| { | |
| "best_metric": 0.7475538160469667, | |
| "best_model_checkpoint": "tiny-bert-sst2-distilled/run-2/checkpoint-1344", | |
| "epoch": 9.0, | |
| "eval_steps": 500, | |
| "global_step": 1728, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.0371571779251099, | |
| "learning_rate": 0.0001305282668790618, | |
| "loss": 0.3216, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5499021526418787, | |
| "eval_f1": 0.23841059602649006, | |
| "eval_loss": 0.2955350875854492, | |
| "eval_mcc": 0.17350853577425898, | |
| "eval_precision": 0.7741935483870968, | |
| "eval_recall": 0.14090019569471623, | |
| "eval_runtime": 67.6225, | |
| "eval_samples_per_second": 15.113, | |
| "eval_steps_per_second": 0.473, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.1835968494415283, | |
| "learning_rate": 0.00011421223351917907, | |
| "loss": 0.2979, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5587084148727984, | |
| "eval_f1": 0.2642740619902121, | |
| "eval_loss": 0.29196101427078247, | |
| "eval_mcc": 0.19586518668589045, | |
| "eval_precision": 0.7941176470588235, | |
| "eval_recall": 0.15851272015655576, | |
| "eval_runtime": 67.5618, | |
| "eval_samples_per_second": 15.127, | |
| "eval_steps_per_second": 0.474, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.4907313585281372, | |
| "learning_rate": 9.789620015929635e-05, | |
| "loss": 0.288, | |
| "step": 576 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6555772994129159, | |
| "eval_f1": 0.5243243243243243, | |
| "eval_loss": 0.28221988677978516, | |
| "eval_mcc": 0.37311506920915233, | |
| "eval_precision": 0.8471615720524017, | |
| "eval_recall": 0.3796477495107632, | |
| "eval_runtime": 67.1649, | |
| "eval_samples_per_second": 15.216, | |
| "eval_steps_per_second": 0.476, | |
| "step": 576 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.161053419113159, | |
| "learning_rate": 8.158016679941362e-05, | |
| "loss": 0.2856, | |
| "step": 768 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5968688845401174, | |
| "eval_f1": 0.3757575757575758, | |
| "eval_loss": 0.2846841514110565, | |
| "eval_mcc": 0.274495316321839, | |
| "eval_precision": 0.8322147651006712, | |
| "eval_recall": 0.24266144814090018, | |
| "eval_runtime": 67.4167, | |
| "eval_samples_per_second": 15.159, | |
| "eval_steps_per_second": 0.475, | |
| "step": 768 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.5259861946105957, | |
| "learning_rate": 6.52641334395309e-05, | |
| "loss": 0.2841, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6829745596868885, | |
| "eval_f1": 0.5759162303664921, | |
| "eval_loss": 0.28138452768325806, | |
| "eval_mcc": 0.42395345222624214, | |
| "eval_precision": 0.8695652173913043, | |
| "eval_recall": 0.43052837573385516, | |
| "eval_runtime": 67.041, | |
| "eval_samples_per_second": 15.244, | |
| "eval_steps_per_second": 0.477, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 1.6680471897125244, | |
| "learning_rate": 4.8948100079648175e-05, | |
| "loss": 0.2809, | |
| "step": 1152 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7270058708414873, | |
| "eval_f1": 0.6674612634088201, | |
| "eval_loss": 0.282234251499176, | |
| "eval_mcc": 0.48626311495705427, | |
| "eval_precision": 0.8536585365853658, | |
| "eval_recall": 0.547945205479452, | |
| "eval_runtime": 67.379, | |
| "eval_samples_per_second": 15.168, | |
| "eval_steps_per_second": 0.475, | |
| "step": 1152 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 1.0517845153808594, | |
| "learning_rate": 3.263206671976545e-05, | |
| "loss": 0.2793, | |
| "step": 1344 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.7475538160469667, | |
| "eval_f1": 0.7041284403669726, | |
| "eval_loss": 0.28309565782546997, | |
| "eval_mcc": 0.5179241840022546, | |
| "eval_precision": 0.850415512465374, | |
| "eval_recall": 0.6007827788649707, | |
| "eval_runtime": 67.566, | |
| "eval_samples_per_second": 15.126, | |
| "eval_steps_per_second": 0.474, | |
| "step": 1344 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 1.6533994674682617, | |
| "learning_rate": 1.6316033359882725e-05, | |
| "loss": 0.2774, | |
| "step": 1536 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7221135029354208, | |
| "eval_f1": 0.6594724220623502, | |
| "eval_loss": 0.282230943441391, | |
| "eval_mcc": 0.4777338430619903, | |
| "eval_precision": 0.8513931888544891, | |
| "eval_recall": 0.538160469667319, | |
| "eval_runtime": 67.39, | |
| "eval_samples_per_second": 15.165, | |
| "eval_steps_per_second": 0.475, | |
| "step": 1536 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 0.621838390827179, | |
| "learning_rate": 0.0, | |
| "loss": 0.2774, | |
| "step": 1728 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.7142857142857143, | |
| "eval_f1": 0.642156862745098, | |
| "eval_loss": 0.2810159921646118, | |
| "eval_mcc": 0.46831127867565187, | |
| "eval_precision": 0.8590163934426229, | |
| "eval_recall": 0.512720156555773, | |
| "eval_runtime": 67.055, | |
| "eval_samples_per_second": 15.241, | |
| "eval_steps_per_second": 0.477, | |
| "step": 1728 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 1728, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9, | |
| "save_steps": 500, | |
| "total_flos": 2121256775520.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.4208671172864604, | |
| "learning_rate": 0.00014684430023894452, | |
| "num_train_epochs": 9, | |
| "per_device_train_batch_size": 16, | |
| "temperature": 48 | |
| } | |
| } | |