| { | |
| "best_metric": 0.32863062620162964, | |
| "best_model_checkpoint": "/kaggle/working/wev2vec-large960-agu-amharic/checkpoint-1500", | |
| "epoch": 3.9893617021276597, | |
| "eval_steps": 500, | |
| "global_step": 3000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.13297872340425532, | |
| "grad_norm": 14.737821578979492, | |
| "learning_rate": 9.736702127659574e-06, | |
| "loss": 1.5308, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.26595744680851063, | |
| "grad_norm": 14.251932144165039, | |
| "learning_rate": 9.470744680851066e-06, | |
| "loss": 1.314, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.39893617021276595, | |
| "grad_norm": 34.605953216552734, | |
| "learning_rate": 9.204787234042554e-06, | |
| "loss": 0.9622, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.5319148936170213, | |
| "grad_norm": 23.588228225708008, | |
| "learning_rate": 8.938829787234043e-06, | |
| "loss": 0.7054, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.6648936170212766, | |
| "grad_norm": 18.631229400634766, | |
| "learning_rate": 8.672872340425533e-06, | |
| "loss": 0.4808, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.6648936170212766, | |
| "eval_accuracy": 0.865470826625824, | |
| "eval_loss": 0.4616002142429352, | |
| "eval_runtime": 81.351, | |
| "eval_samples_per_second": 16.447, | |
| "eval_steps_per_second": 2.065, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.7978723404255319, | |
| "grad_norm": 19.866378784179688, | |
| "learning_rate": 8.409574468085107e-06, | |
| "loss": 0.4552, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.9308510638297872, | |
| "grad_norm": 15.608444213867188, | |
| "learning_rate": 8.143617021276596e-06, | |
| "loss": 0.3512, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.0638297872340425, | |
| "grad_norm": 16.031946182250977, | |
| "learning_rate": 7.877659574468086e-06, | |
| "loss": 0.2862, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.196808510638298, | |
| "grad_norm": 60.58535385131836, | |
| "learning_rate": 7.6117021276595745e-06, | |
| "loss": 0.2642, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.3297872340425532, | |
| "grad_norm": 0.7799405455589294, | |
| "learning_rate": 7.345744680851065e-06, | |
| "loss": 0.2885, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.3297872340425532, | |
| "eval_accuracy": 0.9103139042854309, | |
| "eval_loss": 0.4094645082950592, | |
| "eval_runtime": 81.0216, | |
| "eval_samples_per_second": 16.514, | |
| "eval_steps_per_second": 2.074, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.4627659574468086, | |
| "grad_norm": 27.568416595458984, | |
| "learning_rate": 7.079787234042554e-06, | |
| "loss": 0.2121, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.5957446808510638, | |
| "grad_norm": 5.639551639556885, | |
| "learning_rate": 6.813829787234043e-06, | |
| "loss": 0.1682, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.728723404255319, | |
| "grad_norm": 18.22026252746582, | |
| "learning_rate": 6.547872340425533e-06, | |
| "loss": 0.2316, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.8617021276595744, | |
| "grad_norm": 0.07559215277433395, | |
| "learning_rate": 6.281914893617022e-06, | |
| "loss": 0.1718, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.9946808510638299, | |
| "grad_norm": 44.52543640136719, | |
| "learning_rate": 6.015957446808511e-06, | |
| "loss": 0.1774, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.9946808510638299, | |
| "eval_accuracy": 0.9431988000869751, | |
| "eval_loss": 0.32863062620162964, | |
| "eval_runtime": 81.7868, | |
| "eval_samples_per_second": 16.36, | |
| "eval_steps_per_second": 2.054, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.127659574468085, | |
| "grad_norm": 42.418846130371094, | |
| "learning_rate": 5.75e-06, | |
| "loss": 0.1214, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.2606382978723403, | |
| "grad_norm": 27.88055992126465, | |
| "learning_rate": 5.48404255319149e-06, | |
| "loss": 0.1253, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.393617021276596, | |
| "grad_norm": 66.22528076171875, | |
| "learning_rate": 5.218085106382979e-06, | |
| "loss": 0.1408, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.526595744680851, | |
| "grad_norm": 59.407527923583984, | |
| "learning_rate": 4.952127659574468e-06, | |
| "loss": 0.124, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.6595744680851063, | |
| "grad_norm": 0.06157585233449936, | |
| "learning_rate": 4.686170212765958e-06, | |
| "loss": 0.1086, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.6595744680851063, | |
| "eval_accuracy": 0.9402092695236206, | |
| "eval_loss": 0.44243016839027405, | |
| "eval_runtime": 81.4569, | |
| "eval_samples_per_second": 16.426, | |
| "eval_steps_per_second": 2.062, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.7925531914893615, | |
| "grad_norm": 36.82463455200195, | |
| "learning_rate": 4.420212765957447e-06, | |
| "loss": 0.156, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2.925531914893617, | |
| "grad_norm": 15.869159698486328, | |
| "learning_rate": 4.154255319148937e-06, | |
| "loss": 0.0914, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 3.0585106382978724, | |
| "grad_norm": 83.17898559570312, | |
| "learning_rate": 3.8882978723404255e-06, | |
| "loss": 0.1335, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 3.1914893617021276, | |
| "grad_norm": 6.511538982391357, | |
| "learning_rate": 3.622340425531915e-06, | |
| "loss": 0.1073, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.324468085106383, | |
| "grad_norm": 2.9230759143829346, | |
| "learning_rate": 3.3563829787234047e-06, | |
| "loss": 0.0949, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.324468085106383, | |
| "eval_accuracy": 0.9402092695236206, | |
| "eval_loss": 0.4137694537639618, | |
| "eval_runtime": 81.4988, | |
| "eval_samples_per_second": 16.417, | |
| "eval_steps_per_second": 2.061, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.4574468085106385, | |
| "grad_norm": 0.04569636657834053, | |
| "learning_rate": 3.0930851063829792e-06, | |
| "loss": 0.0781, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 3.5904255319148937, | |
| "grad_norm": 0.057368163019418716, | |
| "learning_rate": 2.8271276595744684e-06, | |
| "loss": 0.0613, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 3.723404255319149, | |
| "grad_norm": 0.0105384960770607, | |
| "learning_rate": 2.563829787234043e-06, | |
| "loss": 0.0774, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 3.8563829787234045, | |
| "grad_norm": 0.411955326795578, | |
| "learning_rate": 2.297872340425532e-06, | |
| "loss": 0.0942, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 3.9893617021276597, | |
| "grad_norm": 0.037852853536605835, | |
| "learning_rate": 2.0319148936170213e-06, | |
| "loss": 0.0856, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.9893617021276597, | |
| "eval_accuracy": 0.9446935653686523, | |
| "eval_loss": 0.41552355885505676, | |
| "eval_runtime": 81.401, | |
| "eval_samples_per_second": 16.437, | |
| "eval_steps_per_second": 2.064, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.9893617021276597, | |
| "step": 3000, | |
| "total_flos": 6.236882326923493e+18, | |
| "train_loss": 0.3066477144559224, | |
| "train_runtime": 5458.1542, | |
| "train_samples_per_second": 11.009, | |
| "train_steps_per_second": 0.689 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 3760, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 3, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6.236882326923493e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |