss": 0.0308, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.108433734939759, | |
| "grad_norm": 0.08918892592191696, | |
| "learning_rate": 1.4458940905602456e-05, | |
| "loss": 0.0174, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.108433734939759, | |
| "eval_accuracy": 0.978895798002638, | |
| "eval_f1": 0.9788908805296799, | |
| "eval_loss": 0.08836409449577332, | |
| "eval_matthews_correlation": 0.9578287746494739, | |
| "eval_precision": 0.9789860333608025, | |
| "eval_recall": 0.9788427520053816, | |
| "eval_runtime": 2.0489, | |
| "eval_samples_per_second": 2590.197, | |
| "eval_steps_per_second": 40.51, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.2590361445783134, | |
| "grad_norm": 0.012705490924417973, | |
| "learning_rate": 1.3307751343054489e-05, | |
| "loss": 0.0106, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.4096385542168672, | |
| "grad_norm": 2.8612990379333496, | |
| "learning_rate": 1.2156561780506524e-05, | |
| "loss": 0.0073, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.4096385542168672, | |
| "eval_accuracy": 0.9800263802524967, | |
| "eval_f1": 0.9800205057769635, | |
| "eval_loss": 0.09836893528699875, | |
| "eval_matthews_correlation": 0.960131261055992, | |
| "eval_precision": 0.9801773617563092, | |
| "eval_recall": 0.9799539252981333, | |
| "eval_runtime": 2.0486, | |
| "eval_samples_per_second": 2590.534, | |
| "eval_steps_per_second": 40.515, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.5602409638554215, | |
| "grad_norm": 0.3261496126651764, | |
| "learning_rate": 1.1005372217958557e-05, | |
| "loss": 0.0107, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.710843373493976, | |
| "grad_norm": 4.168089389801025, | |
| "learning_rate": 9.854182655410591e-06, | |
| "loss": 0.0058, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.710843373493976, | |
| "eval_accuracy": 0.9787073676276615, | |
| "eval_f1": 0.978705903879415, | |
| "eval_loss": 0.10657216608524323, | |
| "eval_matthews_correlation": 0.9574202184236169, | |
| "eval_precision": 0.9786936551154061, | |
| "eval_recall": 0.9787265638737861, | |
| "eval_runtime": 2.0494, | |
| "eval_samples_per_second": 2589.545, | |
| "eval_steps_per_second": 40.5, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.86144578313253, | |
| "grad_norm": 0.016729481518268585, | |
| "learning_rate": 8.702993092862626e-06, | |
| "loss": 0.0088, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 3.0120481927710845, | |
| "grad_norm": 0.018642086535692215, | |
| "learning_rate": 7.5518035303146585e-06, | |
| "loss": 0.0061, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.0120481927710845, | |
| "eval_accuracy": 0.9783305068777087, | |
| "eval_f1": 0.9783205031761819, | |
| "eval_loss": 0.10219753533601761, | |
| "eval_matthews_correlation": 0.9568830909886441, | |
| "eval_precision": 0.9786702504272362, | |
| "eval_recall": 0.9782129498348509, | |
| "eval_runtime": 2.0489, | |
| "eval_samples_per_second": 2590.175, | |
| "eval_steps_per_second": 40.51, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.1626506024096384, | |
| "grad_norm": 0.004093084018677473, | |
| "learning_rate": 6.4006139677666924e-06, | |
| "loss": 0.001, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 3.3132530120481927, | |
| "grad_norm": 0.012232447974383831, | |
| "learning_rate": 5.249424405218726e-06, | |
| "loss": 0.0024, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 3.3132530120481927, | |
| "eval_accuracy": 0.978895798002638, | |
| "eval_f1": 0.9788895910096218, | |
| "eval_loss": 0.11258693039417267, | |
| "eval_matthews_correlation": 0.957869327344797, | |
| "eval_precision": 0.9790461316777106, | |
| "eval_recall": 0.9788232216042883, | |
| "eval_runtime": 2.0486, | |
| "eval_samples_per_second": 2590.571, | |
| "eval_steps_per_second": 40.516, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 3.463855421686747, | |
| "grad_norm": 0.005885094869881868, | |
| "learning_rate": 4.09823484267076e-06, | |
| "loss": 0.0005, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 3.6144578313253013, | |
| "grad_norm": 0.00728481262922287, | |
| "learning_rate": 2.9470452801227938e-06, | |
| "loss": 0.0023, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.6144578313253013, | |
| "eval_accuracy": 0.9790842283776144, | |
| "eval_f1": 0.9790764991424794, | |
| "eval_loss": 0.11937826871871948, | |
| "eval_matthews_correlation": 0.9583067060360234, | |
| "eval_precision": 0.9793165728379114, | |
| "eval_recall": 0.9789901887787265, | |
| "eval_runtime": 2.048, | |
| "eval_samples_per_second": 2591.335, | |
| "eval_steps_per_second": 40.528, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.765060240963855, | |
| "grad_norm": 0.00713815912604332, | |
| "learning_rate": 1.7958557175748275e-06, | |
| "loss": 0.0023, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.9156626506024095, | |
| "grad_norm": 0.007936985231935978, | |
| "learning_rate": 6.446661550268611e-07, | |
| "loss": 0.0058, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 3.9156626506024095, | |
| "eval_accuracy": 0.979272658752591, | |
| "eval_f1": 0.9792660147189736, | |
| "eval_loss": 0.11451391130685806, | |
| "eval_matthews_correlation": 0.9586433697976791, | |
| "eval_precision": 0.979451094051508, | |
| "eval_recall": 0.9791923106751326, | |
| "eval_runtime": 2.0485, | |
| "eval_samples_per_second": 2590.613, | |
| "eval_steps_per_second": 40.516, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 2656, | |
| "total_flos": 2.833046111484432e+16, | |
| "train_loss": 0.04600991761446538, | |
| "train_runtime": 331.0189, | |
| "train_samples_per_second": 512.986, | |
| "train_steps_per_second": 8.024 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 2656, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 200, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.833046111484432e+16, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |