| { | |
| "best_metric": 0.9015122140364482, | |
| "best_model_checkpoint": "tiny-bert-sst2-distilled/run-14/checkpoint-4515", | |
| "epoch": 7.0, | |
| "eval_steps": 500, | |
| "global_step": 4515, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.9803465604782104, | |
| "learning_rate": 0.00027230837357094026, | |
| "loss": 0.327, | |
| "step": 645 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8887165568049632, | |
| "eval_f1": 0.19154929577464788, | |
| "eval_loss": 0.3063199520111084, | |
| "eval_mcc": 0.1990419339488759, | |
| "eval_precision": 0.48226950354609927, | |
| "eval_recall": 0.1195079086115993, | |
| "eval_runtime": 9.9261, | |
| "eval_samples_per_second": 519.642, | |
| "eval_steps_per_second": 16.321, | |
| "step": 645 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.0405207872390747, | |
| "learning_rate": 0.00023340717734652024, | |
| "loss": 0.3025, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8914307871267934, | |
| "eval_f1": 0.14893617021276595, | |
| "eval_loss": 0.3057989478111267, | |
| "eval_mcc": 0.18620749231762276, | |
| "eval_precision": 0.550561797752809, | |
| "eval_recall": 0.08611599297012303, | |
| "eval_runtime": 9.9284, | |
| "eval_samples_per_second": 519.521, | |
| "eval_steps_per_second": 16.317, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.7381914854049683, | |
| "learning_rate": 0.0001945059811221002, | |
| "loss": 0.29, | |
| "step": 1935 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8951143854207056, | |
| "eval_f1": 0.16640986132511557, | |
| "eval_loss": 0.2883821427822113, | |
| "eval_mcc": 0.2262412458871255, | |
| "eval_precision": 0.675, | |
| "eval_recall": 0.09490333919156414, | |
| "eval_runtime": 9.951, | |
| "eval_samples_per_second": 518.339, | |
| "eval_steps_per_second": 16.28, | |
| "step": 1935 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.401355743408203, | |
| "learning_rate": 0.00015560478489768016, | |
| "loss": 0.2831, | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8933695230709577, | |
| "eval_f1": 0.2782152230971129, | |
| "eval_loss": 0.29702696204185486, | |
| "eval_mcc": 0.2762227316232168, | |
| "eval_precision": 0.5492227979274611, | |
| "eval_recall": 0.18629173989455183, | |
| "eval_runtime": 9.9249, | |
| "eval_samples_per_second": 519.701, | |
| "eval_steps_per_second": 16.323, | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.1628026962280273, | |
| "learning_rate": 0.00011670358867326012, | |
| "loss": 0.278, | |
| "step": 3225 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8929817758821249, | |
| "eval_f1": 0.3284671532846716, | |
| "eval_loss": 0.27827948331832886, | |
| "eval_mcc": 0.30685804485965024, | |
| "eval_precision": 0.5335968379446641, | |
| "eval_recall": 0.23725834797891038, | |
| "eval_runtime": 9.9217, | |
| "eval_samples_per_second": 519.869, | |
| "eval_steps_per_second": 16.328, | |
| "step": 3225 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 1.1073917150497437, | |
| "learning_rate": 7.780239244884008e-05, | |
| "loss": 0.272, | |
| "step": 3870 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8997673516867003, | |
| "eval_f1": 0.31158455392809586, | |
| "eval_loss": 0.2726772427558899, | |
| "eval_mcc": 0.3250995883258456, | |
| "eval_precision": 0.6428571428571429, | |
| "eval_recall": 0.2056239015817223, | |
| "eval_runtime": 9.9563, | |
| "eval_samples_per_second": 518.063, | |
| "eval_steps_per_second": 16.271, | |
| "step": 3870 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.8763220310211182, | |
| "learning_rate": 3.890119622442004e-05, | |
| "loss": 0.2672, | |
| "step": 4515 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9015122140364482, | |
| "eval_f1": 0.3665835411471322, | |
| "eval_loss": 0.2725670337677002, | |
| "eval_mcc": 0.36143858618325836, | |
| "eval_precision": 0.630901287553648, | |
| "eval_recall": 0.2583479789103691, | |
| "eval_runtime": 9.9527, | |
| "eval_samples_per_second": 518.25, | |
| "eval_steps_per_second": 16.277, | |
| "step": 4515 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 5160, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 8, | |
| "save_steps": 500, | |
| "total_flos": 11108597728200.0, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.9932188265495786, | |
| "learning_rate": 0.0003112095697953603, | |
| "num_train_epochs": 8, | |
| "temperature": 21 | |
| } | |
| } | |