| { | |
| "best_metric": 0.6226160526275635, | |
| "best_model_checkpoint": "bert_base_lda_50_v1_mrpc/checkpoint-150", | |
| "epoch": 15.0, | |
| "eval_steps": 500, | |
| "global_step": 225, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 22.977022171020508, | |
| "learning_rate": 0.00098, | |
| "loss": 1.9018, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.3161764705882353, | |
| "eval_combined_score": 0.15808823529411764, | |
| "eval_f1": 0.0, | |
| "eval_loss": 1.746949315071106, | |
| "eval_runtime": 0.2773, | |
| "eval_samples_per_second": 1471.484, | |
| "eval_steps_per_second": 7.213, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 5.6883745193481445, | |
| "learning_rate": 0.00096, | |
| "loss": 0.7758, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6838235294117647, | |
| "eval_combined_score": 0.7480253018237863, | |
| "eval_f1": 0.8122270742358079, | |
| "eval_loss": 0.713627815246582, | |
| "eval_runtime": 0.2735, | |
| "eval_samples_per_second": 1491.747, | |
| "eval_steps_per_second": 7.312, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 2.137789249420166, | |
| "learning_rate": 0.00094, | |
| "loss": 0.6464, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6838235294117647, | |
| "eval_combined_score": 0.7480253018237863, | |
| "eval_f1": 0.8122270742358079, | |
| "eval_loss": 0.6401654481887817, | |
| "eval_runtime": 0.3133, | |
| "eval_samples_per_second": 1302.407, | |
| "eval_steps_per_second": 6.384, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.9102729558944702, | |
| "learning_rate": 0.00092, | |
| "loss": 0.6412, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6838235294117647, | |
| "eval_combined_score": 0.7480253018237863, | |
| "eval_f1": 0.8122270742358079, | |
| "eval_loss": 0.6260914206504822, | |
| "eval_runtime": 0.2816, | |
| "eval_samples_per_second": 1449.033, | |
| "eval_steps_per_second": 7.103, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.8048895001411438, | |
| "learning_rate": 0.0009000000000000001, | |
| "loss": 0.6318, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6838235294117647, | |
| "eval_combined_score": 0.7480253018237863, | |
| "eval_f1": 0.8122270742358079, | |
| "eval_loss": 0.6250861883163452, | |
| "eval_runtime": 0.2775, | |
| "eval_samples_per_second": 1470.21, | |
| "eval_steps_per_second": 7.207, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.975186288356781, | |
| "learning_rate": 0.00088, | |
| "loss": 0.6352, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.6838235294117647, | |
| "eval_combined_score": 0.7480253018237863, | |
| "eval_f1": 0.8122270742358079, | |
| "eval_loss": 0.6246122717857361, | |
| "eval_runtime": 0.2709, | |
| "eval_samples_per_second": 1506.067, | |
| "eval_steps_per_second": 7.383, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.4961640536785126, | |
| "learning_rate": 0.00086, | |
| "loss": 0.635, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.6838235294117647, | |
| "eval_combined_score": 0.7480253018237863, | |
| "eval_f1": 0.8122270742358079, | |
| "eval_loss": 0.6256270408630371, | |
| "eval_runtime": 0.2693, | |
| "eval_samples_per_second": 1515.101, | |
| "eval_steps_per_second": 7.427, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.3314763903617859, | |
| "learning_rate": 0.00084, | |
| "loss": 0.634, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.6838235294117647, | |
| "eval_combined_score": 0.7480253018237863, | |
| "eval_f1": 0.8122270742358079, | |
| "eval_loss": 0.6243536472320557, | |
| "eval_runtime": 0.2744, | |
| "eval_samples_per_second": 1487.06, | |
| "eval_steps_per_second": 7.29, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 0.5980406403541565, | |
| "learning_rate": 0.00082, | |
| "loss": 0.6297, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.6838235294117647, | |
| "eval_combined_score": 0.7480253018237863, | |
| "eval_f1": 0.8122270742358079, | |
| "eval_loss": 0.634880542755127, | |
| "eval_runtime": 0.2751, | |
| "eval_samples_per_second": 1482.984, | |
| "eval_steps_per_second": 7.27, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.4592650532722473, | |
| "learning_rate": 0.0008, | |
| "loss": 0.6357, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.6838235294117647, | |
| "eval_combined_score": 0.7480253018237863, | |
| "eval_f1": 0.8122270742358079, | |
| "eval_loss": 0.6226160526275635, | |
| "eval_runtime": 0.2684, | |
| "eval_samples_per_second": 1520.307, | |
| "eval_steps_per_second": 7.452, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 0.7528880834579468, | |
| "learning_rate": 0.0007800000000000001, | |
| "loss": 0.6341, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.6838235294117647, | |
| "eval_combined_score": 0.7480253018237863, | |
| "eval_f1": 0.8122270742358079, | |
| "eval_loss": 0.6255601048469543, | |
| "eval_runtime": 0.2722, | |
| "eval_samples_per_second": 1499.113, | |
| "eval_steps_per_second": 7.349, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 0.380775511264801, | |
| "learning_rate": 0.00076, | |
| "loss": 0.6333, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.6838235294117647, | |
| "eval_combined_score": 0.7480253018237863, | |
| "eval_f1": 0.8122270742358079, | |
| "eval_loss": 0.6249569058418274, | |
| "eval_runtime": 0.2722, | |
| "eval_samples_per_second": 1498.681, | |
| "eval_steps_per_second": 7.346, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 0.6090389490127563, | |
| "learning_rate": 0.00074, | |
| "loss": 0.6309, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.6838235294117647, | |
| "eval_combined_score": 0.7480253018237863, | |
| "eval_f1": 0.8122270742358079, | |
| "eval_loss": 0.6240330338478088, | |
| "eval_runtime": 0.2747, | |
| "eval_samples_per_second": 1485.187, | |
| "eval_steps_per_second": 7.28, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 1.031883716583252, | |
| "learning_rate": 0.0007199999999999999, | |
| "loss": 0.6343, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.6838235294117647, | |
| "eval_combined_score": 0.7480253018237863, | |
| "eval_f1": 0.8122270742358079, | |
| "eval_loss": 0.6258903741836548, | |
| "eval_runtime": 0.2733, | |
| "eval_samples_per_second": 1492.679, | |
| "eval_steps_per_second": 7.317, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.6238628029823303, | |
| "learning_rate": 0.0007, | |
| "loss": 0.6328, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.6838235294117647, | |
| "eval_combined_score": 0.7480253018237863, | |
| "eval_f1": 0.8122270742358079, | |
| "eval_loss": 0.6254164576530457, | |
| "eval_runtime": 0.2698, | |
| "eval_samples_per_second": 1512.061, | |
| "eval_steps_per_second": 7.412, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "step": 225, | |
| "total_flos": 7238185132953600.0, | |
| "train_loss": 0.7287869644165039, | |
| "train_runtime": 125.4342, | |
| "train_samples_per_second": 1462.121, | |
| "train_steps_per_second": 5.979 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 750, | |
| "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": 7238185132953600.0, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |