| { | |
| "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": 4.0894904136657715, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.541, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7293233082706767, | |
| "eval_f1": 0.6396989966555184, | |
| "eval_loss": 0.4985284209251404, | |
| "eval_precision": 0.664766661583041, | |
| "eval_recall": 0.630978359701764, | |
| "eval_runtime": 5.2157, | |
| "eval_samples_per_second": 76.5, | |
| "eval_steps_per_second": 9.586, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 5.95181941986084, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.4477, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7644110275689223, | |
| "eval_f1": 0.7461557203963398, | |
| "eval_loss": 0.44652456045150757, | |
| "eval_precision": 0.7426785714285714, | |
| "eval_recall": 0.7883251500272777, | |
| "eval_runtime": 5.1119, | |
| "eval_samples_per_second": 78.053, | |
| "eval_steps_per_second": 9.781, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 3.869917154312134, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.347, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8646616541353384, | |
| "eval_f1": 0.8255102040816327, | |
| "eval_loss": 0.323697566986084, | |
| "eval_precision": 0.8556293485135991, | |
| "eval_recall": 0.8067375886524822, | |
| "eval_runtime": 5.1398, | |
| "eval_samples_per_second": 77.63, | |
| "eval_steps_per_second": 9.728, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 3.4031054973602295, | |
| "learning_rate": 4e-05, | |
| "loss": 0.3005, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8922305764411027, | |
| "eval_f1": 0.8683279483657071, | |
| "eval_loss": 0.2991793751716614, | |
| "eval_precision": 0.873366724738676, | |
| "eval_recall": 0.863747954173486, | |
| "eval_runtime": 5.1616, | |
| "eval_samples_per_second": 77.302, | |
| "eval_steps_per_second": 9.687, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 3.774945020675659, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.281, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8646616541353384, | |
| "eval_f1": 0.8342105263157895, | |
| "eval_loss": 0.2868594527244568, | |
| "eval_precision": 0.8398085585585586, | |
| "eval_recall": 0.82924168030551, | |
| "eval_runtime": 5.1483, | |
| "eval_samples_per_second": 77.502, | |
| "eval_steps_per_second": 9.712, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.3688924312591553, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.2419, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.87468671679198, | |
| "eval_f1": 0.8537390029325513, | |
| "eval_loss": 0.29453349113464355, | |
| "eval_precision": 0.8442805058676086, | |
| "eval_recall": 0.8663393344244408, | |
| "eval_runtime": 5.0969, | |
| "eval_samples_per_second": 78.284, | |
| "eval_steps_per_second": 9.81, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.5997357964515686, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.2394, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8771929824561403, | |
| "eval_f1": 0.8529524583135901, | |
| "eval_loss": 0.2835337221622467, | |
| "eval_precision": 0.8504480286738352, | |
| "eval_recall": 0.8556101109292599, | |
| "eval_runtime": 5.0928, | |
| "eval_samples_per_second": 78.346, | |
| "eval_steps_per_second": 9.818, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 2.8720760345458984, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2192, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8771929824561403, | |
| "eval_f1": 0.850729517396184, | |
| "eval_loss": 0.28028008341789246, | |
| "eval_precision": 0.8535087719298247, | |
| "eval_recall": 0.8481087470449173, | |
| "eval_runtime": 5.1964, | |
| "eval_samples_per_second": 76.784, | |
| "eval_steps_per_second": 9.622, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 5.954423904418945, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.2144, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.87468671679198, | |
| "eval_f1": 0.8480717680029244, | |
| "eval_loss": 0.28611448407173157, | |
| "eval_precision": 0.8498775260257195, | |
| "eval_recall": 0.8463356973995272, | |
| "eval_runtime": 5.1448, | |
| "eval_samples_per_second": 77.554, | |
| "eval_steps_per_second": 9.719, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 4.511594772338867, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2056, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8922305764411027, | |
| "eval_f1": 0.8696722245432793, | |
| "eval_loss": 0.27244648337364197, | |
| "eval_precision": 0.8706135006701596, | |
| "eval_recall": 0.8687488634297145, | |
| "eval_runtime": 5.104, | |
| "eval_samples_per_second": 78.174, | |
| "eval_steps_per_second": 9.796, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 2.713789463043213, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.1822, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8872180451127819, | |
| "eval_f1": 0.8662440310793597, | |
| "eval_loss": 0.28133633732795715, | |
| "eval_precision": 0.8606158357771261, | |
| "eval_recall": 0.872704128023277, | |
| "eval_runtime": 5.1442, | |
| "eval_samples_per_second": 77.563, | |
| "eval_steps_per_second": 9.72, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 6.766155242919922, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1817, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8872180451127819, | |
| "eval_f1": 0.8584001703456596, | |
| "eval_loss": 0.2900215685367584, | |
| "eval_precision": 0.8759655377302435, | |
| "eval_recall": 0.8451991271140207, | |
| "eval_runtime": 5.1374, | |
| "eval_samples_per_second": 77.665, | |
| "eval_steps_per_second": 9.733, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 0.9034644365310669, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.1621, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8947368421052632, | |
| "eval_f1": 0.8710526315789473, | |
| "eval_loss": 0.29263192415237427, | |
| "eval_precision": 0.8772522522522522, | |
| "eval_recall": 0.8655210038188761, | |
| "eval_runtime": 5.1149, | |
| "eval_samples_per_second": 78.008, | |
| "eval_steps_per_second": 9.775, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 6.05497932434082, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.1577, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8922305764411027, | |
| "eval_f1": 0.8709582797445791, | |
| "eval_loss": 0.29044803977012634, | |
| "eval_precision": 0.8683243727598566, | |
| "eval_recall": 0.8737497726859429, | |
| "eval_runtime": 5.096, | |
| "eval_samples_per_second": 78.297, | |
| "eval_steps_per_second": 9.812, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 3.013443946838379, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.1612, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8847117794486216, | |
| "eval_f1": 0.8587719298245614, | |
| "eval_loss": 0.2996305227279663, | |
| "eval_precision": 0.864771021021021, | |
| "eval_recall": 0.8534278959810875, | |
| "eval_runtime": 5.1189, | |
| "eval_samples_per_second": 77.946, | |
| "eval_steps_per_second": 9.768, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 1.775810718536377, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1496, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8872180451127819, | |
| "eval_f1": 0.8649563392675828, | |
| "eval_loss": 0.29704856872558594, | |
| "eval_precision": 0.8623655913978494, | |
| "eval_recall": 0.8677032187670486, | |
| "eval_runtime": 5.2102, | |
| "eval_samples_per_second": 76.58, | |
| "eval_steps_per_second": 9.596, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 0.5526378154754639, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.149, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.858259325044405, | |
| "eval_loss": 0.29482966661453247, | |
| "eval_precision": 0.8573798178418481, | |
| "eval_recall": 0.8591562102200401, | |
| "eval_runtime": 5.1218, | |
| "eval_samples_per_second": 77.902, | |
| "eval_steps_per_second": 9.762, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 4.444484233856201, | |
| "learning_rate": 5e-06, | |
| "loss": 0.1424, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8847117794486216, | |
| "eval_f1": 0.8609292598654301, | |
| "eval_loss": 0.29769569635391235, | |
| "eval_precision": 0.8609292598654301, | |
| "eval_recall": 0.8609292598654301, | |
| "eval_runtime": 5.0934, | |
| "eval_samples_per_second": 78.336, | |
| "eval_steps_per_second": 9.817, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 3.0727450847625732, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.1383, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8847117794486216, | |
| "eval_f1": 0.8602260265626904, | |
| "eval_loss": 0.2990491986274719, | |
| "eval_precision": 0.8620943049601959, | |
| "eval_recall": 0.8584288052373159, | |
| "eval_runtime": 5.1176, | |
| "eval_samples_per_second": 77.966, | |
| "eval_steps_per_second": 9.77, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 4.4248151779174805, | |
| "learning_rate": 0.0, | |
| "loss": 0.1407, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.858259325044405, | |
| "eval_loss": 0.2988053262233734, | |
| "eval_precision": 0.8573798178418481, | |
| "eval_recall": 0.8591562102200401, | |
| "eval_runtime": 5.0823, | |
| "eval_samples_per_second": 78.508, | |
| "eval_steps_per_second": 9.838, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 8444128359504000.0, | |
| "train_loss": 0.2301350734272941, | |
| "train_runtime": 1951.0131, | |
| "train_samples_per_second": 37.293, | |
| "train_steps_per_second": 1.251 | |
| } | |
| ], | |
| "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 | |
| } | |