diff --git "a/trainer_state.json" "b/trainer_state.json" --- "a/trainer_state.json" +++ "b/trainer_state.json" @@ -1,2544 +1,1640 @@ { - "best_metric": 0.43965517241379304, - "best_model_checkpoint": "car_identified_model_7/checkpoint-165", - "epoch": 228.36363636363637, + "best_metric": 0.3805309734513274, + "best_model_checkpoint": "car_identified_model_7/checkpoint-188", + "epoch": 145.45454545454547, "eval_steps": 500, - "global_step": 314, + "global_step": 200, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.73, - "learning_rate": 1.9950000000000004e-05, - "loss": 0.6943, + "learning_rate": 1.9900000000000003e-05, + "loss": 0.6919, "step": 1 }, { "epoch": 0.73, "eval_accuracy": 0.0, - "eval_f1": 0.09523809523809525, - "eval_loss": 0.6940220594406128, - "eval_roc_auc": 0.4290890269151138, - "eval_runtime": 2.3165, - "eval_samples_per_second": 18.131, - "eval_steps_per_second": 1.295, + "eval_f1": 0.17857142857142858, + "eval_loss": 0.6887282133102417, + "eval_roc_auc": 0.5737577639751552, + "eval_runtime": 2.8648, + "eval_samples_per_second": 14.661, + "eval_steps_per_second": 1.047, "step": 1 }, { "epoch": 1.45, "eval_accuracy": 0.0, - "eval_f1": 0.09126984126984126, - "eval_loss": 0.6954004764556885, - "eval_roc_auc": 0.4314182194616977, - "eval_runtime": 2.1188, - "eval_samples_per_second": 19.823, - "eval_steps_per_second": 1.416, + "eval_f1": 0.1818181818181818, + "eval_loss": 0.6855647563934326, + "eval_roc_auc": 0.5760869565217391, + "eval_runtime": 2.5822, + "eval_samples_per_second": 16.265, + "eval_steps_per_second": 1.162, "step": 2 }, { "epoch": 2.91, "eval_accuracy": 0.0, - "eval_f1": 0.10158730158730159, - "eval_loss": 0.7020308971405029, - "eval_roc_auc": 0.4244306418219462, - "eval_runtime": 2.0566, - "eval_samples_per_second": 20.422, - "eval_steps_per_second": 1.459, + "eval_f1": 0.21164021164021166, + "eval_loss": 0.6802406311035156, + "eval_roc_auc": 0.6066252587991718, + "eval_runtime": 2.1976, + "eval_samples_per_second": 19.112, + "eval_steps_per_second": 1.365, "step": 4 }, { "epoch": 3.64, "eval_accuracy": 0.0, - "eval_f1": 0.1020408163265306, - "eval_loss": 0.7104746103286743, - "eval_roc_auc": 0.43322981366459634, - "eval_runtime": 2.0845, - "eval_samples_per_second": 20.149, - "eval_steps_per_second": 1.439, + "eval_f1": 0.18614718614718614, + "eval_loss": 0.6800228953361511, + "eval_roc_auc": 0.5825569358178053, + "eval_runtime": 2.1512, + "eval_samples_per_second": 19.524, + "eval_steps_per_second": 1.395, "step": 5 }, { "epoch": 4.36, "eval_accuracy": 0.0, - "eval_f1": 0.15476190476190477, - "eval_loss": 0.7156671285629272, - "eval_roc_auc": 0.5320910973084887, - "eval_runtime": 2.0793, - "eval_samples_per_second": 20.199, - "eval_steps_per_second": 1.443, + "eval_f1": 0.1904761904761905, + "eval_loss": 0.6858328580856323, + "eval_roc_auc": 0.5973084886128365, + "eval_runtime": 1.8993, + "eval_samples_per_second": 22.113, + "eval_steps_per_second": 1.58, "step": 6 }, { "epoch": 5.82, "eval_accuracy": 0.0, - "eval_f1": 0.1546218487394958, - "eval_loss": 0.6987839937210083, - "eval_roc_auc": 0.5331262939958592, - "eval_runtime": 2.1617, - "eval_samples_per_second": 19.429, - "eval_steps_per_second": 1.388, + "eval_f1": 0.15492957746478872, + "eval_loss": 0.693770706653595, + "eval_roc_auc": 0.5341614906832298, + "eval_runtime": 1.986, + "eval_samples_per_second": 21.148, + "eval_steps_per_second": 1.511, "step": 8 }, { "epoch": 6.55, "eval_accuracy": 0.0, - "eval_f1": 0.15355805243445694, - "eval_loss": 0.6832717061042786, - "eval_roc_auc": 0.5323498964803313, - "eval_runtime": 2.117, - "eval_samples_per_second": 19.84, - "eval_steps_per_second": 1.417, + "eval_f1": 0.18050541516245489, + "eval_loss": 0.6917278170585632, + "eval_roc_auc": 0.5802277432712216, + "eval_runtime": 1.8331, + "eval_samples_per_second": 22.911, + "eval_steps_per_second": 1.637, "step": 9 }, { "epoch": 8.0, "eval_accuracy": 0.0, - "eval_f1": 0.18217054263565893, - "eval_loss": 0.6745525598526001, - "eval_roc_auc": 0.5804865424430642, - "eval_runtime": 2.0693, - "eval_samples_per_second": 20.297, - "eval_steps_per_second": 1.45, + "eval_f1": 0.19047619047619044, + "eval_loss": 0.6735165119171143, + "eval_roc_auc": 0.593167701863354, + "eval_runtime": 2.0579, + "eval_samples_per_second": 20.409, + "eval_steps_per_second": 1.458, "step": 11 }, { "epoch": 8.73, "eval_accuracy": 0.0, - "eval_f1": 0.2058252427184466, - "eval_loss": 0.6750043034553528, - "eval_roc_auc": 0.619824016563147, - "eval_runtime": 2.0585, - "eval_samples_per_second": 20.403, - "eval_steps_per_second": 1.457, + "eval_f1": 0.1952191235059761, + "eval_loss": 0.6727393865585327, + "eval_roc_auc": 0.6006728778467909, + "eval_runtime": 1.8528, + "eval_samples_per_second": 22.669, + "eval_steps_per_second": 1.619, "step": 12 }, { "epoch": 9.45, "eval_accuracy": 0.0, - "eval_f1": 0.2152641878669276, - "eval_loss": 0.673625648021698, - "eval_roc_auc": 0.6348343685300206, - "eval_runtime": 2.1768, - "eval_samples_per_second": 19.294, - "eval_steps_per_second": 1.378, + "eval_f1": 0.20606060606060606, + "eval_loss": 0.6698008179664612, + "eval_roc_auc": 0.6172360248447204, + "eval_runtime": 2.0897, + "eval_samples_per_second": 20.099, + "eval_steps_per_second": 1.436, "step": 13 }, { "epoch": 10.91, "eval_accuracy": 0.0, - "eval_f1": 0.2222222222222222, - "eval_loss": 0.6704300045967102, - "eval_roc_auc": 0.644927536231884, - "eval_runtime": 2.2148, - "eval_samples_per_second": 18.963, - "eval_steps_per_second": 1.355, + "eval_f1": 0.20080321285140562, + "eval_loss": 0.6671529412269592, + "eval_roc_auc": 0.6092132505175983, + "eval_runtime": 1.8899, + "eval_samples_per_second": 22.224, + "eval_steps_per_second": 1.587, "step": 15 }, { "epoch": 11.64, "eval_accuracy": 0.0, - "eval_f1": 0.2258064516129032, - "eval_loss": 0.6693084836006165, - "eval_roc_auc": 0.6490683229813665, - "eval_runtime": 2.0734, - "eval_samples_per_second": 20.257, - "eval_steps_per_second": 1.447, + "eval_f1": 0.2092050209205021, + "eval_loss": 0.6645169854164124, + "eval_roc_auc": 0.6195652173913044, + "eval_runtime": 1.9211, + "eval_samples_per_second": 21.863, + "eval_steps_per_second": 1.562, "step": 16 }, { "epoch": 12.36, "eval_accuracy": 0.0, - "eval_f1": 0.23293172690763056, - "eval_loss": 0.6673960089683533, - "eval_roc_auc": 0.6609730848861284, - "eval_runtime": 2.1348, - "eval_samples_per_second": 19.674, - "eval_steps_per_second": 1.405, + "eval_f1": 0.20491803278688525, + "eval_loss": 0.6645674705505371, + "eval_roc_auc": 0.6143892339544513, + "eval_runtime": 1.9806, + "eval_samples_per_second": 21.206, + "eval_steps_per_second": 1.515, "step": 17 }, { "epoch": 13.82, "eval_accuracy": 0.0, - "eval_f1": 0.2240325865580448, - "eval_loss": 0.6640763282775879, - "eval_roc_auc": 0.6451863354037266, - "eval_runtime": 2.1666, - "eval_samples_per_second": 19.385, - "eval_steps_per_second": 1.385, + "eval_f1": 0.208067940552017, + "eval_loss": 0.6622694134712219, + "eval_roc_auc": 0.6167184265010353, + "eval_runtime": 2.0813, + "eval_samples_per_second": 20.179, + "eval_steps_per_second": 1.441, "step": 19 }, { "epoch": 14.55, "eval_accuracy": 0.0, - "eval_f1": 0.22085889570552147, - "eval_loss": 0.6631330847740173, - "eval_roc_auc": 0.639751552795031, - "eval_runtime": 2.0574, - "eval_samples_per_second": 20.414, - "eval_steps_per_second": 1.458, + "eval_f1": 0.20779220779220778, + "eval_loss": 0.6606575846672058, + "eval_roc_auc": 0.6149068322981366, + "eval_runtime": 1.8418, + "eval_samples_per_second": 22.803, + "eval_steps_per_second": 1.629, "step": 20 }, { "epoch": 16.0, "eval_accuracy": 0.0, - "eval_f1": 0.22406639004149378, - "eval_loss": 0.6598684191703796, - "eval_roc_auc": 0.6433747412008283, - "eval_runtime": 2.1618, - "eval_samples_per_second": 19.428, - "eval_steps_per_second": 1.388, + "eval_f1": 0.22026431718061673, + "eval_loss": 0.6584673523902893, + "eval_roc_auc": 0.6319875776397516, + "eval_runtime": 2.024, + "eval_samples_per_second": 20.751, + "eval_steps_per_second": 1.482, "step": 22 }, { "epoch": 16.73, "eval_accuracy": 0.0, - "eval_f1": 0.23454157782515989, - "eval_loss": 0.6574246287345886, - "eval_roc_auc": 0.6565734989648033, - "eval_runtime": 2.0709, - "eval_samples_per_second": 20.281, - "eval_steps_per_second": 1.449, + "eval_f1": 0.21559633027522934, + "eval_loss": 0.6561669111251831, + "eval_roc_auc": 0.6218944099378882, + "eval_runtime": 1.8303, + "eval_samples_per_second": 22.947, + "eval_steps_per_second": 1.639, "step": 23 }, { "epoch": 17.45, "eval_accuracy": 0.0, - "eval_f1": 0.24017467248908297, - "eval_loss": 0.6559909582138062, - "eval_roc_auc": 0.6622670807453417, - "eval_runtime": 2.2938, - "eval_samples_per_second": 18.311, - "eval_steps_per_second": 1.308, + "eval_f1": 0.21818181818181817, + "eval_loss": 0.6555350422859192, + "eval_roc_auc": 0.6262939958592133, + "eval_runtime": 1.9854, + "eval_samples_per_second": 21.154, + "eval_steps_per_second": 1.511, "step": 24 }, { "epoch": 18.91, "eval_accuracy": 0.0, - "eval_f1": 0.23399558498896247, - "eval_loss": 0.6524508595466614, - "eval_roc_auc": 0.6519151138716356, - "eval_runtime": 2.1408, - "eval_samples_per_second": 19.619, - "eval_steps_per_second": 1.401, + "eval_f1": 0.2185273159144893, + "eval_loss": 0.652241051197052, + "eval_roc_auc": 0.6231884057971014, + "eval_runtime": 1.9789, + "eval_samples_per_second": 21.224, + "eval_steps_per_second": 1.516, "step": 26 }, { "epoch": 19.64, "eval_accuracy": 0.0, - "eval_f1": 0.24608501118568235, - "eval_loss": 0.650789737701416, - "eval_roc_auc": 0.6679606625258798, - "eval_runtime": 2.1461, - "eval_samples_per_second": 19.571, - "eval_steps_per_second": 1.398, + "eval_f1": 0.22276029055690072, + "eval_loss": 0.6512007713317871, + "eval_roc_auc": 0.6273291925465838, + "eval_runtime": 2.0333, + "eval_samples_per_second": 20.656, + "eval_steps_per_second": 1.475, "step": 27 }, { "epoch": 20.36, "eval_accuracy": 0.0, - "eval_f1": 0.24074074074074073, - "eval_loss": 0.6483226418495178, - "eval_roc_auc": 0.6563146997929606, - "eval_runtime": 2.2028, - "eval_samples_per_second": 19.067, - "eval_steps_per_second": 1.362, + "eval_f1": 0.23558897243107768, + "eval_loss": 0.6501262187957764, + "eval_roc_auc": 0.6410455486542443, + "eval_runtime": 2.1299, + "eval_samples_per_second": 19.719, + "eval_steps_per_second": 1.409, "step": 28 }, { "epoch": 21.82, "eval_accuracy": 0.0, - "eval_f1": 0.24545454545454545, - "eval_loss": 0.6458417177200317, - "eval_roc_auc": 0.6651138716356108, - "eval_runtime": 2.2709, - "eval_samples_per_second": 18.495, - "eval_steps_per_second": 1.321, + "eval_f1": 0.22797927461139897, + "eval_loss": 0.6477265357971191, + "eval_roc_auc": 0.6283643892339545, + "eval_runtime": 1.8983, + "eval_samples_per_second": 22.125, + "eval_steps_per_second": 1.58, "step": 30 }, { "epoch": 22.55, "eval_accuracy": 0.0, - "eval_f1": 0.24601366742596809, - "eval_loss": 0.644438624382019, - "eval_roc_auc": 0.665631469979296, - "eval_runtime": 2.1455, - "eval_samples_per_second": 19.576, - "eval_steps_per_second": 1.398, + "eval_f1": 0.23255813953488372, + "eval_loss": 0.6476041078567505, + "eval_roc_auc": 0.6343167701863355, + "eval_runtime": 2.1591, + "eval_samples_per_second": 19.452, + "eval_steps_per_second": 1.389, "step": 31 }, { "epoch": 24.0, "eval_accuracy": 0.0, - "eval_f1": 0.25116279069767444, - "eval_loss": 0.6407680511474609, - "eval_roc_auc": 0.6702898550724637, - "eval_runtime": 2.3067, - "eval_samples_per_second": 18.208, - "eval_steps_per_second": 1.301, + "eval_f1": 0.24083769633507857, + "eval_loss": 0.6469409465789795, + "eval_roc_auc": 0.6433747412008282, + "eval_runtime": 1.8477, + "eval_samples_per_second": 22.731, + "eval_steps_per_second": 1.624, "step": 33 }, { "epoch": 24.73, "eval_accuracy": 0.0, - "eval_f1": 0.2523809523809524, - "eval_loss": 0.6386719346046448, - "eval_roc_auc": 0.6689958592132504, - "eval_runtime": 2.1261, - "eval_samples_per_second": 19.754, - "eval_steps_per_second": 1.411, + "eval_f1": 0.24089635854341737, + "eval_loss": 0.6432453393936157, + "eval_roc_auc": 0.6369047619047619, + "eval_runtime": 2.1797, + "eval_samples_per_second": 19.268, + "eval_steps_per_second": 1.376, "step": 34 }, { "epoch": 25.45, "eval_accuracy": 0.0, - "eval_f1": 0.2579075425790754, - "eval_loss": 0.6363871693611145, - "eval_roc_auc": 0.6736542443064182, - "eval_runtime": 2.0731, - "eval_samples_per_second": 20.259, - "eval_steps_per_second": 1.447, + "eval_f1": 0.2430939226519337, + "eval_loss": 0.6432364583015442, + "eval_roc_auc": 0.6407867494824016, + "eval_runtime": 1.9386, + "eval_samples_per_second": 21.665, + "eval_steps_per_second": 1.547, "step": 35 }, { "epoch": 26.91, "eval_accuracy": 0.0, - "eval_f1": 0.2706766917293233, - "eval_loss": 0.633211076259613, - "eval_roc_auc": 0.686335403726708, - "eval_runtime": 2.1802, - "eval_samples_per_second": 19.264, - "eval_steps_per_second": 1.376, + "eval_f1": 0.24858757062146897, + "eval_loss": 0.6401556730270386, + "eval_roc_auc": 0.644927536231884, + "eval_runtime": 1.9831, + "eval_samples_per_second": 21.179, + "eval_steps_per_second": 1.513, "step": 37 }, { "epoch": 27.64, "eval_accuracy": 0.0, - "eval_f1": 0.26153846153846155, - "eval_loss": 0.6316937208175659, - "eval_roc_auc": 0.671583850931677, - "eval_runtime": 2.2588, - "eval_samples_per_second": 18.594, - "eval_steps_per_second": 1.328, + "eval_f1": 0.26857142857142857, + "eval_loss": 0.6385589838027954, + "eval_roc_auc": 0.666407867494824, + "eval_runtime": 1.8248, + "eval_samples_per_second": 23.016, + "eval_steps_per_second": 1.644, "step": 38 }, { "epoch": 28.36, "eval_accuracy": 0.0, - "eval_f1": 0.26666666666666666, - "eval_loss": 0.6295819878578186, - "eval_roc_auc": 0.6780538302277432, - "eval_runtime": 2.0803, - "eval_samples_per_second": 20.189, - "eval_steps_per_second": 1.442, + "eval_f1": 0.27624309392265195, + "eval_loss": 0.6375865936279297, + "eval_roc_auc": 0.6796066252587992, + "eval_runtime": 2.0365, + "eval_samples_per_second": 20.624, + "eval_steps_per_second": 1.473, "step": 39 }, { "epoch": 29.82, "eval_accuracy": 0.0, - "eval_f1": 0.2819843342036553, - "eval_loss": 0.6249864101409912, - "eval_roc_auc": 0.6946169772256728, - "eval_runtime": 2.1933, - "eval_samples_per_second": 19.149, - "eval_steps_per_second": 1.368, + "eval_f1": 0.2692307692307692, + "eval_loss": 0.6347113847732544, + "eval_roc_auc": 0.6721014492753624, + "eval_runtime": 2.0248, + "eval_samples_per_second": 20.743, + "eval_steps_per_second": 1.482, "step": 41 }, { "epoch": 30.55, "eval_accuracy": 0.0, - "eval_f1": 0.2872340425531915, - "eval_loss": 0.624562680721283, - "eval_roc_auc": 0.6982401656314701, - "eval_runtime": 2.1386, - "eval_samples_per_second": 19.639, - "eval_steps_per_second": 1.403, + "eval_f1": 0.26553672316384186, + "eval_loss": 0.6338767409324646, + "eval_roc_auc": 0.6643374741200828, + "eval_runtime": 1.912, + "eval_samples_per_second": 21.966, + "eval_steps_per_second": 1.569, "step": 42 }, { "epoch": 32.0, "eval_accuracy": 0.0, - "eval_f1": 0.2936288088642659, - "eval_loss": 0.6203415393829346, - "eval_roc_auc": 0.6995341614906833, - "eval_runtime": 2.0334, - "eval_samples_per_second": 20.655, - "eval_steps_per_second": 1.475, + "eval_f1": 0.2674418604651163, + "eval_loss": 0.6310274004936218, + "eval_roc_auc": 0.6630434782608696, + "eval_runtime": 2.048, + "eval_samples_per_second": 20.508, + "eval_steps_per_second": 1.465, "step": 44 }, { "epoch": 32.73, "eval_accuracy": 0.0, - "eval_f1": 0.27932960893854747, - "eval_loss": 0.619665265083313, - "eval_roc_auc": 0.6816770186335404, - "eval_runtime": 2.2329, - "eval_samples_per_second": 18.809, - "eval_steps_per_second": 1.344, + "eval_f1": 0.27893175074183973, + "eval_loss": 0.630736768245697, + "eval_roc_auc": 0.6731366459627329, + "eval_runtime": 1.9873, + "eval_samples_per_second": 21.134, + "eval_steps_per_second": 1.51, "step": 45 }, { "epoch": 33.45, "eval_accuracy": 0.0, - "eval_f1": 0.2832861189801699, - "eval_loss": 0.6191604733467102, - "eval_roc_auc": 0.6842650103519669, - "eval_runtime": 2.1287, - "eval_samples_per_second": 19.73, - "eval_steps_per_second": 1.409, + "eval_f1": 0.2713864306784661, + "eval_loss": 0.6290633678436279, + "eval_roc_auc": 0.6656314699792961, + "eval_runtime": 1.9917, + "eval_samples_per_second": 21.087, + "eval_steps_per_second": 1.506, "step": 46 }, { "epoch": 34.91, "eval_accuracy": 0.0, - "eval_f1": 0.29059829059829057, - "eval_loss": 0.6188385486602783, - "eval_roc_auc": 0.6917701863354038, - "eval_runtime": 1.9704, - "eval_samples_per_second": 21.316, - "eval_steps_per_second": 1.523, + "eval_f1": 0.2760736196319018, + "eval_loss": 0.6271013617515564, + "eval_roc_auc": 0.6658902691511386, + "eval_runtime": 1.9904, + "eval_samples_per_second": 21.101, + "eval_steps_per_second": 1.507, "step": 48 }, { "epoch": 35.64, "eval_accuracy": 0.0, - "eval_f1": 0.2994350282485876, - "eval_loss": 0.6187217831611633, - "eval_roc_auc": 0.7031573498964803, - "eval_runtime": 2.1103, - "eval_samples_per_second": 19.902, - "eval_steps_per_second": 1.422, + "eval_f1": 0.2686567164179105, + "eval_loss": 0.6270779371261597, + "eval_roc_auc": 0.661231884057971, + "eval_runtime": 2.05, + "eval_samples_per_second": 20.488, + "eval_steps_per_second": 1.463, "step": 49 }, { "epoch": 36.36, "eval_accuracy": 0.0, - "eval_f1": 0.29394812680115273, - "eval_loss": 0.6172403693199158, - "eval_roc_auc": 0.6938405797101449, - "eval_runtime": 2.0512, - "eval_samples_per_second": 20.476, - "eval_steps_per_second": 1.463, + "eval_f1": 0.2606060606060606, + "eval_loss": 0.6277384757995605, + "eval_roc_auc": 0.650879917184265, + "eval_runtime": 1.8738, + "eval_samples_per_second": 22.415, + "eval_steps_per_second": 1.601, "step": 50 }, { "epoch": 37.82, "eval_accuracy": 0.0, - "eval_f1": 0.3072289156626506, - "eval_loss": 0.6139649748802185, - "eval_roc_auc": 0.7016045548654244, - "eval_runtime": 2.0138, - "eval_samples_per_second": 20.857, - "eval_steps_per_second": 1.49, + "eval_f1": 0.2741433021806854, + "eval_loss": 0.6256951689720154, + "eval_roc_auc": 0.6620082815734989, + "eval_runtime": 2.0555, + "eval_samples_per_second": 20.433, + "eval_steps_per_second": 1.459, "step": 52 }, { "epoch": 38.55, "eval_accuracy": 0.0, - "eval_f1": 0.30538922155688625, - "eval_loss": 0.6133411526679993, - "eval_roc_auc": 0.7005693581780539, - "eval_runtime": 2.1136, - "eval_samples_per_second": 19.871, - "eval_steps_per_second": 1.419, + "eval_f1": 0.28923076923076924, + "eval_loss": 0.624409556388855, + "eval_roc_auc": 0.6793478260869565, + "eval_runtime": 1.842, + "eval_samples_per_second": 22.801, + "eval_steps_per_second": 1.629, "step": 53 }, { "epoch": 40.0, "eval_accuracy": 0.0, - "eval_f1": 0.3081570996978852, - "eval_loss": 0.6092615723609924, - "eval_roc_auc": 0.7021221532091098, - "eval_runtime": 2.0132, - "eval_samples_per_second": 20.863, - "eval_steps_per_second": 1.49, + "eval_f1": 0.29677419354838713, + "eval_loss": 0.6203488111495972, + "eval_roc_auc": 0.6806418219461698, + "eval_runtime": 1.9945, + "eval_samples_per_second": 21.058, + "eval_steps_per_second": 1.504, "step": 55 }, { "epoch": 40.73, "eval_accuracy": 0.0, - "eval_f1": 0.3151515151515151, - "eval_loss": 0.6076943278312683, - "eval_roc_auc": 0.7091097308488613, - "eval_runtime": 2.0499, - "eval_samples_per_second": 20.489, - "eval_steps_per_second": 1.463, + "eval_f1": 0.2902208201892744, + "eval_loss": 0.6198378801345825, + "eval_roc_auc": 0.6770186335403728, + "eval_runtime": 1.9168, + "eval_samples_per_second": 21.912, + "eval_steps_per_second": 1.565, "step": 56 }, { "epoch": 41.45, "eval_accuracy": 0.0, - "eval_f1": 0.31097560975609756, - "eval_loss": 0.6047279238700867, - "eval_roc_auc": 0.7036749482401656, - "eval_runtime": 2.2835, - "eval_samples_per_second": 18.392, - "eval_steps_per_second": 1.314, + "eval_f1": 0.3022508038585209, + "eval_loss": 0.6183561682701111, + "eval_roc_auc": 0.6865942028985508, + "eval_runtime": 2.0474, + "eval_samples_per_second": 20.514, + "eval_steps_per_second": 1.465, "step": 57 }, { "epoch": 42.91, "eval_accuracy": 0.0, - "eval_f1": 0.3164556962025316, - "eval_loss": 0.6036221385002136, - "eval_roc_auc": 0.7034161490683231, - "eval_runtime": 1.9891, - "eval_samples_per_second": 21.115, - "eval_steps_per_second": 1.508, + "eval_f1": 0.2977346278317152, + "eval_loss": 0.6163293719291687, + "eval_roc_auc": 0.6811594202898551, + "eval_runtime": 2.0211, + "eval_samples_per_second": 20.781, + "eval_steps_per_second": 1.484, "step": 59 }, { "epoch": 43.64, "eval_accuracy": 0.0, - "eval_f1": 0.33116883116883117, - "eval_loss": 0.6039034128189087, - "eval_roc_auc": 0.7140269151138717, - "eval_runtime": 2.2035, - "eval_samples_per_second": 19.061, - "eval_steps_per_second": 1.361, + "eval_f1": 0.33222591362126247, + "eval_loss": 0.6147304177284241, + "eval_roc_auc": 0.7111801242236025, + "eval_runtime": 1.8612, + "eval_samples_per_second": 22.566, + "eval_steps_per_second": 1.612, "step": 60 }, { "epoch": 44.36, "eval_accuracy": 0.0, - "eval_f1": 0.32258064516129037, - "eval_loss": 0.6021687984466553, - "eval_roc_auc": 0.7065217391304347, - "eval_runtime": 1.9925, - "eval_samples_per_second": 21.079, - "eval_steps_per_second": 1.506, + "eval_f1": 0.3197278911564626, + "eval_loss": 0.6154341697692871, + "eval_roc_auc": 0.6953933747412009, + "eval_runtime": 2.1228, + "eval_samples_per_second": 19.785, + "eval_steps_per_second": 1.413, "step": 61 }, { "epoch": 45.82, "eval_accuracy": 0.0, - "eval_f1": 0.326797385620915, - "eval_loss": 0.6014130711555481, - "eval_roc_auc": 0.708592132505176, - "eval_runtime": 2.0419, - "eval_samples_per_second": 20.569, - "eval_steps_per_second": 1.469, + "eval_f1": 0.30163934426229505, + "eval_loss": 0.6129170060157776, + "eval_roc_auc": 0.6832298136645963, + "eval_runtime": 1.8919, + "eval_samples_per_second": 22.2, + "eval_steps_per_second": 1.586, "step": 63 }, { "epoch": 46.55, "eval_accuracy": 0.0, - "eval_f1": 0.3130990415335463, - "eval_loss": 0.6009289622306824, - "eval_roc_auc": 0.6984989648033127, - "eval_runtime": 1.9923, - "eval_samples_per_second": 21.081, - "eval_steps_per_second": 1.506, + "eval_f1": 0.30201342281879195, + "eval_loss": 0.6111928224563599, + "eval_roc_auc": 0.680383022774327, + "eval_runtime": 2.0579, + "eval_samples_per_second": 20.409, + "eval_steps_per_second": 1.458, "step": 64 }, { "epoch": 48.0, "eval_accuracy": 0.0, - "eval_f1": 0.3411371237458194, - "eval_loss": 0.5962069034576416, - "eval_roc_auc": 0.7186853002070394, - "eval_runtime": 2.2383, - "eval_samples_per_second": 18.764, - "eval_steps_per_second": 1.34, + "eval_f1": 0.29605263157894735, + "eval_loss": 0.60954749584198, + "eval_roc_auc": 0.6772774327122153, + "eval_runtime": 2.0077, + "eval_samples_per_second": 20.92, + "eval_steps_per_second": 1.494, "step": 66 }, { "epoch": 48.73, "eval_accuracy": 0.0, - "eval_f1": 0.34576271186440677, - "eval_loss": 0.5965468287467957, - "eval_roc_auc": 0.7207556935817806, - "eval_runtime": 2.0677, - "eval_samples_per_second": 20.312, - "eval_steps_per_second": 1.451, + "eval_f1": 0.31333333333333335, + "eval_loss": 0.6090860962867737, + "eval_roc_auc": 0.692287784679089, + "eval_runtime": 1.8446, + "eval_samples_per_second": 22.769, + "eval_steps_per_second": 1.626, "step": 67 }, { "epoch": 49.45, "eval_accuracy": 0.0, - "eval_f1": 0.341296928327645, - "eval_loss": 0.5968855023384094, - "eval_roc_auc": 0.7153209109730849, - "eval_runtime": 2.0241, - "eval_samples_per_second": 20.75, - "eval_steps_per_second": 1.482, + "eval_f1": 0.32653061224489793, + "eval_loss": 0.6089531183242798, + "eval_roc_auc": 0.7018633540372671, + "eval_runtime": 2.0701, + "eval_samples_per_second": 20.289, + "eval_steps_per_second": 1.449, "step": 68 }, { "epoch": 50.91, "eval_accuracy": 0.0, - "eval_f1": 0.34, - "eval_loss": 0.593564510345459, - "eval_roc_auc": 0.7181677018633541, - "eval_runtime": 2.0433, - "eval_samples_per_second": 20.555, - "eval_steps_per_second": 1.468, + "eval_f1": 0.30927835051546393, + "eval_loss": 0.6077401638031006, + "eval_roc_auc": 0.6840062111801242, + "eval_runtime": 1.8396, + "eval_samples_per_second": 22.832, + "eval_steps_per_second": 1.631, "step": 70 }, { "epoch": 51.64, "eval_accuracy": 0.0, - "eval_f1": 0.3411371237458194, - "eval_loss": 0.591184139251709, - "eval_roc_auc": 0.7186853002070394, - "eval_runtime": 2.0663, - "eval_samples_per_second": 20.326, - "eval_steps_per_second": 1.452, + "eval_f1": 0.32394366197183105, + "eval_loss": 0.6065036654472351, + "eval_roc_auc": 0.6940993788819875, + "eval_runtime": 2.0429, + "eval_samples_per_second": 20.559, + "eval_steps_per_second": 1.469, "step": 71 }, { "epoch": 52.36, "eval_accuracy": 0.0, - "eval_f1": 0.3310344827586207, - "eval_loss": 0.592108428478241, - "eval_roc_auc": 0.7039337474120083, - "eval_runtime": 2.2694, - "eval_samples_per_second": 18.507, - "eval_steps_per_second": 1.322, + "eval_f1": 0.32374100719424465, + "eval_loss": 0.6057615876197815, + "eval_roc_auc": 0.690734989648033, + "eval_runtime": 1.8836, + "eval_samples_per_second": 22.298, + "eval_steps_per_second": 1.593, "step": 72 }, { "epoch": 53.82, "eval_accuracy": 0.0, - "eval_f1": 0.34657039711191334, - "eval_loss": 0.5902323722839355, - "eval_roc_auc": 0.7106625258799172, - "eval_runtime": 2.0722, - "eval_samples_per_second": 20.269, - "eval_steps_per_second": 1.448, + "eval_f1": 0.3284671532846715, + "eval_loss": 0.6027652621269226, + "eval_roc_auc": 0.6928053830227744, + "eval_runtime": 1.8924, + "eval_samples_per_second": 22.194, + "eval_steps_per_second": 1.585, "step": 74 }, { "epoch": 54.55, - "eval_accuracy": 0.0, - "eval_f1": 0.34306569343065696, - "eval_loss": 0.5893351435661316, - "eval_roc_auc": 0.7057453416149069, - "eval_runtime": 2.055, - "eval_samples_per_second": 20.438, - "eval_steps_per_second": 1.46, + "eval_accuracy": 0.023809523809523808, + "eval_f1": 0.3284671532846715, + "eval_loss": 0.6037685871124268, + "eval_roc_auc": 0.6928053830227744, + "eval_runtime": 1.8574, + "eval_samples_per_second": 22.612, + "eval_steps_per_second": 1.615, "step": 75 }, { "epoch": 56.0, "eval_accuracy": 0.0, - "eval_f1": 0.3404255319148936, - "eval_loss": 0.5895388722419739, - "eval_roc_auc": 0.7080745341614907, - "eval_runtime": 2.0467, - "eval_samples_per_second": 20.52, - "eval_steps_per_second": 1.466, + "eval_f1": 0.31970260223048325, + "eval_loss": 0.6055669188499451, + "eval_roc_auc": 0.6824534161490683, + "eval_runtime": 2.0249, + "eval_samples_per_second": 20.742, + "eval_steps_per_second": 1.482, "step": 77 }, { "epoch": 56.73, "eval_accuracy": 0.0, - "eval_f1": 0.3252595155709343, - "eval_loss": 0.5894604921340942, - "eval_roc_auc": 0.6979813664596274, - "eval_runtime": 2.1145, - "eval_samples_per_second": 19.863, - "eval_steps_per_second": 1.419, + "eval_f1": 0.3249097472924187, + "eval_loss": 0.6073567867279053, + "eval_roc_auc": 0.6912525879917184, + "eval_runtime": 1.9844, + "eval_samples_per_second": 21.165, + "eval_steps_per_second": 1.512, "step": 78 }, { "epoch": 57.45, - "eval_accuracy": 0.0, - "eval_f1": 0.35294117647058826, - "eval_loss": 0.5903995037078857, - "eval_roc_auc": 0.7238612836438924, - "eval_runtime": 2.1187, - "eval_samples_per_second": 19.824, - "eval_steps_per_second": 1.416, + "eval_accuracy": 0.023809523809523808, + "eval_f1": 0.3157894736842105, + "eval_loss": 0.6030458807945251, + "eval_roc_auc": 0.677536231884058, + "eval_runtime": 1.8981, + "eval_samples_per_second": 22.127, + "eval_steps_per_second": 1.581, "step": 79 }, { "epoch": 58.91, - "eval_accuracy": 0.0, - "eval_f1": 0.33793103448275863, - "eval_loss": 0.5916685461997986, - "eval_roc_auc": 0.7104037267080746, - "eval_runtime": 2.1241, - "eval_samples_per_second": 19.773, - "eval_steps_per_second": 1.412, + "eval_accuracy": 0.023809523809523808, + "eval_f1": 0.3358778625954198, + "eval_loss": 0.6000563502311707, + "eval_roc_auc": 0.6925465838509317, + "eval_runtime": 2.137, + "eval_samples_per_second": 19.653, + "eval_steps_per_second": 1.404, "step": 81 }, { "epoch": 59.64, - "eval_accuracy": 0.0, - "eval_f1": 0.3298245614035088, - "eval_loss": 0.5893025994300842, - "eval_roc_auc": 0.7000517598343685, - "eval_runtime": 2.1145, - "eval_samples_per_second": 19.863, - "eval_steps_per_second": 1.419, + "eval_accuracy": 0.023809523809523808, + "eval_f1": 0.3409090909090909, + "eval_loss": 0.5992631316184998, + "eval_roc_auc": 0.6979813664596273, + "eval_runtime": 1.8122, + "eval_samples_per_second": 23.177, + "eval_steps_per_second": 1.655, "step": 82 }, { "epoch": 60.36, - "eval_accuracy": 0.0, - "eval_f1": 0.36823104693140796, - "eval_loss": 0.5839800238609314, - "eval_roc_auc": 0.7300724637681159, - "eval_runtime": 2.0295, - "eval_samples_per_second": 20.695, - "eval_steps_per_second": 1.478, + "eval_accuracy": 0.023809523809523808, + "eval_f1": 0.32592592592592595, + "eval_loss": 0.6017413139343262, + "eval_roc_auc": 0.6884057971014492, + "eval_runtime": 2.0667, + "eval_samples_per_second": 20.323, + "eval_steps_per_second": 1.452, "step": 83 }, { "epoch": 61.82, "eval_accuracy": 0.023809523809523808, - "eval_f1": 0.3636363636363636, - "eval_loss": 0.5821632146835327, - "eval_roc_auc": 0.7246376811594203, - "eval_runtime": 2.1312, - "eval_samples_per_second": 19.707, - "eval_steps_per_second": 1.408, + "eval_f1": 0.3146067415730337, + "eval_loss": 0.6009005308151245, + "eval_roc_auc": 0.6770186335403726, + "eval_runtime": 1.9156, + "eval_samples_per_second": 21.925, + "eval_steps_per_second": 1.566, "step": 85 }, { "epoch": 62.55, "eval_accuracy": 0.023809523809523808, - "eval_f1": 0.35507246376811596, - "eval_loss": 0.5816187858581543, - "eval_roc_auc": 0.7176501035196688, - "eval_runtime": 2.1966, - "eval_samples_per_second": 19.121, - "eval_steps_per_second": 1.366, + "eval_f1": 0.31970260223048325, + "eval_loss": 0.6017952561378479, + "eval_roc_auc": 0.6824534161490683, + "eval_runtime": 1.9734, + "eval_samples_per_second": 21.283, + "eval_steps_per_second": 1.52, "step": 86 }, { "epoch": 64.0, - "eval_accuracy": 0.0, - "eval_f1": 0.34532374100719426, - "eval_loss": 0.5823355317115784, - "eval_roc_auc": 0.7101449275362319, - "eval_runtime": 2.4993, - "eval_samples_per_second": 16.804, - "eval_steps_per_second": 1.2, + "eval_accuracy": 0.023809523809523808, + "eval_f1": 0.3129770992366412, + "eval_loss": 0.5975026488304138, + "eval_roc_auc": 0.6731366459627329, + "eval_runtime": 1.9304, + "eval_samples_per_second": 21.757, + "eval_steps_per_second": 1.554, "step": 88 }, { "epoch": 64.73, - "eval_accuracy": 0.0, - "eval_f1": 0.3537906137184116, - "eval_loss": 0.5836206674575806, - "eval_roc_auc": 0.7171325051759834, - "eval_runtime": 1.9876, - "eval_samples_per_second": 21.131, - "eval_steps_per_second": 1.509, + "eval_accuracy": 0.023809523809523808, + "eval_f1": 0.3271375464684015, + "eval_loss": 0.5978490710258484, + "eval_roc_auc": 0.6889233954451346, + "eval_runtime": 1.993, + "eval_samples_per_second": 21.073, + "eval_steps_per_second": 1.505, "step": 89 }, { "epoch": 65.45, - "eval_accuracy": 0.0, - "eval_f1": 0.36823104693140796, - "eval_loss": 0.5811671018600464, - "eval_roc_auc": 0.7300724637681159, - "eval_runtime": 2.2487, - "eval_samples_per_second": 18.677, - "eval_steps_per_second": 1.334, + "eval_accuracy": 0.023809523809523808, + "eval_f1": 0.34241245136186765, + "eval_loss": 0.5966595411300659, + "eval_roc_auc": 0.6951345755693582, + "eval_runtime": 1.8441, + "eval_samples_per_second": 22.775, + "eval_steps_per_second": 1.627, "step": 90 }, { "epoch": 66.91, "eval_accuracy": 0.023809523809523808, - "eval_f1": 0.35294117647058826, - "eval_loss": 0.5822634100914001, - "eval_roc_auc": 0.7132505175983437, - "eval_runtime": 2.1189, - "eval_samples_per_second": 19.821, - "eval_steps_per_second": 1.416, + "eval_f1": 0.3125, + "eval_loss": 0.5973162055015564, + "eval_roc_auc": 0.6697722567287785, + "eval_runtime": 1.9347, + "eval_samples_per_second": 21.708, + "eval_steps_per_second": 1.551, "step": 92 }, { "epoch": 67.64, "eval_accuracy": 0.023809523809523808, - "eval_f1": 0.3468634686346863, - "eval_loss": 0.5832495093345642, - "eval_roc_auc": 0.7072981366459627, - "eval_runtime": 2.1564, - "eval_samples_per_second": 19.477, - "eval_steps_per_second": 1.391, + "eval_f1": 0.3371647509578544, + "eval_loss": 0.5955641865730286, + "eval_roc_auc": 0.693064182194617, + "eval_runtime": 1.9179, + "eval_samples_per_second": 21.899, + "eval_steps_per_second": 1.564, "step": 93 }, { "epoch": 68.36, "eval_accuracy": 0.023809523809523808, - "eval_f1": 0.34558823529411764, - "eval_loss": 0.5849612355232239, - "eval_roc_auc": 0.7067805383022774, - "eval_runtime": 2.2328, - "eval_samples_per_second": 18.811, - "eval_steps_per_second": 1.344, + "eval_f1": 0.3372549019607843, + "eval_loss": 0.5921858549118042, + "eval_roc_auc": 0.6896997929606625, + "eval_runtime": 1.9017, + "eval_samples_per_second": 22.086, + "eval_steps_per_second": 1.578, "step": 94 }, { "epoch": 69.82, - "eval_accuracy": 0.0, - "eval_f1": 0.35125448028673834, - "eval_loss": 0.5786746144294739, - "eval_roc_auc": 0.7160973084886129, - "eval_runtime": 2.0885, - "eval_samples_per_second": 20.11, - "eval_steps_per_second": 1.436, + "eval_accuracy": 0.047619047619047616, + "eval_f1": 0.33201581027667987, + "eval_loss": 0.5948788523674011, + "eval_roc_auc": 0.6842650103519669, + "eval_runtime": 1.8776, + "eval_samples_per_second": 22.369, + "eval_steps_per_second": 1.598, "step": 96 }, { "epoch": 70.55, - "eval_accuracy": 0.0, - "eval_f1": 0.3695652173913044, - "eval_loss": 0.5768818855285645, - "eval_roc_auc": 0.7305900621118012, - "eval_runtime": 2.1063, - "eval_samples_per_second": 19.94, - "eval_steps_per_second": 1.424, + "eval_accuracy": 0.047619047619047616, + "eval_f1": 0.3412698412698413, + "eval_loss": 0.5959014892578125, + "eval_roc_auc": 0.6912525879917184, + "eval_runtime": 2.0178, + "eval_samples_per_second": 20.814, + "eval_steps_per_second": 1.487, "step": 97 }, { "epoch": 72.0, - "eval_accuracy": 0.0, - "eval_f1": 0.3684210526315789, - "eval_loss": 0.5726543068885803, - "eval_roc_auc": 0.7228260869565218, - "eval_runtime": 2.2057, - "eval_samples_per_second": 19.042, - "eval_steps_per_second": 1.36, + "eval_accuracy": 0.023809523809523808, + "eval_f1": 0.3420074349442379, + "eval_loss": 0.5944311022758484, + "eval_roc_auc": 0.7018633540372671, + "eval_runtime": 1.8838, + "eval_samples_per_second": 22.295, + "eval_steps_per_second": 1.593, "step": 99 }, { "epoch": 72.73, - "eval_accuracy": 0.0, - "eval_f1": 0.3863636363636363, - "eval_loss": 0.5710762739181519, - "eval_roc_auc": 0.736801242236025, - "eval_runtime": 2.1542, - "eval_samples_per_second": 19.496, - "eval_steps_per_second": 1.393, + "eval_accuracy": 0.047619047619047616, + "eval_f1": 0.3333333333333333, + "eval_loss": 0.5955294370651245, + "eval_roc_auc": 0.6881469979296065, + "eval_runtime": 2.0716, + "eval_samples_per_second": 20.275, + "eval_steps_per_second": 1.448, "step": 100 }, { "epoch": 73.45, - "eval_accuracy": 0.0, - "eval_f1": 0.37037037037037035, - "eval_loss": 0.5743962526321411, - "eval_roc_auc": 0.7272256728778468, - "eval_runtime": 2.0601, - "eval_samples_per_second": 20.387, - "eval_steps_per_second": 1.456, + "eval_accuracy": 0.023809523809523808, + "eval_f1": 0.3346303501945525, + "eval_loss": 0.5932918190956116, + "eval_roc_auc": 0.6886645962732919, + "eval_runtime": 1.9689, + "eval_samples_per_second": 21.332, + "eval_steps_per_second": 1.524, "step": 101 }, { "epoch": 74.91, - "eval_accuracy": 0.0, - "eval_f1": 0.36363636363636365, - "eval_loss": 0.5718241333961487, - "eval_roc_auc": 0.7173913043478259, - "eval_runtime": 2.3822, - "eval_samples_per_second": 17.631, - "eval_steps_per_second": 1.259, + "eval_accuracy": 0.023809523809523808, + "eval_f1": 0.3543307086614173, + "eval_loss": 0.5894082188606262, + "eval_roc_auc": 0.7031573498964803, + "eval_runtime": 2.0135, + "eval_samples_per_second": 20.859, + "eval_steps_per_second": 1.49, "step": 103 }, { "epoch": 75.64, - "eval_accuracy": 0.0, - "eval_f1": 0.36296296296296293, - "eval_loss": 0.5745264887809753, - "eval_roc_auc": 0.7207556935817806, - "eval_runtime": 2.1941, - "eval_samples_per_second": 19.143, - "eval_steps_per_second": 1.367, + "eval_accuracy": 0.023809523809523808, + "eval_f1": 0.34241245136186765, + "eval_loss": 0.5903118848800659, + "eval_roc_auc": 0.6951345755693582, + "eval_runtime": 1.8434, + "eval_samples_per_second": 22.784, + "eval_steps_per_second": 1.627, "step": 104 }, { "epoch": 76.36, - "eval_accuracy": 0.0, - "eval_f1": 0.3676470588235294, - "eval_loss": 0.5776281952857971, - "eval_roc_auc": 0.7261904761904763, - "eval_runtime": 2.4317, - "eval_samples_per_second": 17.272, - "eval_steps_per_second": 1.234, + "eval_accuracy": 0.047619047619047616, + "eval_f1": 0.3410852713178295, + "eval_loss": 0.5889719128608704, + "eval_roc_auc": 0.694616977225673, + "eval_runtime": 2.0369, + "eval_samples_per_second": 20.62, + "eval_steps_per_second": 1.473, "step": 105 }, { "epoch": 77.82, - "eval_accuracy": 0.023809523809523808, - "eval_f1": 0.36296296296296293, - "eval_loss": 0.5703898668289185, - "eval_roc_auc": 0.7207556935817806, - "eval_runtime": 2.0787, - "eval_samples_per_second": 20.205, - "eval_steps_per_second": 1.443, + "eval_accuracy": 0.047619047619047616, + "eval_f1": 0.3346303501945525, + "eval_loss": 0.5922066569328308, + "eval_roc_auc": 0.6886645962732919, + "eval_runtime": 1.8672, + "eval_samples_per_second": 22.493, + "eval_steps_per_second": 1.607, "step": 107 }, { "epoch": 78.55, - "eval_accuracy": 0.0, - "eval_f1": 0.37547892720306514, - "eval_loss": 0.5683363676071167, - "eval_roc_auc": 0.7254140786749483, - "eval_runtime": 2.2741, - "eval_samples_per_second": 18.469, - "eval_steps_per_second": 1.319, + "eval_accuracy": 0.047619047619047616, + "eval_f1": 0.32432432432432434, + "eval_loss": 0.5923347473144531, + "eval_roc_auc": 0.6811594202898552, + "eval_runtime": 2.0697, + "eval_samples_per_second": 20.293, + "eval_steps_per_second": 1.449, "step": 108 }, { "epoch": 80.0, - "eval_accuracy": 0.023809523809523808, - "eval_f1": 0.41600000000000004, - "eval_loss": 0.5640491247177124, - "eval_roc_auc": 0.7505175983436853, - "eval_runtime": 2.3409, - "eval_samples_per_second": 17.941, - "eval_steps_per_second": 1.282, + "eval_accuracy": 0.047619047619047616, + "eval_f1": 0.3467741935483871, + "eval_loss": 0.590828537940979, + "eval_roc_auc": 0.6933229813664595, + "eval_runtime": 1.8922, + "eval_samples_per_second": 22.196, + "eval_steps_per_second": 1.585, "step": 110 }, { "epoch": 80.73, - "eval_accuracy": 0.023809523809523808, - "eval_f1": 0.38095238095238093, - "eval_loss": 0.5674084424972534, - "eval_roc_auc": 0.7236024844720497, - "eval_runtime": 2.19, - "eval_samples_per_second": 19.178, - "eval_steps_per_second": 1.37, + "eval_accuracy": 0.047619047619047616, + "eval_f1": 0.328, + "eval_loss": 0.5922424793243408, + "eval_roc_auc": 0.6793478260869565, + "eval_runtime": 2.0683, + "eval_samples_per_second": 20.307, + "eval_steps_per_second": 1.45, "step": 111 }, { "epoch": 81.45, "eval_accuracy": 0.023809523809523808, - "eval_f1": 0.3730158730158731, - "eval_loss": 0.564328670501709, - "eval_roc_auc": 0.7171325051759834, - "eval_runtime": 2.4742, - "eval_samples_per_second": 16.975, - "eval_steps_per_second": 1.213, + "eval_f1": 0.34400000000000003, + "eval_loss": 0.5891572833061218, + "eval_roc_auc": 0.692287784679089, + "eval_runtime": 1.8345, + "eval_samples_per_second": 22.894, + "eval_steps_per_second": 1.635, "step": 112 }, { "epoch": 82.91, "eval_accuracy": 0.023809523809523808, - "eval_f1": 0.3813229571984436, - "eval_loss": 0.5711435675621033, - "eval_roc_auc": 0.7274844720496896, - "eval_runtime": 2.1624, - "eval_samples_per_second": 19.423, - "eval_steps_per_second": 1.387, + "eval_f1": 0.35059760956175307, + "eval_loss": 0.5879591703414917, + "eval_roc_auc": 0.69824016563147, + "eval_runtime": 1.9744, + "eval_samples_per_second": 21.272, + "eval_steps_per_second": 1.519, "step": 114 }, { "epoch": 83.64, - "eval_accuracy": 0.023809523809523808, - "eval_f1": 0.366412213740458, - "eval_loss": 0.5771781206130981, - "eval_roc_auc": 0.7184265010351967, - "eval_runtime": 2.2132, - "eval_samples_per_second": 18.977, - "eval_steps_per_second": 1.356, + "eval_accuracy": 0.047619047619047616, + "eval_f1": 0.34538152610441764, + "eval_loss": 0.5868695378303528, + "eval_roc_auc": 0.6928053830227743, + "eval_runtime": 1.943, + "eval_samples_per_second": 21.616, + "eval_steps_per_second": 1.544, "step": 115 }, { "epoch": 84.36, "eval_accuracy": 0.023809523809523808, - "eval_f1": 0.3582089552238805, - "eval_loss": 0.5758051872253418, - "eval_roc_auc": 0.7153209109730849, - "eval_runtime": 2.2313, - "eval_samples_per_second": 18.823, - "eval_steps_per_second": 1.345, + "eval_f1": 0.3464566929133858, + "eval_loss": 0.5840514302253723, + "eval_roc_auc": 0.6966873706004141, + "eval_runtime": 2.0388, + "eval_samples_per_second": 20.6, + "eval_steps_per_second": 1.471, "step": 116 }, { "epoch": 85.82, - "eval_accuracy": 0.0, - "eval_f1": 0.37209302325581395, - "eval_loss": 0.5687234401702881, - "eval_roc_auc": 0.7204968944099378, - "eval_runtime": 2.117, - "eval_samples_per_second": 19.839, - "eval_steps_per_second": 1.417, + "eval_accuracy": 0.07142857142857142, + "eval_f1": 0.35684647302904565, + "eval_loss": 0.5840616226196289, + "eval_roc_auc": 0.6969461697722568, + "eval_runtime": 1.9152, + "eval_samples_per_second": 21.93, + "eval_steps_per_second": 1.566, "step": 118 }, { "epoch": 86.55, - "eval_accuracy": 0.0, - "eval_f1": 0.3904382470119522, - "eval_loss": 0.565062403678894, - "eval_roc_auc": 0.7305900621118013, - "eval_runtime": 2.145, - "eval_samples_per_second": 19.58, - "eval_steps_per_second": 1.399, + "eval_accuracy": 0.047619047619047616, + "eval_f1": 0.3495934959349593, + "eval_loss": 0.5842622518539429, + "eval_roc_auc": 0.6943581780538302, + "eval_runtime": 2.0014, + "eval_samples_per_second": 20.985, + "eval_steps_per_second": 1.499, "step": 119 }, { "epoch": 88.0, - "eval_accuracy": 0.0, - "eval_f1": 0.4, - "eval_loss": 0.5594702959060669, - "eval_roc_auc": 0.7375776397515528, - "eval_runtime": 2.141, - "eval_samples_per_second": 19.617, - "eval_steps_per_second": 1.401, + "eval_accuracy": 0.047619047619047616, + "eval_f1": 0.3598326359832636, + "eval_loss": 0.5860231518745422, + "eval_roc_auc": 0.6979813664596273, + "eval_runtime": 1.7516, + "eval_samples_per_second": 23.978, + "eval_steps_per_second": 1.713, "step": 121 }, { "epoch": 88.73, - "eval_accuracy": 0.0, - "eval_f1": 0.421875, - "eval_loss": 0.5597259998321533, - "eval_roc_auc": 0.760351966873706, - "eval_runtime": 2.2421, - "eval_samples_per_second": 18.733, - "eval_steps_per_second": 1.338, + "eval_accuracy": 0.047619047619047616, + "eval_f1": 0.34567901234567905, + "eval_loss": 0.5837295055389404, + "eval_roc_auc": 0.6894409937888197, + "eval_runtime": 2.0716, + "eval_samples_per_second": 20.275, + "eval_steps_per_second": 1.448, "step": 122 }, { "epoch": 89.45, - "eval_accuracy": 0.0, - "eval_f1": 0.4156862745098039, - "eval_loss": 0.5603341460227966, - "eval_roc_auc": 0.754399585921325, - "eval_runtime": 2.1509, - "eval_samples_per_second": 19.527, - "eval_steps_per_second": 1.395, + "eval_accuracy": 0.07142857142857142, + "eval_f1": 0.3636363636363636, + "eval_loss": 0.5826032161712646, + "eval_roc_auc": 0.7028985507246376, + "eval_runtime": 2.155, + "eval_samples_per_second": 19.49, + "eval_steps_per_second": 1.392, "step": 123 }, { "epoch": 90.91, - "eval_accuracy": 0.0, - "eval_f1": 0.4108527131782946, - "eval_loss": 0.554810643196106, - "eval_roc_auc": 0.7528467908902691, - "eval_runtime": 2.0949, - "eval_samples_per_second": 20.049, - "eval_steps_per_second": 1.432, + "eval_accuracy": 0.07142857142857142, + "eval_f1": 0.3651452282157676, + "eval_loss": 0.5821616053581238, + "eval_roc_auc": 0.7034161490683231, + "eval_runtime": 1.8665, + "eval_samples_per_second": 22.502, + "eval_steps_per_second": 1.607, "step": 125 }, { "epoch": 91.64, - "eval_accuracy": 0.0, - "eval_f1": 0.41406249999999994, - "eval_loss": 0.55705326795578, - "eval_roc_auc": 0.7538819875776397, - "eval_runtime": 2.1818, - "eval_samples_per_second": 19.25, - "eval_steps_per_second": 1.375, + "eval_accuracy": 0.07142857142857142, + "eval_f1": 0.36065573770491804, + "eval_loss": 0.5813528299331665, + "eval_roc_auc": 0.701863354037267, + "eval_runtime": 2.0722, + "eval_samples_per_second": 20.268, + "eval_steps_per_second": 1.448, "step": 126 }, { "epoch": 92.36, - "eval_accuracy": 0.023809523809523808, - "eval_f1": 0.40476190476190477, - "eval_loss": 0.5568869709968567, - "eval_roc_auc": 0.7430124223602484, - "eval_runtime": 2.1783, - "eval_samples_per_second": 19.281, - "eval_steps_per_second": 1.377, + "eval_accuracy": 0.047619047619047616, + "eval_f1": 0.3629032258064516, + "eval_loss": 0.5813586115837097, + "eval_roc_auc": 0.7062629399585921, + "eval_runtime": 1.8413, + "eval_samples_per_second": 22.809, + "eval_steps_per_second": 1.629, "step": 127 }, { "epoch": 93.82, - "eval_accuracy": 0.023809523809523808, - "eval_f1": 0.41732283464566927, - "eval_loss": 0.5582275986671448, - "eval_roc_auc": 0.7549171842650103, - "eval_runtime": 2.1395, - "eval_samples_per_second": 19.63, - "eval_steps_per_second": 1.402, + "eval_accuracy": 0.07142857142857142, + "eval_f1": 0.37130801687763715, + "eval_loss": 0.5817737579345703, + "eval_roc_auc": 0.7054865424430642, + "eval_runtime": 2.0383, + "eval_samples_per_second": 20.605, + "eval_steps_per_second": 1.472, "step": 129 }, { "epoch": 94.55, - "eval_accuracy": 0.023809523809523808, - "eval_f1": 0.40310077519379844, - "eval_loss": 0.561271071434021, - "eval_roc_auc": 0.7463768115942029, - "eval_runtime": 2.1505, - "eval_samples_per_second": 19.53, - "eval_steps_per_second": 1.395, + "eval_accuracy": 0.07142857142857142, + "eval_f1": 0.3765690376569037, + "eval_loss": 0.5802175402641296, + "eval_roc_auc": 0.7109213250517598, + "eval_runtime": 1.9338, + "eval_samples_per_second": 21.718, + "eval_steps_per_second": 1.551, "step": 130 }, { "epoch": 96.0, - "eval_accuracy": 0.023809523809523808, - "eval_f1": 0.40800000000000003, - "eval_loss": 0.5574062466621399, - "eval_roc_auc": 0.7440476190476191, - "eval_runtime": 2.2873, - "eval_samples_per_second": 18.362, - "eval_steps_per_second": 1.312, + "eval_accuracy": 0.07142857142857142, + "eval_f1": 0.36752136752136755, + "eval_loss": 0.5802630186080933, + "eval_roc_auc": 0.7005693581780539, + "eval_runtime": 1.9997, + "eval_samples_per_second": 21.003, + "eval_steps_per_second": 1.5, "step": 132 }, { "epoch": 96.73, - "eval_accuracy": 0.0, - "eval_f1": 0.421875, - "eval_loss": 0.5570902228355408, - "eval_roc_auc": 0.760351966873706, - "eval_runtime": 2.2328, - "eval_samples_per_second": 18.81, - "eval_steps_per_second": 1.344, + "eval_accuracy": 0.07142857142857142, + "eval_f1": 0.351931330472103, + "eval_loss": 0.5825139880180359, + "eval_roc_auc": 0.6881469979296067, + "eval_runtime": 2.0169, + "eval_samples_per_second": 20.824, + "eval_steps_per_second": 1.487, "step": 133 }, { "epoch": 97.45, - "eval_accuracy": 0.023809523809523808, - "eval_f1": 0.41732283464566927, - "eval_loss": 0.5585192441940308, - "eval_roc_auc": 0.7549171842650103, - "eval_runtime": 2.1462, - "eval_samples_per_second": 19.569, - "eval_steps_per_second": 1.398, + "eval_accuracy": 0.07142857142857142, + "eval_f1": 0.3628691983122363, + "eval_loss": 0.5789888501167297, + "eval_roc_auc": 0.6990165631469979, + "eval_runtime": 2.0766, + "eval_samples_per_second": 20.225, + "eval_steps_per_second": 1.445, "step": 134 }, { "epoch": 98.91, - "eval_accuracy": 0.0, - "eval_f1": 0.40625, - "eval_loss": 0.5598769783973694, - "eval_roc_auc": 0.7474120082815735, - "eval_runtime": 2.1253, - "eval_samples_per_second": 19.762, - "eval_steps_per_second": 1.412, + "eval_accuracy": 0.07142857142857142, + "eval_f1": 0.3765690376569037, + "eval_loss": 0.5794755816459656, + "eval_roc_auc": 0.7109213250517598, + "eval_runtime": 1.8413, + "eval_samples_per_second": 22.81, + "eval_steps_per_second": 1.629, "step": 136 }, { "epoch": 99.64, - "eval_accuracy": 0.023809523809523808, - "eval_f1": 0.40476190476190477, - "eval_loss": 0.5586134195327759, - "eval_roc_auc": 0.7430124223602484, - "eval_runtime": 2.1887, - "eval_samples_per_second": 19.189, - "eval_steps_per_second": 1.371, + "eval_accuracy": 0.07142857142857142, + "eval_f1": 0.3697478991596639, + "eval_loss": 0.5784409642219543, + "eval_roc_auc": 0.7049689440993788, + "eval_runtime": 2.0156, + "eval_samples_per_second": 20.837, + "eval_steps_per_second": 1.488, "step": 137 }, { "epoch": 100.36, - "eval_accuracy": 0.047619047619047616, - "eval_f1": 0.4112903225806452, - "eval_loss": 0.5581173300743103, - "eval_roc_auc": 0.7450828157349897, - "eval_runtime": 2.1216, - "eval_samples_per_second": 19.796, - "eval_steps_per_second": 1.414, + "eval_accuracy": 0.07142857142857142, + "eval_f1": 0.35833333333333334, + "eval_loss": 0.5818943977355957, + "eval_roc_auc": 0.697463768115942, + "eval_runtime": 1.8187, + "eval_samples_per_second": 23.093, + "eval_steps_per_second": 1.65, "step": 138 }, { "epoch": 101.82, "eval_accuracy": 0.047619047619047616, - "eval_f1": 0.4117647058823529, - "eval_loss": 0.554703414440155, - "eval_roc_auc": 0.7373188405797102, - "eval_runtime": 2.199, - "eval_samples_per_second": 19.1, - "eval_steps_per_second": 1.364, + "eval_f1": 0.35245901639344257, + "eval_loss": 0.5833745002746582, + "eval_roc_auc": 0.6953933747412008, + "eval_runtime": 2.073, + "eval_samples_per_second": 20.261, + "eval_steps_per_second": 1.447, "step": 140 }, { "epoch": 102.55, - "eval_accuracy": 0.047619047619047616, - "eval_f1": 0.430379746835443, - "eval_loss": 0.5479743480682373, - "eval_roc_auc": 0.7507763975155279, - "eval_runtime": 2.2214, - "eval_samples_per_second": 18.907, - "eval_steps_per_second": 1.35, + "eval_accuracy": 0.023809523809523808, + "eval_f1": 0.3688524590163934, + "eval_loss": 0.5825251340866089, + "eval_roc_auc": 0.7083333333333333, + "eval_runtime": 1.9912, + "eval_samples_per_second": 21.093, + "eval_steps_per_second": 1.507, "step": 141 }, { "epoch": 104.0, - "eval_accuracy": 0.023809523809523808, - "eval_f1": 0.43333333333333335, - "eval_loss": 0.5521491169929504, - "eval_roc_auc": 0.7556935817805382, - "eval_runtime": 2.0733, - "eval_samples_per_second": 20.258, - "eval_steps_per_second": 1.447, + "eval_accuracy": 0.07142857142857142, + "eval_f1": 0.3459915611814346, + "eval_loss": 0.5839240550994873, + "eval_roc_auc": 0.6860766045548654, + "eval_runtime": 2.1002, + "eval_samples_per_second": 19.998, + "eval_steps_per_second": 1.428, "step": 143 }, { "epoch": 104.73, "eval_accuracy": 0.047619047619047616, - "eval_f1": 0.4193548387096774, - "eval_loss": 0.5539843440055847, - "eval_roc_auc": 0.7515527950310559, - "eval_runtime": 2.1533, - "eval_samples_per_second": 19.505, - "eval_steps_per_second": 1.393, + "eval_f1": 0.3333333333333333, + "eval_loss": 0.5838228464126587, + "eval_roc_auc": 0.6814182194616977, + "eval_runtime": 1.9156, + "eval_samples_per_second": 21.926, + "eval_steps_per_second": 1.566, "step": 144 }, { "epoch": 105.45, - "eval_accuracy": 0.047619047619047616, - "eval_f1": 0.4032921810699588, - "eval_loss": 0.5572003126144409, - "eval_roc_auc": 0.7347308488612837, - "eval_runtime": 2.1589, - "eval_samples_per_second": 19.454, - "eval_steps_per_second": 1.39, + "eval_accuracy": 0.023809523809523808, + "eval_f1": 0.33870967741935487, + "eval_loss": 0.580091655254364, + "eval_roc_auc": 0.6868530020703933, + "eval_runtime": 2.1494, + "eval_samples_per_second": 19.541, + "eval_steps_per_second": 1.396, "step": 145 }, { "epoch": 106.91, - "eval_accuracy": 0.023809523809523808, - "eval_f1": 0.40322580645161293, - "eval_loss": 0.5563521981239319, - "eval_roc_auc": 0.7386128364389234, - "eval_runtime": 2.1472, - "eval_samples_per_second": 19.561, - "eval_steps_per_second": 1.397, + "eval_accuracy": 0.047619047619047616, + "eval_f1": 0.35146443514644354, + "eval_loss": 0.581135630607605, + "eval_roc_auc": 0.6915113871635612, + "eval_runtime": 1.8396, + "eval_samples_per_second": 22.831, + "eval_steps_per_second": 1.631, "step": 147 }, { "epoch": 107.64, "eval_accuracy": 0.047619047619047616, - "eval_f1": 0.39669421487603307, - "eval_loss": 0.5568532943725586, - "eval_roc_auc": 0.7287784679089027, - "eval_runtime": 2.2259, - "eval_samples_per_second": 18.869, - "eval_steps_per_second": 1.348, + "eval_f1": 0.3374485596707819, + "eval_loss": 0.5792709589004517, + "eval_roc_auc": 0.6829710144927537, + "eval_runtime": 2.1328, + "eval_samples_per_second": 19.693, + "eval_steps_per_second": 1.407, "step": 148 }, { "epoch": 108.36, "eval_accuracy": 0.07142857142857142, - "eval_f1": 0.41201716738197425, - "eval_loss": 0.557677149772644, - "eval_roc_auc": 0.7334368530020704, - "eval_runtime": 2.0947, - "eval_samples_per_second": 20.051, - "eval_steps_per_second": 1.432, + "eval_f1": 0.3448275862068966, + "eval_loss": 0.5765911936759949, + "eval_roc_auc": 0.6821946169772257, + "eval_runtime": 2.0548, + "eval_samples_per_second": 20.44, + "eval_steps_per_second": 1.46, "step": 149 }, { "epoch": 109.82, - "eval_accuracy": 0.0, - "eval_f1": 0.39004149377593367, - "eval_loss": 0.552203357219696, - "eval_roc_auc": 0.7228260869565217, - "eval_runtime": 2.2966, - "eval_samples_per_second": 18.288, - "eval_steps_per_second": 1.306, + "eval_accuracy": 0.07142857142857142, + "eval_f1": 0.34453781512605036, + "eval_loss": 0.5759981870651245, + "eval_roc_auc": 0.6855590062111802, + "eval_runtime": 2.1017, + "eval_samples_per_second": 19.984, + "eval_steps_per_second": 1.427, "step": 151 }, { "epoch": 110.55, - "eval_accuracy": 0.0, - "eval_f1": 0.3949579831932773, - "eval_loss": 0.5478394031524658, - "eval_roc_auc": 0.7243788819875777, - "eval_runtime": 2.2413, - "eval_samples_per_second": 18.739, - "eval_steps_per_second": 1.339, + "eval_accuracy": 0.07142857142857142, + "eval_f1": 0.3559322033898305, + "eval_loss": 0.5757153034210205, + "eval_roc_auc": 0.693064182194617, + "eval_runtime": 1.8256, + "eval_samples_per_second": 23.006, + "eval_steps_per_second": 1.643, "step": 152 }, { "epoch": 112.0, - "eval_accuracy": 0.023809523809523808, - "eval_f1": 0.41841004184100417, - "eval_loss": 0.5458911061286926, - "eval_roc_auc": 0.7432712215320911, - "eval_runtime": 2.2404, - "eval_samples_per_second": 18.747, - "eval_steps_per_second": 1.339, + "eval_accuracy": 0.07142857142857142, + "eval_f1": 0.3474576271186441, + "eval_loss": 0.5759687423706055, + "eval_roc_auc": 0.6865942028985509, + "eval_runtime": 1.998, + "eval_samples_per_second": 21.021, + "eval_steps_per_second": 1.502, "step": 154 }, { "epoch": 112.73, - "eval_accuracy": 0.047619047619047616, - "eval_f1": 0.40506329113924044, - "eval_loss": 0.5490586161613464, - "eval_roc_auc": 0.731366459627329, - "eval_runtime": 2.2886, - "eval_samples_per_second": 18.352, - "eval_steps_per_second": 1.311, + "eval_accuracy": 0.07142857142857142, + "eval_f1": 0.3628691983122363, + "eval_loss": 0.5742767453193665, + "eval_roc_auc": 0.6990165631469979, + "eval_runtime": 2.0687, + "eval_samples_per_second": 20.303, + "eval_steps_per_second": 1.45, "step": 155 }, { "epoch": 113.45, - "eval_accuracy": 0.023809523809523808, - "eval_f1": 0.41841004184100417, - "eval_loss": 0.5519034266471863, - "eval_roc_auc": 0.7432712215320911, - "eval_runtime": 2.1727, - "eval_samples_per_second": 19.331, - "eval_steps_per_second": 1.381, + "eval_accuracy": 0.07142857142857142, + "eval_f1": 0.3636363636363636, + "eval_loss": 0.5731971263885498, + "eval_roc_auc": 0.7028985507246376, + "eval_runtime": 1.9431, + "eval_samples_per_second": 21.615, + "eval_steps_per_second": 1.544, "step": 156 }, { "epoch": 114.91, - "eval_accuracy": 0.023809523809523808, - "eval_f1": 0.3950617283950617, - "eval_loss": 0.550848662853241, - "eval_roc_auc": 0.7282608695652175, - "eval_runtime": 2.5249, - "eval_samples_per_second": 16.635, - "eval_steps_per_second": 1.188, + "eval_accuracy": 0.047619047619047616, + "eval_f1": 0.37860082304526754, + "eval_loss": 0.5735821723937988, + "eval_roc_auc": 0.7153209109730849, + "eval_runtime": 2.067, + "eval_samples_per_second": 20.32, + "eval_steps_per_second": 1.451, "step": 158 }, { "epoch": 115.64, - "eval_accuracy": 0.0, - "eval_f1": 0.4032921810699588, - "eval_loss": 0.5499736070632935, - "eval_roc_auc": 0.7347308488612837, - "eval_runtime": 2.203, - "eval_samples_per_second": 19.065, - "eval_steps_per_second": 1.362, + "eval_accuracy": 0.023809523809523808, + "eval_f1": 0.36666666666666664, + "eval_loss": 0.5763630867004395, + "eval_roc_auc": 0.7039337474120083, + "eval_runtime": 1.876, + "eval_samples_per_second": 22.388, + "eval_steps_per_second": 1.599, "step": 159 }, { "epoch": 116.36, - "eval_accuracy": 0.023809523809523808, - "eval_f1": 0.38866396761133604, - "eval_loss": 0.5513898134231567, - "eval_roc_auc": 0.7261904761904762, - "eval_runtime": 2.1649, - "eval_samples_per_second": 19.4, - "eval_steps_per_second": 1.386, + "eval_accuracy": 0.047619047619047616, + "eval_f1": 0.361344537815126, + "eval_loss": 0.5764896273612976, + "eval_roc_auc": 0.6984989648033126, + "eval_runtime": 2.186, + "eval_samples_per_second": 19.213, + "eval_steps_per_second": 1.372, "step": 160 }, { "epoch": 117.82, - "eval_accuracy": 0.047619047619047616, - "eval_f1": 0.4255319148936171, - "eval_loss": 0.544844388961792, - "eval_roc_auc": 0.7453416149068324, - "eval_runtime": 2.2226, - "eval_samples_per_second": 18.897, - "eval_steps_per_second": 1.35, + "eval_accuracy": 0.07142857142857142, + "eval_f1": 0.35744680851063826, + "eval_loss": 0.5749201774597168, + "eval_roc_auc": 0.6935817805383023, + "eval_runtime": 1.8855, + "eval_samples_per_second": 22.276, + "eval_steps_per_second": 1.591, "step": 162 }, { "epoch": 118.55, "eval_accuracy": 0.047619047619047616, - "eval_f1": 0.430379746835443, - "eval_loss": 0.5451788306236267, - "eval_roc_auc": 0.7507763975155279, - "eval_runtime": 2.2158, - "eval_samples_per_second": 18.955, - "eval_steps_per_second": 1.354, + "eval_f1": 0.3591836734693878, + "eval_loss": 0.5753706693649292, + "eval_roc_auc": 0.7013457556935817, + "eval_runtime": 2.0351, + "eval_samples_per_second": 20.637, + "eval_steps_per_second": 1.474, "step": 163 }, { "epoch": 120.0, - "eval_accuracy": 0.07142857142857142, - "eval_f1": 0.43965517241379304, - "eval_loss": 0.5420504212379456, - "eval_roc_auc": 0.7533643892339544, - "eval_runtime": 2.2713, - "eval_samples_per_second": 18.492, - "eval_steps_per_second": 1.321, + "eval_accuracy": 0.047619047619047616, + "eval_f1": 0.3665338645418327, + "eval_loss": 0.5756567120552063, + "eval_roc_auc": 0.7111801242236024, + "eval_runtime": 1.777, + "eval_samples_per_second": 23.636, + "eval_steps_per_second": 1.688, "step": 165 }, { "epoch": 120.73, "eval_accuracy": 0.07142857142857142, - "eval_f1": 0.43478260869565216, - "eval_loss": 0.5413345098495483, - "eval_roc_auc": 0.7479296066252589, - "eval_runtime": 2.1593, - "eval_samples_per_second": 19.451, - "eval_steps_per_second": 1.389, + "eval_f1": 0.3728813559322034, + "eval_loss": 0.5771237015724182, + "eval_roc_auc": 0.7060041407867494, + "eval_runtime": 2.0719, + "eval_samples_per_second": 20.271, + "eval_steps_per_second": 1.448, "step": 166 }, { "epoch": 121.45, "eval_accuracy": 0.07142857142857142, - "eval_f1": 0.42735042735042733, - "eval_loss": 0.5442317724227905, - "eval_roc_auc": 0.7458592132505176, - "eval_runtime": 2.22, - "eval_samples_per_second": 18.919, - "eval_steps_per_second": 1.351, + "eval_f1": 0.3628691983122363, + "eval_loss": 0.5745561122894287, + "eval_roc_auc": 0.6990165631469979, + "eval_runtime": 1.8367, + "eval_samples_per_second": 22.867, + "eval_steps_per_second": 1.633, "step": 167 }, { "epoch": 122.91, - "eval_accuracy": 0.047619047619047616, - "eval_f1": 0.430379746835443, - "eval_loss": 0.5420621037483215, - "eval_roc_auc": 0.7507763975155279, - "eval_runtime": 2.2358, - "eval_samples_per_second": 18.785, - "eval_steps_per_second": 1.342, + "eval_accuracy": 0.07142857142857142, + "eval_f1": 0.36440677966101687, + "eval_loss": 0.5757673382759094, + "eval_roc_auc": 0.6995341614906831, + "eval_runtime": 2.032, + "eval_samples_per_second": 20.669, + "eval_steps_per_second": 1.476, "step": 169 }, { "epoch": 123.64, "eval_accuracy": 0.07142857142857142, - "eval_f1": 0.42372881355932207, - "eval_loss": 0.5403344035148621, - "eval_roc_auc": 0.7448240165631471, - "eval_runtime": 2.1592, - "eval_samples_per_second": 19.451, - "eval_steps_per_second": 1.389, + "eval_f1": 0.3559322033898305, + "eval_loss": 0.5745112299919128, + "eval_roc_auc": 0.693064182194617, + "eval_runtime": 1.8329, + "eval_samples_per_second": 22.914, + "eval_steps_per_second": 1.637, "step": 170 }, { "epoch": 124.36, - "eval_accuracy": 0.09523809523809523, - "eval_f1": 0.43478260869565216, - "eval_loss": 0.5386914014816284, - "eval_roc_auc": 0.7479296066252589, - "eval_runtime": 2.1762, - "eval_samples_per_second": 19.3, - "eval_steps_per_second": 1.379, + "eval_accuracy": 0.07142857142857142, + "eval_f1": 0.35443037974683544, + "eval_loss": 0.5758460164070129, + "eval_roc_auc": 0.6925465838509317, + "eval_runtime": 2.0956, + "eval_samples_per_second": 20.042, + "eval_steps_per_second": 1.432, "step": 171 }, { "epoch": 125.82, - "eval_accuracy": 0.047619047619047616, - "eval_f1": 0.42241379310344834, - "eval_loss": 0.5383176207542419, - "eval_roc_auc": 0.740424430641822, - "eval_runtime": 2.3145, - "eval_samples_per_second": 18.147, - "eval_steps_per_second": 1.296, + "eval_accuracy": 0.07142857142857142, + "eval_f1": 0.3598326359832636, + "eval_loss": 0.5759280323982239, + "eval_roc_auc": 0.6979813664596273, + "eval_runtime": 1.9931, + "eval_samples_per_second": 21.072, + "eval_steps_per_second": 1.505, "step": 173 }, { "epoch": 126.55, "eval_accuracy": 0.07142857142857142, - "eval_f1": 0.42290748898678415, - "eval_loss": 0.5346717238426208, - "eval_roc_auc": 0.7365424430641822, - "eval_runtime": 2.2566, - "eval_samples_per_second": 18.612, - "eval_steps_per_second": 1.329, + "eval_f1": 0.35684647302904565, + "eval_loss": 0.5771880149841309, + "eval_roc_auc": 0.6969461697722568, + "eval_runtime": 1.9762, + "eval_samples_per_second": 21.253, + "eval_steps_per_second": 1.518, "step": 174 }, { "epoch": 128.0, "eval_accuracy": 0.07142857142857142, - "eval_f1": 0.43478260869565216, - "eval_loss": 0.5410609245300293, - "eval_roc_auc": 0.7479296066252589, - "eval_runtime": 2.2479, - "eval_samples_per_second": 18.684, - "eval_steps_per_second": 1.335, + "eval_f1": 0.35833333333333334, + "eval_loss": 0.5747218132019043, + "eval_roc_auc": 0.697463768115942, + "eval_runtime": 2.0367, + "eval_samples_per_second": 20.622, + "eval_steps_per_second": 1.473, "step": 176 }, { "epoch": 128.73, "eval_accuracy": 0.07142857142857142, - "eval_f1": 0.40869565217391307, - "eval_loss": 0.5448868870735168, - "eval_roc_auc": 0.72851966873706, - "eval_runtime": 2.2593, - "eval_samples_per_second": 18.59, - "eval_steps_per_second": 1.328, + "eval_f1": 0.36440677966101687, + "eval_loss": 0.5737968683242798, + "eval_roc_auc": 0.6995341614906831, + "eval_runtime": 1.802, + "eval_samples_per_second": 23.307, + "eval_steps_per_second": 1.665, "step": 177 }, { "epoch": 129.45, "eval_accuracy": 0.07142857142857142, - "eval_f1": 0.4188034188034188, - "eval_loss": 0.5426321029663086, - "eval_roc_auc": 0.7393892339544514, - "eval_runtime": 2.1827, - "eval_samples_per_second": 19.242, - "eval_steps_per_second": 1.374, + "eval_f1": 0.36440677966101687, + "eval_loss": 0.575109601020813, + "eval_roc_auc": 0.6995341614906831, + "eval_runtime": 2.0006, + "eval_samples_per_second": 20.993, + "eval_steps_per_second": 1.5, "step": 178 }, { "epoch": 130.91, - "eval_accuracy": 0.07142857142857142, - "eval_f1": 0.4034334763948498, - "eval_loss": 0.5429388880729675, - "eval_roc_auc": 0.7269668737060041, - "eval_runtime": 2.1992, - "eval_samples_per_second": 19.098, - "eval_steps_per_second": 1.364, + "eval_accuracy": 0.09523809523809523, + "eval_f1": 0.37130801687763715, + "eval_loss": 0.5740981101989746, + "eval_roc_auc": 0.7054865424430642, + "eval_runtime": 1.7241, + "eval_samples_per_second": 24.361, + "eval_steps_per_second": 1.74, "step": 180 }, { "epoch": 131.64, "eval_accuracy": 0.09523809523809523, - "eval_f1": 0.4255319148936171, - "eval_loss": 0.5380597710609436, - "eval_roc_auc": 0.7453416149068324, - "eval_runtime": 2.2546, - "eval_samples_per_second": 18.628, - "eval_steps_per_second": 1.331, + "eval_f1": 0.37130801687763715, + "eval_loss": 0.5748480558395386, + "eval_roc_auc": 0.7054865424430642, + "eval_runtime": 2.2478, + "eval_samples_per_second": 18.685, + "eval_steps_per_second": 1.335, "step": 181 }, { "epoch": 132.36, "eval_accuracy": 0.07142857142857142, - "eval_f1": 0.4279475982532751, - "eval_loss": 0.5344164967536926, - "eval_roc_auc": 0.7419772256728779, - "eval_runtime": 2.2146, - "eval_samples_per_second": 18.965, - "eval_steps_per_second": 1.355, + "eval_f1": 0.3659574468085106, + "eval_loss": 0.5767146944999695, + "eval_roc_auc": 0.7000517598343685, + "eval_runtime": 1.9197, + "eval_samples_per_second": 21.879, + "eval_steps_per_second": 1.563, "step": 182 }, { "epoch": 133.82, - "eval_accuracy": 0.047619047619047616, - "eval_f1": 0.39316239316239315, - "eval_loss": 0.5364577174186707, - "eval_roc_auc": 0.7199792960662525, - "eval_runtime": 2.1956, - "eval_samples_per_second": 19.129, - "eval_steps_per_second": 1.366, + "eval_accuracy": 0.09523809523809523, + "eval_f1": 0.3659574468085106, + "eval_loss": 0.5731571316719055, + "eval_roc_auc": 0.7000517598343685, + "eval_runtime": 1.8071, + "eval_samples_per_second": 23.241, + "eval_steps_per_second": 1.66, "step": 184 }, { "epoch": 134.55, - "eval_accuracy": 0.07142857142857142, - "eval_f1": 0.40869565217391307, - "eval_loss": 0.5361512303352356, - "eval_roc_auc": 0.72851966873706, - "eval_runtime": 2.2016, - "eval_samples_per_second": 19.077, + "eval_accuracy": 0.09523809523809523, + "eval_f1": 0.3771929824561403, + "eval_loss": 0.5742261409759521, + "eval_roc_auc": 0.7036749482401656, + "eval_runtime": 2.2015, + "eval_samples_per_second": 19.078, "eval_steps_per_second": 1.363, "step": 185 }, { "epoch": 136.0, - "eval_accuracy": 0.023809523809523808, - "eval_f1": 0.39999999999999997, - "eval_loss": 0.5336142182350159, - "eval_roc_auc": 0.7142857142857142, - "eval_runtime": 2.176, - "eval_samples_per_second": 19.302, - "eval_steps_per_second": 1.379, + "eval_accuracy": 0.09523809523809523, + "eval_f1": 0.37554585152838427, + "eval_loss": 0.5690290927886963, + "eval_roc_auc": 0.7031573498964804, + "eval_runtime": 2.2207, + "eval_samples_per_second": 18.913, + "eval_steps_per_second": 1.351, "step": 187 }, { "epoch": 136.73, - "eval_accuracy": 0.047619047619047616, - "eval_f1": 0.42533936651583715, - "eval_loss": 0.5356618762016296, - "eval_roc_auc": 0.7331780538302277, - "eval_runtime": 2.3066, - "eval_samples_per_second": 18.209, - "eval_steps_per_second": 1.301, + "eval_accuracy": 0.07142857142857142, + "eval_f1": 0.3805309734513274, + "eval_loss": 0.5698773264884949, + "eval_roc_auc": 0.7047101449275363, + "eval_runtime": 2.491, + "eval_samples_per_second": 16.861, + "eval_steps_per_second": 1.204, "step": 188 }, { "epoch": 137.45, - "eval_accuracy": 0.047619047619047616, - "eval_f1": 0.4266666666666666, - "eval_loss": 0.5409244894981384, - "eval_roc_auc": 0.7375776397515528, - "eval_runtime": 2.269, - "eval_samples_per_second": 18.51, - "eval_steps_per_second": 1.322, + "eval_accuracy": 0.07142857142857142, + "eval_f1": 0.37068965517241376, + "eval_loss": 0.5743153095245361, + "eval_roc_auc": 0.7016045548654244, + "eval_runtime": 2.2065, + "eval_samples_per_second": 19.034, + "eval_steps_per_second": 1.36, "step": 189 }, { "epoch": 138.91, - "eval_accuracy": 0.023809523809523808, - "eval_f1": 0.41558441558441556, - "eval_loss": 0.5381691455841064, - "eval_roc_auc": 0.7344720496894409, - "eval_runtime": 2.2181, - "eval_samples_per_second": 18.935, - "eval_steps_per_second": 1.353, + "eval_accuracy": 0.09523809523809523, + "eval_f1": 0.3529411764705882, + "eval_loss": 0.5739556550979614, + "eval_roc_auc": 0.6920289855072465, + "eval_runtime": 2.1255, + "eval_samples_per_second": 19.76, + "eval_steps_per_second": 1.411, "step": 191 }, { "epoch": 139.64, - "eval_accuracy": 0.047619047619047616, - "eval_f1": 0.41558441558441556, - "eval_loss": 0.5408413410186768, - "eval_roc_auc": 0.7344720496894409, - "eval_runtime": 2.1767, - "eval_samples_per_second": 19.296, - "eval_steps_per_second": 1.378, + "eval_accuracy": 0.07142857142857142, + "eval_f1": 0.3659574468085106, + "eval_loss": 0.5739744901657104, + "eval_roc_auc": 0.7000517598343685, + "eval_runtime": 2.0832, + "eval_samples_per_second": 20.161, + "eval_steps_per_second": 1.44, "step": 192 }, { "epoch": 140.36, - "eval_accuracy": 0.047619047619047616, - "eval_f1": 0.39473684210526316, - "eval_loss": 0.54255610704422, - "eval_roc_auc": 0.716614906832298, - "eval_runtime": 2.3421, - "eval_samples_per_second": 17.933, - "eval_steps_per_second": 1.281, + "eval_accuracy": 0.07142857142857142, + "eval_f1": 0.36440677966101687, + "eval_loss": 0.5733990669250488, + "eval_roc_auc": 0.6995341614906831, + "eval_runtime": 2.021, + "eval_samples_per_second": 20.782, + "eval_steps_per_second": 1.484, "step": 193 }, { "epoch": 141.82, - "eval_accuracy": 0.023809523809523808, - "eval_f1": 0.38260869565217387, - "eval_loss": 0.5436768531799316, - "eval_roc_auc": 0.7091097308488613, - "eval_runtime": 2.2168, - "eval_samples_per_second": 18.946, - "eval_steps_per_second": 1.353, + "eval_accuracy": 0.07142857142857142, + "eval_f1": 0.36752136752136755, + "eval_loss": 0.5739865303039551, + "eval_roc_auc": 0.7005693581780539, + "eval_runtime": 2.0466, + "eval_samples_per_second": 20.522, + "eval_steps_per_second": 1.466, "step": 195 }, { "epoch": 142.55, "eval_accuracy": 0.07142857142857142, - "eval_f1": 0.3876651982378855, - "eval_loss": 0.540955662727356, - "eval_roc_auc": 0.7106625258799172, - "eval_runtime": 2.1746, - "eval_samples_per_second": 19.314, - "eval_steps_per_second": 1.38, + "eval_f1": 0.37068965517241376, + "eval_loss": 0.5721157789230347, + "eval_roc_auc": 0.7016045548654244, + "eval_runtime": 1.8265, + "eval_samples_per_second": 22.994, + "eval_steps_per_second": 1.642, "step": 196 }, { "epoch": 144.0, "eval_accuracy": 0.07142857142857142, - "eval_f1": 0.3982683982683983, - "eval_loss": 0.5399672389030457, - "eval_roc_auc": 0.7215320910973085, - "eval_runtime": 2.2745, - "eval_samples_per_second": 18.466, - "eval_steps_per_second": 1.319, + "eval_f1": 0.3766816143497758, + "eval_loss": 0.5724892020225525, + "eval_roc_auc": 0.6997929606625258, + "eval_runtime": 2.0667, + "eval_samples_per_second": 20.323, + "eval_steps_per_second": 1.452, "step": 198 }, { "epoch": 144.73, - "eval_accuracy": 0.047619047619047616, - "eval_f1": 0.3859649122807018, - "eval_loss": 0.5390969514846802, - "eval_roc_auc": 0.7101449275362318, - "eval_runtime": 2.3353, - "eval_samples_per_second": 17.984, - "eval_steps_per_second": 1.285, + "eval_accuracy": 0.09523809523809523, + "eval_f1": 0.3728813559322034, + "eval_loss": 0.5734347105026245, + "eval_roc_auc": 0.7060041407867494, + "eval_runtime": 1.798, + "eval_samples_per_second": 23.359, + "eval_steps_per_second": 1.669, "step": 199 }, { "epoch": 145.45, "eval_accuracy": 0.07142857142857142, - "eval_f1": 0.42290748898678415, - "eval_loss": 0.5356051325798035, - "eval_roc_auc": 0.7365424430641822, - "eval_runtime": 2.238, - "eval_samples_per_second": 18.767, - "eval_steps_per_second": 1.34, - "step": 200 - }, - { - "epoch": 146.91, - "eval_accuracy": 0.07142857142857142, - "eval_f1": 0.41441441441441446, - "eval_loss": 0.5343185663223267, - "eval_roc_auc": 0.7261904761904762, - "eval_runtime": 2.2389, - "eval_samples_per_second": 18.759, - "eval_steps_per_second": 1.34, - "step": 202 - }, - { - "epoch": 147.64, - "eval_accuracy": 0.09523809523809523, - "eval_f1": 0.42857142857142855, - "eval_loss": 0.5286682844161987, - "eval_roc_auc": 0.7380952380952381, - "eval_runtime": 2.2757, - "eval_samples_per_second": 18.456, - "eval_steps_per_second": 1.318, - "step": 203 - }, - { - "epoch": 148.36, - "eval_accuracy": 0.07142857142857142, - "eval_f1": 0.41071428571428575, - "eval_loss": 0.5306946635246277, - "eval_roc_auc": 0.7251552795031057, - "eval_runtime": 2.1874, - "eval_samples_per_second": 19.201, - "eval_steps_per_second": 1.372, - "step": 204 - }, - { - "epoch": 149.82, - "eval_accuracy": 0.09523809523809523, - "eval_f1": 0.4177777777777778, - "eval_loss": 0.5335448384284973, - "eval_roc_auc": 0.7311076604554866, - "eval_runtime": 2.3633, - "eval_samples_per_second": 17.772, - "eval_steps_per_second": 1.269, - "step": 206 - }, - { - "epoch": 150.55, - "eval_accuracy": 0.07142857142857142, - "eval_f1": 0.4000000000000001, - "eval_loss": 0.535821795463562, - "eval_roc_auc": 0.7220496894409938, - "eval_runtime": 2.1476, - "eval_samples_per_second": 19.557, - "eval_steps_per_second": 1.397, - "step": 207 - }, - { - "epoch": 152.0, - "eval_accuracy": 0.07142857142857142, - "eval_f1": 0.4052863436123348, - "eval_loss": 0.5356825590133667, - "eval_roc_auc": 0.7236024844720497, - "eval_runtime": 2.2782, - "eval_samples_per_second": 18.436, - "eval_steps_per_second": 1.317, - "step": 209 - }, - { - "epoch": 152.73, - "eval_accuracy": 0.09523809523809523, - "eval_f1": 0.4069264069264069, - "eval_loss": 0.5324029922485352, - "eval_roc_auc": 0.7280020703933747, - "eval_runtime": 2.2465, - "eval_samples_per_second": 18.696, - "eval_steps_per_second": 1.335, - "step": 210 - }, - { - "epoch": 153.45, - "eval_accuracy": 0.07142857142857142, - "eval_f1": 0.39316239316239315, - "eval_loss": 0.5348923802375793, - "eval_roc_auc": 0.7199792960662525, - "eval_runtime": 2.1592, - "eval_samples_per_second": 19.452, - "eval_steps_per_second": 1.389, - "step": 211 - }, - { - "epoch": 154.91, - "eval_accuracy": 0.07142857142857142, - "eval_f1": 0.41379310344827586, - "eval_loss": 0.5328140258789062, - "eval_roc_auc": 0.7339544513457557, - "eval_runtime": 2.1367, - "eval_samples_per_second": 19.657, - "eval_steps_per_second": 1.404, - "step": 213 - }, - { - "epoch": 155.64, - "eval_accuracy": 0.07142857142857142, - "eval_f1": 0.412280701754386, - "eval_loss": 0.533903956413269, - "eval_roc_auc": 0.7295548654244306, - "eval_runtime": 2.1392, - "eval_samples_per_second": 19.634, - "eval_steps_per_second": 1.402, - "step": 214 - }, - { - "epoch": 156.36, - "eval_accuracy": 0.09523809523809523, - "eval_f1": 0.43049327354260086, - "eval_loss": 0.5287569761276245, - "eval_roc_auc": 0.7386128364389234, - "eval_runtime": 2.3564, - "eval_samples_per_second": 17.824, - "eval_steps_per_second": 1.273, - "step": 215 - }, - { - "epoch": 157.82, - "eval_accuracy": 0.07142857142857142, - "eval_f1": 0.4272727272727273, - "eval_loss": 0.5261039137840271, - "eval_roc_auc": 0.733695652173913, - "eval_runtime": 2.1573, - "eval_samples_per_second": 19.469, - "eval_steps_per_second": 1.391, - "step": 217 - }, - { - "epoch": 158.55, - "eval_accuracy": 0.07142857142857142, - "eval_f1": 0.42290748898678415, - "eval_loss": 0.5312274694442749, - "eval_roc_auc": 0.7365424430641822, - "eval_runtime": 2.2337, - "eval_samples_per_second": 18.803, - "eval_steps_per_second": 1.343, - "step": 218 - }, - { - "epoch": 160.0, - "eval_accuracy": 0.09523809523809523, - "eval_f1": 0.40358744394618834, - "eval_loss": 0.5341694951057434, - "eval_roc_auc": 0.7192028985507245, - "eval_runtime": 2.2456, - "eval_samples_per_second": 18.703, + "eval_f1": 0.3628691983122363, + "eval_loss": 0.575474202632904, + "eval_roc_auc": 0.6990165631469979, + "eval_runtime": 2.2459, + "eval_samples_per_second": 18.701, "eval_steps_per_second": 1.336, - "step": 220 - }, - { - "epoch": 160.73, - "eval_accuracy": 0.09523809523809523, - "eval_f1": 0.3981900452488688, - "eval_loss": 0.5362971425056458, - "eval_roc_auc": 0.7137681159420289, - "eval_runtime": 2.212, - "eval_samples_per_second": 18.988, - "eval_steps_per_second": 1.356, - "step": 221 - }, - { - "epoch": 161.45, - "eval_accuracy": 0.09523809523809523, - "eval_f1": 0.41409691629955947, - "eval_loss": 0.532296895980835, - "eval_roc_auc": 0.730072463768116, - "eval_runtime": 2.3611, - "eval_samples_per_second": 17.788, - "eval_steps_per_second": 1.271, - "step": 222 - }, - { - "epoch": 162.91, - "eval_accuracy": 0.09523809523809523, - "eval_f1": 0.41071428571428575, - "eval_loss": 0.5253220796585083, - "eval_roc_auc": 0.7251552795031057, - "eval_runtime": 2.1376, - "eval_samples_per_second": 19.648, - "eval_steps_per_second": 1.403, - "step": 224 - }, - { - "epoch": 163.64, - "eval_accuracy": 0.09523809523809523, - "eval_f1": 0.4234234234234235, - "eval_loss": 0.5222101807594299, - "eval_roc_auc": 0.7326604554865425, - "eval_runtime": 2.2698, - "eval_samples_per_second": 18.504, - "eval_steps_per_second": 1.322, - "step": 225 - }, - { - "epoch": 164.36, - "eval_accuracy": 0.07142857142857142, - "eval_f1": 0.412280701754386, - "eval_loss": 0.5278355479240417, - "eval_roc_auc": 0.7295548654244306, - "eval_runtime": 2.2788, - "eval_samples_per_second": 18.431, - "eval_steps_per_second": 1.316, - "step": 226 - }, - { - "epoch": 165.82, - "eval_accuracy": 0.11904761904761904, - "eval_f1": 0.415929203539823, - "eval_loss": 0.5236978530883789, - "eval_roc_auc": 0.7305900621118012, - "eval_runtime": 2.3596, - "eval_samples_per_second": 17.8, - "eval_steps_per_second": 1.271, - "step": 228 - }, - { - "epoch": 166.55, - "eval_accuracy": 0.11904761904761904, - "eval_f1": 0.42290748898678415, - "eval_loss": 0.5267902612686157, - "eval_roc_auc": 0.7365424430641822, - "eval_runtime": 2.2183, - "eval_samples_per_second": 18.933, - "eval_steps_per_second": 1.352, - "step": 229 - }, - { - "epoch": 168.0, - "eval_accuracy": 0.09523809523809523, - "eval_f1": 0.39639639639639634, - "eval_loss": 0.5275822877883911, - "eval_roc_auc": 0.7132505175983436, - "eval_runtime": 2.1555, - "eval_samples_per_second": 19.485, - "eval_steps_per_second": 1.392, - "step": 231 - }, - { - "epoch": 168.73, - "eval_accuracy": 0.11904761904761904, - "eval_f1": 0.41409691629955947, - "eval_loss": 0.522292971611023, - "eval_roc_auc": 0.730072463768116, - "eval_runtime": 2.5244, - "eval_samples_per_second": 16.638, - "eval_steps_per_second": 1.188, - "step": 232 - }, - { - "epoch": 169.45, - "eval_accuracy": 0.11904761904761904, - "eval_f1": 0.4177777777777778, - "eval_loss": 0.5228198170661926, - "eval_roc_auc": 0.7311076604554866, - "eval_runtime": 2.2486, - "eval_samples_per_second": 18.678, - "eval_steps_per_second": 1.334, - "step": 233 - }, - { - "epoch": 170.91, - "eval_accuracy": 0.11904761904761904, - "eval_f1": 0.4234234234234235, - "eval_loss": 0.5201071500778198, - "eval_roc_auc": 0.7326604554865425, - "eval_runtime": 2.2664, - "eval_samples_per_second": 18.531, - "eval_steps_per_second": 1.324, - "step": 235 - }, - { - "epoch": 171.64, - "eval_accuracy": 0.11904761904761904, - "eval_f1": 0.42152466367713004, - "eval_loss": 0.5248059034347534, - "eval_roc_auc": 0.7321428571428572, - "eval_runtime": 2.1597, - "eval_samples_per_second": 19.447, - "eval_steps_per_second": 1.389, - "step": 236 - }, - { - "epoch": 172.36, - "eval_accuracy": 0.11904761904761904, - "eval_f1": 0.4125560538116592, - "eval_loss": 0.5245119333267212, - "eval_roc_auc": 0.7256728778467909, - "eval_runtime": 2.2008, - "eval_samples_per_second": 19.084, - "eval_steps_per_second": 1.363, - "step": 237 - }, - { - "epoch": 173.82, - "eval_accuracy": 0.11904761904761904, - "eval_f1": 0.4177777777777778, - "eval_loss": 0.5223248600959778, - "eval_roc_auc": 0.7311076604554866, - "eval_runtime": 2.2306, - "eval_samples_per_second": 18.829, - "eval_steps_per_second": 1.345, - "step": 239 - }, - { - "epoch": 174.55, - "eval_accuracy": 0.11904761904761904, - "eval_f1": 0.409090909090909, - "eval_loss": 0.5221878886222839, - "eval_roc_auc": 0.7207556935817805, - "eval_runtime": 2.2812, - "eval_samples_per_second": 18.411, - "eval_steps_per_second": 1.315, - "step": 240 - }, - { - "epoch": 176.0, - "eval_accuracy": 0.11904761904761904, - "eval_f1": 0.41628959276018096, - "eval_loss": 0.5217894315719604, - "eval_roc_auc": 0.7267080745341615, - "eval_runtime": 2.3942, - "eval_samples_per_second": 17.542, - "eval_steps_per_second": 1.253, - "step": 242 - }, - { - "epoch": 176.73, - "eval_accuracy": 0.09523809523809523, - "eval_f1": 0.40540540540540543, - "eval_loss": 0.5226414203643799, - "eval_roc_auc": 0.7197204968944099, - "eval_runtime": 2.1776, - "eval_samples_per_second": 19.287, - "eval_steps_per_second": 1.378, - "step": 243 - }, - { - "epoch": 177.45, - "eval_accuracy": 0.09523809523809523, - "eval_f1": 0.39823008849557523, - "eval_loss": 0.5248770713806152, - "eval_roc_auc": 0.7176501035196686, - "eval_runtime": 2.26, - "eval_samples_per_second": 18.584, - "eval_steps_per_second": 1.327, - "step": 244 - }, - { - "epoch": 178.91, - "eval_accuracy": 0.09523809523809523, - "eval_f1": 0.4017857142857143, - "eval_loss": 0.5241516828536987, - "eval_roc_auc": 0.7186853002070392, - "eval_runtime": 2.1743, - "eval_samples_per_second": 19.317, - "eval_steps_per_second": 1.38, - "step": 246 - }, - { - "epoch": 179.64, - "eval_accuracy": 0.09523809523809523, - "eval_f1": 0.38914027149321273, - "eval_loss": 0.5245065093040466, - "eval_roc_auc": 0.7072981366459626, - "eval_runtime": 2.2384, - "eval_samples_per_second": 18.763, - "eval_steps_per_second": 1.34, - "step": 247 - }, - { - "epoch": 180.36, - "eval_accuracy": 0.11904761904761904, - "eval_f1": 0.39449541284403666, - "eval_loss": 0.5249381065368652, - "eval_roc_auc": 0.7088509316770186, - "eval_runtime": 2.2201, - "eval_samples_per_second": 18.918, - "eval_steps_per_second": 1.351, - "step": 248 - }, - { - "epoch": 181.82, - "eval_accuracy": 0.09523809523809523, - "eval_f1": 0.4234234234234235, - "eval_loss": 0.52437824010849, - "eval_roc_auc": 0.7326604554865425, - "eval_runtime": 2.2705, - "eval_samples_per_second": 18.498, - "eval_steps_per_second": 1.321, - "step": 250 - }, - { - "epoch": 182.55, - "eval_accuracy": 0.07142857142857142, - "eval_f1": 0.409090909090909, - "eval_loss": 0.5216240286827087, - "eval_roc_auc": 0.7207556935817805, - "eval_runtime": 2.238, - "eval_samples_per_second": 18.767, - "eval_steps_per_second": 1.34, - "step": 251 - }, - { - "epoch": 184.0, - "eval_accuracy": 0.09523809523809523, - "eval_f1": 0.40358744394618834, - "eval_loss": 0.5211586952209473, - "eval_roc_auc": 0.7192028985507245, - "eval_runtime": 2.244, - "eval_samples_per_second": 18.716, - "eval_steps_per_second": 1.337, - "step": 253 - }, - { - "epoch": 184.73, - "eval_accuracy": 0.09523809523809523, - "eval_f1": 0.41628959276018096, - "eval_loss": 0.5235975384712219, - "eval_roc_auc": 0.7267080745341615, - "eval_runtime": 2.1767, - "eval_samples_per_second": 19.295, - "eval_steps_per_second": 1.378, - "step": 254 - }, - { - "epoch": 185.45, - "eval_accuracy": 0.11904761904761904, - "eval_f1": 0.41441441441441446, - "eval_loss": 0.5226503610610962, - "eval_roc_auc": 0.7261904761904762, - "eval_runtime": 2.3845, - "eval_samples_per_second": 17.614, - "eval_steps_per_second": 1.258, - "step": 255 - }, - { - "epoch": 186.91, - "eval_accuracy": 0.11904761904761904, - "eval_f1": 0.40540540540540543, - "eval_loss": 0.5200766324996948, - "eval_roc_auc": 0.7197204968944099, - "eval_runtime": 2.2501, - "eval_samples_per_second": 18.666, - "eval_steps_per_second": 1.333, - "step": 257 - }, - { - "epoch": 187.64, - "eval_accuracy": 0.14285714285714285, - "eval_f1": 0.4181818181818182, - "eval_loss": 0.5195832848548889, - "eval_roc_auc": 0.7272256728778468, - "eval_runtime": 2.2154, - "eval_samples_per_second": 18.958, - "eval_steps_per_second": 1.354, - "step": 258 - }, - { - "epoch": 188.36, - "eval_accuracy": 0.14285714285714285, - "eval_f1": 0.4177777777777778, - "eval_loss": 0.5187467336654663, - "eval_roc_auc": 0.7311076604554866, - "eval_runtime": 2.2882, - "eval_samples_per_second": 18.355, - "eval_steps_per_second": 1.311, - "step": 259 - }, - { - "epoch": 189.82, - "eval_accuracy": 0.11904761904761904, - "eval_f1": 0.4125560538116592, - "eval_loss": 0.5165961384773254, - "eval_roc_auc": 0.7256728778467909, - "eval_runtime": 2.2388, - "eval_samples_per_second": 18.76, - "eval_steps_per_second": 1.34, - "step": 261 - }, - { - "epoch": 190.55, - "eval_accuracy": 0.11904761904761904, - "eval_f1": 0.3873873873873873, - "eval_loss": 0.5241953730583191, - "eval_roc_auc": 0.7067805383022774, - "eval_runtime": 2.4908, - "eval_samples_per_second": 16.862, - "eval_steps_per_second": 1.204, - "step": 262 - }, - { - "epoch": 192.0, - "eval_accuracy": 0.11904761904761904, - "eval_f1": 0.42533936651583715, - "eval_loss": 0.5164801478385925, - "eval_roc_auc": 0.7331780538302277, - "eval_runtime": 2.2626, - "eval_samples_per_second": 18.563, - "eval_steps_per_second": 1.326, - "step": 264 - }, - { - "epoch": 192.73, - "eval_accuracy": 0.11904761904761904, - "eval_f1": 0.43518518518518523, - "eval_loss": 0.5178862810134888, - "eval_roc_auc": 0.7357660455486543, - "eval_runtime": 2.5128, - "eval_samples_per_second": 16.714, - "eval_steps_per_second": 1.194, - "step": 265 - }, - { - "epoch": 193.45, - "eval_accuracy": 0.11904761904761904, - "eval_f1": 0.42452830188679247, - "eval_loss": 0.5133179426193237, - "eval_roc_auc": 0.7248964803312629, - "eval_runtime": 2.5865, - "eval_samples_per_second": 16.238, - "eval_steps_per_second": 1.16, - "step": 266 - }, - { - "epoch": 194.91, - "eval_accuracy": 0.11904761904761904, - "eval_f1": 0.42790697674418604, - "eval_loss": 0.5165557861328125, - "eval_roc_auc": 0.7298136645962733, - "eval_runtime": 2.2181, - "eval_samples_per_second": 18.935, - "eval_steps_per_second": 1.353, - "step": 268 - }, - { - "epoch": 195.64, - "eval_accuracy": 0.09523809523809523, - "eval_f1": 0.4125560538116592, - "eval_loss": 0.5179272294044495, - "eval_roc_auc": 0.7256728778467909, - "eval_runtime": 2.1768, - "eval_samples_per_second": 19.294, - "eval_steps_per_second": 1.378, - "step": 269 - }, - { - "epoch": 196.36, - "eval_accuracy": 0.11904761904761904, - "eval_f1": 0.4311926605504587, - "eval_loss": 0.5189908146858215, - "eval_roc_auc": 0.7347308488612837, - "eval_runtime": 2.1519, - "eval_samples_per_second": 19.517, - "eval_steps_per_second": 1.394, - "step": 270 - }, - { - "epoch": 197.82, - "eval_accuracy": 0.11904761904761904, - "eval_f1": 0.40552995391705066, - "eval_loss": 0.5177509784698486, - "eval_roc_auc": 0.7158385093167701, - "eval_runtime": 2.24, - "eval_samples_per_second": 18.75, - "eval_steps_per_second": 1.339, - "step": 272 - }, - { - "epoch": 198.55, - "eval_accuracy": 0.11904761904761904, - "eval_f1": 0.40740740740740733, - "eval_loss": 0.5172959566116333, - "eval_roc_auc": 0.7163561076604554, - "eval_runtime": 2.2769, - "eval_samples_per_second": 18.447, - "eval_steps_per_second": 1.318, - "step": 273 - }, - { - "epoch": 200.0, - "eval_accuracy": 0.11904761904761904, - "eval_f1": 0.39639639639639634, - "eval_loss": 0.5173150897026062, - "eval_roc_auc": 0.7132505175983436, - "eval_runtime": 2.4473, - "eval_samples_per_second": 17.162, - "eval_steps_per_second": 1.226, - "step": 275 - }, - { - "epoch": 200.73, - "eval_accuracy": 0.11904761904761904, - "eval_f1": 0.41666666666666663, - "eval_loss": 0.5198870897293091, - "eval_roc_auc": 0.7228260869565216, - "eval_runtime": 2.2209, - "eval_samples_per_second": 18.911, - "eval_steps_per_second": 1.351, - "step": 276 - }, - { - "epoch": 201.45, - "eval_accuracy": 0.11904761904761904, - "eval_f1": 0.3981900452488688, - "eval_loss": 0.5189902186393738, - "eval_roc_auc": 0.7137681159420289, - "eval_runtime": 2.2872, - "eval_samples_per_second": 18.363, - "eval_steps_per_second": 1.312, - "step": 277 - }, - { - "epoch": 202.91, - "eval_accuracy": 0.14285714285714285, - "eval_f1": 0.42452830188679247, - "eval_loss": 0.5163199305534363, - "eval_roc_auc": 0.7248964803312629, - "eval_runtime": 2.2888, - "eval_samples_per_second": 18.35, - "eval_steps_per_second": 1.311, - "step": 279 - }, - { - "epoch": 203.64, - "eval_accuracy": 0.11904761904761904, - "eval_f1": 0.4170616113744076, - "eval_loss": 0.5172598958015442, - "eval_roc_auc": 0.7189440993788819, - "eval_runtime": 2.2409, - "eval_samples_per_second": 18.743, - "eval_steps_per_second": 1.339, - "step": 280 - }, - { - "epoch": 204.36, - "eval_accuracy": 0.14285714285714285, - "eval_f1": 0.41314553990610325, - "eval_loss": 0.5206424593925476, - "eval_roc_auc": 0.7179089026915114, - "eval_runtime": 2.2107, - "eval_samples_per_second": 18.999, - "eval_steps_per_second": 1.357, - "step": 281 - }, - { - "epoch": 205.82, - "eval_accuracy": 0.11904761904761904, - "eval_f1": 0.411214953271028, - "eval_loss": 0.5185695886611938, - "eval_roc_auc": 0.717391304347826, - "eval_runtime": 2.2741, - "eval_samples_per_second": 18.468, - "eval_steps_per_second": 1.319, - "step": 283 - }, - { - "epoch": 206.55, - "eval_accuracy": 0.11904761904761904, - "eval_f1": 0.43317972350230416, - "eval_loss": 0.5186063647270203, - "eval_roc_auc": 0.735248447204969, - "eval_runtime": 2.2252, - "eval_samples_per_second": 18.875, - "eval_steps_per_second": 1.348, - "step": 284 - }, - { - "epoch": 208.0, - "eval_accuracy": 0.09523809523809523, - "eval_f1": 0.39639639639639634, - "eval_loss": 0.5183053016662598, - "eval_roc_auc": 0.7132505175983436, - "eval_runtime": 2.2662, - "eval_samples_per_second": 18.533, - "eval_steps_per_second": 1.324, - "step": 286 - }, - { - "epoch": 208.73, - "eval_accuracy": 0.09523809523809523, - "eval_f1": 0.3946188340807175, - "eval_loss": 0.5180411338806152, - "eval_roc_auc": 0.7127329192546583, - "eval_runtime": 2.1744, - "eval_samples_per_second": 19.316, - "eval_steps_per_second": 1.38, - "step": 287 - }, - { - "epoch": 209.45, - "eval_accuracy": 0.09523809523809523, - "eval_f1": 0.40930232558139534, - "eval_loss": 0.5190138220787048, - "eval_roc_auc": 0.7168737060041408, - "eval_runtime": 2.2594, - "eval_samples_per_second": 18.589, - "eval_steps_per_second": 1.328, - "step": 288 - }, - { - "epoch": 210.91, - "eval_accuracy": 0.09523809523809523, - "eval_f1": 0.3761467889908257, - "eval_loss": 0.5196405053138733, - "eval_roc_auc": 0.6959109730848861, - "eval_runtime": 2.2692, - "eval_samples_per_second": 18.509, - "eval_steps_per_second": 1.322, - "step": 290 - }, - { - "epoch": 211.64, - "eval_accuracy": 0.09523809523809523, - "eval_f1": 0.3981900452488688, - "eval_loss": 0.5149042010307312, - "eval_roc_auc": 0.7137681159420289, - "eval_runtime": 2.2833, - "eval_samples_per_second": 18.395, - "eval_steps_per_second": 1.314, - "step": 291 - }, - { - "epoch": 212.36, - "eval_accuracy": 0.09523809523809523, - "eval_f1": 0.39285714285714285, - "eval_loss": 0.5198233723640442, - "eval_roc_auc": 0.7122153209109731, - "eval_runtime": 2.5159, - "eval_samples_per_second": 16.694, - "eval_steps_per_second": 1.192, - "step": 292 - }, - { - "epoch": 213.82, - "eval_accuracy": 0.09523809523809523, - "eval_f1": 0.3888888888888889, - "eval_loss": 0.5179316401481628, - "eval_roc_auc": 0.703416149068323, - "eval_runtime": 2.3615, - "eval_samples_per_second": 17.786, - "eval_steps_per_second": 1.27, - "step": 294 - }, - { - "epoch": 214.55, - "eval_accuracy": 0.09523809523809523, - "eval_f1": 0.3981900452488688, - "eval_loss": 0.5147101879119873, - "eval_roc_auc": 0.7137681159420289, - "eval_runtime": 2.2382, - "eval_samples_per_second": 18.765, - "eval_steps_per_second": 1.34, - "step": 295 - }, - { - "epoch": 216.0, - "eval_accuracy": 0.09523809523809523, - "eval_f1": 0.3870967741935484, - "eval_loss": 0.5126305818557739, - "eval_roc_auc": 0.7028985507246377, - "eval_runtime": 2.2182, - "eval_samples_per_second": 18.934, - "eval_steps_per_second": 1.352, - "step": 297 - }, - { - "epoch": 216.73, - "eval_accuracy": 0.09523809523809523, - "eval_f1": 0.40740740740740733, - "eval_loss": 0.5115792751312256, - "eval_roc_auc": 0.7163561076604554, - "eval_runtime": 2.1407, - "eval_samples_per_second": 19.619, - "eval_steps_per_second": 1.401, - "step": 298 - }, - { - "epoch": 217.45, - "eval_accuracy": 0.11904761904761904, - "eval_f1": 0.4170616113744076, - "eval_loss": 0.5107331275939941, - "eval_roc_auc": 0.7189440993788819, - "eval_runtime": 2.5127, - "eval_samples_per_second": 16.715, - "eval_steps_per_second": 1.194, - "step": 299 - }, - { - "epoch": 218.91, - "eval_accuracy": 0.14285714285714285, - "eval_f1": 0.41095890410958896, - "eval_loss": 0.511676549911499, - "eval_roc_auc": 0.7212732919254659, - "eval_runtime": 2.3542, - "eval_samples_per_second": 17.84, - "eval_steps_per_second": 1.274, - "step": 301 - }, - { - "epoch": 219.64, - "eval_accuracy": 0.14285714285714285, - "eval_f1": 0.41666666666666663, - "eval_loss": 0.5091940760612488, - "eval_roc_auc": 0.7228260869565216, - "eval_runtime": 2.1376, - "eval_samples_per_second": 19.648, - "eval_steps_per_second": 1.403, - "step": 302 - }, - { - "epoch": 220.36, - "eval_accuracy": 0.09523809523809523, - "eval_f1": 0.3873873873873873, - "eval_loss": 0.5167934894561768, - "eval_roc_auc": 0.7067805383022774, - "eval_runtime": 2.2065, - "eval_samples_per_second": 19.035, - "eval_steps_per_second": 1.36, - "step": 303 - }, - { - "epoch": 221.82, - "eval_accuracy": 0.09523809523809523, - "eval_f1": 0.39069767441860465, - "eval_loss": 0.5202212333679199, - "eval_roc_auc": 0.7039337474120083, - "eval_runtime": 2.483, - "eval_samples_per_second": 16.915, - "eval_steps_per_second": 1.208, - "step": 305 + "step": 200 }, { - "epoch": 222.55, - "eval_accuracy": 0.09523809523809523, - "eval_f1": 0.40740740740740733, - "eval_loss": 0.5138672590255737, - "eval_roc_auc": 0.7163561076604554, - "eval_runtime": 2.2543, - "eval_samples_per_second": 18.631, - "eval_steps_per_second": 1.331, - "step": 306 - }, - { - "epoch": 224.0, - "eval_accuracy": 0.11904761904761904, - "eval_f1": 0.40552995391705066, - "eval_loss": 0.5207112431526184, - "eval_roc_auc": 0.7158385093167701, - "eval_runtime": 2.3791, - "eval_samples_per_second": 17.654, - "eval_steps_per_second": 1.261, - "step": 308 - }, - { - "epoch": 224.73, - "eval_accuracy": 0.11904761904761904, - "eval_f1": 0.4292237442922375, - "eval_loss": 0.5156222581863403, - "eval_roc_auc": 0.7342132505175983, - "eval_runtime": 2.6317, - "eval_samples_per_second": 15.96, - "eval_steps_per_second": 1.14, - "step": 309 - }, - { - "epoch": 225.45, - "eval_accuracy": 0.11904761904761904, - "eval_f1": 0.4181818181818182, - "eval_loss": 0.5122308731079102, - "eval_roc_auc": 0.7272256728778468, - "eval_runtime": 2.6565, - "eval_samples_per_second": 15.81, - "eval_steps_per_second": 1.129, - "step": 310 - }, - { - "epoch": 226.91, - "eval_accuracy": 0.11904761904761904, - "eval_f1": 0.40540540540540543, - "eval_loss": 0.511935830116272, - "eval_roc_auc": 0.7197204968944099, - "eval_runtime": 2.4688, - "eval_samples_per_second": 17.012, - "eval_steps_per_second": 1.215, - "step": 312 - }, - { - "epoch": 227.64, - "eval_accuracy": 0.11904761904761904, - "eval_f1": 0.41441441441441446, - "eval_loss": 0.510082483291626, - "eval_roc_auc": 0.7261904761904762, - "eval_runtime": 2.2331, - "eval_samples_per_second": 18.808, - "eval_steps_per_second": 1.343, - "step": 313 - }, - { - "epoch": 228.36, - "eval_accuracy": 0.14285714285714285, - "eval_f1": 0.4186046511627907, - "eval_loss": 0.5066081285476685, - "eval_roc_auc": 0.7233436853002071, - "eval_runtime": 2.3108, - "eval_samples_per_second": 18.176, - "eval_steps_per_second": 1.298, - "step": 314 + "epoch": 145.45, + "step": 200, + "total_flos": 1.357171875053568e+17, + "train_loss": 0.5662997841835022, + "train_runtime": 1746.9239, + "train_samples_per_second": 19.005, + "train_steps_per_second": 0.114 } ], "logging_steps": 500, - "max_steps": 400, - "num_train_epochs": 400, + "max_steps": 200, + "num_train_epochs": 200, "save_steps": 500, - "total_flos": 2.136295306761339e+17, + "total_flos": 1.357171875053568e+17, "trial_name": null, "trial_params": null }