| { | |
| "best_global_step": 1497, | |
| "best_metric": 0.7373271889400922, | |
| "best_model_checkpoint": "./bert-tiny-heading-classifier\\checkpoint-1497", | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 1497, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.033400133600534405, | |
| "grad_norm": 4.833153247833252, | |
| "learning_rate": 2.45e-05, | |
| "loss": 0.6188, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.06680026720106881, | |
| "grad_norm": 1.177682876586914, | |
| "learning_rate": 4.9500000000000004e-05, | |
| "loss": 0.412, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.10020040080160321, | |
| "grad_norm": 1.452412724494934, | |
| "learning_rate": 4.979370158302459e-05, | |
| "loss": 0.2571, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.13360053440213762, | |
| "grad_norm": 1.0122822523117065, | |
| "learning_rate": 4.958319299427417e-05, | |
| "loss": 0.2227, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.16700066800267202, | |
| "grad_norm": 2.77670955657959, | |
| "learning_rate": 4.937268440552375e-05, | |
| "loss": 0.1801, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.20040080160320642, | |
| "grad_norm": 1.1728415489196777, | |
| "learning_rate": 4.9162175816773325e-05, | |
| "loss": 0.214, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.23380093520374082, | |
| "grad_norm": 1.3484389781951904, | |
| "learning_rate": 4.8951667228022907e-05, | |
| "loss": 0.183, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.26720106880427524, | |
| "grad_norm": 1.534583330154419, | |
| "learning_rate": 4.874115863927248e-05, | |
| "loss": 0.2521, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.30060120240480964, | |
| "grad_norm": 1.8582364320755005, | |
| "learning_rate": 4.853065005052206e-05, | |
| "loss": 0.2323, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.33400133600534404, | |
| "grad_norm": 1.7614449262619019, | |
| "learning_rate": 4.832014146177164e-05, | |
| "loss": 0.1383, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.36740146960587844, | |
| "grad_norm": 3.3362653255462646, | |
| "learning_rate": 4.810963287302122e-05, | |
| "loss": 0.1536, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.40080160320641284, | |
| "grad_norm": 0.37206974625587463, | |
| "learning_rate": 4.78991242842708e-05, | |
| "loss": 0.1457, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.43420173680694724, | |
| "grad_norm": 4.668218612670898, | |
| "learning_rate": 4.768861569552038e-05, | |
| "loss": 0.1306, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.46760187040748163, | |
| "grad_norm": 0.343142569065094, | |
| "learning_rate": 4.747810710676996e-05, | |
| "loss": 0.1527, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.501002004008016, | |
| "grad_norm": 1.6645559072494507, | |
| "learning_rate": 4.726759851801954e-05, | |
| "loss": 0.2129, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.5344021376085505, | |
| "grad_norm": 0.23233528435230255, | |
| "learning_rate": 4.7057089929269117e-05, | |
| "loss": 0.109, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.5678022712090849, | |
| "grad_norm": 0.19723999500274658, | |
| "learning_rate": 4.68465813405187e-05, | |
| "loss": 0.0968, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.6012024048096193, | |
| "grad_norm": 0.19006595015525818, | |
| "learning_rate": 4.663607275176827e-05, | |
| "loss": 0.1053, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.6346025384101537, | |
| "grad_norm": 0.1779528111219406, | |
| "learning_rate": 4.642556416301785e-05, | |
| "loss": 0.1444, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.6680026720106881, | |
| "grad_norm": 0.2541033923625946, | |
| "learning_rate": 4.621505557426743e-05, | |
| "loss": 0.1217, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.7014028056112225, | |
| "grad_norm": 0.24320191144943237, | |
| "learning_rate": 4.600454698551701e-05, | |
| "loss": 0.1758, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.7348029392117569, | |
| "grad_norm": 0.15954767167568207, | |
| "learning_rate": 4.579403839676659e-05, | |
| "loss": 0.0936, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.7682030728122913, | |
| "grad_norm": 0.17993752658367157, | |
| "learning_rate": 4.5583529808016164e-05, | |
| "loss": 0.098, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.8016032064128257, | |
| "grad_norm": 0.22059965133666992, | |
| "learning_rate": 4.5373021219265746e-05, | |
| "loss": 0.1256, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.8350033400133601, | |
| "grad_norm": 0.3228837549686432, | |
| "learning_rate": 4.516251263051533e-05, | |
| "loss": 0.1258, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.8684034736138945, | |
| "grad_norm": 7.371569633483887, | |
| "learning_rate": 4.495200404176491e-05, | |
| "loss": 0.1075, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.9018036072144289, | |
| "grad_norm": 0.17234943807125092, | |
| "learning_rate": 4.474149545301449e-05, | |
| "loss": 0.1442, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.9352037408149633, | |
| "grad_norm": 6.882818222045898, | |
| "learning_rate": 4.453098686426406e-05, | |
| "loss": 0.0995, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.9686038744154977, | |
| "grad_norm": 0.19052863121032715, | |
| "learning_rate": 4.4320478275513645e-05, | |
| "loss": 0.1307, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9730241362991008, | |
| "eval_f1": 0.7373271889400922, | |
| "eval_loss": 0.07805395871400833, | |
| "eval_runtime": 1.8035, | |
| "eval_samples_per_second": 1171.611, | |
| "eval_steps_per_second": 146.937, | |
| "step": 1497 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 11976, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 8, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7605135421440.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |