| { | |
| "best_metric": 0.5861960041064519, | |
| "best_model_checkpoint": "/scratch/mrahma45/pixel/finetuned_models/bert/bert-base-finetuned-pos-ud-Chinese-GSD/checkpoint-500", | |
| "epoch": 24.0, | |
| "global_step": 3000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 4.9500000000000004e-05, | |
| "loss": 1.8173, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 4.966778523489933e-05, | |
| "loss": 1.3257, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 4.933221476510068e-05, | |
| "loss": 1.2541, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 4.8996644295302016e-05, | |
| "loss": 1.2279, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.8661073825503355e-05, | |
| "loss": 1.1842, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5861960041064519, | |
| "eval_loss": 1.2214140892028809, | |
| "eval_runtime": 2.5478, | |
| "eval_samples_per_second": 196.251, | |
| "eval_steps_per_second": 24.728, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 4.83255033557047e-05, | |
| "loss": 1.1457, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "learning_rate": 4.798993288590604e-05, | |
| "loss": 1.1089, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "learning_rate": 4.765436241610739e-05, | |
| "loss": 1.0922, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 7.2, | |
| "learning_rate": 4.7318791946308726e-05, | |
| "loss": 1.0397, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 4.698322147651007e-05, | |
| "loss": 1.0256, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.5806680881307748, | |
| "eval_loss": 1.2595834732055664, | |
| "eval_runtime": 2.5388, | |
| "eval_samples_per_second": 196.945, | |
| "eval_steps_per_second": 24.815, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 8.8, | |
| "learning_rate": 4.664765100671141e-05, | |
| "loss": 0.9657, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "learning_rate": 4.631208053691276e-05, | |
| "loss": 0.9376, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 10.4, | |
| "learning_rate": 4.5976510067114097e-05, | |
| "loss": 0.901, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 11.2, | |
| "learning_rate": 4.564093959731544e-05, | |
| "loss": 0.8583, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 4.5305369127516775e-05, | |
| "loss": 0.832, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.5698491668640923, | |
| "eval_loss": 1.4014437198638916, | |
| "eval_runtime": 2.5467, | |
| "eval_samples_per_second": 196.333, | |
| "eval_steps_per_second": 24.738, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 12.8, | |
| "learning_rate": 4.496979865771812e-05, | |
| "loss": 0.7809, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 13.6, | |
| "learning_rate": 4.463422818791946e-05, | |
| "loss": 0.7387, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 14.4, | |
| "learning_rate": 4.4298657718120806e-05, | |
| "loss": 0.7135, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 15.2, | |
| "learning_rate": 4.3963087248322146e-05, | |
| "loss": 0.6757, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 4.362751677852349e-05, | |
| "loss": 0.6519, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.5552396746426597, | |
| "eval_loss": 1.6267313957214355, | |
| "eval_runtime": 2.5372, | |
| "eval_samples_per_second": 197.067, | |
| "eval_steps_per_second": 24.83, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 16.8, | |
| "learning_rate": 4.329194630872484e-05, | |
| "loss": 0.6038, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 17.6, | |
| "learning_rate": 4.295637583892618e-05, | |
| "loss": 0.5758, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 18.4, | |
| "learning_rate": 4.262080536912752e-05, | |
| "loss": 0.545, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 19.2, | |
| "learning_rate": 4.228523489932886e-05, | |
| "loss": 0.5239, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 4.194966442953021e-05, | |
| "loss": 0.5023, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.550817341862118, | |
| "eval_loss": 1.8601970672607422, | |
| "eval_runtime": 2.5314, | |
| "eval_samples_per_second": 197.52, | |
| "eval_steps_per_second": 24.887, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 20.8, | |
| "learning_rate": 4.161409395973155e-05, | |
| "loss": 0.4627, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 21.6, | |
| "learning_rate": 4.127852348993289e-05, | |
| "loss": 0.4396, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 22.4, | |
| "learning_rate": 4.0942953020134226e-05, | |
| "loss": 0.4187, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 23.2, | |
| "learning_rate": 4.060738255033557e-05, | |
| "loss": 0.4056, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "learning_rate": 4.027181208053691e-05, | |
| "loss": 0.3811, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.5463160388533523, | |
| "eval_loss": 2.0793230533599854, | |
| "eval_runtime": 2.5286, | |
| "eval_samples_per_second": 197.741, | |
| "eval_steps_per_second": 24.915, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "step": 3000, | |
| "total_flos": 1.2534537269366784e+16, | |
| "train_loss": 0.8378373018900553, | |
| "train_runtime": 774.849, | |
| "train_samples_per_second": 619.476, | |
| "train_steps_per_second": 19.359 | |
| } | |
| ], | |
| "max_steps": 15000, | |
| "num_train_epochs": 120, | |
| "total_flos": 1.2534537269366784e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |