| { | |
| "best_metric": 0.9506395189108336, | |
| "best_model_checkpoint": "./output//25_roberta-large_nace_5__5e-6_0.01_0.06_07-21-22_10-40/checkpoint-54000", | |
| "epoch": 3.01255230125523, | |
| "global_step": 54000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 1.8594272963927112e-06, | |
| "loss": 0.2318, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "eval_accuracy": 0.876702441473671, | |
| "eval_f1": 0.90677030703459, | |
| "eval_loss": 0.08175961673259735, | |
| "eval_roc_auc": 0.9387159436934914, | |
| "eval_runtime": 942.0949, | |
| "eval_samples_per_second": 33.825, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 3.7188545927854224e-06, | |
| "loss": 0.0732, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_accuracy": 0.8966296366032762, | |
| "eval_f1": 0.9234467455621302, | |
| "eval_loss": 0.060932476073503494, | |
| "eval_roc_auc": 0.9514088251746582, | |
| "eval_runtime": 942.3301, | |
| "eval_samples_per_second": 33.816, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 4.963084738922455e-06, | |
| "loss": 0.0584, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eval_accuracy": 0.8969434506998054, | |
| "eval_f1": 0.9278133577310156, | |
| "eval_loss": 0.05285193771123886, | |
| "eval_roc_auc": 0.9578889777085196, | |
| "eval_runtime": 943.2116, | |
| "eval_samples_per_second": 33.785, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 4.844386150248674e-06, | |
| "loss": 0.0521, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "eval_accuracy": 0.9035649281365719, | |
| "eval_f1": 0.931070919922204, | |
| "eval_loss": 0.049892134964466095, | |
| "eval_roc_auc": 0.9581022789802229, | |
| "eval_runtime": 943.267, | |
| "eval_samples_per_second": 33.783, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 4.7256875615748935e-06, | |
| "loss": 0.0488, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "eval_accuracy": 0.9092449632837507, | |
| "eval_f1": 0.9369276016030965, | |
| "eval_loss": 0.046968165785074234, | |
| "eval_roc_auc": 0.9633660743801509, | |
| "eval_runtime": 943.3238, | |
| "eval_samples_per_second": 33.781, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 4.606988972901112e-06, | |
| "loss": 0.0469, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "eval_accuracy": 0.913512834996548, | |
| "eval_f1": 0.941321897843637, | |
| "eval_loss": 0.04364275932312012, | |
| "eval_roc_auc": 0.9653120464218385, | |
| "eval_runtime": 943.108, | |
| "eval_samples_per_second": 33.788, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 4.4882903842273315e-06, | |
| "loss": 0.0447, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "eval_accuracy": 0.9181886650348333, | |
| "eval_f1": 0.9415976522286926, | |
| "eval_loss": 0.04153330251574516, | |
| "eval_roc_auc": 0.9628170726293566, | |
| "eval_runtime": 943.4861, | |
| "eval_samples_per_second": 33.775, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 4.369591795553551e-06, | |
| "loss": 0.0444, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "eval_accuracy": 0.9147994727923179, | |
| "eval_f1": 0.9419169981558534, | |
| "eval_loss": 0.042323991656303406, | |
| "eval_roc_auc": 0.9661668065711105, | |
| "eval_runtime": 943.053, | |
| "eval_samples_per_second": 33.79, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.25089320687977e-06, | |
| "loss": 0.0428, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.92073055921672, | |
| "eval_f1": 0.945392908407904, | |
| "eval_loss": 0.040571972727775574, | |
| "eval_roc_auc": 0.967396285229951, | |
| "eval_runtime": 943.6862, | |
| "eval_samples_per_second": 33.768, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 4.13219461820599e-06, | |
| "loss": 0.038, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "eval_accuracy": 0.9201343124333144, | |
| "eval_f1": 0.9434847253943194, | |
| "eval_loss": 0.04074249416589737, | |
| "eval_roc_auc": 0.9648247119998007, | |
| "eval_runtime": 943.7404, | |
| "eval_samples_per_second": 33.766, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 4.013496029532209e-06, | |
| "loss": 0.0365, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "eval_accuracy": 0.9186280047699743, | |
| "eval_f1": 0.9448325191374088, | |
| "eval_loss": 0.04050755128264427, | |
| "eval_roc_auc": 0.968414370803744, | |
| "eval_runtime": 944.2502, | |
| "eval_samples_per_second": 33.747, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 3.8947974408584285e-06, | |
| "loss": 0.0363, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "eval_accuracy": 0.9201343124333144, | |
| "eval_f1": 0.9443637975031708, | |
| "eval_loss": 0.0401209257543087, | |
| "eval_roc_auc": 0.9665495515278828, | |
| "eval_runtime": 943.3154, | |
| "eval_samples_per_second": 33.781, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 3.7760988521846475e-06, | |
| "loss": 0.0369, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "eval_accuracy": 0.9228331136634658, | |
| "eval_f1": 0.9453032652246665, | |
| "eval_loss": 0.03975052386522293, | |
| "eval_roc_auc": 0.9652725030917011, | |
| "eval_runtime": 943.3891, | |
| "eval_samples_per_second": 33.778, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 3.6574002635108673e-06, | |
| "loss": 0.0367, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "eval_accuracy": 0.924308039917153, | |
| "eval_f1": 0.9483727178359894, | |
| "eval_loss": 0.03769804537296295, | |
| "eval_roc_auc": 0.9692338448531075, | |
| "eval_runtime": 943.731, | |
| "eval_samples_per_second": 33.766, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 3.5387016748370867e-06, | |
| "loss": 0.0357, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "eval_accuracy": 0.9256888219418816, | |
| "eval_f1": 0.9485330073349634, | |
| "eval_loss": 0.03771040216088295, | |
| "eval_roc_auc": 0.9689151700049551, | |
| "eval_runtime": 943.5531, | |
| "eval_samples_per_second": 33.772, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 3.420003086163306e-06, | |
| "loss": 0.0353, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "eval_accuracy": 0.9218289085545722, | |
| "eval_f1": 0.945624456015514, | |
| "eval_loss": 0.03922228887677193, | |
| "eval_roc_auc": 0.9675467885174716, | |
| "eval_runtime": 943.7902, | |
| "eval_samples_per_second": 33.764, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 3.3013044974895255e-06, | |
| "loss": 0.0361, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "eval_accuracy": 0.9247159982426411, | |
| "eval_f1": 0.9477583383080007, | |
| "eval_loss": 0.03780689090490341, | |
| "eval_roc_auc": 0.9678151508441292, | |
| "eval_runtime": 943.8287, | |
| "eval_samples_per_second": 33.762, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 3.1826059088157445e-06, | |
| "loss": 0.0352, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "eval_accuracy": 0.9262536873156342, | |
| "eval_f1": 0.9494783086216365, | |
| "eval_loss": 0.03738059848546982, | |
| "eval_roc_auc": 0.970156686689613, | |
| "eval_runtime": 944.0794, | |
| "eval_samples_per_second": 33.754, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 3.063907320141964e-06, | |
| "loss": 0.0304, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "eval_accuracy": 0.9259084918094521, | |
| "eval_f1": 0.948512218605291, | |
| "eval_loss": 0.03789151459932327, | |
| "eval_roc_auc": 0.9688610378201717, | |
| "eval_runtime": 943.994, | |
| "eval_samples_per_second": 33.757, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 2.9452087314681833e-06, | |
| "loss": 0.03, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "eval_accuracy": 0.9245904726040294, | |
| "eval_f1": 0.9488459898909932, | |
| "eval_loss": 0.03766411170363426, | |
| "eval_roc_auc": 0.9705674249639594, | |
| "eval_runtime": 943.9278, | |
| "eval_samples_per_second": 33.759, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 2.8265101427944027e-06, | |
| "loss": 0.0293, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "eval_accuracy": 0.9265675014121635, | |
| "eval_f1": 0.9496503709571762, | |
| "eval_loss": 0.03824329748749733, | |
| "eval_roc_auc": 0.9707690753304716, | |
| "eval_runtime": 943.7438, | |
| "eval_samples_per_second": 33.766, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 2.7078115541206216e-06, | |
| "loss": 0.0295, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "eval_accuracy": 0.9280110462561978, | |
| "eval_f1": 0.9498280736608847, | |
| "eval_loss": 0.03822626546025276, | |
| "eval_roc_auc": 0.9696253499219528, | |
| "eval_runtime": 944.0868, | |
| "eval_samples_per_second": 33.753, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 2.589112965446841e-06, | |
| "loss": 0.0293, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "eval_accuracy": 0.9273520366534864, | |
| "eval_f1": 0.950136518250736, | |
| "eval_loss": 0.0382731668651104, | |
| "eval_roc_auc": 0.9705518129264687, | |
| "eval_runtime": 944.2173, | |
| "eval_samples_per_second": 33.749, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "learning_rate": 2.4704143767730604e-06, | |
| "loss": 0.0289, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "eval_accuracy": 0.9288583443168267, | |
| "eval_f1": 0.9496914100188371, | |
| "eval_loss": 0.038598936051130295, | |
| "eval_roc_auc": 0.9687129340225382, | |
| "eval_runtime": 943.702, | |
| "eval_samples_per_second": 33.767, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 2.35171578809928e-06, | |
| "loss": 0.0298, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "eval_accuracy": 0.9273206552438336, | |
| "eval_f1": 0.9502127205355373, | |
| "eval_loss": 0.037168510258197784, | |
| "eval_roc_auc": 0.9707164609217411, | |
| "eval_runtime": 943.7733, | |
| "eval_samples_per_second": 33.764, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 2.233017199425499e-06, | |
| "loss": 0.0287, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "eval_accuracy": 0.9278855206175861, | |
| "eval_f1": 0.950306102196913, | |
| "eval_loss": 0.03820985183119774, | |
| "eval_roc_auc": 0.970297989008143, | |
| "eval_runtime": 944.0102, | |
| "eval_samples_per_second": 33.756, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 2.1143186107517182e-06, | |
| "loss": 0.0286, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "eval_accuracy": 0.9286072930396033, | |
| "eval_f1": 0.9506395189108336, | |
| "eval_loss": 0.03879757598042488, | |
| "eval_roc_auc": 0.9705917028008494, | |
| "eval_runtime": 943.6083, | |
| "eval_samples_per_second": 33.77, | |
| "step": 54000 | |
| } | |
| ], | |
| "max_steps": 89625, | |
| "num_train_epochs": 5, | |
| "total_flos": 6.986708904362803e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |