diff --git "a/trainer_state.json" "b/trainer_state.json" new file mode 100644--- /dev/null +++ "b/trainer_state.json" @@ -0,0 +1,4250 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 2.766735570042932, + "global_step": 14500, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.02, + "eval_accuracy": 0.3251428571428571, + "eval_avg_bleuScore": 0.0745733853493418, + "eval_loss": 2.1220545768737793, + "eval_rouge1_fmeasure": 0.2992808520793915, + "eval_rouge1_precision": 0.2812275290489197, + "eval_rouge1_recall": 0.3760206699371338, + "eval_rouge2_fmeasure": 0.08637792617082596, + "eval_rouge2_precision": 0.08594733476638794, + "eval_rouge2_recall": 0.08828301727771759, + "eval_rougeL_fmeasure": 0.29808688163757324, + "eval_rougeL_precision": 0.27988532185554504, + "eval_rougeL_recall": 0.3748208284378052, + "eval_rougeLsum_fmeasure": 0.2981035113334656, + "eval_rougeLsum_precision": 0.27989742159843445, + "eval_rougeLsum_recall": 0.3748476803302765, + "eval_runtime": 913.6102, + "eval_samples_per_second": 3.831, + "eval_steps_per_second": 0.16, + "step": 100 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.3251428571428571, + "epoch": 0.02, + "step": 100 + }, + { + "epoch": 0.04, + "eval_accuracy": 0.31942857142857145, + "eval_avg_bleuScore": 0.0738210571195398, + "eval_loss": 1.942696452140808, + "eval_rouge1_fmeasure": 0.31272730231285095, + "eval_rouge1_precision": 0.30013591051101685, + "eval_rouge1_recall": 0.36964982748031616, + "eval_rouge2_fmeasure": 0.08618804067373276, + "eval_rouge2_precision": 0.08643926680088043, + "eval_rouge2_recall": 0.08688555657863617, + "eval_rougeL_fmeasure": 0.31108295917510986, + "eval_rougeL_precision": 0.29842445254325867, + "eval_rougeL_recall": 0.367925226688385, + "eval_rougeLsum_fmeasure": 0.3111913502216339, + "eval_rougeLsum_precision": 0.29851189255714417, + "eval_rougeLsum_recall": 0.3680727183818817, + "eval_runtime": 898.1759, + "eval_samples_per_second": 3.897, + "eval_steps_per_second": 0.163, + "step": 200 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.3251428571428571, + "epoch": 0.04, + "step": 200 + }, + { + "epoch": 0.06, + "eval_accuracy": 0.32057142857142856, + "eval_avg_bleuScore": 0.07543847053391593, + "eval_loss": 1.844632863998413, + "eval_rouge1_fmeasure": 0.3260011374950409, + "eval_rouge1_precision": 0.31697535514831543, + "eval_rouge1_recall": 0.3668860197067261, + "eval_rouge2_fmeasure": 0.08678946644067764, + "eval_rouge2_precision": 0.08774472773075104, + "eval_rouge2_recall": 0.08616892993450165, + "eval_rougeL_fmeasure": 0.3238302767276764, + "eval_rougeL_precision": 0.3146527409553528, + "eval_rougeL_recall": 0.36478349566459656, + "eval_rougeLsum_fmeasure": 0.32413598895072937, + "eval_rougeLsum_precision": 0.31493040919303894, + "eval_rougeLsum_recall": 0.3651273250579834, + "eval_runtime": 890.2932, + "eval_samples_per_second": 3.931, + "eval_steps_per_second": 0.164, + "step": 300 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.3251428571428571, + "epoch": 0.06, + "step": 300 + }, + { + "epoch": 0.08, + "eval_accuracy": 0.32142857142857145, + "eval_avg_bleuScore": 0.07643436065103326, + "eval_loss": 1.8006651401519775, + "eval_rouge1_fmeasure": 0.33292585611343384, + "eval_rouge1_precision": 0.3243910074234009, + "eval_rouge1_recall": 0.3675489127635956, + "eval_rouge2_fmeasure": 0.08797147870063782, + "eval_rouge2_precision": 0.08911019563674927, + "eval_rouge2_recall": 0.08713296055793762, + "eval_rougeL_fmeasure": 0.3302447199821472, + "eval_rougeL_precision": 0.32162585854530334, + "eval_rougeL_recall": 0.3648756742477417, + "eval_rougeLsum_fmeasure": 0.3307594954967499, + "eval_rougeLsum_precision": 0.3220929801464081, + "eval_rougeLsum_recall": 0.3654501140117645, + "eval_runtime": 798.3322, + "eval_samples_per_second": 4.384, + "eval_steps_per_second": 0.183, + "step": 400 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.3251428571428571, + "epoch": 0.08, + "step": 400 + }, + { + "epoch": 0.1, + "learning_rate": 7.716893438351779e-05, + "loss": 1.5246, + "step": 500 + }, + { + "epoch": 0.1, + "eval_accuracy": 0.32285714285714284, + "eval_avg_bleuScore": 0.07462256556749344, + "eval_loss": 1.7648615837097168, + "eval_rouge1_fmeasure": 0.33761346340179443, + "eval_rouge1_precision": 0.3302350640296936, + "eval_rouge1_recall": 0.366239458322525, + "eval_rouge2_fmeasure": 0.08622510731220245, + "eval_rouge2_precision": 0.08811099827289581, + "eval_rouge2_recall": 0.08490706235170364, + "eval_rougeL_fmeasure": 0.33533230423927307, + "eval_rougeL_precision": 0.3278222680091858, + "eval_rougeL_recall": 0.36401718854904175, + "eval_rougeLsum_fmeasure": 0.3356483578681946, + "eval_rougeLsum_precision": 0.328112930059433, + "eval_rougeLsum_recall": 0.3643641471862793, + "eval_runtime": 748.3362, + "eval_samples_per_second": 4.677, + "eval_steps_per_second": 0.195, + "step": 500 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.3251428571428571, + "epoch": 0.1, + "step": 500 + }, + { + "epoch": 0.11, + "eval_accuracy": 0.32657142857142857, + "eval_avg_bleuScore": 0.07433064450323582, + "eval_loss": 1.714306354522705, + "eval_rouge1_fmeasure": 0.3448028862476349, + "eval_rouge1_precision": 0.338601678609848, + "eval_rouge1_recall": 0.37148910760879517, + "eval_rouge2_fmeasure": 0.08587684482336044, + "eval_rouge2_precision": 0.08747506886720657, + "eval_rouge2_recall": 0.08492149412631989, + "eval_rougeL_fmeasure": 0.34224826097488403, + "eval_rougeL_precision": 0.33595019578933716, + "eval_rougeL_recall": 0.36895322799682617, + "eval_rougeLsum_fmeasure": 0.34255078434944153, + "eval_rougeLsum_precision": 0.3362291157245636, + "eval_rougeLsum_recall": 0.3692845404148102, + "eval_runtime": 724.7443, + "eval_samples_per_second": 4.829, + "eval_steps_per_second": 0.201, + "step": 600 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.32657142857142857, + "epoch": 0.11, + "step": 600 + }, + { + "epoch": 0.13, + "eval_accuracy": 0.334, + "eval_avg_bleuScore": 0.08131330903513091, + "eval_loss": 1.671054720878601, + "eval_rouge1_fmeasure": 0.3516198396682739, + "eval_rouge1_precision": 0.34505701065063477, + "eval_rouge1_recall": 0.3760612905025482, + "eval_rouge2_fmeasure": 0.09178104251623154, + "eval_rouge2_precision": 0.09262540936470032, + "eval_rouge2_recall": 0.09127616137266159, + "eval_rougeL_fmeasure": 0.3492134213447571, + "eval_rougeL_precision": 0.3426218628883362, + "eval_rougeL_recall": 0.3736189305782318, + "eval_rougeLsum_fmeasure": 0.3495814800262451, + "eval_rougeLsum_precision": 0.3429575264453888, + "eval_rougeLsum_recall": 0.3740270435810089, + "eval_runtime": 698.442, + "eval_samples_per_second": 5.011, + "eval_steps_per_second": 0.209, + "step": 700 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.334, + "epoch": 0.13, + "step": 700 + }, + { + "epoch": 0.15, + "eval_accuracy": 0.32971428571428574, + "eval_avg_bleuScore": 0.07740221605556352, + "eval_loss": 1.651564121246338, + "eval_rouge1_fmeasure": 0.3513752222061157, + "eval_rouge1_precision": 0.3461254835128784, + "eval_rouge1_recall": 0.3743011951446533, + "eval_rouge2_fmeasure": 0.08896287530660629, + "eval_rouge2_precision": 0.09063904732465744, + "eval_rouge2_recall": 0.0877472534775734, + "eval_rougeL_fmeasure": 0.349230021238327, + "eval_rougeL_precision": 0.3438914716243744, + "eval_rougeL_recall": 0.37217843532562256, + "eval_rougeLsum_fmeasure": 0.34958502650260925, + "eval_rougeLsum_precision": 0.3442164361476898, + "eval_rougeLsum_recall": 0.3725700378417969, + "eval_runtime": 703.9958, + "eval_samples_per_second": 4.972, + "eval_steps_per_second": 0.207, + "step": 800 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.334, + "epoch": 0.15, + "step": 800 + }, + { + "epoch": 0.17, + "eval_accuracy": 0.32771428571428574, + "eval_avg_bleuScore": 0.07535144642421178, + "eval_loss": 1.6252684593200684, + "eval_rouge1_fmeasure": 0.3528688848018646, + "eval_rouge1_precision": 0.3490174412727356, + "eval_rouge1_recall": 0.3739897608757019, + "eval_rouge2_fmeasure": 0.08777336776256561, + "eval_rouge2_precision": 0.09018763899803162, + "eval_rouge2_recall": 0.08610080182552338, + "eval_rougeL_fmeasure": 0.350429505109787, + "eval_rougeL_precision": 0.3465721309185028, + "eval_rougeL_recall": 0.37148576974868774, + "eval_rougeLsum_fmeasure": 0.3509392738342285, + "eval_rougeLsum_precision": 0.3470383882522583, + "eval_rougeLsum_recall": 0.3720492422580719, + "eval_runtime": 695.6653, + "eval_samples_per_second": 5.031, + "eval_steps_per_second": 0.21, + "step": 900 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.334, + "epoch": 0.17, + "step": 900 + }, + { + "epoch": 0.19, + "learning_rate": 8.577598223715977e-05, + "loss": 1.3823, + "step": 1000 + }, + { + "epoch": 0.19, + "eval_accuracy": 0.3374285714285714, + "eval_avg_bleuScore": 0.0865415023139545, + "eval_loss": 1.5922160148620605, + "eval_rouge1_fmeasure": 0.3594245910644531, + "eval_rouge1_precision": 0.3542421758174896, + "eval_rouge1_recall": 0.380466103553772, + "eval_rouge2_fmeasure": 0.09720273315906525, + "eval_rouge2_precision": 0.0980989933013916, + "eval_rouge2_recall": 0.09667235612869263, + "eval_rougeL_fmeasure": 0.35733601450920105, + "eval_rougeL_precision": 0.3521427512168884, + "eval_rougeL_recall": 0.3783356249332428, + "eval_rougeLsum_fmeasure": 0.35757553577423096, + "eval_rougeLsum_precision": 0.3523651659488678, + "eval_rougeLsum_recall": 0.37859562039375305, + "eval_runtime": 708.5097, + "eval_samples_per_second": 4.94, + "eval_steps_per_second": 0.206, + "step": 1000 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.3374285714285714, + "epoch": 0.19, + "step": 1000 + }, + { + "epoch": 0.21, + "eval_accuracy": 0.344, + "eval_avg_bleuScore": 0.08740193068555423, + "eval_loss": 1.564757227897644, + "eval_rouge1_fmeasure": 0.3652290403842926, + "eval_rouge1_precision": 0.35990938544273376, + "eval_rouge1_recall": 0.38584136962890625, + "eval_rouge2_fmeasure": 0.09822201728820801, + "eval_rouge2_precision": 0.09913189709186554, + "eval_rouge2_recall": 0.09776327013969421, + "eval_rougeL_fmeasure": 0.3631420135498047, + "eval_rougeL_precision": 0.35781964659690857, + "eval_rougeL_recall": 0.38370072841644287, + "eval_rougeLsum_fmeasure": 0.36354386806488037, + "eval_rougeLsum_precision": 0.35819050669670105, + "eval_rougeLsum_recall": 0.38414037227630615, + "eval_runtime": 699.3996, + "eval_samples_per_second": 5.004, + "eval_steps_per_second": 0.209, + "step": 1100 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.344, + "epoch": 0.21, + "step": 1100 + }, + { + "epoch": 0.23, + "eval_accuracy": 0.33057142857142857, + "eval_avg_bleuScore": 0.0753437274992466, + "eval_loss": 1.5405296087265015, + "eval_rouge1_fmeasure": 0.36006495356559753, + "eval_rouge1_precision": 0.3572610318660736, + "eval_rouge1_recall": 0.3769795000553131, + "eval_rouge2_fmeasure": 0.08717846870422363, + "eval_rouge2_precision": 0.08941493183374405, + "eval_rouge2_recall": 0.0856637954711914, + "eval_rougeL_fmeasure": 0.3579935133457184, + "eval_rougeL_precision": 0.35501858592033386, + "eval_rougeL_recall": 0.37497374415397644, + "eval_rougeLsum_fmeasure": 0.35818058252334595, + "eval_rougeLsum_precision": 0.3551930785179138, + "eval_rougeLsum_recall": 0.3751756548881531, + "eval_runtime": 699.4377, + "eval_samples_per_second": 5.004, + "eval_steps_per_second": 0.209, + "step": 1200 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.344, + "epoch": 0.23, + "step": 1200 + }, + { + "epoch": 0.25, + "eval_accuracy": 0.33285714285714285, + "eval_avg_bleuScore": 0.07532266126360213, + "eval_loss": 1.5064581632614136, + "eval_rouge1_fmeasure": 0.3675118088722229, + "eval_rouge1_precision": 0.365093469619751, + "eval_rouge1_recall": 0.3846178948879242, + "eval_rouge2_fmeasure": 0.08729325979948044, + "eval_rouge2_precision": 0.0897003710269928, + "eval_rouge2_recall": 0.08560299873352051, + "eval_rougeL_fmeasure": 0.36565306782722473, + "eval_rougeL_precision": 0.3631473183631897, + "eval_rougeL_recall": 0.38278815150260925, + "eval_rougeLsum_fmeasure": 0.36586666107177734, + "eval_rougeLsum_precision": 0.36334457993507385, + "eval_rougeLsum_recall": 0.3830215334892273, + "eval_runtime": 704.3112, + "eval_samples_per_second": 4.969, + "eval_steps_per_second": 0.207, + "step": 1300 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.344, + "epoch": 0.25, + "step": 1300 + }, + { + "epoch": 0.27, + "eval_accuracy": 0.3394285714285714, + "eval_avg_bleuScore": 0.08493705511944634, + "eval_loss": 1.481276273727417, + "eval_rouge1_fmeasure": 0.37872248888015747, + "eval_rouge1_precision": 0.3743116557598114, + "eval_rouge1_recall": 0.3956756293773651, + "eval_rouge2_fmeasure": 0.0944737046957016, + "eval_rouge2_precision": 0.09505806863307953, + "eval_rouge2_recall": 0.09495003521442413, + "eval_rougeL_fmeasure": 0.3768906593322754, + "eval_rougeL_precision": 0.37237071990966797, + "eval_rougeL_recall": 0.39389413595199585, + "eval_rougeLsum_fmeasure": 0.37710797786712646, + "eval_rougeLsum_precision": 0.37257373332977295, + "eval_rougeLsum_recall": 0.3941282033920288, + "eval_runtime": 714.0662, + "eval_samples_per_second": 4.902, + "eval_steps_per_second": 0.204, + "step": 1400 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.344, + "epoch": 0.27, + "step": 1400 + }, + { + "epoch": 0.29, + "learning_rate": 9.081078247345285e-05, + "loss": 1.3285, + "step": 1500 + }, + { + "epoch": 0.29, + "eval_accuracy": 0.3362857142857143, + "eval_avg_bleuScore": 0.07449912045257433, + "eval_loss": 1.4618046283721924, + "eval_rouge1_fmeasure": 0.3828262686729431, + "eval_rouge1_precision": 0.380904883146286, + "eval_rouge1_recall": 0.3967396020889282, + "eval_rouge2_fmeasure": 0.08693035691976547, + "eval_rouge2_precision": 0.0893305316567421, + "eval_rouge2_recall": 0.08521973341703415, + "eval_rougeL_fmeasure": 0.3809206187725067, + "eval_rougeL_precision": 0.3789350986480713, + "eval_rougeL_recall": 0.3948442339897156, + "eval_rougeLsum_fmeasure": 0.3812638223171234, + "eval_rougeLsum_precision": 0.3792624771595001, + "eval_rougeLsum_recall": 0.39520859718322754, + "eval_runtime": 701.2173, + "eval_samples_per_second": 4.991, + "eval_steps_per_second": 0.208, + "step": 1500 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.344, + "epoch": 0.29, + "step": 1500 + }, + { + "epoch": 0.31, + "eval_accuracy": 0.3477142857142857, + "eval_avg_bleuScore": 0.08140425328697477, + "eval_loss": 1.4373503923416138, + "eval_rouge1_fmeasure": 0.39635929465293884, + "eval_rouge1_precision": 0.39322328567504883, + "eval_rouge1_recall": 0.4119550883769989, + "eval_rouge2_fmeasure": 0.09226943552494049, + "eval_rouge2_precision": 0.09366623312234879, + "eval_rouge2_recall": 0.09146234393119812, + "eval_rougeL_fmeasure": 0.3947158455848694, + "eval_rougeL_precision": 0.39147716760635376, + "eval_rougeL_recall": 0.4103486239910126, + "eval_rougeLsum_fmeasure": 0.3948381841182709, + "eval_rougeLsum_precision": 0.3915906548500061, + "eval_rougeLsum_recall": 0.4104819595813751, + "eval_runtime": 696.5191, + "eval_samples_per_second": 5.025, + "eval_steps_per_second": 0.21, + "step": 1600 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.3477142857142857, + "epoch": 0.31, + "step": 1600 + }, + { + "epoch": 0.32, + "eval_accuracy": 0.3457142857142857, + "eval_avg_bleuScore": 0.07487628586803163, + "eval_loss": 1.4143986701965332, + "eval_rouge1_fmeasure": 0.399248331785202, + "eval_rouge1_precision": 0.3977319896221161, + "eval_rouge1_recall": 0.41258272528648376, + "eval_rouge2_fmeasure": 0.08611094951629639, + "eval_rouge2_precision": 0.08842268586158752, + "eval_rouge2_recall": 0.08461552113294601, + "eval_rougeL_fmeasure": 0.3975357115268707, + "eval_rougeL_precision": 0.3959176540374756, + "eval_rougeL_recall": 0.4109116792678833, + "eval_rougeLsum_fmeasure": 0.39778974652290344, + "eval_rougeLsum_precision": 0.3961593210697174, + "eval_rougeLsum_recall": 0.4111817181110382, + "eval_runtime": 713.1959, + "eval_samples_per_second": 4.907, + "eval_steps_per_second": 0.205, + "step": 1700 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.3477142857142857, + "epoch": 0.32, + "step": 1700 + }, + { + "epoch": 0.34, + "eval_accuracy": 0.3485714285714286, + "eval_avg_bleuScore": 0.07690940167861325, + "eval_loss": 1.3987131118774414, + "eval_rouge1_fmeasure": 0.40655869245529175, + "eval_rouge1_precision": 0.4051145315170288, + "eval_rouge1_recall": 0.4184306561946869, + "eval_rouge2_fmeasure": 0.08818119019269943, + "eval_rouge2_precision": 0.09028911590576172, + "eval_rouge2_recall": 0.08680981397628784, + "eval_rougeL_fmeasure": 0.40484708547592163, + "eval_rougeL_precision": 0.4033145308494568, + "eval_rougeL_recall": 0.41675934195518494, + "eval_rougeLsum_fmeasure": 0.4049142003059387, + "eval_rougeLsum_precision": 0.4033772647380829, + "eval_rougeLsum_recall": 0.4168316423892975, + "eval_runtime": 701.4816, + "eval_samples_per_second": 4.989, + "eval_steps_per_second": 0.208, + "step": 1800 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.3485714285714286, + "epoch": 0.34, + "step": 1800 + }, + { + "epoch": 0.36, + "eval_accuracy": 0.3517142857142857, + "eval_avg_bleuScore": 0.07360557946562767, + "eval_loss": 1.3808749914169312, + "eval_rouge1_fmeasure": 0.4142747223377228, + "eval_rouge1_precision": 0.41384708881378174, + "eval_rouge1_recall": 0.424259752035141, + "eval_rouge2_fmeasure": 0.08621610701084137, + "eval_rouge2_precision": 0.08913641422986984, + "eval_rouge2_recall": 0.08391223102807999, + "eval_rougeL_fmeasure": 0.41260817646980286, + "eval_rougeL_precision": 0.41208595037460327, + "eval_rougeL_recall": 0.42260509729385376, + "eval_rougeLsum_fmeasure": 0.41275304555892944, + "eval_rougeLsum_precision": 0.4122185707092285, + "eval_rougeLsum_recall": 0.42276477813720703, + "eval_runtime": 694.5533, + "eval_samples_per_second": 5.039, + "eval_steps_per_second": 0.21, + "step": 1900 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.3517142857142857, + "epoch": 0.36, + "step": 1900 + }, + { + "epoch": 0.38, + "learning_rate": 9.438303009080175e-05, + "loss": 1.2994, + "step": 2000 + }, + { + "epoch": 0.38, + "eval_accuracy": 0.3517142857142857, + "eval_avg_bleuScore": 0.07286149430700711, + "eval_loss": 1.3714731931686401, + "eval_rouge1_fmeasure": 0.4117237627506256, + "eval_rouge1_precision": 0.4120018184185028, + "eval_rouge1_recall": 0.42169326543807983, + "eval_rouge2_fmeasure": 0.08532802760601044, + "eval_rouge2_precision": 0.08820176124572754, + "eval_rouge2_recall": 0.08304416388273239, + "eval_rougeL_fmeasure": 0.41048464179039, + "eval_rougeL_precision": 0.4106236696243286, + "eval_rougeL_recall": 0.42048653960227966, + "eval_rougeLsum_fmeasure": 0.41061776876449585, + "eval_rougeLsum_precision": 0.41074612736701965, + "eval_rougeLsum_recall": 0.42063260078430176, + "eval_runtime": 692.9843, + "eval_samples_per_second": 5.051, + "eval_steps_per_second": 0.211, + "step": 2000 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.3517142857142857, + "epoch": 0.38, + "step": 2000 + }, + { + "epoch": 0.4, + "eval_accuracy": 0.36742857142857144, + "eval_avg_bleuScore": 0.08312786594884736, + "eval_loss": 1.3592321872711182, + "eval_rouge1_fmeasure": 0.4221250116825104, + "eval_rouge1_precision": 0.4204561710357666, + "eval_rouge1_recall": 0.4345695674419403, + "eval_rouge2_fmeasure": 0.09401486068964005, + "eval_rouge2_precision": 0.09581853449344635, + "eval_rouge2_recall": 0.09320167452096939, + "eval_rougeL_fmeasure": 0.42061400413513184, + "eval_rougeL_precision": 0.41885581612586975, + "eval_rougeL_recall": 0.43308743834495544, + "eval_rougeLsum_fmeasure": 0.42070651054382324, + "eval_rougeLsum_precision": 0.41894257068634033, + "eval_rougeLsum_recall": 0.43318700790405273, + "eval_runtime": 709.0944, + "eval_samples_per_second": 4.936, + "eval_steps_per_second": 0.206, + "step": 2100 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.36742857142857144, + "epoch": 0.4, + "step": 2100 + }, + { + "epoch": 0.42, + "eval_accuracy": 0.354, + "eval_avg_bleuScore": 0.07440906418221338, + "eval_loss": 1.3535100221633911, + "eval_rouge1_fmeasure": 0.4181794226169586, + "eval_rouge1_precision": 0.41823887825012207, + "eval_rouge1_recall": 0.4276091754436493, + "eval_rouge2_fmeasure": 0.08696828782558441, + "eval_rouge2_precision": 0.09004759043455124, + "eval_rouge2_recall": 0.08465278148651123, + "eval_rougeL_fmeasure": 0.4165239632129669, + "eval_rougeL_precision": 0.41647037863731384, + "eval_rougeL_recall": 0.4260002672672272, + "eval_rougeLsum_fmeasure": 0.41665026545524597, + "eval_rougeLsum_precision": 0.4165937900543213, + "eval_rougeLsum_recall": 0.42613115906715393, + "eval_runtime": 709.0675, + "eval_samples_per_second": 4.936, + "eval_steps_per_second": 0.206, + "step": 2200 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.36742857142857144, + "epoch": 0.42, + "step": 2200 + }, + { + "epoch": 0.44, + "eval_accuracy": 0.3505714285714286, + "eval_avg_bleuScore": 0.07958767581624644, + "eval_loss": 1.3441429138183594, + "eval_rouge1_fmeasure": 0.4174051284790039, + "eval_rouge1_precision": 0.41696277260780334, + "eval_rouge1_recall": 0.42647552490234375, + "eval_rouge2_fmeasure": 0.09096981585025787, + "eval_rouge2_precision": 0.09321748465299606, + "eval_rouge2_recall": 0.08939649909734726, + "eval_rougeL_fmeasure": 0.4159727990627289, + "eval_rougeL_precision": 0.41545334458351135, + "eval_rougeL_recall": 0.42506295442581177, + "eval_rougeLsum_fmeasure": 0.4161238968372345, + "eval_rougeLsum_precision": 0.4155937135219574, + "eval_rougeLsum_recall": 0.42522698640823364, + "eval_runtime": 699.093, + "eval_samples_per_second": 5.006, + "eval_steps_per_second": 0.209, + "step": 2300 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.36742857142857144, + "epoch": 0.44, + "step": 2300 + }, + { + "epoch": 0.46, + "eval_accuracy": 0.3525714285714286, + "eval_avg_bleuScore": 0.07383340352560792, + "eval_loss": 1.3369330167770386, + "eval_rouge1_fmeasure": 0.41869544982910156, + "eval_rouge1_precision": 0.4197627007961273, + "eval_rouge1_recall": 0.42565464973449707, + "eval_rouge2_fmeasure": 0.08611708134412766, + "eval_rouge2_precision": 0.08915538340806961, + "eval_rouge2_recall": 0.08372662216424942, + "eval_rougeL_fmeasure": 0.41737088561058044, + "eval_rougeL_precision": 0.41823625564575195, + "eval_rougeL_recall": 0.424407422542572, + "eval_rougeLsum_fmeasure": 0.4174564480781555, + "eval_rougeLsum_precision": 0.4183203876018524, + "eval_rougeLsum_recall": 0.42449623346328735, + "eval_runtime": 690.9783, + "eval_samples_per_second": 5.065, + "eval_steps_per_second": 0.211, + "step": 2400 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.36742857142857144, + "epoch": 0.46, + "step": 2400 + }, + { + "epoch": 0.48, + "learning_rate": 9.715388060892908e-05, + "loss": 1.2715, + "step": 2500 + }, + { + "epoch": 0.48, + "eval_accuracy": 0.3534285714285714, + "eval_avg_bleuScore": 0.08200813191703388, + "eval_loss": 1.3224023580551147, + "eval_rouge1_fmeasure": 0.42352792620658875, + "eval_rouge1_precision": 0.42280080914497375, + "eval_rouge1_recall": 0.4321795403957367, + "eval_rouge2_fmeasure": 0.09303068369626999, + "eval_rouge2_precision": 0.0950784906744957, + "eval_rouge2_recall": 0.0917518362402916, + "eval_rougeL_fmeasure": 0.42219287157058716, + "eval_rougeL_precision": 0.4213830530643463, + "eval_rougeL_recall": 0.4308737516403198, + "eval_rougeLsum_fmeasure": 0.42229074239730835, + "eval_rougeLsum_precision": 0.42147621512413025, + "eval_rougeLsum_recall": 0.43097832798957825, + "eval_runtime": 691.1523, + "eval_samples_per_second": 5.064, + "eval_steps_per_second": 0.211, + "step": 2500 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.36742857142857144, + "epoch": 0.48, + "step": 2500 + }, + { + "epoch": 0.5, + "eval_accuracy": 0.356, + "eval_avg_bleuScore": 0.07903471415383476, + "eval_loss": 1.313102126121521, + "eval_rouge1_fmeasure": 0.4285103976726532, + "eval_rouge1_precision": 0.42894691228866577, + "eval_rouge1_recall": 0.43571338057518005, + "eval_rouge2_fmeasure": 0.0918598547577858, + "eval_rouge2_precision": 0.09501548856496811, + "eval_rouge2_recall": 0.08941777795553207, + "eval_rougeL_fmeasure": 0.4274981915950775, + "eval_rougeL_precision": 0.42778754234313965, + "eval_rougeL_recall": 0.43475398421287537, + "eval_rougeLsum_fmeasure": 0.42751333117485046, + "eval_rougeLsum_precision": 0.4278022050857544, + "eval_rougeLsum_recall": 0.43476974964141846, + "eval_runtime": 693.4522, + "eval_samples_per_second": 5.047, + "eval_steps_per_second": 0.211, + "step": 2600 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.36742857142857144, + "epoch": 0.5, + "step": 2600 + }, + { + "epoch": 0.52, + "eval_accuracy": 0.36314285714285716, + "eval_avg_bleuScore": 0.0845105659365654, + "eval_loss": 1.31019926071167, + "eval_rouge1_fmeasure": 0.4313008785247803, + "eval_rouge1_precision": 0.4301567077636719, + "eval_rouge1_recall": 0.4409593343734741, + "eval_rouge2_fmeasure": 0.09496873617172241, + "eval_rouge2_precision": 0.09683161228895187, + "eval_rouge2_recall": 0.09365107119083405, + "eval_rougeL_fmeasure": 0.4300004541873932, + "eval_rougeL_precision": 0.4287497103214264, + "eval_rougeL_recall": 0.43968743085861206, + "eval_rougeLsum_fmeasure": 0.43011200428009033, + "eval_rougeLsum_precision": 0.42885690927505493, + "eval_rougeLsum_recall": 0.4398055374622345, + "eval_runtime": 697.8753, + "eval_samples_per_second": 5.015, + "eval_steps_per_second": 0.209, + "step": 2700 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.36742857142857144, + "epoch": 0.52, + "step": 2700 + }, + { + "epoch": 0.53, + "eval_accuracy": 0.3585714285714286, + "eval_avg_bleuScore": 0.09339015859791211, + "eval_loss": 1.2990689277648926, + "eval_rouge1_fmeasure": 0.43968042731285095, + "eval_rouge1_precision": 0.43758952617645264, + "eval_rouge1_recall": 0.4486676752567291, + "eval_rouge2_fmeasure": 0.10271972417831421, + "eval_rouge2_precision": 0.1033976599574089, + "eval_rouge2_recall": 0.10252416133880615, + "eval_rougeL_fmeasure": 0.4387567937374115, + "eval_rougeL_precision": 0.43660858273506165, + "eval_rougeL_recall": 0.4477545917034149, + "eval_rougeLsum_fmeasure": 0.4389358460903168, + "eval_rougeLsum_precision": 0.4367820918560028, + "eval_rougeLsum_recall": 0.4479435682296753, + "eval_runtime": 709.4176, + "eval_samples_per_second": 4.934, + "eval_steps_per_second": 0.206, + "step": 2800 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.36742857142857144, + "epoch": 0.53, + "step": 2800 + }, + { + "epoch": 0.55, + "eval_accuracy": 0.36857142857142855, + "eval_avg_bleuScore": 0.0916646334188325, + "eval_loss": 1.2878822088241577, + "eval_rouge1_fmeasure": 0.441276878118515, + "eval_rouge1_precision": 0.440507173538208, + "eval_rouge1_recall": 0.4485040009021759, + "eval_rouge2_fmeasure": 0.10225692391395569, + "eval_rouge2_precision": 0.10385410487651825, + "eval_rouge2_recall": 0.10115861147642136, + "eval_rougeL_fmeasure": 0.4400864839553833, + "eval_rougeL_precision": 0.4392184019088745, + "eval_rougeL_recall": 0.4473549723625183, + "eval_rougeLsum_fmeasure": 0.4402983486652374, + "eval_rougeLsum_precision": 0.4394386112689972, + "eval_rougeLsum_recall": 0.44756269454956055, + "eval_runtime": 705.9775, + "eval_samples_per_second": 4.958, + "eval_steps_per_second": 0.207, + "step": 2900 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.36857142857142855, + "epoch": 0.55, + "step": 2900 + }, + { + "epoch": 0.57, + "learning_rate": 9.94178303270948e-05, + "loss": 1.2595, + "step": 3000 + }, + { + "epoch": 0.57, + "eval_accuracy": 0.36457142857142855, + "eval_avg_bleuScore": 0.08460148728745324, + "eval_loss": 1.2883292436599731, + "eval_rouge1_fmeasure": 0.43844008445739746, + "eval_rouge1_precision": 0.43918582797050476, + "eval_rouge1_recall": 0.4443702697753906, + "eval_rouge2_fmeasure": 0.0959305688738823, + "eval_rouge2_precision": 0.09856921434402466, + "eval_rouge2_recall": 0.09386628121137619, + "eval_rougeL_fmeasure": 0.43742692470550537, + "eval_rougeL_precision": 0.4380800724029541, + "eval_rougeL_recall": 0.4433978497982025, + "eval_rougeLsum_fmeasure": 0.43755075335502625, + "eval_rougeLsum_precision": 0.43819552659988403, + "eval_rougeLsum_recall": 0.4435324966907501, + "eval_runtime": 692.4557, + "eval_samples_per_second": 5.054, + "eval_steps_per_second": 0.211, + "step": 3000 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.36857142857142855, + "epoch": 0.57, + "step": 3000 + }, + { + "epoch": 0.59, + "eval_accuracy": 0.36228571428571427, + "eval_avg_bleuScore": 0.08632448217698506, + "eval_loss": 1.2798010110855103, + "eval_rouge1_fmeasure": 0.4405716359615326, + "eval_rouge1_precision": 0.4409351348876953, + "eval_rouge1_recall": 0.4467078745365143, + "eval_rouge2_fmeasure": 0.09655094146728516, + "eval_rouge2_precision": 0.0983826145529747, + "eval_rouge2_recall": 0.09521066397428513, + "eval_rougeL_fmeasure": 0.43951985239982605, + "eval_rougeL_precision": 0.4398261606693268, + "eval_rougeL_recall": 0.44567814469337463, + "eval_rougeLsum_fmeasure": 0.4396824538707733, + "eval_rougeLsum_precision": 0.43997839093208313, + "eval_rougeLsum_recall": 0.44585421681404114, + "eval_runtime": 709.5546, + "eval_samples_per_second": 4.933, + "eval_steps_per_second": 0.206, + "step": 3100 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.36857142857142855, + "epoch": 0.59, + "step": 3100 + }, + { + "epoch": 0.61, + "eval_accuracy": 0.36114285714285715, + "eval_avg_bleuScore": 0.07876428666710854, + "eval_loss": 1.2813605070114136, + "eval_rouge1_fmeasure": 0.44014546275138855, + "eval_rouge1_precision": 0.4417456090450287, + "eval_rouge1_recall": 0.44410258531570435, + "eval_rouge2_fmeasure": 0.091702900826931, + "eval_rouge2_precision": 0.0948430597782135, + "eval_rouge2_recall": 0.08914921432733536, + "eval_rougeL_fmeasure": 0.4393033981323242, + "eval_rougeL_precision": 0.44087451696395874, + "eval_rougeL_recall": 0.4432714283466339, + "eval_rougeLsum_fmeasure": 0.43946728110313416, + "eval_rougeLsum_precision": 0.44102901220321655, + "eval_rougeLsum_recall": 0.44344621896743774, + "eval_runtime": 704.1348, + "eval_samples_per_second": 4.971, + "eval_steps_per_second": 0.207, + "step": 3200 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.36857142857142855, + "epoch": 0.61, + "step": 3200 + }, + { + "epoch": 0.63, + "eval_accuracy": 0.3688571428571429, + "eval_avg_bleuScore": 0.08849955462557929, + "eval_loss": 1.2695039510726929, + "eval_rouge1_fmeasure": 0.44859549403190613, + "eval_rouge1_precision": 0.44912123680114746, + "eval_rouge1_recall": 0.453075647354126, + "eval_rouge2_fmeasure": 0.09941935539245605, + "eval_rouge2_precision": 0.10121257603168488, + "eval_rouge2_recall": 0.0978400707244873, + "eval_rougeL_fmeasure": 0.44765880703926086, + "eval_rougeL_precision": 0.44812893867492676, + "eval_rougeL_recall": 0.4521603286266327, + "eval_rougeLsum_fmeasure": 0.44789251685142517, + "eval_rougeLsum_precision": 0.44836610555648804, + "eval_rougeLsum_recall": 0.4523959755897522, + "eval_runtime": 696.8838, + "eval_samples_per_second": 5.022, + "eval_steps_per_second": 0.21, + "step": 3300 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.3688571428571429, + "epoch": 0.63, + "step": 3300 + }, + { + "epoch": 0.65, + "eval_accuracy": 0.3697142857142857, + "eval_avg_bleuScore": 0.08374038383364678, + "eval_loss": 1.2647489309310913, + "eval_rouge1_fmeasure": 0.4532695412635803, + "eval_rouge1_precision": 0.45522966980934143, + "eval_rouge1_recall": 0.4574224352836609, + "eval_rouge2_fmeasure": 0.09654012322425842, + "eval_rouge2_precision": 0.09985741972923279, + "eval_rouge2_recall": 0.09369153529405594, + "eval_rougeL_fmeasure": 0.45258432626724243, + "eval_rougeL_precision": 0.4544983506202698, + "eval_rougeL_recall": 0.45675867795944214, + "eval_rougeLsum_fmeasure": 0.4527440369129181, + "eval_rougeLsum_precision": 0.45465853810310364, + "eval_rougeLsum_recall": 0.4569220244884491, + "eval_runtime": 702.7928, + "eval_samples_per_second": 4.98, + "eval_steps_per_second": 0.208, + "step": 3400 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.3697142857142857, + "epoch": 0.65, + "step": 3400 + }, + { + "epoch": 0.67, + "learning_rate": 0.0001, + "loss": 1.2468, + "step": 3500 + }, + { + "epoch": 0.67, + "eval_accuracy": 0.3722857142857143, + "eval_avg_bleuScore": 0.0898344544172287, + "eval_loss": 1.2631494998931885, + "eval_rouge1_fmeasure": 0.4526388943195343, + "eval_rouge1_precision": 0.45332252979278564, + "eval_rouge1_recall": 0.45762139558792114, + "eval_rouge2_fmeasure": 0.10130273550748825, + "eval_rouge2_precision": 0.10334417968988419, + "eval_rouge2_recall": 0.09988858550786972, + "eval_rougeL_fmeasure": 0.4517468512058258, + "eval_rougeL_precision": 0.4523561894893646, + "eval_rougeL_recall": 0.45676282048225403, + "eval_rougeLsum_fmeasure": 0.45191681385040283, + "eval_rougeLsum_precision": 0.4525180757045746, + "eval_rougeLsum_recall": 0.4569425880908966, + "eval_runtime": 690.6871, + "eval_samples_per_second": 5.067, + "eval_steps_per_second": 0.211, + "step": 3500 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.3722857142857143, + "epoch": 0.67, + "step": 3500 + }, + { + "epoch": 0.69, + "eval_accuracy": 0.36114285714285715, + "eval_avg_bleuScore": 0.08248263702222279, + "eval_loss": 1.2628397941589355, + "eval_rouge1_fmeasure": 0.4450514316558838, + "eval_rouge1_precision": 0.44653281569480896, + "eval_rouge1_recall": 0.44836175441741943, + "eval_rouge2_fmeasure": 0.09449034184217453, + "eval_rouge2_precision": 0.09710904955863953, + "eval_rouge2_recall": 0.09215880185365677, + "eval_rougeL_fmeasure": 0.4441673159599304, + "eval_rougeL_precision": 0.4455670714378357, + "eval_rougeL_recall": 0.4475208818912506, + "eval_rougeLsum_fmeasure": 0.44438573718070984, + "eval_rougeLsum_precision": 0.445785254240036, + "eval_rougeLsum_recall": 0.44774359464645386, + "eval_runtime": 700.9702, + "eval_samples_per_second": 4.993, + "eval_steps_per_second": 0.208, + "step": 3600 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.3722857142857143, + "epoch": 0.69, + "step": 3600 + }, + { + "epoch": 0.71, + "eval_accuracy": 0.36542857142857144, + "eval_avg_bleuScore": 0.0821412101643426, + "eval_loss": 1.258082628250122, + "eval_rouge1_fmeasure": 0.4495663046836853, + "eval_rouge1_precision": 0.45089733600616455, + "eval_rouge1_recall": 0.4526377022266388, + "eval_rouge2_fmeasure": 0.0941174328327179, + "eval_rouge2_precision": 0.0965518206357956, + "eval_rouge2_recall": 0.09197410196065903, + "eval_rougeL_fmeasure": 0.44863811135292053, + "eval_rougeL_precision": 0.4499381184577942, + "eval_rougeL_recall": 0.45171183347702026, + "eval_rougeLsum_fmeasure": 0.44875702261924744, + "eval_rougeLsum_precision": 0.4500638544559479, + "eval_rougeLsum_recall": 0.4518292546272278, + "eval_runtime": 681.9596, + "eval_samples_per_second": 5.132, + "eval_steps_per_second": 0.214, + "step": 3700 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.3722857142857143, + "epoch": 0.71, + "step": 3700 + }, + { + "epoch": 0.73, + "eval_accuracy": 0.36914285714285716, + "eval_avg_bleuScore": 0.08507979455164501, + "eval_loss": 1.2532260417938232, + "eval_rouge1_fmeasure": 0.45211511850357056, + "eval_rouge1_precision": 0.4531807005405426, + "eval_rouge1_recall": 0.4561786651611328, + "eval_rouge2_fmeasure": 0.0958511233329773, + "eval_rouge2_precision": 0.09806417673826218, + "eval_rouge2_recall": 0.09390494227409363, + "eval_rougeL_fmeasure": 0.4513184726238251, + "eval_rougeL_precision": 0.4523225426673889, + "eval_rougeL_recall": 0.45541298389434814, + "eval_rougeLsum_fmeasure": 0.4513740837574005, + "eval_rougeLsum_precision": 0.4523768723011017, + "eval_rougeLsum_recall": 0.4554702937602997, + "eval_runtime": 689.1966, + "eval_samples_per_second": 5.078, + "eval_steps_per_second": 0.212, + "step": 3800 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.3722857142857143, + "epoch": 0.73, + "step": 3800 + }, + { + "epoch": 0.74, + "eval_accuracy": 0.3677142857142857, + "eval_avg_bleuScore": 0.08453243540440288, + "eval_loss": 1.2511509656906128, + "eval_rouge1_fmeasure": 0.4539650082588196, + "eval_rouge1_precision": 0.4542441666126251, + "eval_rouge1_recall": 0.45889800786972046, + "eval_rouge2_fmeasure": 0.09482467174530029, + "eval_rouge2_precision": 0.09644889086484909, + "eval_rouge2_recall": 0.09386061877012253, + "eval_rougeL_fmeasure": 0.4531556963920593, + "eval_rougeL_precision": 0.453296035528183, + "eval_rougeL_recall": 0.4581603407859802, + "eval_rougeLsum_fmeasure": 0.4532601535320282, + "eval_rougeLsum_precision": 0.45340654253959656, + "eval_rougeLsum_recall": 0.4582621157169342, + "eval_runtime": 687.7981, + "eval_samples_per_second": 5.089, + "eval_steps_per_second": 0.212, + "step": 3900 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.3722857142857143, + "epoch": 0.74, + "step": 3900 + }, + { + "epoch": 0.76, + "learning_rate": 0.0001, + "loss": 1.2388, + "step": 4000 + }, + { + "epoch": 0.76, + "eval_accuracy": 0.36657142857142855, + "eval_avg_bleuScore": 0.08306138198290552, + "eval_loss": 1.2480295896530151, + "eval_rouge1_fmeasure": 0.4550284147262573, + "eval_rouge1_precision": 0.45570093393325806, + "eval_rouge1_recall": 0.45861899852752686, + "eval_rouge2_fmeasure": 0.09431128203868866, + "eval_rouge2_precision": 0.09617506712675095, + "eval_rouge2_recall": 0.09272296726703644, + "eval_rougeL_fmeasure": 0.4542606472969055, + "eval_rougeL_precision": 0.45486485958099365, + "eval_rougeL_recall": 0.45787984132766724, + "eval_rougeLsum_fmeasure": 0.4544673264026642, + "eval_rougeLsum_precision": 0.45507797598838806, + "eval_rougeLsum_recall": 0.45808762311935425, + "eval_runtime": 709.3533, + "eval_samples_per_second": 4.934, + "eval_steps_per_second": 0.206, + "step": 4000 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.3722857142857143, + "epoch": 0.76, + "step": 4000 + }, + { + "epoch": 0.78, + "eval_accuracy": 0.3717142857142857, + "eval_avg_bleuScore": 0.09331397067649023, + "eval_loss": 1.241654872894287, + "eval_rouge1_fmeasure": 0.4610021412372589, + "eval_rouge1_precision": 0.46171820163726807, + "eval_rouge1_recall": 0.46422144770622253, + "eval_rouge2_fmeasure": 0.10355381667613983, + "eval_rouge2_precision": 0.10518115013837814, + "eval_rouge2_recall": 0.10221449285745621, + "eval_rougeL_fmeasure": 0.4605155885219574, + "eval_rougeL_precision": 0.4611774981021881, + "eval_rougeL_recall": 0.46375998854637146, + "eval_rougeLsum_fmeasure": 0.46060290932655334, + "eval_rougeLsum_precision": 0.46126237511634827, + "eval_rougeLsum_recall": 0.46385031938552856, + "eval_runtime": 689.3096, + "eval_samples_per_second": 5.078, + "eval_steps_per_second": 0.212, + "step": 4100 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.3722857142857143, + "epoch": 0.78, + "step": 4100 + }, + { + "epoch": 0.8, + "eval_accuracy": 0.37514285714285717, + "eval_avg_bleuScore": 0.08929537396771567, + "eval_loss": 1.2388948202133179, + "eval_rouge1_fmeasure": 0.45842626690864563, + "eval_rouge1_precision": 0.4595540165901184, + "eval_rouge1_recall": 0.46154847741127014, + "eval_rouge2_fmeasure": 0.09942089766263962, + "eval_rouge2_precision": 0.1010797843337059, + "eval_rouge2_recall": 0.0979815125465393, + "eval_rougeL_fmeasure": 0.45771992206573486, + "eval_rougeL_precision": 0.45877012610435486, + "eval_rougeL_recall": 0.46088099479675293, + "eval_rougeLsum_fmeasure": 0.4578106999397278, + "eval_rougeLsum_precision": 0.45886218547821045, + "eval_rougeLsum_recall": 0.46097204089164734, + "eval_runtime": 689.1816, + "eval_samples_per_second": 5.078, + "eval_steps_per_second": 0.212, + "step": 4200 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.37514285714285717, + "epoch": 0.8, + "step": 4200 + }, + { + "epoch": 0.82, + "eval_accuracy": 0.37457142857142856, + "eval_avg_bleuScore": 0.08391665462936673, + "eval_loss": 1.2351411581039429, + "eval_rouge1_fmeasure": 0.46283605694770813, + "eval_rouge1_precision": 0.46500271558761597, + "eval_rouge1_recall": 0.4642695188522339, + "eval_rouge2_fmeasure": 0.09591429680585861, + "eval_rouge2_precision": 0.09872540086507797, + "eval_rouge2_recall": 0.09339157491922379, + "eval_rougeL_fmeasure": 0.46226710081100464, + "eval_rougeL_precision": 0.4643747806549072, + "eval_rougeL_recall": 0.46372970938682556, + "eval_rougeLsum_fmeasure": 0.46237149834632874, + "eval_rougeLsum_precision": 0.464464008808136, + "eval_rougeLsum_recall": 0.46384570002555847, + "eval_runtime": 678.572, + "eval_samples_per_second": 5.158, + "eval_steps_per_second": 0.215, + "step": 4300 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.37514285714285717, + "epoch": 0.82, + "step": 4300 + }, + { + "epoch": 0.84, + "eval_accuracy": 0.3802857142857143, + "eval_avg_bleuScore": 0.08860891361321722, + "eval_loss": 1.2320390939712524, + "eval_rouge1_fmeasure": 0.4631531834602356, + "eval_rouge1_precision": 0.4639800488948822, + "eval_rouge1_recall": 0.46552175283432007, + "eval_rouge2_fmeasure": 0.0995183140039444, + "eval_rouge2_precision": 0.10100917518138885, + "eval_rouge2_recall": 0.0982545018196106, + "eval_rougeL_fmeasure": 0.4624533951282501, + "eval_rougeL_precision": 0.46323657035827637, + "eval_rougeL_recall": 0.4648307263851166, + "eval_rougeLsum_fmeasure": 0.4626549482345581, + "eval_rougeLsum_precision": 0.46342751383781433, + "eval_rougeLsum_recall": 0.46504610776901245, + "eval_runtime": 685.9681, + "eval_samples_per_second": 5.102, + "eval_steps_per_second": 0.213, + "step": 4400 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.3802857142857143, + "epoch": 0.84, + "step": 4400 + }, + { + "epoch": 0.86, + "learning_rate": 0.0001, + "loss": 1.2246, + "step": 4500 + }, + { + "epoch": 0.86, + "eval_accuracy": 0.37257142857142855, + "eval_avg_bleuScore": 0.08456097078323364, + "eval_loss": 1.2283884286880493, + "eval_rouge1_fmeasure": 0.4615192115306854, + "eval_rouge1_precision": 0.4634266495704651, + "eval_rouge1_recall": 0.46272367238998413, + "eval_rouge2_fmeasure": 0.09663666784763336, + "eval_rouge2_precision": 0.0992717370390892, + "eval_rouge2_recall": 0.09429265558719635, + "eval_rougeL_fmeasure": 0.46088704466819763, + "eval_rougeL_precision": 0.4627465605735779, + "eval_rougeL_recall": 0.4620998203754425, + "eval_rougeLsum_fmeasure": 0.46105268597602844, + "eval_rougeLsum_precision": 0.4629073441028595, + "eval_rougeLsum_recall": 0.46227604150772095, + "eval_runtime": 681.4522, + "eval_samples_per_second": 5.136, + "eval_steps_per_second": 0.214, + "step": 4500 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.3802857142857143, + "epoch": 0.86, + "step": 4500 + }, + { + "epoch": 0.88, + "eval_accuracy": 0.37285714285714283, + "eval_avg_bleuScore": 0.08733673360518046, + "eval_loss": 1.2267086505889893, + "eval_rouge1_fmeasure": 0.4655407965183258, + "eval_rouge1_precision": 0.4668717086315155, + "eval_rouge1_recall": 0.46693241596221924, + "eval_rouge2_fmeasure": 0.09869344532489777, + "eval_rouge2_precision": 0.10078585147857666, + "eval_rouge2_recall": 0.09682551771402359, + "eval_rougeL_fmeasure": 0.46528565883636475, + "eval_rougeL_precision": 0.4666202664375305, + "eval_rougeL_recall": 0.4666643738746643, + "eval_rougeLsum_fmeasure": 0.46536117792129517, + "eval_rougeLsum_precision": 0.4666873812675476, + "eval_rougeLsum_recall": 0.4667511582374573, + "eval_runtime": 677.5144, + "eval_samples_per_second": 5.166, + "eval_steps_per_second": 0.215, + "step": 4600 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.3802857142857143, + "epoch": 0.88, + "step": 4600 + }, + { + "epoch": 0.9, + "eval_accuracy": 0.3811428571428571, + "eval_avg_bleuScore": 0.09408873009681702, + "eval_loss": 1.2224647998809814, + "eval_rouge1_fmeasure": 0.4698290228843689, + "eval_rouge1_precision": 0.4714408814907074, + "eval_rouge1_recall": 0.4712493121623993, + "eval_rouge2_fmeasure": 0.1037345677614212, + "eval_rouge2_precision": 0.1054706871509552, + "eval_rouge2_recall": 0.10228389501571655, + "eval_rougeL_fmeasure": 0.46944233775138855, + "eval_rougeL_precision": 0.4710277020931244, + "eval_rougeL_recall": 0.47087541222572327, + "eval_rougeLsum_fmeasure": 0.4694858193397522, + "eval_rougeLsum_precision": 0.47107043862342834, + "eval_rougeLsum_recall": 0.4709205627441406, + "eval_runtime": 693.1449, + "eval_samples_per_second": 5.049, + "eval_steps_per_second": 0.211, + "step": 4700 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.3811428571428571, + "epoch": 0.9, + "step": 4700 + }, + { + "epoch": 0.92, + "eval_accuracy": 0.3802857142857143, + "eval_avg_bleuScore": 0.08957303579364505, + "eval_loss": 1.2195558547973633, + "eval_rouge1_fmeasure": 0.4704902470111847, + "eval_rouge1_precision": 0.47286683320999146, + "eval_rouge1_recall": 0.47120949625968933, + "eval_rouge2_fmeasure": 0.10144046694040298, + "eval_rouge2_precision": 0.10431737452745438, + "eval_rouge2_recall": 0.09885827451944351, + "eval_rougeL_fmeasure": 0.47009220719337463, + "eval_rougeL_precision": 0.47242221236228943, + "eval_rougeL_recall": 0.4708355963230133, + "eval_rougeLsum_fmeasure": 0.47012150287628174, + "eval_rougeLsum_precision": 0.4724501967430115, + "eval_rougeLsum_recall": 0.4708665609359741, + "eval_runtime": 709.2326, + "eval_samples_per_second": 4.935, + "eval_steps_per_second": 0.206, + "step": 4800 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.3811428571428571, + "epoch": 0.92, + "step": 4800 + }, + { + "epoch": 0.93, + "eval_accuracy": 0.38, + "eval_avg_bleuScore": 0.08428653080974306, + "eval_loss": 1.2192410230636597, + "eval_rouge1_fmeasure": 0.47025272250175476, + "eval_rouge1_precision": 0.4731932580471039, + "eval_rouge1_recall": 0.46974682807922363, + "eval_rouge2_fmeasure": 0.09642089903354645, + "eval_rouge2_precision": 0.09943059086799622, + "eval_rouge2_recall": 0.09372298419475555, + "eval_rougeL_fmeasure": 0.4698809087276459, + "eval_rougeL_precision": 0.4727679193019867, + "eval_rougeL_recall": 0.46940457820892334, + "eval_rougeLsum_fmeasure": 0.4698815941810608, + "eval_rougeLsum_precision": 0.4727679193019867, + "eval_rougeLsum_recall": 0.4694061577320099, + "eval_runtime": 708.23, + "eval_samples_per_second": 4.942, + "eval_steps_per_second": 0.206, + "step": 4900 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.3811428571428571, + "epoch": 0.93, + "step": 4900 + }, + { + "epoch": 0.95, + "learning_rate": 0.0001, + "loss": 1.2194, + "step": 5000 + }, + { + "epoch": 0.95, + "eval_accuracy": 0.3822857142857143, + "eval_avg_bleuScore": 0.08732458365815027, + "eval_loss": 1.2133744955062866, + "eval_rouge1_fmeasure": 0.473276823759079, + "eval_rouge1_precision": 0.4759586453437805, + "eval_rouge1_recall": 0.47319260239601135, + "eval_rouge2_fmeasure": 0.09946362674236298, + "eval_rouge2_precision": 0.10251562297344208, + "eval_rouge2_recall": 0.09668624401092529, + "eval_rougeL_fmeasure": 0.4728994071483612, + "eval_rougeL_precision": 0.4755318760871887, + "eval_rougeL_recall": 0.4728360176086426, + "eval_rougeLsum_fmeasure": 0.4729423522949219, + "eval_rougeLsum_precision": 0.4755731523036957, + "eval_rougeLsum_recall": 0.47288116812705994, + "eval_runtime": 689.8699, + "eval_samples_per_second": 5.073, + "eval_steps_per_second": 0.212, + "step": 5000 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.3822857142857143, + "epoch": 0.95, + "step": 5000 + }, + { + "epoch": 0.97, + "eval_accuracy": 0.384, + "eval_avg_bleuScore": 0.08920761404718672, + "eval_loss": 1.2110172510147095, + "eval_rouge1_fmeasure": 0.47299057245254517, + "eval_rouge1_precision": 0.4750961661338806, + "eval_rouge1_recall": 0.4734393060207367, + "eval_rouge2_fmeasure": 0.10076272487640381, + "eval_rouge2_precision": 0.10314268618822098, + "eval_rouge2_recall": 0.09872191399335861, + "eval_rougeL_fmeasure": 0.4725087881088257, + "eval_rougeL_precision": 0.4745262861251831, + "eval_rougeL_recall": 0.47299158573150635, + "eval_rougeLsum_fmeasure": 0.47250983119010925, + "eval_rougeLsum_precision": 0.47452741861343384, + "eval_rougeLsum_recall": 0.4729924201965332, + "eval_runtime": 710.2747, + "eval_samples_per_second": 4.928, + "eval_steps_per_second": 0.206, + "step": 5100 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.384, + "epoch": 0.97, + "step": 5100 + }, + { + "epoch": 0.99, + "eval_accuracy": 0.38085714285714284, + "eval_avg_bleuScore": 0.08709148204326629, + "eval_loss": 1.206809163093567, + "eval_rouge1_fmeasure": 0.47192901372909546, + "eval_rouge1_precision": 0.4742652177810669, + "eval_rouge1_recall": 0.47203734517097473, + "eval_rouge2_fmeasure": 0.09854426980018616, + "eval_rouge2_precision": 0.10090313106775284, + "eval_rouge2_recall": 0.09638763964176178, + "eval_rougeL_fmeasure": 0.4714796543121338, + "eval_rougeL_precision": 0.47378063201904297, + "eval_rougeL_recall": 0.4716048836708069, + "eval_rougeLsum_fmeasure": 0.4715076684951782, + "eval_rougeLsum_precision": 0.4738079011440277, + "eval_rougeLsum_recall": 0.4716342091560364, + "eval_runtime": 719.374, + "eval_samples_per_second": 4.865, + "eval_steps_per_second": 0.203, + "step": 5200 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.384, + "epoch": 0.99, + "step": 5200 + }, + { + "epoch": 1.01, + "eval_accuracy": 0.38285714285714284, + "eval_avg_bleuScore": 0.09228691346304757, + "eval_loss": 1.2042845487594604, + "eval_rouge1_fmeasure": 0.47944971919059753, + "eval_rouge1_precision": 0.48242396116256714, + "eval_rouge1_recall": 0.47879064083099365, + "eval_rouge2_fmeasure": 0.10499687492847443, + "eval_rouge2_precision": 0.10838944464921951, + "eval_rouge2_recall": 0.10190355032682419, + "eval_rougeL_fmeasure": 0.4793473780155182, + "eval_rougeL_precision": 0.4823038876056671, + "eval_rougeL_recall": 0.4787004292011261, + "eval_rougeLsum_fmeasure": 0.4793815016746521, + "eval_rougeLsum_precision": 0.4823433458805084, + "eval_rougeLsum_recall": 0.4787304997444153, + "eval_runtime": 713.4678, + "eval_samples_per_second": 4.906, + "eval_steps_per_second": 0.205, + "step": 5300 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.384, + "epoch": 1.01, + "step": 5300 + }, + { + "epoch": 1.03, + "eval_accuracy": 0.3822857142857143, + "eval_avg_bleuScore": 0.08736944925785065, + "eval_loss": 1.2020999193191528, + "eval_rouge1_fmeasure": 0.4769892692565918, + "eval_rouge1_precision": 0.4802291691303253, + "eval_rouge1_recall": 0.47634607553482056, + "eval_rouge2_fmeasure": 0.10028692334890366, + "eval_rouge2_precision": 0.10376148670911789, + "eval_rouge2_recall": 0.09721053391695023, + "eval_rougeL_fmeasure": 0.4766009449958801, + "eval_rougeL_precision": 0.47980427742004395, + "eval_rougeL_recall": 0.47597748041152954, + "eval_rougeLsum_fmeasure": 0.476613849401474, + "eval_rougeLsum_precision": 0.4798152446746826, + "eval_rougeLsum_recall": 0.4759925305843353, + "eval_runtime": 699.2903, + "eval_samples_per_second": 5.005, + "eval_steps_per_second": 0.209, + "step": 5400 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.384, + "epoch": 1.03, + "step": 5400 + }, + { + "epoch": 1.05, + "learning_rate": 0.0001, + "loss": 1.1943, + "step": 5500 + }, + { + "epoch": 1.05, + "eval_accuracy": 0.384, + "eval_avg_bleuScore": 0.08876867526769638, + "eval_loss": 1.1995388269424438, + "eval_rouge1_fmeasure": 0.4784497320652008, + "eval_rouge1_precision": 0.4813612997531891, + "eval_rouge1_recall": 0.4777953326702118, + "eval_rouge2_fmeasure": 0.10139812529087067, + "eval_rouge2_precision": 0.10456015169620514, + "eval_rouge2_recall": 0.09853238612413406, + "eval_rougeL_fmeasure": 0.4780391454696655, + "eval_rougeL_precision": 0.48090261220932007, + "eval_rougeL_recall": 0.47740626335144043, + "eval_rougeLsum_fmeasure": 0.47806599736213684, + "eval_rougeLsum_precision": 0.4809248447418213, + "eval_rougeLsum_recall": 0.47743871808052063, + "eval_runtime": 695.0658, + "eval_samples_per_second": 5.035, + "eval_steps_per_second": 0.21, + "step": 5500 + }, + { + "best_epoch": 0, + "best_eval_accuracy": 0.384, + "epoch": 1.05, + "step": 5500 + }, + { + "epoch": 1.07, + "eval_accuracy": 0.38542857142857145, + "eval_avg_bleuScore": 0.0921421125446047, + "eval_loss": 1.196988821029663, + "eval_rouge1_fmeasure": 0.4810657799243927, + "eval_rouge1_precision": 0.48325568437576294, + "eval_rouge1_recall": 0.4812372922897339, + "eval_rouge2_fmeasure": 0.10458412021398544, + "eval_rouge2_precision": 0.1072564348578453, + "eval_rouge2_recall": 0.10226050764322281, + "eval_rougeL_fmeasure": 0.48091500997543335, + "eval_rougeL_precision": 0.4830835163593292, + "eval_rougeL_recall": 0.4811002314090729, + "eval_rougeLsum_fmeasure": 0.48094817996025085, + "eval_rougeLsum_precision": 0.4831198453903198, + "eval_rougeLsum_recall": 0.48113110661506653, + "eval_runtime": 701.6728, + "eval_samples_per_second": 4.988, + "eval_steps_per_second": 0.208, + "step": 5600 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.38542857142857145, + "epoch": 1.07, + "step": 5600 + }, + { + "epoch": 1.09, + "eval_accuracy": 0.38571428571428573, + "eval_avg_bleuScore": 0.10434803967816489, + "eval_loss": 1.1926367282867432, + "eval_rouge1_fmeasure": 0.48613592982292175, + "eval_rouge1_precision": 0.4863094091415405, + "eval_rouge1_recall": 0.4882674515247345, + "eval_rouge2_fmeasure": 0.1133088693022728, + "eval_rouge2_precision": 0.11394620686769485, + "eval_rouge2_recall": 0.1128883957862854, + "eval_rougeL_fmeasure": 0.48604506254196167, + "eval_rougeL_precision": 0.4862136244773865, + "eval_rougeL_recall": 0.4881778955459595, + "eval_rougeLsum_fmeasure": 0.48604506254196167, + "eval_rougeLsum_precision": 0.4862136244773865, + "eval_rougeLsum_recall": 0.4881778955459595, + "eval_runtime": 700.0051, + "eval_samples_per_second": 5.0, + "eval_steps_per_second": 0.209, + "step": 5700 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.38571428571428573, + "epoch": 1.09, + "step": 5700 + }, + { + "epoch": 1.11, + "eval_accuracy": 0.39085714285714285, + "eval_avg_bleuScore": 0.09774087310688837, + "eval_loss": 1.192405104637146, + "eval_rouge1_fmeasure": 0.4822241961956024, + "eval_rouge1_precision": 0.48336881399154663, + "eval_rouge1_recall": 0.4831463098526001, + "eval_rouge2_fmeasure": 0.10871667414903641, + "eval_rouge2_precision": 0.11029361188411713, + "eval_rouge2_recall": 0.10732249170541763, + "eval_rougeL_fmeasure": 0.48192843794822693, + "eval_rougeL_precision": 0.48306867480278015, + "eval_rougeL_recall": 0.48284775018692017, + "eval_rougeLsum_fmeasure": 0.48195597529411316, + "eval_rougeLsum_precision": 0.48309409618377686, + "eval_rougeLsum_recall": 0.4828778803348541, + "eval_runtime": 703.0299, + "eval_samples_per_second": 4.978, + "eval_steps_per_second": 0.208, + "step": 5800 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.39085714285714285, + "epoch": 1.11, + "step": 5800 + }, + { + "epoch": 1.13, + "eval_accuracy": 0.38771428571428573, + "eval_avg_bleuScore": 0.09932618699754987, + "eval_loss": 1.1922338008880615, + "eval_rouge1_fmeasure": 0.4852292239665985, + "eval_rouge1_precision": 0.48615825176239014, + "eval_rouge1_recall": 0.4863511323928833, + "eval_rouge2_fmeasure": 0.1089785248041153, + "eval_rouge2_precision": 0.11030538380146027, + "eval_rouge2_recall": 0.10779362916946411, + "eval_rougeL_fmeasure": 0.4850394129753113, + "eval_rougeL_precision": 0.48594945669174194, + "eval_rougeL_recall": 0.48617222905158997, + "eval_rougeLsum_fmeasure": 0.4850572645664215, + "eval_rougeLsum_precision": 0.4859714210033417, + "eval_rougeLsum_recall": 0.48618724942207336, + "eval_runtime": 717.9475, + "eval_samples_per_second": 4.875, + "eval_steps_per_second": 0.203, + "step": 5900 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.39085714285714285, + "epoch": 1.13, + "step": 5900 + }, + { + "epoch": 1.14, + "learning_rate": 0.0001, + "loss": 1.1829, + "step": 6000 + }, + { + "epoch": 1.14, + "eval_accuracy": 0.3882857142857143, + "eval_avg_bleuScore": 0.09516796236378806, + "eval_loss": 1.1876760721206665, + "eval_rouge1_fmeasure": 0.4868794083595276, + "eval_rouge1_precision": 0.48939061164855957, + "eval_rouge1_recall": 0.48638010025024414, + "eval_rouge2_fmeasure": 0.1067165732383728, + "eval_rouge2_precision": 0.10926699638366699, + "eval_rouge2_recall": 0.104483962059021, + "eval_rougeL_fmeasure": 0.48677870631217957, + "eval_rougeL_precision": 0.4892902374267578, + "eval_rougeL_recall": 0.486273854970932, + "eval_rougeLsum_fmeasure": 0.48683223128318787, + "eval_rougeLsum_precision": 0.48933741450309753, + "eval_rougeLsum_recall": 0.48633575439453125, + "eval_runtime": 701.5427, + "eval_samples_per_second": 4.989, + "eval_steps_per_second": 0.208, + "step": 6000 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.39085714285714285, + "epoch": 1.14, + "step": 6000 + }, + { + "epoch": 1.16, + "eval_accuracy": 0.38742857142857146, + "eval_avg_bleuScore": 0.09795815995761327, + "eval_loss": 1.1858559846878052, + "eval_rouge1_fmeasure": 0.4874568283557892, + "eval_rouge1_precision": 0.4888821244239807, + "eval_rouge1_recall": 0.4880567193031311, + "eval_rouge2_fmeasure": 0.10779689252376556, + "eval_rouge2_precision": 0.1093817800283432, + "eval_rouge2_recall": 0.10647542774677277, + "eval_rougeL_fmeasure": 0.48734065890312195, + "eval_rougeL_precision": 0.48874735832214355, + "eval_rougeL_recall": 0.48795217275619507, + "eval_rougeLsum_fmeasure": 0.48734065890312195, + "eval_rougeLsum_precision": 0.48874735832214355, + "eval_rougeLsum_recall": 0.48795217275619507, + "eval_runtime": 702.3435, + "eval_samples_per_second": 4.983, + "eval_steps_per_second": 0.208, + "step": 6100 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.39085714285714285, + "epoch": 1.16, + "step": 6100 + }, + { + "epoch": 1.18, + "eval_accuracy": 0.382, + "eval_avg_bleuScore": 0.09201604807376862, + "eval_loss": 1.18765127658844, + "eval_rouge1_fmeasure": 0.48146504163742065, + "eval_rouge1_precision": 0.4835813343524933, + "eval_rouge1_recall": 0.4810131788253784, + "eval_rouge2_fmeasure": 0.10354610532522202, + "eval_rouge2_precision": 0.10569047182798386, + "eval_rouge2_recall": 0.10160320997238159, + "eval_rougeL_fmeasure": 0.4813587963581085, + "eval_rougeL_precision": 0.48344969749450684, + "eval_rougeL_recall": 0.48092371225357056, + "eval_rougeLsum_fmeasure": 0.4813587963581085, + "eval_rougeLsum_precision": 0.48344969749450684, + "eval_rougeLsum_recall": 0.48092371225357056, + "eval_runtime": 698.2581, + "eval_samples_per_second": 5.012, + "eval_steps_per_second": 0.209, + "step": 6200 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.39085714285714285, + "epoch": 1.18, + "step": 6200 + }, + { + "epoch": 1.2, + "eval_accuracy": 0.38885714285714285, + "eval_avg_bleuScore": 0.09650462246792657, + "eval_loss": 1.183226227760315, + "eval_rouge1_fmeasure": 0.4896017014980316, + "eval_rouge1_precision": 0.4917184114456177, + "eval_rouge1_recall": 0.48921307921409607, + "eval_rouge2_fmeasure": 0.10720385611057281, + "eval_rouge2_precision": 0.1093168631196022, + "eval_rouge2_recall": 0.10527023673057556, + "eval_rougeL_fmeasure": 0.4895344376564026, + "eval_rougeL_precision": 0.4916393458843231, + "eval_rougeL_recall": 0.4891536831855774, + "eval_rougeLsum_fmeasure": 0.4895344376564026, + "eval_rougeLsum_precision": 0.4916393458843231, + "eval_rougeLsum_recall": 0.4891536831855774, + "eval_runtime": 689.3635, + "eval_samples_per_second": 5.077, + "eval_steps_per_second": 0.212, + "step": 6300 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.39085714285714285, + "epoch": 1.2, + "step": 6300 + }, + { + "epoch": 1.22, + "eval_accuracy": 0.38657142857142857, + "eval_avg_bleuScore": 0.09086202553766114, + "eval_loss": 1.1810081005096436, + "eval_rouge1_fmeasure": 0.48292866349220276, + "eval_rouge1_precision": 0.4856628477573395, + "eval_rouge1_recall": 0.48179009556770325, + "eval_rouge2_fmeasure": 0.1026853546500206, + "eval_rouge2_precision": 0.10534290224313736, + "eval_rouge2_recall": 0.10025645792484283, + "eval_rougeL_fmeasure": 0.48265212774276733, + "eval_rougeL_precision": 0.48533371090888977, + "eval_rougeL_recall": 0.4815382659435272, + "eval_rougeLsum_fmeasure": 0.48265212774276733, + "eval_rougeLsum_precision": 0.48533371090888977, + "eval_rougeLsum_recall": 0.4815382659435272, + "eval_runtime": 684.3289, + "eval_samples_per_second": 5.114, + "eval_steps_per_second": 0.213, + "step": 6400 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.39085714285714285, + "epoch": 1.22, + "step": 6400 + }, + { + "epoch": 1.24, + "learning_rate": 0.0001, + "loss": 1.1645, + "step": 6500 + }, + { + "epoch": 1.24, + "eval_accuracy": 0.394, + "eval_avg_bleuScore": 0.09758192454491343, + "eval_loss": 1.1781606674194336, + "eval_rouge1_fmeasure": 0.4877890944480896, + "eval_rouge1_precision": 0.4900844097137451, + "eval_rouge1_recall": 0.4870730936527252, + "eval_rouge2_fmeasure": 0.10900319367647171, + "eval_rouge2_precision": 0.11137502640485764, + "eval_rouge2_recall": 0.10684141516685486, + "eval_rougeL_fmeasure": 0.4875901937484741, + "eval_rougeL_precision": 0.489857017993927, + "eval_rougeL_recall": 0.4868933856487274, + "eval_rougeLsum_fmeasure": 0.48765110969543457, + "eval_rougeLsum_precision": 0.48991936445236206, + "eval_rougeLsum_recall": 0.48695358633995056, + "eval_runtime": 711.2919, + "eval_samples_per_second": 4.921, + "eval_steps_per_second": 0.205, + "step": 6500 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.394, + "epoch": 1.24, + "step": 6500 + }, + { + "epoch": 1.26, + "eval_accuracy": 0.39057142857142857, + "eval_avg_bleuScore": 0.09796631112268993, + "eval_loss": 1.177560567855835, + "eval_rouge1_fmeasure": 0.4866205155849457, + "eval_rouge1_precision": 0.489122211933136, + "eval_rouge1_recall": 0.4860360622406006, + "eval_rouge2_fmeasure": 0.10959078371524811, + "eval_rouge2_precision": 0.11217135936021805, + "eval_rouge2_recall": 0.10722710192203522, + "eval_rougeL_fmeasure": 0.48647359013557434, + "eval_rougeL_precision": 0.4889563322067261, + "eval_rougeL_recall": 0.48590072989463806, + "eval_rougeLsum_fmeasure": 0.4865030348300934, + "eval_rougeLsum_precision": 0.4889856278896332, + "eval_rougeLsum_recall": 0.48593080043792725, + "eval_runtime": 681.3501, + "eval_samples_per_second": 5.137, + "eval_steps_per_second": 0.214, + "step": 6600 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.394, + "epoch": 1.26, + "step": 6600 + }, + { + "epoch": 1.28, + "eval_accuracy": 0.3914285714285714, + "eval_avg_bleuScore": 0.0928525439798832, + "eval_loss": 1.175042986869812, + "eval_rouge1_fmeasure": 0.4890231490135193, + "eval_rouge1_precision": 0.4924193024635315, + "eval_rouge1_recall": 0.48778238892555237, + "eval_rouge2_fmeasure": 0.10552280396223068, + "eval_rouge2_precision": 0.1088574156165123, + "eval_rouge2_recall": 0.10246119648218155, + "eval_rougeL_fmeasure": 0.48884162306785583, + "eval_rougeL_precision": 0.4922114908695221, + "eval_rougeL_recall": 0.48761773109436035, + "eval_rougeLsum_fmeasure": 0.4888552129268646, + "eval_rougeLsum_precision": 0.49222391843795776, + "eval_rougeLsum_recall": 0.48763278126716614, + "eval_runtime": 678.0154, + "eval_samples_per_second": 5.162, + "eval_steps_per_second": 0.215, + "step": 6700 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.394, + "epoch": 1.28, + "step": 6700 + }, + { + "epoch": 1.3, + "eval_accuracy": 0.3894285714285714, + "eval_avg_bleuScore": 0.09422704415661948, + "eval_loss": 1.1733677387237549, + "eval_rouge1_fmeasure": 0.48861464858055115, + "eval_rouge1_precision": 0.4916938543319702, + "eval_rouge1_recall": 0.4870302081108093, + "eval_rouge2_fmeasure": 0.10575228184461594, + "eval_rouge2_precision": 0.10856050997972488, + "eval_rouge2_recall": 0.10320048034191132, + "eval_rougeL_fmeasure": 0.4884510934352875, + "eval_rougeL_precision": 0.4915034770965576, + "eval_rougeL_recall": 0.48688212037086487, + "eval_rougeLsum_fmeasure": 0.4884510934352875, + "eval_rougeLsum_precision": 0.4915034770965576, + "eval_rougeLsum_recall": 0.48688212037086487, + "eval_runtime": 669.9302, + "eval_samples_per_second": 5.224, + "eval_steps_per_second": 0.218, + "step": 6800 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.394, + "epoch": 1.3, + "step": 6800 + }, + { + "epoch": 1.32, + "eval_accuracy": 0.3862857142857143, + "eval_avg_bleuScore": 0.09501611430304391, + "eval_loss": 1.1741470098495483, + "eval_rouge1_fmeasure": 0.48634904623031616, + "eval_rouge1_precision": 0.4894244968891144, + "eval_rouge1_recall": 0.48504891991615295, + "eval_rouge2_fmeasure": 0.10691665858030319, + "eval_rouge2_precision": 0.10964075475931168, + "eval_rouge2_recall": 0.10442051291465759, + "eval_rougeL_fmeasure": 0.48623180389404297, + "eval_rougeL_precision": 0.48928794264793396, + "eval_rougeL_recall": 0.4849444329738617, + "eval_rougeLsum_fmeasure": 0.48623180389404297, + "eval_rougeLsum_precision": 0.48928794264793396, + "eval_rougeLsum_recall": 0.4849444329738617, + "eval_runtime": 678.5521, + "eval_samples_per_second": 5.158, + "eval_steps_per_second": 0.215, + "step": 6900 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.394, + "epoch": 1.32, + "step": 6900 + }, + { + "epoch": 1.34, + "learning_rate": 0.0001, + "loss": 1.1708, + "step": 7000 + }, + { + "epoch": 1.34, + "eval_accuracy": 0.38571428571428573, + "eval_avg_bleuScore": 0.08925226952348436, + "eval_loss": 1.1751573085784912, + "eval_rouge1_fmeasure": 0.4845730662345886, + "eval_rouge1_precision": 0.48776912689208984, + "eval_rouge1_recall": 0.48277971148490906, + "eval_rouge2_fmeasure": 0.10183030366897583, + "eval_rouge2_precision": 0.10492262244224548, + "eval_rouge2_recall": 0.0989968404173851, + "eval_rougeL_fmeasure": 0.484536737203598, + "eval_rougeL_precision": 0.4877227246761322, + "eval_rougeL_recall": 0.4827496409416199, + "eval_rougeLsum_fmeasure": 0.484536737203598, + "eval_rougeLsum_precision": 0.4877227246761322, + "eval_rougeLsum_recall": 0.4827496409416199, + "eval_runtime": 675.9775, + "eval_samples_per_second": 5.178, + "eval_steps_per_second": 0.216, + "step": 7000 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.394, + "epoch": 1.34, + "step": 7000 + }, + { + "epoch": 1.35, + "eval_accuracy": 0.38742857142857146, + "eval_avg_bleuScore": 0.08925152574266706, + "eval_loss": 1.1729258298873901, + "eval_rouge1_fmeasure": 0.48528826236724854, + "eval_rouge1_precision": 0.4884517192840576, + "eval_rouge1_recall": 0.4838833808898926, + "eval_rouge2_fmeasure": 0.10141944885253906, + "eval_rouge2_precision": 0.1043931171298027, + "eval_rouge2_recall": 0.09871000796556473, + "eval_rougeL_fmeasure": 0.48515835404396057, + "eval_rougeL_precision": 0.48830461502075195, + "eval_rougeL_recall": 0.4837645888328552, + "eval_rougeLsum_fmeasure": 0.48515835404396057, + "eval_rougeLsum_precision": 0.48830461502075195, + "eval_rougeLsum_recall": 0.4837645888328552, + "eval_runtime": 681.0713, + "eval_samples_per_second": 5.139, + "eval_steps_per_second": 0.214, + "step": 7100 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.394, + "epoch": 1.35, + "step": 7100 + }, + { + "epoch": 1.37, + "eval_accuracy": 0.386, + "eval_avg_bleuScore": 0.09471559107303619, + "eval_loss": 1.1688710451126099, + "eval_rouge1_fmeasure": 0.4882533848285675, + "eval_rouge1_precision": 0.489577054977417, + "eval_rouge1_recall": 0.48825153708457947, + "eval_rouge2_fmeasure": 0.10504159331321716, + "eval_rouge2_precision": 0.1062684953212738, + "eval_rouge2_recall": 0.10397736728191376, + "eval_rougeL_fmeasure": 0.4881993234157562, + "eval_rougeL_precision": 0.48950305581092834, + "eval_rougeL_recall": 0.48820725083351135, + "eval_rougeLsum_fmeasure": 0.4881993234157562, + "eval_rougeLsum_precision": 0.48950305581092834, + "eval_rougeLsum_recall": 0.48820725083351135, + "eval_runtime": 683.2016, + "eval_samples_per_second": 5.123, + "eval_steps_per_second": 0.214, + "step": 7200 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.394, + "epoch": 1.37, + "step": 7200 + }, + { + "epoch": 1.39, + "eval_accuracy": 0.3891428571428571, + "eval_avg_bleuScore": 0.0915304719039372, + "eval_loss": 1.164903998374939, + "eval_rouge1_fmeasure": 0.48901432752609253, + "eval_rouge1_precision": 0.4904957413673401, + "eval_rouge1_recall": 0.489198237657547, + "eval_rouge2_fmeasure": 0.10308346897363663, + "eval_rouge2_precision": 0.10479623824357986, + "eval_rouge2_recall": 0.10155932605266571, + "eval_rougeL_fmeasure": 0.48899751901626587, + "eval_rougeL_precision": 0.49047669768333435, + "eval_rougeL_recall": 0.4891831874847412, + "eval_rougeLsum_fmeasure": 0.48899751901626587, + "eval_rougeLsum_precision": 0.49047669768333435, + "eval_rougeLsum_recall": 0.4891831874847412, + "eval_runtime": 678.1152, + "eval_samples_per_second": 5.161, + "eval_steps_per_second": 0.215, + "step": 7300 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.394, + "epoch": 1.39, + "step": 7300 + }, + { + "epoch": 1.41, + "eval_accuracy": 0.3994285714285714, + "eval_avg_bleuScore": 0.10238692969935281, + "eval_loss": 1.1616579294204712, + "eval_rouge1_fmeasure": 0.49463626742362976, + "eval_rouge1_precision": 0.4962061643600464, + "eval_rouge1_recall": 0.4944556653499603, + "eval_rouge2_fmeasure": 0.11282491683959961, + "eval_rouge2_precision": 0.11456984281539917, + "eval_rouge2_recall": 0.1112387403845787, + "eval_rougeL_fmeasure": 0.4945405125617981, + "eval_rougeL_precision": 0.49610191583633423, + "eval_rougeL_recall": 0.4943661689758301, + "eval_rougeLsum_fmeasure": 0.4945405125617981, + "eval_rougeLsum_precision": 0.49610191583633423, + "eval_rougeLsum_recall": 0.4943661689758301, + "eval_runtime": 669.0119, + "eval_samples_per_second": 5.232, + "eval_steps_per_second": 0.218, + "step": 7400 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.3994285714285714, + "epoch": 1.41, + "step": 7400 + }, + { + "epoch": 1.43, + "learning_rate": 0.0001, + "loss": 1.1764, + "step": 7500 + }, + { + "epoch": 1.43, + "eval_accuracy": 0.39714285714285713, + "eval_avg_bleuScore": 0.09719528065408979, + "eval_loss": 1.1652673482894897, + "eval_rouge1_fmeasure": 0.49130886793136597, + "eval_rouge1_precision": 0.49369290471076965, + "eval_rouge1_recall": 0.49046799540519714, + "eval_rouge2_fmeasure": 0.10847558826208115, + "eval_rouge2_precision": 0.11078434437513351, + "eval_rouge2_recall": 0.10637179017066956, + "eval_rougeL_fmeasure": 0.4912373721599579, + "eval_rougeL_precision": 0.4935903251171112, + "eval_rougeL_recall": 0.4904085397720337, + "eval_rougeLsum_fmeasure": 0.4912373721599579, + "eval_rougeLsum_precision": 0.4935903251171112, + "eval_rougeLsum_recall": 0.4904085397720337, + "eval_runtime": 681.5502, + "eval_samples_per_second": 5.135, + "eval_steps_per_second": 0.214, + "step": 7500 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.3994285714285714, + "epoch": 1.43, + "step": 7500 + }, + { + "epoch": 1.45, + "eval_accuracy": 0.3911428571428571, + "eval_avg_bleuScore": 0.0990878015416009, + "eval_loss": 1.1646941900253296, + "eval_rouge1_fmeasure": 0.48855194449424744, + "eval_rouge1_precision": 0.48997461795806885, + "eval_rouge1_recall": 0.4884979724884033, + "eval_rouge2_fmeasure": 0.10910660028457642, + "eval_rouge2_precision": 0.11063293367624283, + "eval_rouge2_recall": 0.1077328771352768, + "eval_rougeL_fmeasure": 0.48841577768325806, + "eval_rougeL_precision": 0.4898097813129425, + "eval_rougeL_recall": 0.48837992548942566, + "eval_rougeLsum_fmeasure": 0.48841577768325806, + "eval_rougeLsum_precision": 0.4898097813129425, + "eval_rougeLsum_recall": 0.48837992548942566, + "eval_runtime": 675.0303, + "eval_samples_per_second": 5.185, + "eval_steps_per_second": 0.216, + "step": 7600 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.3994285714285714, + "epoch": 1.45, + "step": 7600 + }, + { + "epoch": 1.47, + "eval_accuracy": 0.386, + "eval_avg_bleuScore": 0.08874689975806645, + "eval_loss": 1.1637684106826782, + "eval_rouge1_fmeasure": 0.4841983914375305, + "eval_rouge1_precision": 0.4875686764717102, + "eval_rouge1_recall": 0.48199939727783203, + "eval_rouge2_fmeasure": 0.10131097584962845, + "eval_rouge2_precision": 0.10435326397418976, + "eval_rouge2_recall": 0.09852544218301773, + "eval_rougeL_fmeasure": 0.48416340351104736, + "eval_rougeL_precision": 0.4875248670578003, + "eval_rougeL_recall": 0.4819701015949249, + "eval_rougeLsum_fmeasure": 0.48416340351104736, + "eval_rougeLsum_precision": 0.4875248670578003, + "eval_rougeLsum_recall": 0.4819701015949249, + "eval_runtime": 673.2048, + "eval_samples_per_second": 5.199, + "eval_steps_per_second": 0.217, + "step": 7700 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.3994285714285714, + "epoch": 1.47, + "step": 7700 + }, + { + "epoch": 1.49, + "eval_accuracy": 0.3891428571428571, + "eval_avg_bleuScore": 0.09112303535427366, + "eval_loss": 1.1632740497589111, + "eval_rouge1_fmeasure": 0.48752155900001526, + "eval_rouge1_precision": 0.4900144636631012, + "eval_rouge1_recall": 0.48648855090141296, + "eval_rouge2_fmeasure": 0.10284367948770523, + "eval_rouge2_precision": 0.10531271249055862, + "eval_rouge2_recall": 0.10058772563934326, + "eval_rougeL_fmeasure": 0.48748520016670227, + "eval_rougeL_precision": 0.48996809124946594, + "eval_rougeL_recall": 0.48645851016044617, + "eval_rougeLsum_fmeasure": 0.48748520016670227, + "eval_rougeLsum_precision": 0.48996809124946594, + "eval_rougeLsum_recall": 0.48645851016044617, + "eval_runtime": 675.1855, + "eval_samples_per_second": 5.184, + "eval_steps_per_second": 0.216, + "step": 7800 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.3994285714285714, + "epoch": 1.49, + "step": 7800 + }, + { + "epoch": 1.51, + "eval_accuracy": 0.3882857142857143, + "eval_avg_bleuScore": 0.09189968245370048, + "eval_loss": 1.1620914936065674, + "eval_rouge1_fmeasure": 0.4892265200614929, + "eval_rouge1_precision": 0.4931372106075287, + "eval_rouge1_recall": 0.48663127422332764, + "eval_rouge2_fmeasure": 0.10468500852584839, + "eval_rouge2_precision": 0.10827896744012833, + "eval_rouge2_recall": 0.10138659924268723, + "eval_rougeL_fmeasure": 0.4891880452632904, + "eval_rougeL_precision": 0.49308106303215027, + "eval_rougeL_recall": 0.48660120368003845, + "eval_rougeLsum_fmeasure": 0.4891880452632904, + "eval_rougeLsum_precision": 0.49308106303215027, + "eval_rougeLsum_recall": 0.48660120368003845, + "eval_runtime": 670.8779, + "eval_samples_per_second": 5.217, + "eval_steps_per_second": 0.218, + "step": 7900 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.3994285714285714, + "epoch": 1.51, + "step": 7900 + }, + { + "epoch": 1.53, + "learning_rate": 0.0001, + "loss": 1.164, + "step": 8000 + }, + { + "epoch": 1.53, + "eval_accuracy": 0.38742857142857146, + "eval_avg_bleuScore": 0.09163561352661678, + "eval_loss": 1.159679651260376, + "eval_rouge1_fmeasure": 0.48750895261764526, + "eval_rouge1_precision": 0.4905746281147003, + "eval_rouge1_recall": 0.4857390522956848, + "eval_rouge2_fmeasure": 0.10373055934906006, + "eval_rouge2_precision": 0.1064901053905487, + "eval_rouge2_recall": 0.10120783001184464, + "eval_rougeL_fmeasure": 0.487440824508667, + "eval_rougeL_precision": 0.4904926121234894, + "eval_rougeL_recall": 0.48568034172058105, + "eval_rougeLsum_fmeasure": 0.487440824508667, + "eval_rougeLsum_precision": 0.4904926121234894, + "eval_rougeLsum_recall": 0.48568034172058105, + "eval_runtime": 673.4305, + "eval_samples_per_second": 5.197, + "eval_steps_per_second": 0.217, + "step": 8000 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.3994285714285714, + "epoch": 1.53, + "step": 8000 + }, + { + "epoch": 1.55, + "eval_accuracy": 0.3911428571428571, + "eval_avg_bleuScore": 0.09753262519836425, + "eval_loss": 1.1540312767028809, + "eval_rouge1_fmeasure": 0.4915134310722351, + "eval_rouge1_precision": 0.49452635645866394, + "eval_rouge1_recall": 0.48958027362823486, + "eval_rouge2_fmeasure": 0.10915620625019073, + "eval_rouge2_precision": 0.11188501119613647, + "eval_rouge2_recall": 0.1066582128405571, + "eval_rougeL_fmeasure": 0.49149662256240845, + "eval_rougeL_precision": 0.4945073127746582, + "eval_rougeL_recall": 0.48956525325775146, + "eval_rougeLsum_fmeasure": 0.49149662256240845, + "eval_rougeLsum_precision": 0.4945073127746582, + "eval_rougeLsum_recall": 0.48956525325775146, + "eval_runtime": 668.2826, + "eval_samples_per_second": 5.237, + "eval_steps_per_second": 0.218, + "step": 8100 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.3994285714285714, + "epoch": 1.55, + "step": 8100 + }, + { + "epoch": 1.56, + "eval_accuracy": 0.3882857142857143, + "eval_avg_bleuScore": 0.09108461868337223, + "eval_loss": 1.1555140018463135, + "eval_rouge1_fmeasure": 0.48951366543769836, + "eval_rouge1_precision": 0.4930928349494934, + "eval_rouge1_recall": 0.4871842563152313, + "eval_rouge2_fmeasure": 0.10390307754278183, + "eval_rouge2_precision": 0.10718970000743866, + "eval_rouge2_recall": 0.10089369118213654, + "eval_rougeL_fmeasure": 0.48946547508239746, + "eval_rougeL_precision": 0.49304071068763733, + "eval_rougeL_recall": 0.48713913559913635, + "eval_rougeLsum_fmeasure": 0.48946547508239746, + "eval_rougeLsum_precision": 0.49304071068763733, + "eval_rougeLsum_recall": 0.48713913559913635, + "eval_runtime": 671.7801, + "eval_samples_per_second": 5.21, + "eval_steps_per_second": 0.217, + "step": 8200 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.3994285714285714, + "epoch": 1.56, + "step": 8200 + }, + { + "epoch": 1.58, + "eval_accuracy": 0.39485714285714285, + "eval_avg_bleuScore": 0.09901149128590311, + "eval_loss": 1.156792402267456, + "eval_rouge1_fmeasure": 0.4921184480190277, + "eval_rouge1_precision": 0.4940776824951172, + "eval_rouge1_recall": 0.4915772080421448, + "eval_rouge2_fmeasure": 0.10985264927148819, + "eval_rouge2_precision": 0.11188966035842896, + "eval_rouge2_recall": 0.10799801349639893, + "eval_rougeL_fmeasure": 0.4920693039894104, + "eval_rougeL_precision": 0.49402162432670593, + "eval_rougeL_recall": 0.4915321171283722, + "eval_rougeLsum_fmeasure": 0.4920693039894104, + "eval_rougeLsum_precision": 0.49402162432670593, + "eval_rougeLsum_recall": 0.4915321171283722, + "eval_runtime": 694.0094, + "eval_samples_per_second": 5.043, + "eval_steps_per_second": 0.21, + "step": 8300 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.3994285714285714, + "epoch": 1.58, + "step": 8300 + }, + { + "epoch": 1.6, + "eval_accuracy": 0.38885714285714285, + "eval_avg_bleuScore": 0.0932955071585519, + "eval_loss": 1.154080867767334, + "eval_rouge1_fmeasure": 0.49065786600112915, + "eval_rouge1_precision": 0.493596613407135, + "eval_rouge1_recall": 0.489065021276474, + "eval_rouge2_fmeasure": 0.1053086519241333, + "eval_rouge2_precision": 0.10811422020196915, + "eval_rouge2_recall": 0.10273490846157074, + "eval_rougeL_fmeasure": 0.49060845375061035, + "eval_rougeL_precision": 0.4935418665409088, + "eval_rougeL_recall": 0.48901987075805664, + "eval_rougeLsum_fmeasure": 0.49060845375061035, + "eval_rougeLsum_precision": 0.4935418665409088, + "eval_rougeLsum_recall": 0.48901987075805664, + "eval_runtime": 667.222, + "eval_samples_per_second": 5.246, + "eval_steps_per_second": 0.219, + "step": 8400 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.3994285714285714, + "epoch": 1.6, + "step": 8400 + }, + { + "epoch": 1.62, + "learning_rate": 0.0001, + "loss": 1.1425, + "step": 8500 + }, + { + "epoch": 1.62, + "eval_accuracy": 0.38542857142857145, + "eval_avg_bleuScore": 0.0923562583412443, + "eval_loss": 1.1554484367370605, + "eval_rouge1_fmeasure": 0.48639845848083496, + "eval_rouge1_precision": 0.48961731791496277, + "eval_rouge1_recall": 0.48460787534713745, + "eval_rouge2_fmeasure": 0.10548852384090424, + "eval_rouge2_precision": 0.10863995552062988, + "eval_rouge2_recall": 0.10259844362735748, + "eval_rougeL_fmeasure": 0.48631876707077026, + "eval_rougeL_precision": 0.48953062295913696, + "eval_rougeL_recall": 0.48453348875045776, + "eval_rougeLsum_fmeasure": 0.48631876707077026, + "eval_rougeLsum_precision": 0.48953062295913696, + "eval_rougeLsum_recall": 0.48453348875045776, + "eval_runtime": 680.2252, + "eval_samples_per_second": 5.145, + "eval_steps_per_second": 0.215, + "step": 8500 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.3994285714285714, + "epoch": 1.62, + "step": 8500 + }, + { + "epoch": 1.64, + "eval_accuracy": 0.38971428571428574, + "eval_avg_bleuScore": 0.09810057600906917, + "eval_loss": 1.1496975421905518, + "eval_rouge1_fmeasure": 0.489268958568573, + "eval_rouge1_precision": 0.49140146374702454, + "eval_rouge1_recall": 0.4884556829929352, + "eval_rouge2_fmeasure": 0.10957295447587967, + "eval_rouge2_precision": 0.11167951673269272, + "eval_rouge2_recall": 0.10766546428203583, + "eval_rougeL_fmeasure": 0.48921525478363037, + "eval_rougeL_precision": 0.49133482575416565, + "eval_rougeL_recall": 0.4884105324745178, + "eval_rougeLsum_fmeasure": 0.48921525478363037, + "eval_rougeLsum_precision": 0.49133482575416565, + "eval_rougeLsum_recall": 0.4884105324745178, + "eval_runtime": 674.9718, + "eval_samples_per_second": 5.185, + "eval_steps_per_second": 0.216, + "step": 8600 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.3994285714285714, + "epoch": 1.64, + "step": 8600 + }, + { + "epoch": 1.66, + "eval_accuracy": 0.388, + "eval_avg_bleuScore": 0.09322598004341126, + "eval_loss": 1.1533132791519165, + "eval_rouge1_fmeasure": 0.48855796456336975, + "eval_rouge1_precision": 0.49131739139556885, + "eval_rouge1_recall": 0.48707839846611023, + "eval_rouge2_fmeasure": 0.10511361062526703, + "eval_rouge2_precision": 0.10767119377851486, + "eval_rouge2_recall": 0.10277401655912399, + "eval_rougeL_fmeasure": 0.4884869456291199, + "eval_rougeL_precision": 0.4912281632423401, + "eval_rougeL_recall": 0.48701897263526917, + "eval_rougeLsum_fmeasure": 0.4884869456291199, + "eval_rougeLsum_precision": 0.4912281632423401, + "eval_rougeLsum_recall": 0.48701897263526917, + "eval_runtime": 657.1006, + "eval_samples_per_second": 5.326, + "eval_steps_per_second": 0.222, + "step": 8700 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.3994285714285714, + "epoch": 1.66, + "step": 8700 + }, + { + "epoch": 1.68, + "eval_accuracy": 0.38542857142857145, + "eval_avg_bleuScore": 0.09028857551302229, + "eval_loss": 1.151167392730713, + "eval_rouge1_fmeasure": 0.48631584644317627, + "eval_rouge1_precision": 0.48982492089271545, + "eval_rouge1_recall": 0.483847439289093, + "eval_rouge2_fmeasure": 0.10330528020858765, + "eval_rouge2_precision": 0.10646063834428787, + "eval_rouge2_recall": 0.10040877759456635, + "eval_rougeL_fmeasure": 0.48624518513679504, + "eval_rougeL_precision": 0.48973792791366577, + "eval_rougeL_recall": 0.48378732800483704, + "eval_rougeLsum_fmeasure": 0.48624518513679504, + "eval_rougeLsum_precision": 0.48973792791366577, + "eval_rougeLsum_recall": 0.48378732800483704, + "eval_runtime": 670.9252, + "eval_samples_per_second": 5.217, + "eval_steps_per_second": 0.218, + "step": 8800 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.3994285714285714, + "epoch": 1.68, + "step": 8800 + }, + { + "epoch": 1.7, + "eval_accuracy": 0.38857142857142857, + "eval_avg_bleuScore": 0.0925539482661656, + "eval_loss": 1.1473448276519775, + "eval_rouge1_fmeasure": 0.4882241189479828, + "eval_rouge1_precision": 0.4912709891796112, + "eval_rouge1_recall": 0.4861782193183899, + "eval_rouge2_fmeasure": 0.10489356517791748, + "eval_rouge2_precision": 0.1077188029885292, + "eval_rouge2_recall": 0.10229746997356415, + "eval_rougeL_fmeasure": 0.48820778727531433, + "eval_rougeL_precision": 0.491253137588501, + "eval_rougeL_recall": 0.4861631989479065, + "eval_rougeLsum_fmeasure": 0.48820778727531433, + "eval_rougeLsum_precision": 0.491253137588501, + "eval_rougeLsum_recall": 0.4861631989479065, + "eval_runtime": 656.9834, + "eval_samples_per_second": 5.327, + "eval_steps_per_second": 0.222, + "step": 8900 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.3994285714285714, + "epoch": 1.7, + "step": 8900 + }, + { + "epoch": 1.72, + "learning_rate": 0.0001, + "loss": 1.1652, + "step": 9000 + }, + { + "epoch": 1.72, + "eval_accuracy": 0.3934285714285714, + "eval_avg_bleuScore": 0.10051175594329834, + "eval_loss": 1.1473743915557861, + "eval_rouge1_fmeasure": 0.49160248041152954, + "eval_rouge1_precision": 0.4928930997848511, + "eval_rouge1_recall": 0.4918086528778076, + "eval_rouge2_fmeasure": 0.11095060408115387, + "eval_rouge2_precision": 0.11229802668094635, + "eval_rouge2_recall": 0.10975246131420135, + "eval_rougeL_fmeasure": 0.49160248041152954, + "eval_rougeL_precision": 0.4928930997848511, + "eval_rougeL_recall": 0.4918086528778076, + "eval_rougeLsum_fmeasure": 0.49160248041152954, + "eval_rougeLsum_precision": 0.4928930997848511, + "eval_rougeLsum_recall": 0.4918086528778076, + "eval_runtime": 666.4965, + "eval_samples_per_second": 5.251, + "eval_steps_per_second": 0.219, + "step": 9000 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.3994285714285714, + "epoch": 1.72, + "step": 9000 + }, + { + "epoch": 1.74, + "eval_accuracy": 0.406, + "eval_avg_bleuScore": 0.10549074566364289, + "eval_loss": 1.1458275318145752, + "eval_rouge1_fmeasure": 0.495516836643219, + "eval_rouge1_precision": 0.49655869603157043, + "eval_rouge1_recall": 0.49581125378608704, + "eval_rouge2_fmeasure": 0.11509491503238678, + "eval_rouge2_precision": 0.11617732793092728, + "eval_rouge2_recall": 0.11414265632629395, + "eval_rougeL_fmeasure": 0.495516836643219, + "eval_rougeL_precision": 0.49655869603157043, + "eval_rougeL_recall": 0.49581125378608704, + "eval_rougeLsum_fmeasure": 0.495516836643219, + "eval_rougeLsum_precision": 0.49655869603157043, + "eval_rougeLsum_recall": 0.49581125378608704, + "eval_runtime": 667.9924, + "eval_samples_per_second": 5.24, + "eval_steps_per_second": 0.219, + "step": 9100 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.406, + "epoch": 1.74, + "step": 9100 + }, + { + "epoch": 1.76, + "eval_accuracy": 0.3922857142857143, + "eval_avg_bleuScore": 0.09280328852789743, + "eval_loss": 1.146837592124939, + "eval_rouge1_fmeasure": 0.4908609092235565, + "eval_rouge1_precision": 0.49369746446609497, + "eval_rouge1_recall": 0.4891278147697449, + "eval_rouge2_fmeasure": 0.1050875261425972, + "eval_rouge2_precision": 0.10781942307949066, + "eval_rouge2_recall": 0.10259181261062622, + "eval_rougeL_fmeasure": 0.4908272922039032, + "eval_rougeL_precision": 0.4936593472957611, + "eval_rougeL_recall": 0.4890977740287781, + "eval_rougeLsum_fmeasure": 0.4908272922039032, + "eval_rougeLsum_precision": 0.4936593472957611, + "eval_rougeLsum_recall": 0.4890977740287781, + "eval_runtime": 679.8225, + "eval_samples_per_second": 5.148, + "eval_steps_per_second": 0.215, + "step": 9200 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.406, + "epoch": 1.76, + "step": 9200 + }, + { + "epoch": 1.77, + "eval_accuracy": 0.388, + "eval_avg_bleuScore": 0.09181915475640978, + "eval_loss": 1.1440588235855103, + "eval_rouge1_fmeasure": 0.4871406555175781, + "eval_rouge1_precision": 0.4905088245868683, + "eval_rouge1_recall": 0.48519039154052734, + "eval_rouge2_fmeasure": 0.10430341958999634, + "eval_rouge2_precision": 0.10743512958288193, + "eval_rouge2_recall": 0.10151927173137665, + "eval_rougeL_fmeasure": 0.48693567514419556, + "eval_rougeL_precision": 0.49028563499450684, + "eval_rougeL_recall": 0.4849971532821655, + "eval_rougeLsum_fmeasure": 0.4869495630264282, + "eval_rougeLsum_precision": 0.49029862880706787, + "eval_rougeLsum_recall": 0.4850122034549713, + "eval_runtime": 666.5052, + "eval_samples_per_second": 5.251, + "eval_steps_per_second": 0.219, + "step": 9300 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.406, + "epoch": 1.77, + "step": 9300 + }, + { + "epoch": 1.79, + "eval_accuracy": 0.3997142857142857, + "eval_avg_bleuScore": 0.09969659684385572, + "eval_loss": 1.139668583869934, + "eval_rouge1_fmeasure": 0.49307823181152344, + "eval_rouge1_precision": 0.49522465467453003, + "eval_rouge1_recall": 0.49222007393836975, + "eval_rouge2_fmeasure": 0.11044353246688843, + "eval_rouge2_precision": 0.11249065399169922, + "eval_rouge2_recall": 0.10860251635313034, + "eval_rougeL_fmeasure": 0.4930614233016968, + "eval_rougeL_precision": 0.4952056109905243, + "eval_rougeL_recall": 0.49220508337020874, + "eval_rougeLsum_fmeasure": 0.4930614233016968, + "eval_rougeLsum_precision": 0.4952056109905243, + "eval_rougeLsum_recall": 0.49220508337020874, + "eval_runtime": 662.5593, + "eval_samples_per_second": 5.283, + "eval_steps_per_second": 0.22, + "step": 9400 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.406, + "epoch": 1.79, + "step": 9400 + }, + { + "epoch": 1.81, + "learning_rate": 0.0001, + "loss": 1.1671, + "step": 9500 + }, + { + "epoch": 1.81, + "eval_accuracy": 0.3974285714285714, + "eval_avg_bleuScore": 0.101736744591168, + "eval_loss": 1.1373085975646973, + "eval_rouge1_fmeasure": 0.4946697950363159, + "eval_rouge1_precision": 0.49600887298583984, + "eval_rouge1_recall": 0.4947906732559204, + "eval_rouge2_fmeasure": 0.11145281046628952, + "eval_rouge2_precision": 0.11275575309991837, + "eval_rouge2_recall": 0.11030512303113937, + "eval_rougeL_fmeasure": 0.4946697950363159, + "eval_rougeL_precision": 0.49600887298583984, + "eval_rougeL_recall": 0.4947906732559204, + "eval_rougeLsum_fmeasure": 0.4946697950363159, + "eval_rougeLsum_precision": 0.49600887298583984, + "eval_rougeLsum_recall": 0.4947906732559204, + "eval_runtime": 661.9677, + "eval_samples_per_second": 5.287, + "eval_steps_per_second": 0.221, + "step": 9500 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.406, + "epoch": 1.81, + "step": 9500 + }, + { + "epoch": 1.83, + "eval_accuracy": 0.39571428571428574, + "eval_avg_bleuScore": 0.09729854326588767, + "eval_loss": 1.1410714387893677, + "eval_rouge1_fmeasure": 0.4920606315135956, + "eval_rouge1_precision": 0.49451515078544617, + "eval_rouge1_recall": 0.4908415973186493, + "eval_rouge2_fmeasure": 0.108421191573143, + "eval_rouge2_precision": 0.11078125238418579, + "eval_rouge2_recall": 0.10627485066652298, + "eval_rougeL_fmeasure": 0.4920283555984497, + "eval_rougeL_precision": 0.494479238986969, + "eval_rougeL_recall": 0.4908123016357422, + "eval_rougeLsum_fmeasure": 0.4920283555984497, + "eval_rougeLsum_precision": 0.494479238986969, + "eval_rougeLsum_recall": 0.4908123016357422, + "eval_runtime": 668.4134, + "eval_samples_per_second": 5.236, + "eval_steps_per_second": 0.218, + "step": 9600 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.406, + "epoch": 1.83, + "step": 9600 + }, + { + "epoch": 1.85, + "eval_accuracy": 0.4114285714285714, + "eval_avg_bleuScore": 0.1059767484494618, + "eval_loss": 1.137070894241333, + "eval_rouge1_fmeasure": 0.49964040517807007, + "eval_rouge1_precision": 0.5011536478996277, + "eval_rouge1_recall": 0.49912384152412415, + "eval_rouge2_fmeasure": 0.11548753082752228, + "eval_rouge2_precision": 0.11680314689874649, + "eval_rouge2_recall": 0.11431842297315598, + "eval_rougeL_fmeasure": 0.49964040517807007, + "eval_rougeL_precision": 0.5011536478996277, + "eval_rougeL_recall": 0.49912384152412415, + "eval_rougeLsum_fmeasure": 0.49964040517807007, + "eval_rougeLsum_precision": 0.5011536478996277, + "eval_rougeLsum_recall": 0.49912384152412415, + "eval_runtime": 659.9935, + "eval_samples_per_second": 5.303, + "eval_steps_per_second": 0.221, + "step": 9700 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.4114285714285714, + "epoch": 1.85, + "step": 9700 + }, + { + "epoch": 1.87, + "eval_accuracy": 0.4105714285714286, + "eval_avg_bleuScore": 0.10547940979685103, + "eval_loss": 1.1376601457595825, + "eval_rouge1_fmeasure": 0.49725162982940674, + "eval_rouge1_precision": 0.49904465675354004, + "eval_rouge1_recall": 0.49639928340911865, + "eval_rouge2_fmeasure": 0.11582674831151962, + "eval_rouge2_precision": 0.1175522729754448, + "eval_rouge2_recall": 0.11427413672208786, + "eval_rougeL_fmeasure": 0.49725162982940674, + "eval_rougeL_precision": 0.49904465675354004, + "eval_rougeL_recall": 0.49639928340911865, + "eval_rougeLsum_fmeasure": 0.49725162982940674, + "eval_rougeLsum_precision": 0.49904465675354004, + "eval_rougeLsum_recall": 0.49639928340911865, + "eval_runtime": 665.3024, + "eval_samples_per_second": 5.261, + "eval_steps_per_second": 0.219, + "step": 9800 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.4114285714285714, + "epoch": 1.87, + "step": 9800 + }, + { + "epoch": 1.89, + "eval_accuracy": 0.4074285714285714, + "eval_avg_bleuScore": 0.10238381918839046, + "eval_loss": 1.1363214254379272, + "eval_rouge1_fmeasure": 0.4969761371612549, + "eval_rouge1_precision": 0.49917930364608765, + "eval_rouge1_recall": 0.49593454599380493, + "eval_rouge2_fmeasure": 0.11284031718969345, + "eval_rouge2_precision": 0.11483848094940186, + "eval_rouge2_recall": 0.11104827374219894, + "eval_rougeL_fmeasure": 0.49687933921813965, + "eval_rougeL_precision": 0.4990689754486084, + "eval_rougeL_recall": 0.49584585428237915, + "eval_rougeLsum_fmeasure": 0.49687933921813965, + "eval_rougeLsum_precision": 0.4990689754486084, + "eval_rougeLsum_recall": 0.49584585428237915, + "eval_runtime": 662.3457, + "eval_samples_per_second": 5.284, + "eval_steps_per_second": 0.22, + "step": 9900 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.4114285714285714, + "epoch": 1.89, + "step": 9900 + }, + { + "epoch": 1.91, + "learning_rate": 0.0001, + "loss": 1.1591, + "step": 10000 + }, + { + "epoch": 1.91, + "eval_accuracy": 0.3882857142857143, + "eval_avg_bleuScore": 0.09409862802709852, + "eval_loss": 1.1363805532455444, + "eval_rouge1_fmeasure": 0.4896745979785919, + "eval_rouge1_precision": 0.492841899394989, + "eval_rouge1_recall": 0.48764321208000183, + "eval_rouge2_fmeasure": 0.1059836894273758, + "eval_rouge2_precision": 0.1089235469698906, + "eval_rouge2_recall": 0.10328426957130432, + "eval_rougeL_fmeasure": 0.4895933270454407, + "eval_rougeL_precision": 0.492752343416214, + "eval_rougeL_recall": 0.48756882548332214, + "eval_rougeLsum_fmeasure": 0.4895933270454407, + "eval_rougeLsum_precision": 0.492752343416214, + "eval_rougeLsum_recall": 0.48756882548332214, + "eval_runtime": 658.0356, + "eval_samples_per_second": 5.319, + "eval_steps_per_second": 0.222, + "step": 10000 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.4114285714285714, + "epoch": 1.91, + "step": 10000 + }, + { + "epoch": 1.93, + "eval_accuracy": 0.38885714285714285, + "eval_avg_bleuScore": 0.0918497473682676, + "eval_loss": 1.1335597038269043, + "eval_rouge1_fmeasure": 0.4899066984653473, + "eval_rouge1_precision": 0.49359604716300964, + "eval_rouge1_recall": 0.48732754588127136, + "eval_rouge2_fmeasure": 0.10395392775535583, + "eval_rouge2_precision": 0.10719601064920425, + "eval_rouge2_recall": 0.10097425431013107, + "eval_rougeL_fmeasure": 0.48988988995552063, + "eval_rougeL_precision": 0.4935770034790039, + "eval_rougeL_recall": 0.4873124957084656, + "eval_rougeLsum_fmeasure": 0.48988988995552063, + "eval_rougeLsum_precision": 0.4935770034790039, + "eval_rougeLsum_recall": 0.4873124957084656, + "eval_runtime": 665.8554, + "eval_samples_per_second": 5.256, + "eval_steps_per_second": 0.219, + "step": 10100 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.4114285714285714, + "epoch": 1.93, + "step": 10100 + }, + { + "epoch": 1.95, + "eval_accuracy": 0.39085714285714285, + "eval_avg_bleuScore": 0.09279956887449536, + "eval_loss": 1.1355457305908203, + "eval_rouge1_fmeasure": 0.4911603629589081, + "eval_rouge1_precision": 0.4945644438266754, + "eval_rouge1_recall": 0.4889014661312103, + "eval_rouge2_fmeasure": 0.10449952632188797, + "eval_rouge2_precision": 0.10756177455186844, + "eval_rouge2_recall": 0.10168877243995667, + "eval_rougeL_fmeasure": 0.4911603629589081, + "eval_rougeL_precision": 0.4945644438266754, + "eval_rougeL_recall": 0.4889014661312103, + "eval_rougeLsum_fmeasure": 0.4911603629589081, + "eval_rougeLsum_precision": 0.4945644438266754, + "eval_rougeLsum_recall": 0.4889014661312103, + "eval_runtime": 663.1659, + "eval_samples_per_second": 5.278, + "eval_steps_per_second": 0.22, + "step": 10200 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.4114285714285714, + "epoch": 1.95, + "step": 10200 + }, + { + "epoch": 1.97, + "eval_accuracy": 0.39371428571428574, + "eval_avg_bleuScore": 0.09612749903542654, + "eval_loss": 1.1300697326660156, + "eval_rouge1_fmeasure": 0.49530211091041565, + "eval_rouge1_precision": 0.49904003739356995, + "eval_rouge1_recall": 0.4928312599658966, + "eval_rouge2_fmeasure": 0.10829079896211624, + "eval_rouge2_precision": 0.11166734993457794, + "eval_rouge2_recall": 0.10519128292798996, + "eval_rougeL_fmeasure": 0.49530211091041565, + "eval_rougeL_precision": 0.49904003739356995, + "eval_rougeL_recall": 0.4928312599658966, + "eval_rougeLsum_fmeasure": 0.49530211091041565, + "eval_rougeLsum_precision": 0.49904003739356995, + "eval_rougeLsum_recall": 0.4928312599658966, + "eval_runtime": 669.1616, + "eval_samples_per_second": 5.23, + "eval_steps_per_second": 0.218, + "step": 10300 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.4114285714285714, + "epoch": 1.97, + "step": 10300 + }, + { + "epoch": 1.98, + "eval_accuracy": 0.39, + "eval_avg_bleuScore": 0.09225563905920302, + "eval_loss": 1.1352628469467163, + "eval_rouge1_fmeasure": 0.49202075600624084, + "eval_rouge1_precision": 0.4954337477684021, + "eval_rouge1_recall": 0.4897514283657074, + "eval_rouge2_fmeasure": 0.10506647825241089, + "eval_rouge2_precision": 0.1082741990685463, + "eval_rouge2_recall": 0.10211516171693802, + "eval_rougeL_fmeasure": 0.49202075600624084, + "eval_rougeL_precision": 0.4954337477684021, + "eval_rougeL_recall": 0.4897514283657074, + "eval_rougeLsum_fmeasure": 0.49202075600624084, + "eval_rougeLsum_precision": 0.4954337477684021, + "eval_rougeLsum_recall": 0.4897514283657074, + "eval_runtime": 675.2424, + "eval_samples_per_second": 5.183, + "eval_steps_per_second": 0.216, + "step": 10400 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.4114285714285714, + "epoch": 1.98, + "step": 10400 + }, + { + "epoch": 2.0, + "learning_rate": 0.0001, + "loss": 1.1422, + "step": 10500 + }, + { + "epoch": 2.0, + "eval_accuracy": 0.3945714285714286, + "eval_avg_bleuScore": 0.09838531419209072, + "eval_loss": 1.1325803995132446, + "eval_rouge1_fmeasure": 0.49576354026794434, + "eval_rouge1_precision": 0.4981629252433777, + "eval_rouge1_recall": 0.4945514500141144, + "eval_rouge2_fmeasure": 0.10962875932455063, + "eval_rouge2_precision": 0.11185833066701889, + "eval_rouge2_recall": 0.10759438574314117, + "eval_rougeL_fmeasure": 0.4957461953163147, + "eval_rougeL_precision": 0.4981424808502197, + "eval_rougeL_recall": 0.4945363998413086, + "eval_rougeLsum_fmeasure": 0.4957461953163147, + "eval_rougeLsum_precision": 0.4981424808502197, + "eval_rougeLsum_recall": 0.4945363998413086, + "eval_runtime": 660.6385, + "eval_samples_per_second": 5.298, + "eval_steps_per_second": 0.221, + "step": 10500 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.4114285714285714, + "epoch": 2.0, + "step": 10500 + }, + { + "epoch": 2.02, + "eval_accuracy": 0.39257142857142857, + "eval_avg_bleuScore": 0.09484845500332968, + "eval_loss": 1.1336746215820312, + "eval_rouge1_fmeasure": 0.49416378140449524, + "eval_rouge1_precision": 0.497199684381485, + "eval_rouge1_recall": 0.49220919609069824, + "eval_rouge2_fmeasure": 0.10701490938663483, + "eval_rouge2_precision": 0.10974805057048798, + "eval_rouge2_recall": 0.10450366139411926, + "eval_rougeL_fmeasure": 0.494112491607666, + "eval_rougeL_precision": 0.49713876843452454, + "eval_rougeL_recall": 0.49216407537460327, + "eval_rougeLsum_fmeasure": 0.494112491607666, + "eval_rougeLsum_precision": 0.49713876843452454, + "eval_rougeLsum_recall": 0.49216407537460327, + "eval_runtime": 662.5713, + "eval_samples_per_second": 5.282, + "eval_steps_per_second": 0.22, + "step": 10600 + }, + { + "best_epoch": 1, + "best_eval_accuracy": 0.4114285714285714, + "epoch": 2.02, + "step": 10600 + }, + { + "epoch": 2.04, + "eval_accuracy": 0.42314285714285715, + "eval_avg_bleuScore": 0.11079574659892491, + "eval_loss": 1.1268057823181152, + "eval_rouge1_fmeasure": 0.5006508231163025, + "eval_rouge1_precision": 0.5017867684364319, + "eval_rouge1_recall": 0.5005566477775574, + "eval_rouge2_fmeasure": 0.11987101286649704, + "eval_rouge2_precision": 0.12097515165805817, + "eval_rouge2_recall": 0.11888349056243896, + "eval_rougeL_fmeasure": 0.5006508231163025, + "eval_rougeL_precision": 0.5017867684364319, + "eval_rougeL_recall": 0.5005566477775574, + "eval_rougeLsum_fmeasure": 0.5006508231163025, + "eval_rougeLsum_precision": 0.5017867684364319, + "eval_rougeLsum_recall": 0.5005566477775574, + "eval_runtime": 657.2961, + "eval_samples_per_second": 5.325, + "eval_steps_per_second": 0.222, + "step": 10700 + }, + { + "best_epoch": 2, + "best_eval_accuracy": 0.42314285714285715, + "epoch": 2.04, + "step": 10700 + }, + { + "epoch": 2.06, + "eval_accuracy": 0.4114285714285714, + "eval_avg_bleuScore": 0.10423517337867191, + "eval_loss": 1.1289868354797363, + "eval_rouge1_fmeasure": 0.5011691451072693, + "eval_rouge1_precision": 0.5032759308815002, + "eval_rouge1_recall": 0.5003928542137146, + "eval_rouge2_fmeasure": 0.11485806107521057, + "eval_rouge2_precision": 0.11676233261823654, + "eval_rouge2_recall": 0.11312901973724365, + "eval_rougeL_fmeasure": 0.5011257529258728, + "eval_rougeL_precision": 0.5032330751419067, + "eval_rougeL_recall": 0.5003477334976196, + "eval_rougeLsum_fmeasure": 0.5011257529258728, + "eval_rougeLsum_precision": 0.5032330751419067, + "eval_rougeLsum_recall": 0.5003477334976196, + "eval_runtime": 651.6684, + "eval_samples_per_second": 5.371, + "eval_steps_per_second": 0.224, + "step": 10800 + }, + { + "best_epoch": 2, + "best_eval_accuracy": 0.42314285714285715, + "epoch": 2.06, + "step": 10800 + }, + { + "epoch": 2.08, + "eval_accuracy": 0.4177142857142857, + "eval_avg_bleuScore": 0.10932397210597992, + "eval_loss": 1.1277879476547241, + "eval_rouge1_fmeasure": 0.5011088848114014, + "eval_rouge1_precision": 0.5019851326942444, + "eval_rouge1_recall": 0.50115966796875, + "eval_rouge2_fmeasure": 0.11832311749458313, + "eval_rouge2_precision": 0.11913478374481201, + "eval_rouge2_recall": 0.11763662099838257, + "eval_rougeL_fmeasure": 0.5010904669761658, + "eval_rougeL_precision": 0.5019612908363342, + "eval_rougeL_recall": 0.5011445879936218, + "eval_rougeLsum_fmeasure": 0.5011088848114014, + "eval_rougeLsum_precision": 0.5019851326942444, + "eval_rougeLsum_recall": 0.50115966796875, + "eval_runtime": 649.4217, + "eval_samples_per_second": 5.389, + "eval_steps_per_second": 0.225, + "step": 10900 + }, + { + "best_epoch": 2, + "best_eval_accuracy": 0.42314285714285715, + "epoch": 2.08, + "step": 10900 + }, + { + "epoch": 2.1, + "learning_rate": 0.0001, + "loss": 1.109, + "step": 11000 + }, + { + "epoch": 2.1, + "eval_accuracy": 0.41628571428571426, + "eval_avg_bleuScore": 0.10696110429082598, + "eval_loss": 1.124712586402893, + "eval_rouge1_fmeasure": 0.5031704306602478, + "eval_rouge1_precision": 0.5047944188117981, + "eval_rouge1_recall": 0.5027408003807068, + "eval_rouge2_fmeasure": 0.11679600924253464, + "eval_rouge2_precision": 0.11825795471668243, + "eval_rouge2_recall": 0.1154724657535553, + "eval_rougeL_fmeasure": 0.5031704306602478, + "eval_rougeL_precision": 0.5047944188117981, + "eval_rougeL_recall": 0.5027408003807068, + "eval_rougeLsum_fmeasure": 0.5031704306602478, + "eval_rougeLsum_precision": 0.5047944188117981, + "eval_rougeLsum_recall": 0.5027408003807068, + "eval_runtime": 660.3507, + "eval_samples_per_second": 5.3, + "eval_steps_per_second": 0.221, + "step": 11000 + }, + { + "best_epoch": 2, + "best_eval_accuracy": 0.42314285714285715, + "epoch": 2.1, + "step": 11000 + }, + { + "epoch": 2.12, + "eval_accuracy": 0.3965714285714286, + "eval_avg_bleuScore": 0.10580109832968031, + "eval_loss": 1.1263281106948853, + "eval_rouge1_fmeasure": 0.49867135286331177, + "eval_rouge1_precision": 0.5000426173210144, + "eval_rouge1_recall": 0.49840638041496277, + "eval_rouge2_fmeasure": 0.11486261337995529, + "eval_rouge2_precision": 0.11607719957828522, + "eval_rouge2_recall": 0.11376763135194778, + "eval_rougeL_fmeasure": 0.4986545443534851, + "eval_rougeL_precision": 0.500023603439331, + "eval_rougeL_recall": 0.498391330242157, + "eval_rougeLsum_fmeasure": 0.4986545443534851, + "eval_rougeLsum_precision": 0.500023603439331, + "eval_rougeLsum_recall": 0.498391330242157, + "eval_runtime": 662.0225, + "eval_samples_per_second": 5.287, + "eval_steps_per_second": 0.221, + "step": 11100 + }, + { + "best_epoch": 2, + "best_eval_accuracy": 0.42314285714285715, + "epoch": 2.12, + "step": 11100 + }, + { + "epoch": 2.14, + "eval_accuracy": 0.39085714285714285, + "eval_avg_bleuScore": 0.09182067048549652, + "eval_loss": 1.1284575462341309, + "eval_rouge1_fmeasure": 0.49182596802711487, + "eval_rouge1_precision": 0.4957812428474426, + "eval_rouge1_recall": 0.4890452027320862, + "eval_rouge2_fmeasure": 0.10434707999229431, + "eval_rouge2_precision": 0.10772871971130371, + "eval_rouge2_recall": 0.10124608874320984, + "eval_rougeL_fmeasure": 0.4918091595172882, + "eval_rougeL_precision": 0.4957621991634369, + "eval_rougeL_recall": 0.4890301823616028, + "eval_rougeLsum_fmeasure": 0.4918091595172882, + "eval_rougeLsum_precision": 0.4957621991634369, + "eval_rougeLsum_recall": 0.4890301823616028, + "eval_runtime": 673.213, + "eval_samples_per_second": 5.199, + "eval_steps_per_second": 0.217, + "step": 11200 + }, + { + "best_epoch": 2, + "best_eval_accuracy": 0.42314285714285715, + "epoch": 2.14, + "step": 11200 + }, + { + "epoch": 2.16, + "eval_accuracy": 0.3931428571428571, + "eval_avg_bleuScore": 0.092410079904965, + "eval_loss": 1.1278029680252075, + "eval_rouge1_fmeasure": 0.49491363763809204, + "eval_rouge1_precision": 0.4984244108200073, + "eval_rouge1_recall": 0.4926760494709015, + "eval_rouge2_fmeasure": 0.1051405742764473, + "eval_rouge2_precision": 0.1083112582564354, + "eval_rouge2_recall": 0.10222513228654861, + "eval_rougeL_fmeasure": 0.4948609173297882, + "eval_rougeL_precision": 0.4983600080013275, + "eval_rougeL_recall": 0.4926309287548065, + "eval_rougeLsum_fmeasure": 0.4948609173297882, + "eval_rougeLsum_precision": 0.4983600080013275, + "eval_rougeLsum_recall": 0.4926309287548065, + "eval_runtime": 673.1405, + "eval_samples_per_second": 5.2, + "eval_steps_per_second": 0.217, + "step": 11300 + }, + { + "best_epoch": 2, + "best_eval_accuracy": 0.42314285714285715, + "epoch": 2.16, + "step": 11300 + }, + { + "epoch": 2.18, + "eval_accuracy": 0.3902857142857143, + "eval_avg_bleuScore": 0.09023959563459669, + "eval_loss": 1.1281388998031616, + "eval_rouge1_fmeasure": 0.49190571904182434, + "eval_rouge1_precision": 0.49583691358566284, + "eval_rouge1_recall": 0.4891955256462097, + "eval_rouge2_fmeasure": 0.1032390147447586, + "eval_rouge2_precision": 0.1067647710442543, + "eval_rouge2_recall": 0.10000596195459366, + "eval_rougeL_fmeasure": 0.49190571904182434, + "eval_rougeL_precision": 0.49583691358566284, + "eval_rougeL_recall": 0.4891955256462097, + "eval_rougeLsum_fmeasure": 0.49190571904182434, + "eval_rougeLsum_precision": 0.49583691358566284, + "eval_rougeLsum_recall": 0.4891955256462097, + "eval_runtime": 669.4234, + "eval_samples_per_second": 5.228, + "eval_steps_per_second": 0.218, + "step": 11400 + }, + { + "best_epoch": 2, + "best_eval_accuracy": 0.42314285714285715, + "epoch": 2.18, + "step": 11400 + }, + { + "epoch": 2.19, + "learning_rate": 0.0001, + "loss": 1.1087, + "step": 11500 + }, + { + "epoch": 2.19, + "eval_accuracy": 0.392, + "eval_avg_bleuScore": 0.09493656938416617, + "eval_loss": 1.124393105506897, + "eval_rouge1_fmeasure": 0.4959202706813812, + "eval_rouge1_precision": 0.49967092275619507, + "eval_rouge1_recall": 0.49329763650894165, + "eval_rouge2_fmeasure": 0.10782264918088913, + "eval_rouge2_precision": 0.11131300032138824, + "eval_rouge2_recall": 0.10460935533046722, + "eval_rougeL_fmeasure": 0.4958876669406891, + "eval_rougeL_precision": 0.49963507056236267, + "eval_rougeL_recall": 0.4932675361633301, + "eval_rougeLsum_fmeasure": 0.4958876669406891, + "eval_rougeLsum_precision": 0.49963507056236267, + "eval_rougeLsum_recall": 0.4932675361633301, + "eval_runtime": 675.2374, + "eval_samples_per_second": 5.183, + "eval_steps_per_second": 0.216, + "step": 11500 + }, + { + "best_epoch": 2, + "best_eval_accuracy": 0.42314285714285715, + "epoch": 2.19, + "step": 11500 + }, + { + "epoch": 2.21, + "eval_accuracy": 0.3902857142857143, + "eval_avg_bleuScore": 0.09425990174497877, + "eval_loss": 1.1322588920593262, + "eval_rouge1_fmeasure": 0.49463632702827454, + "eval_rouge1_precision": 0.4986339509487152, + "eval_rouge1_recall": 0.491852343082428, + "eval_rouge2_fmeasure": 0.10734687000513077, + "eval_rouge2_precision": 0.11072012782096863, + "eval_rouge2_recall": 0.1042417362332344, + "eval_rougeL_fmeasure": 0.49463632702827454, + "eval_rougeL_precision": 0.4986339509487152, + "eval_rougeL_recall": 0.491852343082428, + "eval_rougeLsum_fmeasure": 0.49463632702827454, + "eval_rougeLsum_precision": 0.4986339509487152, + "eval_rougeLsum_recall": 0.491852343082428, + "eval_runtime": 671.9841, + "eval_samples_per_second": 5.208, + "eval_steps_per_second": 0.217, + "step": 11600 + }, + { + "best_epoch": 2, + "best_eval_accuracy": 0.42314285714285715, + "epoch": 2.21, + "step": 11600 + }, + { + "epoch": 2.23, + "eval_accuracy": 0.4025714285714286, + "eval_avg_bleuScore": 0.10102391575063978, + "eval_loss": 1.1322745084762573, + "eval_rouge1_fmeasure": 0.49643146991729736, + "eval_rouge1_precision": 0.4987451136112213, + "eval_rouge1_recall": 0.4951358735561371, + "eval_rouge2_fmeasure": 0.11220090091228485, + "eval_rouge2_precision": 0.11423654109239578, + "eval_rouge2_recall": 0.11034509539604187, + "eval_rougeL_fmeasure": 0.49643146991729736, + "eval_rougeL_precision": 0.4987451136112213, + "eval_rougeL_recall": 0.4951358735561371, + "eval_rougeLsum_fmeasure": 0.49643146991729736, + "eval_rougeLsum_precision": 0.4987451136112213, + "eval_rougeLsum_recall": 0.4951358735561371, + "eval_runtime": 681.1333, + "eval_samples_per_second": 5.138, + "eval_steps_per_second": 0.214, + "step": 11700 + }, + { + "best_epoch": 2, + "best_eval_accuracy": 0.42314285714285715, + "epoch": 2.23, + "step": 11700 + }, + { + "epoch": 2.25, + "eval_accuracy": 0.3891428571428571, + "eval_avg_bleuScore": 0.09059217444487981, + "eval_loss": 1.1340641975402832, + "eval_rouge1_fmeasure": 0.4908812344074249, + "eval_rouge1_precision": 0.4947715103626251, + "eval_rouge1_recall": 0.4882884919643402, + "eval_rouge2_fmeasure": 0.10454118996858597, + "eval_rouge2_precision": 0.10805298388004303, + "eval_rouge2_recall": 0.10131365805864334, + "eval_rougeL_fmeasure": 0.4908812344074249, + "eval_rougeL_precision": 0.4947715103626251, + "eval_rougeL_recall": 0.4882884919643402, + "eval_rougeLsum_fmeasure": 0.4908812344074249, + "eval_rougeLsum_precision": 0.4947715103626251, + "eval_rougeLsum_recall": 0.4882884919643402, + "eval_runtime": 658.4563, + "eval_samples_per_second": 5.315, + "eval_steps_per_second": 0.222, + "step": 11800 + }, + { + "best_epoch": 2, + "best_eval_accuracy": 0.42314285714285715, + "epoch": 2.25, + "step": 11800 + }, + { + "epoch": 2.27, + "eval_accuracy": 0.3945714285714286, + "eval_avg_bleuScore": 0.09734489480086736, + "eval_loss": 1.1317287683486938, + "eval_rouge1_fmeasure": 0.4939056932926178, + "eval_rouge1_precision": 0.4966863691806793, + "eval_rouge1_recall": 0.49237895011901855, + "eval_rouge2_fmeasure": 0.10915359854698181, + "eval_rouge2_precision": 0.11189696937799454, + "eval_rouge2_recall": 0.10663413256406784, + "eval_rougeL_fmeasure": 0.4939056932926178, + "eval_rougeL_precision": 0.4966863691806793, + "eval_rougeL_recall": 0.49237895011901855, + "eval_rougeLsum_fmeasure": 0.4939056932926178, + "eval_rougeLsum_precision": 0.4966863691806793, + "eval_rougeLsum_recall": 0.49237895011901855, + "eval_runtime": 672.6239, + "eval_samples_per_second": 5.204, + "eval_steps_per_second": 0.217, + "step": 11900 + }, + { + "best_epoch": 2, + "best_eval_accuracy": 0.42314285714285715, + "epoch": 2.27, + "step": 11900 + }, + { + "epoch": 2.29, + "learning_rate": 0.0001, + "loss": 1.1115, + "step": 12000 + }, + { + "epoch": 2.29, + "eval_accuracy": 0.40885714285714286, + "eval_avg_bleuScore": 0.10408217024803161, + "eval_loss": 1.1265759468078613, + "eval_rouge1_fmeasure": 0.4989066421985626, + "eval_rouge1_precision": 0.5013710260391235, + "eval_rouge1_recall": 0.49781954288482666, + "eval_rouge2_fmeasure": 0.1148609071969986, + "eval_rouge2_precision": 0.11715953797101974, + "eval_rouge2_recall": 0.11275454610586166, + "eval_rougeL_fmeasure": 0.4989066421985626, + "eval_rougeL_precision": 0.5013710260391235, + "eval_rougeL_recall": 0.49781954288482666, + "eval_rougeLsum_fmeasure": 0.4989066421985626, + "eval_rougeLsum_precision": 0.5013710260391235, + "eval_rougeLsum_recall": 0.49781954288482666, + "eval_runtime": 659.6387, + "eval_samples_per_second": 5.306, + "eval_steps_per_second": 0.221, + "step": 12000 + }, + { + "best_epoch": 2, + "best_eval_accuracy": 0.42314285714285715, + "epoch": 2.29, + "step": 12000 + }, + { + "epoch": 2.31, + "eval_accuracy": 0.4014285714285714, + "eval_avg_bleuScore": 0.1046858798776354, + "eval_loss": 1.126979947090149, + "eval_rouge1_fmeasure": 0.5006938576698303, + "eval_rouge1_precision": 0.5019481778144836, + "eval_rouge1_recall": 0.500682532787323, + "eval_rouge2_fmeasure": 0.11389509588479996, + "eval_rouge2_precision": 0.11497559398412704, + "eval_rouge2_recall": 0.11293908953666687, + "eval_rougeL_fmeasure": 0.5006569623947144, + "eval_rougeL_precision": 0.5019006133079529, + "eval_rougeL_recall": 0.5006524324417114, + "eval_rougeLsum_fmeasure": 0.5006569623947144, + "eval_rougeLsum_precision": 0.5019006133079529, + "eval_rougeLsum_recall": 0.5006524324417114, + "eval_runtime": 674.1788, + "eval_samples_per_second": 5.192, + "eval_steps_per_second": 0.217, + "step": 12100 + }, + { + "best_epoch": 2, + "best_eval_accuracy": 0.42314285714285715, + "epoch": 2.31, + "step": 12100 + }, + { + "epoch": 2.33, + "eval_accuracy": 0.3977142857142857, + "eval_avg_bleuScore": 0.09774469310896737, + "eval_loss": 1.1287332773208618, + "eval_rouge1_fmeasure": 0.4993576407432556, + "eval_rouge1_precision": 0.5026358962059021, + "eval_rouge1_recall": 0.49733662605285645, + "eval_rouge2_fmeasure": 0.1099507212638855, + "eval_rouge2_precision": 0.1128658801317215, + "eval_rouge2_recall": 0.10726912319660187, + "eval_rougeL_fmeasure": 0.49932077527046204, + "eval_rougeL_precision": 0.5025882720947266, + "eval_rougeL_recall": 0.4973064959049225, + "eval_rougeLsum_fmeasure": 0.49932077527046204, + "eval_rougeLsum_precision": 0.5025882720947266, + "eval_rougeLsum_recall": 0.4973064959049225, + "eval_runtime": 673.9245, + "eval_samples_per_second": 5.193, + "eval_steps_per_second": 0.217, + "step": 12200 + }, + { + "best_epoch": 2, + "best_eval_accuracy": 0.42314285714285715, + "epoch": 2.33, + "step": 12200 + }, + { + "epoch": 2.35, + "eval_accuracy": 0.42514285714285716, + "eval_avg_bleuScore": 0.11020775973796844, + "eval_loss": 1.1253113746643066, + "eval_rouge1_fmeasure": 0.5042111277580261, + "eval_rouge1_precision": 0.5053219795227051, + "eval_rouge1_recall": 0.5044164061546326, + "eval_rouge2_fmeasure": 0.11928899586200714, + "eval_rouge2_precision": 0.12029542028903961, + "eval_rouge2_recall": 0.11840342730283737, + "eval_rougeL_fmeasure": 0.5042111277580261, + "eval_rougeL_precision": 0.5053219795227051, + "eval_rougeL_recall": 0.5044164061546326, + "eval_rougeLsum_fmeasure": 0.5042111277580261, + "eval_rougeLsum_precision": 0.5053219795227051, + "eval_rougeLsum_recall": 0.5044164061546326, + "eval_runtime": 672.5105, + "eval_samples_per_second": 5.204, + "eval_steps_per_second": 0.217, + "step": 12300 + }, + { + "best_epoch": 2, + "best_eval_accuracy": 0.42514285714285716, + "epoch": 2.35, + "step": 12300 + }, + { + "epoch": 2.37, + "eval_accuracy": 0.3977142857142857, + "eval_avg_bleuScore": 0.09610247135162353, + "eval_loss": 1.125049114227295, + "eval_rouge1_fmeasure": 0.49996352195739746, + "eval_rouge1_precision": 0.5034509897232056, + "eval_rouge1_recall": 0.4977772533893585, + "eval_rouge2_fmeasure": 0.10930398106575012, + "eval_rouge2_precision": 0.11255652457475662, + "eval_rouge2_recall": 0.10630861669778824, + "eval_rougeL_fmeasure": 0.49996352195739746, + "eval_rougeL_precision": 0.5034509897232056, + "eval_rougeL_recall": 0.4977772533893585, + "eval_rougeLsum_fmeasure": 0.49996352195739746, + "eval_rougeLsum_precision": 0.5034509897232056, + "eval_rougeLsum_recall": 0.4977772533893585, + "eval_runtime": 667.9324, + "eval_samples_per_second": 5.24, + "eval_steps_per_second": 0.219, + "step": 12400 + }, + { + "best_epoch": 2, + "best_eval_accuracy": 0.42514285714285716, + "epoch": 2.37, + "step": 12400 + }, + { + "epoch": 2.39, + "learning_rate": 0.0001, + "loss": 1.1123, + "step": 12500 + }, + { + "epoch": 2.39, + "eval_accuracy": 0.3934285714285714, + "eval_avg_bleuScore": 0.09337386521271297, + "eval_loss": 1.1253979206085205, + "eval_rouge1_fmeasure": 0.4958665668964386, + "eval_rouge1_precision": 0.49966517090797424, + "eval_rouge1_recall": 0.4933442175388336, + "eval_rouge2_fmeasure": 0.1065107062458992, + "eval_rouge2_precision": 0.11001263558864594, + "eval_rouge2_recall": 0.10328928381204605, + "eval_rougeL_fmeasure": 0.4958665668964386, + "eval_rougeL_precision": 0.49966517090797424, + "eval_rougeL_recall": 0.4933442175388336, + "eval_rougeLsum_fmeasure": 0.4958665668964386, + "eval_rougeLsum_precision": 0.49966517090797424, + "eval_rougeLsum_recall": 0.4933442175388336, + "eval_runtime": 674.2467, + "eval_samples_per_second": 5.191, + "eval_steps_per_second": 0.217, + "step": 12500 + }, + { + "best_epoch": 2, + "best_eval_accuracy": 0.42514285714285716, + "epoch": 2.39, + "step": 12500 + }, + { + "epoch": 2.4, + "eval_accuracy": 0.39514285714285713, + "eval_avg_bleuScore": 0.09321406442778452, + "eval_loss": 1.123453140258789, + "eval_rouge1_fmeasure": 0.49691084027290344, + "eval_rouge1_precision": 0.5000061988830566, + "eval_rouge1_recall": 0.49535050988197327, + "eval_rouge2_fmeasure": 0.10573318600654602, + "eval_rouge2_precision": 0.10876386612653732, + "eval_rouge2_recall": 0.10294327884912491, + "eval_rougeL_fmeasure": 0.4968945384025574, + "eval_rougeL_precision": 0.4999883472919464, + "eval_rougeL_recall": 0.4953354597091675, + "eval_rougeLsum_fmeasure": 0.4968945384025574, + "eval_rougeLsum_precision": 0.4999883472919464, + "eval_rougeLsum_recall": 0.4953354597091675, + "eval_runtime": 669.4001, + "eval_samples_per_second": 5.229, + "eval_steps_per_second": 0.218, + "step": 12600 + }, + { + "best_epoch": 2, + "best_eval_accuracy": 0.42514285714285716, + "epoch": 2.4, + "step": 12600 + }, + { + "epoch": 2.42, + "eval_accuracy": 0.4085714285714286, + "eval_avg_bleuScore": 0.10311075380870274, + "eval_loss": 1.1204102039337158, + "eval_rouge1_fmeasure": 0.4997433125972748, + "eval_rouge1_precision": 0.5017807483673096, + "eval_rouge1_recall": 0.49872148036956787, + "eval_rouge2_fmeasure": 0.11368373036384583, + "eval_rouge2_precision": 0.115488201379776, + "eval_rouge2_recall": 0.11205478012561798, + "eval_rougeL_fmeasure": 0.499693363904953, + "eval_rougeL_precision": 0.5017248392105103, + "eval_rougeL_recall": 0.4986763596534729, + "eval_rougeLsum_fmeasure": 0.499693363904953, + "eval_rougeLsum_precision": 0.5017248392105103, + "eval_rougeLsum_recall": 0.4986763596534729, + "eval_runtime": 671.2854, + "eval_samples_per_second": 5.214, + "eval_steps_per_second": 0.217, + "step": 12700 + }, + { + "best_epoch": 2, + "best_eval_accuracy": 0.42514285714285716, + "epoch": 2.42, + "step": 12700 + }, + { + "epoch": 2.44, + "eval_accuracy": 0.3945714285714286, + "eval_avg_bleuScore": 0.09467888181550162, + "eval_loss": 1.1208487749099731, + "eval_rouge1_fmeasure": 0.4955655634403229, + "eval_rouge1_precision": 0.49855026602745056, + "eval_rouge1_recall": 0.49366599321365356, + "eval_rouge2_fmeasure": 0.10684741288423538, + "eval_rouge2_precision": 0.10964684933423996, + "eval_rouge2_recall": 0.10428454726934433, + "eval_rougeL_fmeasure": 0.4955487549304962, + "eval_rougeL_precision": 0.4985312223434448, + "eval_rougeL_recall": 0.4936509132385254, + "eval_rougeLsum_fmeasure": 0.4955487549304962, + "eval_rougeLsum_precision": 0.4985312223434448, + "eval_rougeLsum_recall": 0.4936509132385254, + "eval_runtime": 666.6811, + "eval_samples_per_second": 5.25, + "eval_steps_per_second": 0.219, + "step": 12800 + }, + { + "best_epoch": 2, + "best_eval_accuracy": 0.42514285714285716, + "epoch": 2.44, + "step": 12800 + }, + { + "epoch": 2.46, + "eval_accuracy": 0.39485714285714285, + "eval_avg_bleuScore": 0.09453405777045659, + "eval_loss": 1.1211049556732178, + "eval_rouge1_fmeasure": 0.4953373670578003, + "eval_rouge1_precision": 0.49838489294052124, + "eval_rouge1_recall": 0.4933258593082428, + "eval_rouge2_fmeasure": 0.10639544576406479, + "eval_rouge2_precision": 0.10912305861711502, + "eval_rouge2_recall": 0.10390234738588333, + "eval_rougeL_fmeasure": 0.4953032433986664, + "eval_rougeL_precision": 0.4983454644680023, + "eval_rougeL_recall": 0.4932957887649536, + "eval_rougeLsum_fmeasure": 0.4953032433986664, + "eval_rougeLsum_precision": 0.4983454644680023, + "eval_rougeLsum_recall": 0.4932957887649536, + "eval_runtime": 665.9257, + "eval_samples_per_second": 5.256, + "eval_steps_per_second": 0.219, + "step": 12900 + }, + { + "best_epoch": 2, + "best_eval_accuracy": 0.42514285714285716, + "epoch": 2.46, + "step": 12900 + }, + { + "epoch": 2.48, + "learning_rate": 0.0001, + "loss": 1.1084, + "step": 13000 + }, + { + "epoch": 2.48, + "eval_accuracy": 0.4054285714285714, + "eval_avg_bleuScore": 0.10425920934336526, + "eval_loss": 1.1208699941635132, + "eval_rouge1_fmeasure": 0.5006346702575684, + "eval_rouge1_precision": 0.5025137662887573, + "eval_rouge1_recall": 0.4996601939201355, + "eval_rouge2_fmeasure": 0.11404173821210861, + "eval_rouge2_precision": 0.11560704559087753, + "eval_rouge2_recall": 0.11261889338493347, + "eval_rougeL_fmeasure": 0.5006346702575684, + "eval_rougeL_precision": 0.5025137662887573, + "eval_rougeL_recall": 0.4996601939201355, + "eval_rougeLsum_fmeasure": 0.5006346702575684, + "eval_rougeLsum_precision": 0.5025137662887573, + "eval_rougeLsum_recall": 0.4996601939201355, + "eval_runtime": 668.7515, + "eval_samples_per_second": 5.234, + "eval_steps_per_second": 0.218, + "step": 13000 + }, + { + "best_epoch": 2, + "best_eval_accuracy": 0.42514285714285716, + "epoch": 2.48, + "step": 13000 + }, + { + "epoch": 2.5, + "eval_accuracy": 0.40485714285714286, + "eval_avg_bleuScore": 0.10105134565489633, + "eval_loss": 1.1170507669448853, + "eval_rouge1_fmeasure": 0.49920791387557983, + "eval_rouge1_precision": 0.5012545585632324, + "eval_rouge1_recall": 0.49810245633125305, + "eval_rouge2_fmeasure": 0.11170776188373566, + "eval_rouge2_precision": 0.11356569081544876, + "eval_rouge2_recall": 0.1100294440984726, + "eval_rougeL_fmeasure": 0.49920791387557983, + "eval_rougeL_precision": 0.5012545585632324, + "eval_rougeL_recall": 0.49810245633125305, + "eval_rougeLsum_fmeasure": 0.49920791387557983, + "eval_rougeLsum_precision": 0.5012545585632324, + "eval_rougeLsum_recall": 0.49810245633125305, + "eval_runtime": 668.4709, + "eval_samples_per_second": 5.236, + "eval_steps_per_second": 0.218, + "step": 13100 + }, + { + "best_epoch": 2, + "best_eval_accuracy": 0.42514285714285716, + "epoch": 2.5, + "step": 13100 + }, + { + "epoch": 2.52, + "eval_accuracy": 0.4017142857142857, + "eval_avg_bleuScore": 0.09799285956791469, + "eval_loss": 1.1140613555908203, + "eval_rouge1_fmeasure": 0.5012453198432922, + "eval_rouge1_precision": 0.5045880079269409, + "eval_rouge1_recall": 0.49885624647140503, + "eval_rouge2_fmeasure": 0.11020597070455551, + "eval_rouge2_precision": 0.11312974244356155, + "eval_rouge2_recall": 0.10753265023231506, + "eval_rougeL_fmeasure": 0.5012289881706238, + "eval_rougeL_precision": 0.5045701861381531, + "eval_rougeL_recall": 0.49884122610092163, + "eval_rougeLsum_fmeasure": 0.5012289881706238, + "eval_rougeLsum_precision": 0.5045701861381531, + "eval_rougeLsum_recall": 0.49884122610092163, + "eval_runtime": 656.3794, + "eval_samples_per_second": 5.332, + "eval_steps_per_second": 0.222, + "step": 13200 + }, + { + "best_epoch": 2, + "best_eval_accuracy": 0.42514285714285716, + "epoch": 2.52, + "step": 13200 + }, + { + "epoch": 2.54, + "eval_accuracy": 0.4, + "eval_avg_bleuScore": 0.09729664065156664, + "eval_loss": 1.113106608390808, + "eval_rouge1_fmeasure": 0.5022633671760559, + "eval_rouge1_precision": 0.5054683089256287, + "eval_rouge1_recall": 0.5001609921455383, + "eval_rouge2_fmeasure": 0.10972665250301361, + "eval_rouge2_precision": 0.11277198791503906, + "eval_rouge2_recall": 0.10692593455314636, + "eval_rougeL_fmeasure": 0.5022633671760559, + "eval_rougeL_precision": 0.5054683089256287, + "eval_rougeL_recall": 0.5001609921455383, + "eval_rougeLsum_fmeasure": 0.5022633671760559, + "eval_rougeLsum_precision": 0.5054683089256287, + "eval_rougeLsum_recall": 0.5001609921455383, + "eval_runtime": 671.7883, + "eval_samples_per_second": 5.21, + "eval_steps_per_second": 0.217, + "step": 13300 + }, + { + "best_epoch": 2, + "best_eval_accuracy": 0.42514285714285716, + "epoch": 2.54, + "step": 13300 + }, + { + "epoch": 2.56, + "eval_accuracy": 0.3931428571428571, + "eval_avg_bleuScore": 0.09372351804801396, + "eval_loss": 1.1136925220489502, + "eval_rouge1_fmeasure": 0.49636325240135193, + "eval_rouge1_precision": 0.5002996921539307, + "eval_rouge1_recall": 0.49348852038383484, + "eval_rouge2_fmeasure": 0.10639674961566925, + "eval_rouge2_precision": 0.10986518859863281, + "eval_rouge2_recall": 0.10321289300918579, + "eval_rougeL_fmeasure": 0.49632638692855835, + "eval_rougeL_precision": 0.5002520680427551, + "eval_rougeL_recall": 0.49345844984054565, + "eval_rougeLsum_fmeasure": 0.49632638692855835, + "eval_rougeLsum_precision": 0.5002520680427551, + "eval_rougeLsum_recall": 0.49345844984054565, + "eval_runtime": 664.1082, + "eval_samples_per_second": 5.27, + "eval_steps_per_second": 0.22, + "step": 13400 + }, + { + "best_epoch": 2, + "best_eval_accuracy": 0.42514285714285716, + "epoch": 2.56, + "step": 13400 + }, + { + "epoch": 2.58, + "learning_rate": 0.0001, + "loss": 1.1009, + "step": 13500 + }, + { + "epoch": 2.58, + "eval_accuracy": 0.3922857142857143, + "eval_avg_bleuScore": 0.09417361571107592, + "eval_loss": 1.112511157989502, + "eval_rouge1_fmeasure": 0.4943287968635559, + "eval_rouge1_precision": 0.4975851774215698, + "eval_rouge1_recall": 0.4921298921108246, + "eval_rouge2_fmeasure": 0.10641402006149292, + "eval_rouge2_precision": 0.10936450958251953, + "eval_rouge2_recall": 0.10371486097574234, + "eval_rougeL_fmeasure": 0.4942919611930847, + "eval_rougeL_precision": 0.4975375533103943, + "eval_rougeL_recall": 0.4920998215675354, + "eval_rougeLsum_fmeasure": 0.4942919611930847, + "eval_rougeLsum_precision": 0.4975375533103943, + "eval_rougeLsum_recall": 0.4920998215675354, + "eval_runtime": 674.7989, + "eval_samples_per_second": 5.187, + "eval_steps_per_second": 0.216, + "step": 13500 + }, + { + "best_epoch": 2, + "best_eval_accuracy": 0.42514285714285716, + "epoch": 2.58, + "step": 13500 + }, + { + "epoch": 2.6, + "eval_accuracy": 0.3954285714285714, + "eval_avg_bleuScore": 0.0928648316860199, + "eval_loss": 1.111234426498413, + "eval_rouge1_fmeasure": 0.4980577528476715, + "eval_rouge1_precision": 0.5015146136283875, + "eval_rouge1_recall": 0.4957538843154907, + "eval_rouge2_fmeasure": 0.1053348034620285, + "eval_rouge2_precision": 0.10856404900550842, + "eval_rouge2_recall": 0.10236277431249619, + "eval_rougeL_fmeasure": 0.4980577528476715, + "eval_rougeL_precision": 0.5015146136283875, + "eval_rougeL_recall": 0.4957538843154907, + "eval_rougeLsum_fmeasure": 0.4980577528476715, + "eval_rougeLsum_precision": 0.5015146136283875, + "eval_rougeLsum_recall": 0.4957538843154907, + "eval_runtime": 686.0008, + "eval_samples_per_second": 5.102, + "eval_steps_per_second": 0.213, + "step": 13600 + }, + { + "best_epoch": 2, + "best_eval_accuracy": 0.42514285714285716, + "epoch": 2.6, + "step": 13600 + }, + { + "epoch": 2.61, + "eval_accuracy": 0.3965714285714286, + "eval_avg_bleuScore": 0.09074489026410239, + "eval_loss": 1.1137773990631104, + "eval_rouge1_fmeasure": 0.4967907965183258, + "eval_rouge1_precision": 0.5002678632736206, + "eval_rouge1_recall": 0.4944820702075958, + "eval_rouge2_fmeasure": 0.10332436114549637, + "eval_rouge2_precision": 0.10665764659643173, + "eval_rouge2_recall": 0.100257009267807, + "eval_rougeL_fmeasure": 0.496753990650177, + "eval_rougeL_precision": 0.5002202391624451, + "eval_rougeL_recall": 0.49445199966430664, + "eval_rougeLsum_fmeasure": 0.496753990650177, + "eval_rougeLsum_precision": 0.5002202391624451, + "eval_rougeLsum_recall": 0.49445199966430664, + "eval_runtime": 675.186, + "eval_samples_per_second": 5.184, + "eval_steps_per_second": 0.216, + "step": 13700 + }, + { + "best_epoch": 2, + "best_eval_accuracy": 0.42514285714285716, + "epoch": 2.61, + "step": 13700 + }, + { + "epoch": 2.63, + "eval_accuracy": 0.39485714285714285, + "eval_avg_bleuScore": 0.09137427430493492, + "eval_loss": 1.1150407791137695, + "eval_rouge1_fmeasure": 0.4954349100589752, + "eval_rouge1_precision": 0.4987472593784332, + "eval_rouge1_recall": 0.493621289730072, + "eval_rouge2_fmeasure": 0.10367072373628616, + "eval_rouge2_precision": 0.106794573366642, + "eval_rouge2_recall": 0.10080065578222275, + "eval_rougeL_fmeasure": 0.4954349100589752, + "eval_rougeL_precision": 0.4987472593784332, + "eval_rougeL_recall": 0.493621289730072, + "eval_rougeLsum_fmeasure": 0.4954349100589752, + "eval_rougeLsum_precision": 0.4987472593784332, + "eval_rougeLsum_recall": 0.493621289730072, + "eval_runtime": 660.8722, + "eval_samples_per_second": 5.296, + "eval_steps_per_second": 0.221, + "step": 13800 + }, + { + "best_epoch": 2, + "best_eval_accuracy": 0.42514285714285716, + "epoch": 2.63, + "step": 13800 + }, + { + "epoch": 2.65, + "eval_accuracy": 0.4102857142857143, + "eval_avg_bleuScore": 0.10118092659541539, + "eval_loss": 1.110621690750122, + "eval_rouge1_fmeasure": 0.5003848075866699, + "eval_rouge1_precision": 0.5024074912071228, + "eval_rouge1_recall": 0.49945124983787537, + "eval_rouge2_fmeasure": 0.11233095824718475, + "eval_rouge2_precision": 0.1143016517162323, + "eval_rouge2_recall": 0.11054433137178421, + "eval_rougeL_fmeasure": 0.5003848075866699, + "eval_rougeL_precision": 0.5024074912071228, + "eval_rougeL_recall": 0.49945124983787537, + "eval_rougeLsum_fmeasure": 0.5003848075866699, + "eval_rougeLsum_precision": 0.5024074912071228, + "eval_rougeLsum_recall": 0.49945124983787537, + "eval_runtime": 658.2845, + "eval_samples_per_second": 5.317, + "eval_steps_per_second": 0.222, + "step": 13900 + }, + { + "best_epoch": 2, + "best_eval_accuracy": 0.42514285714285716, + "epoch": 2.65, + "step": 13900 + }, + { + "epoch": 2.67, + "learning_rate": 0.0001, + "loss": 1.1073, + "step": 14000 + }, + { + "epoch": 2.67, + "eval_accuracy": 0.3962857142857143, + "eval_avg_bleuScore": 0.09082218527793884, + "eval_loss": 1.1148605346679688, + "eval_rouge1_fmeasure": 0.49622249603271484, + "eval_rouge1_precision": 0.4997093081474304, + "eval_rouge1_recall": 0.4939694106578827, + "eval_rouge2_fmeasure": 0.10334985703229904, + "eval_rouge2_precision": 0.10656334459781647, + "eval_rouge2_recall": 0.10039636492729187, + "eval_rougeL_fmeasure": 0.4961855709552765, + "eval_rougeL_precision": 0.49966171383857727, + "eval_rougeL_recall": 0.4939393401145935, + "eval_rougeLsum_fmeasure": 0.4961855709552765, + "eval_rougeLsum_precision": 0.49966171383857727, + "eval_rougeLsum_recall": 0.4939393401145935, + "eval_runtime": 668.9152, + "eval_samples_per_second": 5.232, + "eval_steps_per_second": 0.218, + "step": 14000 + }, + { + "best_epoch": 2, + "best_eval_accuracy": 0.42514285714285716, + "epoch": 2.67, + "step": 14000 + }, + { + "epoch": 2.69, + "eval_accuracy": 0.3985714285714286, + "eval_avg_bleuScore": 0.09508898271833148, + "eval_loss": 1.1150691509246826, + "eval_rouge1_fmeasure": 0.4991874694824219, + "eval_rouge1_precision": 0.5020511746406555, + "eval_rouge1_recall": 0.49760565161705017, + "eval_rouge2_fmeasure": 0.10719644278287888, + "eval_rouge2_precision": 0.11002182960510254, + "eval_rouge2_recall": 0.10459007322788239, + "eval_rougeL_fmeasure": 0.4991874694824219, + "eval_rougeL_precision": 0.5020511746406555, + "eval_rougeL_recall": 0.49760565161705017, + "eval_rougeLsum_fmeasure": 0.4991874694824219, + "eval_rougeLsum_precision": 0.5020511746406555, + "eval_rougeLsum_recall": 0.49760565161705017, + "eval_runtime": 679.119, + "eval_samples_per_second": 5.154, + "eval_steps_per_second": 0.215, + "step": 14100 + }, + { + "best_epoch": 2, + "best_eval_accuracy": 0.42514285714285716, + "epoch": 2.69, + "step": 14100 + }, + { + "epoch": 2.71, + "eval_accuracy": 0.39514285714285713, + "eval_avg_bleuScore": 0.093787245120321, + "eval_loss": 1.1157784461975098, + "eval_rouge1_fmeasure": 0.4964507818222046, + "eval_rouge1_precision": 0.4999375343322754, + "eval_rouge1_recall": 0.4945111870765686, + "eval_rouge2_fmeasure": 0.10612804442644119, + "eval_rouge2_precision": 0.1093730553984642, + "eval_rouge2_recall": 0.10315121710300446, + "eval_rougeL_fmeasure": 0.4964507818222046, + "eval_rougeL_precision": 0.4999375343322754, + "eval_rougeL_recall": 0.4945111870765686, + "eval_rougeLsum_fmeasure": 0.4964507818222046, + "eval_rougeLsum_precision": 0.4999375343322754, + "eval_rougeLsum_recall": 0.4945111870765686, + "eval_runtime": 673.2127, + "eval_samples_per_second": 5.199, + "eval_steps_per_second": 0.217, + "step": 14200 + }, + { + "best_epoch": 2, + "best_eval_accuracy": 0.42514285714285716, + "epoch": 2.71, + "step": 14200 + }, + { + "epoch": 2.73, + "eval_accuracy": 0.41085714285714287, + "eval_avg_bleuScore": 0.10571601830210005, + "eval_loss": 1.1125943660736084, + "eval_rouge1_fmeasure": 0.5019577145576477, + "eval_rouge1_precision": 0.5039337277412415, + "eval_rouge1_recall": 0.5009273886680603, + "eval_rouge2_fmeasure": 0.1154671311378479, + "eval_rouge2_precision": 0.11716865748167038, + "eval_rouge2_recall": 0.11393246799707413, + "eval_rougeL_fmeasure": 0.5019413828849792, + "eval_rougeL_precision": 0.5039158463478088, + "eval_rougeL_recall": 0.5009123086929321, + "eval_rougeLsum_fmeasure": 0.5019577145576477, + "eval_rougeLsum_precision": 0.5039337277412415, + "eval_rougeLsum_recall": 0.5009273886680603, + "eval_runtime": 659.3512, + "eval_samples_per_second": 5.308, + "eval_steps_per_second": 0.221, + "step": 14300 + }, + { + "best_epoch": 2, + "best_eval_accuracy": 0.42514285714285716, + "epoch": 2.73, + "step": 14300 + }, + { + "epoch": 2.75, + "eval_accuracy": 0.39285714285714285, + "eval_avg_bleuScore": 0.09230625669445311, + "eval_loss": 1.1112126111984253, + "eval_rouge1_fmeasure": 0.4946015477180481, + "eval_rouge1_precision": 0.4977358281612396, + "eval_rouge1_recall": 0.49260008335113525, + "eval_rouge2_fmeasure": 0.1045902892947197, + "eval_rouge2_precision": 0.10746891051530838, + "eval_rouge2_recall": 0.10196827352046967, + "eval_rougeL_fmeasure": 0.49458473920822144, + "eval_rougeL_precision": 0.4977167844772339, + "eval_rougeL_recall": 0.49258506298065186, + "eval_rougeLsum_fmeasure": 0.4946015477180481, + "eval_rougeLsum_precision": 0.4977358281612396, + "eval_rougeLsum_recall": 0.49260008335113525, + "eval_runtime": 659.833, + "eval_samples_per_second": 5.304, + "eval_steps_per_second": 0.221, + "step": 14400 + }, + { + "best_epoch": 2, + "best_eval_accuracy": 0.42514285714285716, + "epoch": 2.75, + "step": 14400 + }, + { + "epoch": 2.77, + "learning_rate": 0.0001, + "loss": 1.1076, + "step": 14500 + }, + { + "epoch": 2.77, + "eval_accuracy": 0.39485714285714285, + "eval_avg_bleuScore": 0.09671461480004447, + "eval_loss": 1.1113131046295166, + "eval_rouge1_fmeasure": 0.49711933732032776, + "eval_rouge1_precision": 0.5003589391708374, + "eval_rouge1_recall": 0.4954087734222412, + "eval_rouge2_fmeasure": 0.10890911519527435, + "eval_rouge2_precision": 0.11197753995656967, + "eval_rouge2_recall": 0.10609560459852219, + "eval_rougeL_fmeasure": 0.49711933732032776, + "eval_rougeL_precision": 0.5003589391708374, + "eval_rougeL_recall": 0.4954087734222412, + "eval_rougeLsum_fmeasure": 0.49711933732032776, + "eval_rougeLsum_precision": 0.5003589391708374, + "eval_rougeLsum_recall": 0.4954087734222412, + "eval_runtime": 663.2197, + "eval_samples_per_second": 5.277, + "eval_steps_per_second": 0.22, + "step": 14500 + }, + { + "best_epoch": 2, + "best_eval_accuracy": 0.42514285714285716, + "epoch": 2.77, + "step": 14500 + } + ], + "max_steps": 15720, + "num_train_epochs": 3, + "total_flos": 1.1256249423821247e+23, + "trial_name": null, + "trial_params": null +}