| { | |
| "best_metric": 0.3108769357204437, | |
| "best_model_checkpoint": "./checkpoint-xlm-v-base/checkpoint-15000", | |
| "epoch": 0.6640694173897644, | |
| "eval_steps": 1000, | |
| "global_step": 15000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 46.0612678527832, | |
| "learning_rate": 1.9873510587163855e-05, | |
| "loss": 0.943, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "eval_LOC_f1": 0.6007898672649095, | |
| "eval_ORG_f1": 0.5205864729691723, | |
| "eval_PER_f1": 0.6537866457692849, | |
| "eval_loss": 0.5755352973937988, | |
| "eval_overall_accuracy": 0.813493507822672, | |
| "eval_overall_f1": 0.5959796923883924, | |
| "eval_overall_precision": 0.571592844668358, | |
| "eval_overall_recall": 0.622540194436182, | |
| "eval_runtime": 909.7639, | |
| "eval_samples_per_second": 72.217, | |
| "eval_steps_per_second": 0.282, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 5.543514251708984, | |
| "learning_rate": 1.974702117432771e-05, | |
| "loss": 0.5825, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "eval_LOC_f1": 0.6761863812260971, | |
| "eval_ORG_f1": 0.573783382357224, | |
| "eval_PER_f1": 0.7231398018028106, | |
| "eval_loss": 0.5157074332237244, | |
| "eval_overall_accuracy": 0.844698117312631, | |
| "eval_overall_f1": 0.664661446599853, | |
| "eval_overall_precision": 0.690302943433134, | |
| "eval_overall_recall": 0.640856646367237, | |
| "eval_runtime": 884.4485, | |
| "eval_samples_per_second": 74.284, | |
| "eval_steps_per_second": 0.291, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 30.680952072143555, | |
| "learning_rate": 1.9620531761491565e-05, | |
| "loss": 0.5153, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "eval_LOC_f1": 0.6448332585613877, | |
| "eval_ORG_f1": 0.5780655943179445, | |
| "eval_PER_f1": 0.749264457627936, | |
| "eval_loss": 0.4841216504573822, | |
| "eval_overall_accuracy": 0.8415159100197845, | |
| "eval_overall_f1": 0.6599932560127353, | |
| "eval_overall_precision": 0.6612013701212998, | |
| "eval_overall_recall": 0.6587895486638383, | |
| "eval_runtime": 887.5751, | |
| "eval_samples_per_second": 74.022, | |
| "eval_steps_per_second": 0.29, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 14.460062026977539, | |
| "learning_rate": 1.949404234865542e-05, | |
| "loss": 0.4744, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "eval_LOC_f1": 0.7135048963789569, | |
| "eval_ORG_f1": 0.6044746860257756, | |
| "eval_PER_f1": 0.7562344421814889, | |
| "eval_loss": 0.4284209907054901, | |
| "eval_overall_accuracy": 0.8650805108611886, | |
| "eval_overall_f1": 0.6945276906141412, | |
| "eval_overall_precision": 0.6951840204528166, | |
| "eval_overall_recall": 0.6938725989010129, | |
| "eval_runtime": 884.4505, | |
| "eval_samples_per_second": 74.283, | |
| "eval_steps_per_second": 0.291, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 2.3655357360839844, | |
| "learning_rate": 1.9367552935819272e-05, | |
| "loss": 0.4385, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_LOC_f1": 0.7295629535257298, | |
| "eval_ORG_f1": 0.6200475216886777, | |
| "eval_PER_f1": 0.7408250910983861, | |
| "eval_loss": 0.4239448308944702, | |
| "eval_overall_accuracy": 0.8635053351958905, | |
| "eval_overall_f1": 0.7025319684063077, | |
| "eval_overall_precision": 0.7043199823770524, | |
| "eval_overall_recall": 0.7007530096904989, | |
| "eval_runtime": 907.5971, | |
| "eval_samples_per_second": 72.389, | |
| "eval_steps_per_second": 0.283, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 9.867854118347168, | |
| "learning_rate": 1.924106352298313e-05, | |
| "loss": 0.4279, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "eval_LOC_f1": 0.7370099725835874, | |
| "eval_ORG_f1": 0.6403813434199981, | |
| "eval_PER_f1": 0.7776639577500056, | |
| "eval_loss": 0.38233184814453125, | |
| "eval_overall_accuracy": 0.8816618781055326, | |
| "eval_overall_f1": 0.7238370468534203, | |
| "eval_overall_precision": 0.7385234418271267, | |
| "eval_overall_recall": 0.7097233746105797, | |
| "eval_runtime": 948.2573, | |
| "eval_samples_per_second": 69.285, | |
| "eval_steps_per_second": 0.271, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 25.877347946166992, | |
| "learning_rate": 1.9114574110146982e-05, | |
| "loss": 0.4099, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eval_LOC_f1": 0.7575684397708062, | |
| "eval_ORG_f1": 0.630874803840732, | |
| "eval_PER_f1": 0.7838224767358626, | |
| "eval_loss": 0.38040244579315186, | |
| "eval_overall_accuracy": 0.883136741379065, | |
| "eval_overall_f1": 0.7288917006049582, | |
| "eval_overall_precision": 0.7364569017865703, | |
| "eval_overall_recall": 0.7214803450381201, | |
| "eval_runtime": 907.1265, | |
| "eval_samples_per_second": 72.427, | |
| "eval_steps_per_second": 0.283, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 30.637121200561523, | |
| "learning_rate": 1.8988084697310836e-05, | |
| "loss": 0.3874, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "eval_LOC_f1": 0.7474734456723695, | |
| "eval_ORG_f1": 0.6407748343462335, | |
| "eval_PER_f1": 0.7854664027017585, | |
| "eval_loss": 0.37021398544311523, | |
| "eval_overall_accuracy": 0.8872800498308584, | |
| "eval_overall_f1": 0.7295473133392094, | |
| "eval_overall_precision": 0.74022719948437, | |
| "eval_overall_recall": 0.7191712196878376, | |
| "eval_runtime": 884.9913, | |
| "eval_samples_per_second": 74.238, | |
| "eval_steps_per_second": 0.29, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 20.109619140625, | |
| "learning_rate": 1.8861595284474693e-05, | |
| "loss": 0.3841, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_LOC_f1": 0.7684967782745274, | |
| "eval_ORG_f1": 0.655326947582435, | |
| "eval_PER_f1": 0.789217873159736, | |
| "eval_loss": 0.3808096945285797, | |
| "eval_overall_accuracy": 0.8879015799879489, | |
| "eval_overall_f1": 0.743771496693436, | |
| "eval_overall_precision": 0.76602787456446, | |
| "eval_overall_recall": 0.7227718897255663, | |
| "eval_runtime": 885.0879, | |
| "eval_samples_per_second": 74.23, | |
| "eval_steps_per_second": 0.29, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 10.265982627868652, | |
| "learning_rate": 1.8735105871638546e-05, | |
| "loss": 0.3764, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_LOC_f1": 0.7831821749367751, | |
| "eval_ORG_f1": 0.6622161847467495, | |
| "eval_PER_f1": 0.7948864849077164, | |
| "eval_loss": 0.34247785806655884, | |
| "eval_overall_accuracy": 0.8911041208737209, | |
| "eval_overall_f1": 0.7509862429761675, | |
| "eval_overall_precision": 0.7436605881991772, | |
| "eval_overall_recall": 0.7584576608168825, | |
| "eval_runtime": 885.3889, | |
| "eval_samples_per_second": 74.205, | |
| "eval_steps_per_second": 0.29, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 4.401586055755615, | |
| "learning_rate": 1.86086164588024e-05, | |
| "loss": 0.3564, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "eval_LOC_f1": 0.789610444706057, | |
| "eval_ORG_f1": 0.6701892389389907, | |
| "eval_PER_f1": 0.8054954166474735, | |
| "eval_loss": 0.35062676668167114, | |
| "eval_overall_accuracy": 0.8928101093201735, | |
| "eval_overall_f1": 0.7598466310260445, | |
| "eval_overall_precision": 0.7565214692509428, | |
| "eval_overall_recall": 0.7632011522144121, | |
| "eval_runtime": 885.0018, | |
| "eval_samples_per_second": 74.237, | |
| "eval_steps_per_second": 0.29, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 1.0011909008026123, | |
| "learning_rate": 1.8482127045966253e-05, | |
| "loss": 0.3484, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "eval_LOC_f1": 0.7712024123633622, | |
| "eval_ORG_f1": 0.6809512535185331, | |
| "eval_PER_f1": 0.8100081183474653, | |
| "eval_loss": 0.37064051628112793, | |
| "eval_overall_accuracy": 0.8851477337194005, | |
| "eval_overall_f1": 0.758193057536852, | |
| "eval_overall_precision": 0.7375241450255696, | |
| "eval_overall_recall": 0.7800538534996947, | |
| "eval_runtime": 950.4008, | |
| "eval_samples_per_second": 69.129, | |
| "eval_steps_per_second": 0.27, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 13.891754150390625, | |
| "learning_rate": 1.8355637633130106e-05, | |
| "loss": 0.3563, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_LOC_f1": 0.7934942596408595, | |
| "eval_ORG_f1": 0.6686377545091862, | |
| "eval_PER_f1": 0.8152987398240509, | |
| "eval_loss": 0.3389296531677246, | |
| "eval_overall_accuracy": 0.8935766857734662, | |
| "eval_overall_f1": 0.7638085016673694, | |
| "eval_overall_precision": 0.7483700877298401, | |
| "eval_overall_recall": 0.7798973026284891, | |
| "eval_runtime": 970.0425, | |
| "eval_samples_per_second": 67.729, | |
| "eval_steps_per_second": 0.265, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 19.86951446533203, | |
| "learning_rate": 1.8229148220293963e-05, | |
| "loss": 0.3396, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "eval_LOC_f1": 0.7943446440452429, | |
| "eval_ORG_f1": 0.6860717813631874, | |
| "eval_PER_f1": 0.808841180333809, | |
| "eval_loss": 0.33801111578941345, | |
| "eval_overall_accuracy": 0.8965501348456104, | |
| "eval_overall_f1": 0.767065352823492, | |
| "eval_overall_precision": 0.770362767931157, | |
| "eval_overall_recall": 0.7637960455249934, | |
| "eval_runtime": 892.9487, | |
| "eval_samples_per_second": 73.576, | |
| "eval_steps_per_second": 0.288, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 6.008892059326172, | |
| "learning_rate": 1.8102658807457817e-05, | |
| "loss": 0.3513, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_LOC_f1": 0.796750172086523, | |
| "eval_ORG_f1": 0.6869723599718148, | |
| "eval_PER_f1": 0.8222321051851345, | |
| "eval_loss": 0.3108769357204437, | |
| "eval_overall_accuracy": 0.900560597156957, | |
| "eval_overall_f1": 0.773478314631055, | |
| "eval_overall_precision": 0.7637860773210824, | |
| "eval_overall_recall": 0.7834196972306151, | |
| "eval_runtime": 885.3039, | |
| "eval_samples_per_second": 74.212, | |
| "eval_steps_per_second": 0.29, | |
| "step": 15000 | |
| } | |
| ], | |
| "logging_steps": 1000, | |
| "max_steps": 158116, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "save_steps": 500, | |
| "total_flos": 4767734068578096.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |