| { | |
| "best_global_step": 1584, | |
| "best_metric": 0.6938297152519226, | |
| "best_model_checkpoint": "bert_base_train_book_ent_15p_s_init_sst2/checkpoint-1584", | |
| "epoch": 11.0, | |
| "eval_steps": 500, | |
| "global_step": 2904, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 2.0667312145233154, | |
| "learning_rate": 4.9003787878787885e-05, | |
| "loss": 0.6916, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.4908256880733945, | |
| "eval_loss": 0.6944389939308167, | |
| "eval_runtime": 0.7443, | |
| "eval_samples_per_second": 1171.607, | |
| "eval_steps_per_second": 5.374, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.3363434076309204, | |
| "learning_rate": 4.800378787878788e-05, | |
| "loss": 0.6901, | |
| "step": 528 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5091743119266054, | |
| "eval_loss": 0.7226853966712952, | |
| "eval_runtime": 0.7329, | |
| "eval_samples_per_second": 1189.719, | |
| "eval_steps_per_second": 5.457, | |
| "step": 528 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.1986719369888306, | |
| "learning_rate": 4.7003787878787884e-05, | |
| "loss": 0.6894, | |
| "step": 792 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5091743119266054, | |
| "eval_loss": 0.7004013657569885, | |
| "eval_runtime": 0.7378, | |
| "eval_samples_per_second": 1181.875, | |
| "eval_steps_per_second": 5.421, | |
| "step": 792 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.7241930961608887, | |
| "learning_rate": 4.600378787878788e-05, | |
| "loss": 0.6897, | |
| "step": 1056 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5091743119266054, | |
| "eval_loss": 0.6942732334136963, | |
| "eval_runtime": 0.7283, | |
| "eval_samples_per_second": 1197.235, | |
| "eval_steps_per_second": 5.492, | |
| "step": 1056 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 2.013526439666748, | |
| "learning_rate": 4.5003787878787876e-05, | |
| "loss": 0.6891, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5091743119266054, | |
| "eval_loss": 0.6941343545913696, | |
| "eval_runtime": 0.7365, | |
| "eval_samples_per_second": 1183.902, | |
| "eval_steps_per_second": 5.431, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.3852760791778564, | |
| "learning_rate": 4.4003787878787885e-05, | |
| "loss": 0.6892, | |
| "step": 1584 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5091743119266054, | |
| "eval_loss": 0.6938297152519226, | |
| "eval_runtime": 0.7444, | |
| "eval_samples_per_second": 1171.473, | |
| "eval_steps_per_second": 5.374, | |
| "step": 1584 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 2.510197162628174, | |
| "learning_rate": 4.300378787878788e-05, | |
| "loss": 0.6886, | |
| "step": 1848 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.5091743119266054, | |
| "eval_loss": 0.7039402723312378, | |
| "eval_runtime": 0.7422, | |
| "eval_samples_per_second": 1174.886, | |
| "eval_steps_per_second": 5.389, | |
| "step": 1848 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 1.4384307861328125, | |
| "learning_rate": 4.2003787878787884e-05, | |
| "loss": 0.6882, | |
| "step": 2112 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.5091743119266054, | |
| "eval_loss": 0.7006253600120544, | |
| "eval_runtime": 0.7388, | |
| "eval_samples_per_second": 1180.256, | |
| "eval_steps_per_second": 5.414, | |
| "step": 2112 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 2.5491933822631836, | |
| "learning_rate": 4.100378787878788e-05, | |
| "loss": 0.6891, | |
| "step": 2376 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.5091743119266054, | |
| "eval_loss": 0.6943807601928711, | |
| "eval_runtime": 0.7307, | |
| "eval_samples_per_second": 1193.338, | |
| "eval_steps_per_second": 5.474, | |
| "step": 2376 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 1.1896880865097046, | |
| "learning_rate": 4.0003787878787876e-05, | |
| "loss": 0.6884, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.5091743119266054, | |
| "eval_loss": 0.7145212292671204, | |
| "eval_runtime": 0.7376, | |
| "eval_samples_per_second": 1182.238, | |
| "eval_steps_per_second": 5.423, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 1.1106557846069336, | |
| "learning_rate": 3.900378787878788e-05, | |
| "loss": 0.6884, | |
| "step": 2904 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.5091743119266054, | |
| "eval_loss": 0.7038775682449341, | |
| "eval_runtime": 0.7302, | |
| "eval_samples_per_second": 1194.216, | |
| "eval_steps_per_second": 5.478, | |
| "step": 2904 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "step": 2904, | |
| "total_flos": 9.746146557092352e+16, | |
| "train_loss": 0.6892486603792048, | |
| "train_runtime": 1339.7783, | |
| "train_samples_per_second": 2513.438, | |
| "train_steps_per_second": 9.852 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 13200, | |
| "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": 9.746146557092352e+16, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |