| { | |
| "best_metric": 0.31690406799316406, | |
| "best_model_checkpoint": "/kaggle/working/hubert-agum960-amharic/checkpoint-3500", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 3760, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.13297872340425532, | |
| "grad_norm": 2.615074396133423, | |
| "learning_rate": 9.73404255319149e-06, | |
| "loss": 1.5743, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.26595744680851063, | |
| "grad_norm": 2.4918744564056396, | |
| "learning_rate": 9.46808510638298e-06, | |
| "loss": 1.3461, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.39893617021276595, | |
| "grad_norm": 2.8701541423797607, | |
| "learning_rate": 9.20212765957447e-06, | |
| "loss": 1.0072, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.5319148936170213, | |
| "grad_norm": 12.351414680480957, | |
| "learning_rate": 8.938829787234043e-06, | |
| "loss": 0.7362, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.6648936170212766, | |
| "grad_norm": 8.005541801452637, | |
| "learning_rate": 8.672872340425533e-06, | |
| "loss": 0.5275, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.6648936170212766, | |
| "eval_accuracy": 0.8482810258865356, | |
| "eval_loss": 0.4739724397659302, | |
| "eval_runtime": 120.9505, | |
| "eval_samples_per_second": 11.062, | |
| "eval_steps_per_second": 1.389, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.7978723404255319, | |
| "grad_norm": 4.672068119049072, | |
| "learning_rate": 8.406914893617022e-06, | |
| "loss": 0.4446, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.9308510638297872, | |
| "grad_norm": 7.660475254058838, | |
| "learning_rate": 8.140957446808512e-06, | |
| "loss": 0.3683, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.0638297872340425, | |
| "grad_norm": 3.2801108360290527, | |
| "learning_rate": 7.877659574468086e-06, | |
| "loss": 0.3025, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.196808510638298, | |
| "grad_norm": 11.70166015625, | |
| "learning_rate": 7.61436170212766e-06, | |
| "loss": 0.2542, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.3297872340425532, | |
| "grad_norm": 10.051458358764648, | |
| "learning_rate": 7.348404255319149e-06, | |
| "loss": 0.2923, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.3297872340425532, | |
| "eval_accuracy": 0.8998505473136902, | |
| "eval_loss": 0.3727114200592041, | |
| "eval_runtime": 121.4022, | |
| "eval_samples_per_second": 11.021, | |
| "eval_steps_per_second": 1.384, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.4627659574468086, | |
| "grad_norm": 2.711136817932129, | |
| "learning_rate": 7.0824468085106394e-06, | |
| "loss": 0.2151, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.5957446808510638, | |
| "grad_norm": 3.505892038345337, | |
| "learning_rate": 6.816489361702127e-06, | |
| "loss": 0.2303, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.728723404255319, | |
| "grad_norm": 0.5282111763954163, | |
| "learning_rate": 6.550531914893618e-06, | |
| "loss": 0.2323, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.8617021276595744, | |
| "grad_norm": 6.366661071777344, | |
| "learning_rate": 6.284574468085107e-06, | |
| "loss": 0.2029, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.9946808510638299, | |
| "grad_norm": 17.721065521240234, | |
| "learning_rate": 6.018617021276596e-06, | |
| "loss": 0.2168, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.9946808510638299, | |
| "eval_accuracy": 0.9073243737220764, | |
| "eval_loss": 0.41751202940940857, | |
| "eval_runtime": 120.4054, | |
| "eval_samples_per_second": 11.112, | |
| "eval_steps_per_second": 1.395, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.127659574468085, | |
| "grad_norm": 29.569236755371094, | |
| "learning_rate": 5.752659574468086e-06, | |
| "loss": 0.181, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.2606382978723403, | |
| "grad_norm": 3.7976977825164795, | |
| "learning_rate": 5.4867021276595745e-06, | |
| "loss": 0.1902, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.393617021276596, | |
| "grad_norm": 5.339010715484619, | |
| "learning_rate": 5.220744680851064e-06, | |
| "loss": 0.1711, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.526595744680851, | |
| "grad_norm": 20.87778091430664, | |
| "learning_rate": 4.954787234042554e-06, | |
| "loss": 0.1889, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.6595744680851063, | |
| "grad_norm": 8.609652519226074, | |
| "learning_rate": 4.6888297872340425e-06, | |
| "loss": 0.1442, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.6595744680851063, | |
| "eval_accuracy": 0.9312406778335571, | |
| "eval_loss": 0.3470732569694519, | |
| "eval_runtime": 121.4455, | |
| "eval_samples_per_second": 11.017, | |
| "eval_steps_per_second": 1.383, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.7925531914893615, | |
| "grad_norm": 10.679058074951172, | |
| "learning_rate": 4.422872340425532e-06, | |
| "loss": 0.1706, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2.925531914893617, | |
| "grad_norm": 0.32079407572746277, | |
| "learning_rate": 4.156914893617022e-06, | |
| "loss": 0.1333, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 3.0585106382978724, | |
| "grad_norm": 0.19917072355747223, | |
| "learning_rate": 3.890957446808511e-06, | |
| "loss": 0.1414, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 3.1914893617021276, | |
| "grad_norm": 3.178858757019043, | |
| "learning_rate": 3.625e-06, | |
| "loss": 0.1217, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.324468085106383, | |
| "grad_norm": 4.0875020027160645, | |
| "learning_rate": 3.3590425531914896e-06, | |
| "loss": 0.1341, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.324468085106383, | |
| "eval_accuracy": 0.927503764629364, | |
| "eval_loss": 0.3641144931316376, | |
| "eval_runtime": 120.3344, | |
| "eval_samples_per_second": 11.119, | |
| "eval_steps_per_second": 1.396, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.4574468085106385, | |
| "grad_norm": 2.2575433254241943, | |
| "learning_rate": 3.0957446808510637e-06, | |
| "loss": 0.1179, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 3.5904255319148937, | |
| "grad_norm": 2.2771668434143066, | |
| "learning_rate": 2.8297872340425537e-06, | |
| "loss": 0.1247, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 3.723404255319149, | |
| "grad_norm": 0.28244930505752563, | |
| "learning_rate": 2.563829787234043e-06, | |
| "loss": 0.1186, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 3.8563829787234045, | |
| "grad_norm": 10.92738151550293, | |
| "learning_rate": 2.297872340425532e-06, | |
| "loss": 0.1487, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 3.9893617021276597, | |
| "grad_norm": 3.5330650806427, | |
| "learning_rate": 2.0319148936170213e-06, | |
| "loss": 0.1136, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.9893617021276597, | |
| "eval_accuracy": 0.9431988000869751, | |
| "eval_loss": 0.3285652995109558, | |
| "eval_runtime": 121.4946, | |
| "eval_samples_per_second": 11.013, | |
| "eval_steps_per_second": 1.383, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.122340425531915, | |
| "grad_norm": 0.24657931923866272, | |
| "learning_rate": 1.7659574468085109e-06, | |
| "loss": 0.1157, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 4.25531914893617, | |
| "grad_norm": 4.8149800300598145, | |
| "learning_rate": 1.5e-06, | |
| "loss": 0.1127, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 4.388297872340425, | |
| "grad_norm": 12.910594940185547, | |
| "learning_rate": 1.2340425531914894e-06, | |
| "loss": 0.1159, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 4.5212765957446805, | |
| "grad_norm": 0.07863473147153854, | |
| "learning_rate": 9.680851063829788e-07, | |
| "loss": 0.1177, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 4.654255319148936, | |
| "grad_norm": 8.249691009521484, | |
| "learning_rate": 7.021276595744682e-07, | |
| "loss": 0.1102, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 4.654255319148936, | |
| "eval_accuracy": 0.939461886882782, | |
| "eval_loss": 0.31690406799316406, | |
| "eval_runtime": 121.4032, | |
| "eval_samples_per_second": 11.021, | |
| "eval_steps_per_second": 1.384, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 4.787234042553192, | |
| "grad_norm": 0.02278612181544304, | |
| "learning_rate": 4.361702127659575e-07, | |
| "loss": 0.0957, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 4.920212765957447, | |
| "grad_norm": 11.251379013061523, | |
| "learning_rate": 1.7021276595744683e-07, | |
| "loss": 0.1042, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 3760, | |
| "total_flos": 7.814523952420126e+18, | |
| "train_loss": 0.29202749221882923, | |
| "train_runtime": 8842.7154, | |
| "train_samples_per_second": 6.795, | |
| "train_steps_per_second": 0.425 | |
| } | |
| ], | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.814523952420126e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |