| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 2440, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 5.120229244232178, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.5609, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7192982456140351, | |
| "eval_f1": 0.6543102914784331, | |
| "eval_loss": 0.5086308717727661, | |
| "eval_precision": 0.6580196140461879, | |
| "eval_recall": 0.6513911620294599, | |
| "eval_runtime": 5.1386, | |
| "eval_samples_per_second": 77.647, | |
| "eval_steps_per_second": 9.73, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.929361581802368, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.4986, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7493734335839599, | |
| "eval_f1": 0.7201178451178452, | |
| "eval_loss": 0.485486775636673, | |
| "eval_precision": 0.7127371273712737, | |
| "eval_recall": 0.7426804873613384, | |
| "eval_runtime": 5.073, | |
| "eval_samples_per_second": 78.652, | |
| "eval_steps_per_second": 9.856, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 4.046614646911621, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.4593, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7694235588972431, | |
| "eval_f1": 0.7308797653958945, | |
| "eval_loss": 0.42381197214126587, | |
| "eval_precision": 0.7249487296342714, | |
| "eval_recall": 0.7393617021276595, | |
| "eval_runtime": 5.0448, | |
| "eval_samples_per_second": 79.092, | |
| "eval_steps_per_second": 9.911, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.3827860355377197, | |
| "learning_rate": 4e-05, | |
| "loss": 0.3957, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8070175438596491, | |
| "eval_f1": 0.7700391464135747, | |
| "eval_loss": 0.3916189670562744, | |
| "eval_precision": 0.7669852636562704, | |
| "eval_recall": 0.7734588106928533, | |
| "eval_runtime": 5.0667, | |
| "eval_samples_per_second": 78.749, | |
| "eval_steps_per_second": 9.868, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.527370810508728, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.3658, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7994987468671679, | |
| "eval_f1": 0.7744360902255639, | |
| "eval_loss": 0.4266420900821686, | |
| "eval_precision": 0.7640805369127517, | |
| "eval_recall": 0.7981451172940535, | |
| "eval_runtime": 5.0699, | |
| "eval_samples_per_second": 78.7, | |
| "eval_steps_per_second": 9.862, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.4858503341674805, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.3345, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8370927318295739, | |
| "eval_f1": 0.8049369344976196, | |
| "eval_loss": 0.36663180589675903, | |
| "eval_precision": 0.8027777777777778, | |
| "eval_recall": 0.8072376795781051, | |
| "eval_runtime": 5.0518, | |
| "eval_samples_per_second": 78.982, | |
| "eval_steps_per_second": 9.898, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 1.170096516609192, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.3237, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8395989974937343, | |
| "eval_f1": 0.8136058394160584, | |
| "eval_loss": 0.3713766634464264, | |
| "eval_precision": 0.8045112781954887, | |
| "eval_recall": 0.8265139116202946, | |
| "eval_runtime": 5.0706, | |
| "eval_samples_per_second": 78.689, | |
| "eval_steps_per_second": 9.861, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 4.0918803215026855, | |
| "learning_rate": 3e-05, | |
| "loss": 0.304, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8421052631578947, | |
| "eval_f1": 0.8118502107020158, | |
| "eval_loss": 0.35368239879608154, | |
| "eval_precision": 0.8083091673078061, | |
| "eval_recall": 0.8157846881251136, | |
| "eval_runtime": 5.0556, | |
| "eval_samples_per_second": 78.922, | |
| "eval_steps_per_second": 9.89, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 5.677598476409912, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.3027, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8446115288220551, | |
| "eval_f1": 0.8152777777777778, | |
| "eval_loss": 0.3530685007572174, | |
| "eval_precision": 0.8110639802050195, | |
| "eval_recall": 0.8200581923986179, | |
| "eval_runtime": 5.0512, | |
| "eval_samples_per_second": 78.992, | |
| "eval_steps_per_second": 9.899, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 6.779963493347168, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2962, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8521303258145363, | |
| "eval_f1": 0.8211781685593832, | |
| "eval_loss": 0.33822622895240784, | |
| "eval_precision": 0.8219964664310955, | |
| "eval_recall": 0.8203764320785598, | |
| "eval_runtime": 5.0587, | |
| "eval_samples_per_second": 78.873, | |
| "eval_steps_per_second": 9.884, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 4.037176132202148, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.2721, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.849624060150376, | |
| "eval_f1": 0.8228975557791324, | |
| "eval_loss": 0.3489574193954468, | |
| "eval_precision": 0.8162488420565077, | |
| "eval_recall": 0.8311056555737406, | |
| "eval_runtime": 5.072, | |
| "eval_samples_per_second": 78.668, | |
| "eval_steps_per_second": 9.858, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 10.375617027282715, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2693, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8546365914786967, | |
| "eval_f1": 0.8288009705864946, | |
| "eval_loss": 0.35016030073165894, | |
| "eval_precision": 0.822024085224641, | |
| "eval_recall": 0.837152209492635, | |
| "eval_runtime": 5.0553, | |
| "eval_samples_per_second": 78.926, | |
| "eval_steps_per_second": 9.891, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 1.8974263668060303, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.2745, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8571428571428571, | |
| "eval_f1": 0.8263588263588264, | |
| "eval_loss": 0.3283708095550537, | |
| "eval_precision": 0.8289473684210527, | |
| "eval_recall": 0.8239225313693399, | |
| "eval_runtime": 5.0639, | |
| "eval_samples_per_second": 78.793, | |
| "eval_steps_per_second": 9.874, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 11.152755737304688, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.2712, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8596491228070176, | |
| "eval_f1": 0.8315338681464504, | |
| "eval_loss": 0.3297080397605896, | |
| "eval_precision": 0.8299369747899159, | |
| "eval_recall": 0.8331969448990726, | |
| "eval_runtime": 5.0624, | |
| "eval_samples_per_second": 78.816, | |
| "eval_steps_per_second": 9.877, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 1.1235237121582031, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.256, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8646616541353384, | |
| "eval_f1": 0.8391129032258065, | |
| "eval_loss": 0.3356616199016571, | |
| "eval_precision": 0.8345705196182396, | |
| "eval_recall": 0.8442444080741953, | |
| "eval_runtime": 5.0578, | |
| "eval_samples_per_second": 78.888, | |
| "eval_steps_per_second": 9.886, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 6.625916004180908, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2504, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8571428571428571, | |
| "eval_f1": 0.8305757727005222, | |
| "eval_loss": 0.3345828056335449, | |
| "eval_precision": 0.8255131964809383, | |
| "eval_recall": 0.8364248045099109, | |
| "eval_runtime": 5.0718, | |
| "eval_samples_per_second": 78.67, | |
| "eval_steps_per_second": 9.858, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 7.8184356689453125, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.2487, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8571428571428571, | |
| "eval_f1": 0.8272399255573702, | |
| "eval_loss": 0.3242071270942688, | |
| "eval_precision": 0.8280735957109784, | |
| "eval_recall": 0.8264229859974541, | |
| "eval_runtime": 5.0798, | |
| "eval_samples_per_second": 78.547, | |
| "eval_steps_per_second": 9.843, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 4.3523454666137695, | |
| "learning_rate": 5e-06, | |
| "loss": 0.2514, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8365204824303285, | |
| "eval_loss": 0.33086374402046204, | |
| "eval_precision": 0.8313636363636363, | |
| "eval_recall": 0.8424713584288053, | |
| "eval_runtime": 5.0535, | |
| "eval_samples_per_second": 78.956, | |
| "eval_steps_per_second": 9.894, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 3.3547356128692627, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.2451, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8341332527115377, | |
| "eval_loss": 0.32434186339378357, | |
| "eval_precision": 0.8333132275770553, | |
| "eval_recall": 0.8349699945444626, | |
| "eval_runtime": 5.0515, | |
| "eval_samples_per_second": 78.986, | |
| "eval_steps_per_second": 9.898, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 9.069262504577637, | |
| "learning_rate": 0.0, | |
| "loss": 0.2461, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8357422474382676, | |
| "eval_loss": 0.3260349631309509, | |
| "eval_precision": 0.8319228265372551, | |
| "eval_recall": 0.8399709038006911, | |
| "eval_runtime": 5.0498, | |
| "eval_samples_per_second": 79.013, | |
| "eval_steps_per_second": 9.901, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 7609911792720000.0, | |
| "train_loss": 0.3213141730574311, | |
| "train_runtime": 1951.6183, | |
| "train_samples_per_second": 37.282, | |
| "train_steps_per_second": 1.25 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 7609911792720000.0, | |
| "train_batch_size": 30, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |