| { | |
| "best_metric": 0.43151227236737927, | |
| "best_model_checkpoint": "/scratch/mrahma45/pixel/finetuned_models/bert/bert-base-finetuned-pos-ud-Tamil-TTB/checkpoint-500", | |
| "epoch": 230.76923076923077, | |
| "global_step": 3000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 7.69, | |
| "learning_rate": 4.9500000000000004e-05, | |
| "loss": 1.8535, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 15.38, | |
| "learning_rate": 4.966778523489933e-05, | |
| "loss": 1.3671, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 23.08, | |
| "learning_rate": 4.933221476510068e-05, | |
| "loss": 1.0908, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 30.77, | |
| "learning_rate": 4.8996644295302016e-05, | |
| "loss": 0.8496, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 38.46, | |
| "learning_rate": 4.8661073825503355e-05, | |
| "loss": 0.6718, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 38.46, | |
| "eval_accuracy": 0.43151227236737927, | |
| "eval_loss": 2.5977065563201904, | |
| "eval_runtime": 0.4117, | |
| "eval_samples_per_second": 194.297, | |
| "eval_steps_per_second": 24.287, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 46.15, | |
| "learning_rate": 4.83255033557047e-05, | |
| "loss": 0.5325, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 53.85, | |
| "learning_rate": 4.798993288590604e-05, | |
| "loss": 0.433, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 61.54, | |
| "learning_rate": 4.765436241610739e-05, | |
| "loss": 0.3592, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 69.23, | |
| "learning_rate": 4.7318791946308726e-05, | |
| "loss": 0.3163, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 76.92, | |
| "learning_rate": 4.698322147651007e-05, | |
| "loss": 0.2805, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 76.92, | |
| "eval_accuracy": 0.4014251781472684, | |
| "eval_loss": 3.6307425498962402, | |
| "eval_runtime": 0.4118, | |
| "eval_samples_per_second": 194.246, | |
| "eval_steps_per_second": 24.281, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 84.62, | |
| "learning_rate": 4.664765100671141e-05, | |
| "loss": 0.2562, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 92.31, | |
| "learning_rate": 4.631208053691276e-05, | |
| "loss": 0.2323, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "learning_rate": 4.5976510067114097e-05, | |
| "loss": 0.2213, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 107.69, | |
| "learning_rate": 4.564093959731544e-05, | |
| "loss": 0.2116, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 115.38, | |
| "learning_rate": 4.5305369127516775e-05, | |
| "loss": 0.2013, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 115.38, | |
| "eval_accuracy": 0.38717339667458434, | |
| "eval_loss": 4.366728782653809, | |
| "eval_runtime": 0.4129, | |
| "eval_samples_per_second": 193.772, | |
| "eval_steps_per_second": 24.222, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 123.08, | |
| "learning_rate": 4.496979865771812e-05, | |
| "loss": 0.1972, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 130.77, | |
| "learning_rate": 4.463422818791946e-05, | |
| "loss": 0.1935, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 138.46, | |
| "learning_rate": 4.4298657718120806e-05, | |
| "loss": 0.1867, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 146.15, | |
| "learning_rate": 4.3963087248322146e-05, | |
| "loss": 0.183, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 153.85, | |
| "learning_rate": 4.362751677852349e-05, | |
| "loss": 0.1832, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 153.85, | |
| "eval_accuracy": 0.38796516231195566, | |
| "eval_loss": 4.474318504333496, | |
| "eval_runtime": 0.4157, | |
| "eval_samples_per_second": 192.457, | |
| "eval_steps_per_second": 24.057, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 161.54, | |
| "learning_rate": 4.329194630872484e-05, | |
| "loss": 0.1777, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 169.23, | |
| "learning_rate": 4.295637583892618e-05, | |
| "loss": 0.1753, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 176.92, | |
| "learning_rate": 4.262080536912752e-05, | |
| "loss": 0.1754, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 184.62, | |
| "learning_rate": 4.228523489932886e-05, | |
| "loss": 0.1717, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 192.31, | |
| "learning_rate": 4.194966442953021e-05, | |
| "loss": 0.1687, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 192.31, | |
| "eval_accuracy": 0.3784639746634996, | |
| "eval_loss": 4.800501346588135, | |
| "eval_runtime": 0.4165, | |
| "eval_samples_per_second": 192.1, | |
| "eval_steps_per_second": 24.012, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 200.0, | |
| "learning_rate": 4.161409395973155e-05, | |
| "loss": 0.1673, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 207.69, | |
| "learning_rate": 4.127852348993289e-05, | |
| "loss": 0.1673, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 215.38, | |
| "learning_rate": 4.0942953020134226e-05, | |
| "loss": 0.1659, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 223.08, | |
| "learning_rate": 4.060738255033557e-05, | |
| "loss": 0.1675, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 230.77, | |
| "learning_rate": 4.027181208053691e-05, | |
| "loss": 0.1634, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 230.77, | |
| "eval_accuracy": 0.38954869358669836, | |
| "eval_loss": 4.857724666595459, | |
| "eval_runtime": 0.4142, | |
| "eval_samples_per_second": 193.136, | |
| "eval_steps_per_second": 24.142, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 230.77, | |
| "step": 3000, | |
| "total_flos": 1.206309399453696e+16, | |
| "train_loss": 0.38402640787760417, | |
| "train_runtime": 743.2887, | |
| "train_samples_per_second": 645.779, | |
| "train_steps_per_second": 20.181 | |
| } | |
| ], | |
| "max_steps": 15000, | |
| "num_train_epochs": 1154, | |
| "total_flos": 1.206309399453696e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |