| { | |
| "best_metric": 0.311672180891037, | |
| "best_model_checkpoint": "/kaggle/working/wev2vec-base960-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": 2.105905532836914, | |
| "learning_rate": 9.73404255319149e-06, | |
| "loss": 1.6043, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.26595744680851063, | |
| "grad_norm": 4.957370281219482, | |
| "learning_rate": 9.46808510638298e-06, | |
| "loss": 1.4442, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.39893617021276595, | |
| "grad_norm": 4.0505690574646, | |
| "learning_rate": 9.204787234042554e-06, | |
| "loss": 1.2199, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.5319148936170213, | |
| "grad_norm": 3.2562971115112305, | |
| "learning_rate": 8.941489361702127e-06, | |
| "loss": 1.0649, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.6648936170212766, | |
| "grad_norm": 3.3451857566833496, | |
| "learning_rate": 8.675531914893619e-06, | |
| "loss": 0.8682, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.6648936170212766, | |
| "eval_accuracy": 0.7563527822494507, | |
| "eval_loss": 0.7632076144218445, | |
| "eval_runtime": 59.9292, | |
| "eval_samples_per_second": 22.326, | |
| "eval_steps_per_second": 2.803, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.7978723404255319, | |
| "grad_norm": 7.531016826629639, | |
| "learning_rate": 8.409574468085107e-06, | |
| "loss": 0.7386, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.9308510638297872, | |
| "grad_norm": 14.800432205200195, | |
| "learning_rate": 8.143617021276596e-06, | |
| "loss": 0.6506, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.0638297872340425, | |
| "grad_norm": 18.182409286499023, | |
| "learning_rate": 7.877659574468086e-06, | |
| "loss": 0.5668, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.196808510638298, | |
| "grad_norm": 11.308631896972656, | |
| "learning_rate": 7.6117021276595745e-06, | |
| "loss": 0.4825, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.3297872340425532, | |
| "grad_norm": 3.6028597354888916, | |
| "learning_rate": 7.348404255319149e-06, | |
| "loss": 0.4482, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.3297872340425532, | |
| "eval_accuracy": 0.9103139042854309, | |
| "eval_loss": 0.35010650753974915, | |
| "eval_runtime": 60.1508, | |
| "eval_samples_per_second": 22.244, | |
| "eval_steps_per_second": 2.793, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.4627659574468086, | |
| "grad_norm": 13.081503868103027, | |
| "learning_rate": 7.0824468085106394e-06, | |
| "loss": 0.3877, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.5957446808510638, | |
| "grad_norm": 0.6290038228034973, | |
| "learning_rate": 6.816489361702127e-06, | |
| "loss": 0.3846, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.728723404255319, | |
| "grad_norm": 5.16023063659668, | |
| "learning_rate": 6.550531914893618e-06, | |
| "loss": 0.3553, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.8617021276595744, | |
| "grad_norm": 1.3795862197875977, | |
| "learning_rate": 6.284574468085107e-06, | |
| "loss": 0.3397, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.9946808510638299, | |
| "grad_norm": 17.843442916870117, | |
| "learning_rate": 6.018617021276596e-06, | |
| "loss": 0.2724, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.9946808510638299, | |
| "eval_accuracy": 0.9230194091796875, | |
| "eval_loss": 0.311672180891037, | |
| "eval_runtime": 60.1853, | |
| "eval_samples_per_second": 22.231, | |
| "eval_steps_per_second": 2.791, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.127659574468085, | |
| "grad_norm": 0.7912893295288086, | |
| "learning_rate": 5.755319148936171e-06, | |
| "loss": 0.228, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.2606382978723403, | |
| "grad_norm": 14.161015510559082, | |
| "learning_rate": 5.48936170212766e-06, | |
| "loss": 0.2633, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.393617021276596, | |
| "grad_norm": 12.629347801208496, | |
| "learning_rate": 5.223404255319149e-06, | |
| "loss": 0.2218, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.526595744680851, | |
| "grad_norm": 2.9135618209838867, | |
| "learning_rate": 4.957446808510639e-06, | |
| "loss": 0.2379, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.6595744680851063, | |
| "grad_norm": 14.890237808227539, | |
| "learning_rate": 4.691489361702128e-06, | |
| "loss": 0.2269, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.6595744680851063, | |
| "eval_accuracy": 0.9267563819885254, | |
| "eval_loss": 0.34563255310058594, | |
| "eval_runtime": 60.1272, | |
| "eval_samples_per_second": 22.253, | |
| "eval_steps_per_second": 2.794, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.7925531914893615, | |
| "grad_norm": 11.785533905029297, | |
| "learning_rate": 4.425531914893617e-06, | |
| "loss": 0.2319, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2.925531914893617, | |
| "grad_norm": 12.325530052185059, | |
| "learning_rate": 4.1595744680851066e-06, | |
| "loss": 0.1945, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 3.0585106382978724, | |
| "grad_norm": 27.31170082092285, | |
| "learning_rate": 3.893617021276596e-06, | |
| "loss": 0.1832, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 3.1914893617021276, | |
| "grad_norm": 21.935894012451172, | |
| "learning_rate": 3.6276595744680853e-06, | |
| "loss": 0.1982, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.324468085106383, | |
| "grad_norm": 3.217214822769165, | |
| "learning_rate": 3.3617021276595745e-06, | |
| "loss": 0.1663, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.324468085106383, | |
| "eval_accuracy": 0.927503764629364, | |
| "eval_loss": 0.37431710958480835, | |
| "eval_runtime": 60.1697, | |
| "eval_samples_per_second": 22.237, | |
| "eval_steps_per_second": 2.792, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.4574468085106385, | |
| "grad_norm": 0.2889520823955536, | |
| "learning_rate": 3.0957446808510637e-06, | |
| "loss": 0.1712, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 3.5904255319148937, | |
| "grad_norm": 19.394306182861328, | |
| "learning_rate": 2.8297872340425537e-06, | |
| "loss": 0.1615, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 3.723404255319149, | |
| "grad_norm": 0.08302600681781769, | |
| "learning_rate": 2.563829787234043e-06, | |
| "loss": 0.1709, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 3.8563829787234045, | |
| "grad_norm": 59.50064468383789, | |
| "learning_rate": 2.297872340425532e-06, | |
| "loss": 0.1614, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 3.9893617021276597, | |
| "grad_norm": 1.0584908723831177, | |
| "learning_rate": 2.0319148936170213e-06, | |
| "loss": 0.1737, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.9893617021276597, | |
| "eval_accuracy": 0.9327354431152344, | |
| "eval_loss": 0.37135931849479675, | |
| "eval_runtime": 60.1293, | |
| "eval_samples_per_second": 22.252, | |
| "eval_steps_per_second": 2.794, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.9893617021276597, | |
| "step": 3000, | |
| "total_flos": 1.8714792090048768e+18, | |
| "train_loss": 0.4606133778889974, | |
| "train_runtime": 3208.6756, | |
| "train_samples_per_second": 18.727, | |
| "train_steps_per_second": 1.172 | |
| } | |
| ], | |
| "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": 1.8714792090048768e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |