| { | |
| "best_metric": 0.6871697902679443, | |
| "best_model_checkpoint": "bert_base_lda_5_v1_wnli/checkpoint-12", | |
| "epoch": 9.0, | |
| "eval_steps": 500, | |
| "global_step": 27, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 12.35181713104248, | |
| "learning_rate": 0.00098, | |
| "loss": 1.1014, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5633802816901409, | |
| "eval_loss": 1.8102580308914185, | |
| "eval_runtime": 0.052, | |
| "eval_samples_per_second": 1365.859, | |
| "eval_steps_per_second": 19.237, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 26.631643295288086, | |
| "learning_rate": 0.00096, | |
| "loss": 1.4998, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5633802816901409, | |
| "eval_loss": 2.1869497299194336, | |
| "eval_runtime": 0.0545, | |
| "eval_samples_per_second": 1302.778, | |
| "eval_steps_per_second": 18.349, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 14.211752891540527, | |
| "learning_rate": 0.00094, | |
| "loss": 1.6341, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.43661971830985913, | |
| "eval_loss": 2.1260569095611572, | |
| "eval_runtime": 0.0529, | |
| "eval_samples_per_second": 1342.153, | |
| "eval_steps_per_second": 18.904, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 10.683871269226074, | |
| "learning_rate": 0.00092, | |
| "loss": 1.492, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5633802816901409, | |
| "eval_loss": 0.6871697902679443, | |
| "eval_runtime": 0.0527, | |
| "eval_samples_per_second": 1347.485, | |
| "eval_steps_per_second": 18.979, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 2.779585361480713, | |
| "learning_rate": 0.0009000000000000001, | |
| "loss": 0.7047, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.43661971830985913, | |
| "eval_loss": 0.7811673283576965, | |
| "eval_runtime": 0.1151, | |
| "eval_samples_per_second": 616.72, | |
| "eval_steps_per_second": 8.686, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.9340969324111938, | |
| "learning_rate": 0.00088, | |
| "loss": 0.7143, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5633802816901409, | |
| "eval_loss": 0.6881052851676941, | |
| "eval_runtime": 0.0525, | |
| "eval_samples_per_second": 1352.184, | |
| "eval_steps_per_second": 19.045, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 4.257674217224121, | |
| "learning_rate": 0.00086, | |
| "loss": 0.7199, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.43661971830985913, | |
| "eval_loss": 0.8079062104225159, | |
| "eval_runtime": 0.0527, | |
| "eval_samples_per_second": 1348.443, | |
| "eval_steps_per_second": 18.992, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 4.863579273223877, | |
| "learning_rate": 0.00084, | |
| "loss": 0.7574, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.5633802816901409, | |
| "eval_loss": 0.7103047370910645, | |
| "eval_runtime": 0.0522, | |
| "eval_samples_per_second": 1359.325, | |
| "eval_steps_per_second": 19.145, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 1.058039903640747, | |
| "learning_rate": 0.00082, | |
| "loss": 0.7273, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.43661971830985913, | |
| "eval_loss": 0.7278280854225159, | |
| "eval_runtime": 0.0525, | |
| "eval_samples_per_second": 1351.337, | |
| "eval_steps_per_second": 19.033, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "step": 27, | |
| "total_flos": 751839840691200.0, | |
| "train_loss": 1.0389759805467393, | |
| "train_runtime": 30.5356, | |
| "train_samples_per_second": 1039.771, | |
| "train_steps_per_second": 4.912 | |
| } | |
| ], | |
| "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": 751839840691200.0, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |