| { | |
| "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": 4.234379768371582, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.5459, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7393483709273183, | |
| "eval_f1": 0.6458660476548099, | |
| "eval_loss": 0.47629594802856445, | |
| "eval_precision": 0.6804065499717673, | |
| "eval_recall": 0.63557010365521, | |
| "eval_runtime": 4.7443, | |
| "eval_samples_per_second": 84.101, | |
| "eval_steps_per_second": 10.539, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 4.299708843231201, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.4528, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7844611528822055, | |
| "eval_f1": 0.7677594888732471, | |
| "eval_loss": 0.43055105209350586, | |
| "eval_precision": 0.7630357142857143, | |
| "eval_recall": 0.8125113657028551, | |
| "eval_runtime": 5.0592, | |
| "eval_samples_per_second": 78.866, | |
| "eval_steps_per_second": 9.883, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 3.729501962661743, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.3653, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8217317487266553, | |
| "eval_loss": 0.3334697186946869, | |
| "eval_precision": 0.853331681065005, | |
| "eval_recall": 0.802464084378978, | |
| "eval_runtime": 5.0519, | |
| "eval_samples_per_second": 78.98, | |
| "eval_steps_per_second": 9.897, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.0170559883117676, | |
| "learning_rate": 4e-05, | |
| "loss": 0.2987, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8546365914786967, | |
| "eval_f1": 0.8246499363520641, | |
| "eval_loss": 0.335675448179245, | |
| "eval_precision": 0.8246499363520641, | |
| "eval_recall": 0.8246499363520641, | |
| "eval_runtime": 5.0598, | |
| "eval_samples_per_second": 78.858, | |
| "eval_steps_per_second": 9.882, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.7316147089004517, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.2746, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8546365914786967, | |
| "eval_f1": 0.8338931955211025, | |
| "eval_loss": 0.3400849401950836, | |
| "eval_precision": 0.8216641926439057, | |
| "eval_recall": 0.8546553918894344, | |
| "eval_runtime": 5.0556, | |
| "eval_samples_per_second": 78.923, | |
| "eval_steps_per_second": 9.89, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 1.9782907962799072, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.2477, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.849624060150376, | |
| "eval_f1": 0.8308270676691729, | |
| "eval_loss": 0.33233991265296936, | |
| "eval_precision": 0.8176375838926174, | |
| "eval_recall": 0.858610656482997, | |
| "eval_runtime": 5.0483, | |
| "eval_samples_per_second": 79.037, | |
| "eval_steps_per_second": 9.904, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 1.8530523777008057, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.24, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8646616541353384, | |
| "eval_f1": 0.8447157518450185, | |
| "eval_loss": 0.3171246647834778, | |
| "eval_precision": 0.8325401217487549, | |
| "eval_recall": 0.864248045099109, | |
| "eval_runtime": 5.2354, | |
| "eval_samples_per_second": 76.211, | |
| "eval_steps_per_second": 9.55, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 3.7897450923919678, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2069, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8922305764411027, | |
| "eval_f1": 0.8683279483657071, | |
| "eval_loss": 0.27698931097984314, | |
| "eval_precision": 0.873366724738676, | |
| "eval_recall": 0.863747954173486, | |
| "eval_runtime": 5.0519, | |
| "eval_samples_per_second": 78.98, | |
| "eval_steps_per_second": 9.897, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 5.382436275482178, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.2197, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8671679197994987, | |
| "eval_f1": 0.8491192020377148, | |
| "eval_loss": 0.3091033399105072, | |
| "eval_precision": 0.8356209150326797, | |
| "eval_recall": 0.8735224586288416, | |
| "eval_runtime": 5.0573, | |
| "eval_samples_per_second": 78.896, | |
| "eval_steps_per_second": 9.887, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 3.9839751720428467, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2005, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9022556390977443, | |
| "eval_f1": 0.8811928811928812, | |
| "eval_loss": 0.2552439570426941, | |
| "eval_precision": 0.8842105263157894, | |
| "eval_recall": 0.878341516639389, | |
| "eval_runtime": 5.0478, | |
| "eval_samples_per_second": 79.044, | |
| "eval_steps_per_second": 9.905, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 0.8527041673660278, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.1867, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.9047619047619048, | |
| "eval_f1": 0.8867831541218638, | |
| "eval_loss": 0.2726534903049469, | |
| "eval_precision": 0.88158359844468, | |
| "eval_recall": 0.8926168394253501, | |
| "eval_runtime": 5.0594, | |
| "eval_samples_per_second": 78.863, | |
| "eval_steps_per_second": 9.883, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 2.9348244667053223, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1722, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8922305764411027, | |
| "eval_f1": 0.8727838950061173, | |
| "eval_loss": 0.27391675114631653, | |
| "eval_precision": 0.8656898656898657, | |
| "eval_recall": 0.8812511365702855, | |
| "eval_runtime": 5.0516, | |
| "eval_samples_per_second": 78.984, | |
| "eval_steps_per_second": 9.898, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 1.8993942737579346, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.161, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.899749373433584, | |
| "eval_f1": 0.8765393898137261, | |
| "eval_loss": 0.2714451551437378, | |
| "eval_precision": 0.8852261942423283, | |
| "eval_recall": 0.8690671031096563, | |
| "eval_runtime": 5.0831, | |
| "eval_samples_per_second": 78.495, | |
| "eval_steps_per_second": 9.836, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 5.6604695320129395, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.1684, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8972431077694235, | |
| "eval_f1": 0.8781334505389722, | |
| "eval_loss": 0.27736207842826843, | |
| "eval_precision": 0.872316715542522, | |
| "eval_recall": 0.8847972358610656, | |
| "eval_runtime": 5.0631, | |
| "eval_samples_per_second": 78.805, | |
| "eval_steps_per_second": 9.875, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.4771318733692169, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.1548, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.899749373433584, | |
| "eval_f1": 0.8802521008403361, | |
| "eval_loss": 0.27421677112579346, | |
| "eval_precision": 0.8767168083714847, | |
| "eval_recall": 0.8840698308783415, | |
| "eval_runtime": 5.0573, | |
| "eval_samples_per_second": 78.896, | |
| "eval_steps_per_second": 9.887, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.8061890006065369, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1526, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8872180451127819, | |
| "eval_f1": 0.8703282417939551, | |
| "eval_loss": 0.29698580503463745, | |
| "eval_precision": 0.8574462768615692, | |
| "eval_recall": 0.8902073104200764, | |
| "eval_runtime": 5.0603, | |
| "eval_samples_per_second": 78.848, | |
| "eval_steps_per_second": 9.881, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 2.3830015659332275, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.1467, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8897243107769424, | |
| "eval_f1": 0.8707140332272888, | |
| "eval_loss": 0.2729242146015167, | |
| "eval_precision": 0.8618432385874246, | |
| "eval_recall": 0.8819785415530097, | |
| "eval_runtime": 5.0444, | |
| "eval_samples_per_second": 79.097, | |
| "eval_steps_per_second": 9.912, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 4.170547962188721, | |
| "learning_rate": 5e-06, | |
| "loss": 0.1484, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8972431077694235, | |
| "eval_f1": 0.8781334505389722, | |
| "eval_loss": 0.27389636635780334, | |
| "eval_precision": 0.872316715542522, | |
| "eval_recall": 0.8847972358610656, | |
| "eval_runtime": 5.0888, | |
| "eval_samples_per_second": 78.407, | |
| "eval_steps_per_second": 9.825, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 3.1922006607055664, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.1434, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.899749373433584, | |
| "eval_f1": 0.879667048676036, | |
| "eval_loss": 0.27286645770072937, | |
| "eval_precision": 0.8778361344537815, | |
| "eval_recall": 0.8815693762502272, | |
| "eval_runtime": 5.0573, | |
| "eval_samples_per_second": 78.895, | |
| "eval_steps_per_second": 9.887, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 1.1650974750518799, | |
| "learning_rate": 0.0, | |
| "loss": 0.1354, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.9022556390977443, | |
| "eval_f1": 0.8829621606985718, | |
| "eval_loss": 0.2765507400035858, | |
| "eval_precision": 0.8802419354838709, | |
| "eval_recall": 0.8858428805237315, | |
| "eval_runtime": 5.0471, | |
| "eval_samples_per_second": 79.055, | |
| "eval_steps_per_second": 9.907, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 8444128359504000.0, | |
| "train_loss": 0.23107414167435442, | |
| "train_runtime": 1893.6576, | |
| "train_samples_per_second": 38.423, | |
| "train_steps_per_second": 1.289 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 8444128359504000.0, | |
| "train_batch_size": 30, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |