| { | |
| "best_metric": 0.6904709935188293, | |
| "best_model_checkpoint": "bert_base_lda_100_v1_wnli/checkpoint-15", | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 30, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 2.2745001316070557, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.8315, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.43661971830985913, | |
| "eval_loss": 0.8521126508712769, | |
| "eval_runtime": 0.0533, | |
| "eval_samples_per_second": 1331.293, | |
| "eval_steps_per_second": 18.751, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.3648812770843506, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.7469, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5633802816901409, | |
| "eval_loss": 0.6963852643966675, | |
| "eval_runtime": 0.0539, | |
| "eval_samples_per_second": 1316.061, | |
| "eval_steps_per_second": 18.536, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.9801316857337952, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.7109, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.4647887323943662, | |
| "eval_loss": 0.7104423642158508, | |
| "eval_runtime": 0.0544, | |
| "eval_samples_per_second": 1304.828, | |
| "eval_steps_per_second": 18.378, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.6583738923072815, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.707, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.49295774647887325, | |
| "eval_loss": 0.6994937658309937, | |
| "eval_runtime": 0.0541, | |
| "eval_samples_per_second": 1313.46, | |
| "eval_steps_per_second": 18.499, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 2.0036449432373047, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.7035, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5633802816901409, | |
| "eval_loss": 0.6904709935188293, | |
| "eval_runtime": 0.0544, | |
| "eval_samples_per_second": 1306.075, | |
| "eval_steps_per_second": 18.395, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 1.6977219581604004, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.6961, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.4647887323943662, | |
| "eval_loss": 0.709176778793335, | |
| "eval_runtime": 0.0553, | |
| "eval_samples_per_second": 1282.999, | |
| "eval_steps_per_second": 18.07, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.5806699395179749, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.6981, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.5211267605633803, | |
| "eval_loss": 0.6949273347854614, | |
| "eval_runtime": 0.0538, | |
| "eval_samples_per_second": 1319.273, | |
| "eval_steps_per_second": 18.581, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 2.4361250400543213, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.692, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.39436619718309857, | |
| "eval_loss": 0.702409565448761, | |
| "eval_runtime": 0.0534, | |
| "eval_samples_per_second": 1330.805, | |
| "eval_steps_per_second": 18.744, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 0.9071682095527649, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.6974, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.4084507042253521, | |
| "eval_loss": 0.705875813961029, | |
| "eval_runtime": 0.0538, | |
| "eval_samples_per_second": 1319.01, | |
| "eval_steps_per_second": 18.578, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 1.0903393030166626, | |
| "learning_rate": 4e-05, | |
| "loss": 0.685, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.4788732394366197, | |
| "eval_loss": 0.6992462873458862, | |
| "eval_runtime": 0.0537, | |
| "eval_samples_per_second": 1321.891, | |
| "eval_steps_per_second": 18.618, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 30, | |
| "total_flos": 835377600768000.0, | |
| "train_loss": 0.7168338775634766, | |
| "train_runtime": 36.7707, | |
| "train_samples_per_second": 863.458, | |
| "train_steps_per_second": 4.079 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 150, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 5, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 5 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 835377600768000.0, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |