{ "best_metric": 1.6879340410232544, "best_model_checkpoint": "/home3/s5431786/nlp-final-project/results/google-flan-t5-small-e-snli-generation-explanation_use_prompt_label-selected-b64/checkpoint-28000", "epoch": 5.125815470643057, "global_step": 44000, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.23, "learning_rate": 0.0004659832246039143, "loss": 1.6051, "step": 2000 }, { "epoch": 0.23, "eval_accuracy": null, "eval_bertscore_f1": 0.9312730214807033, "eval_bleu": 0.39782836950988876, "eval_f1": null, "eval_loss": 1.7780297994613647, "eval_rouge1": 0.5911787258201253, "eval_rouge2": 0.38303308637957223, "eval_rougeL": 0.5377264257708202, "eval_rougeLsum": 0.5395006217014338, "eval_runtime": 168.0702, "eval_samples_per_second": 58.559, "eval_steps_per_second": 0.916, "step": 2000 }, { "epoch": 0.47, "learning_rate": 0.0009319664492078286, "loss": 1.4342, "step": 4000 }, { "epoch": 0.47, "eval_accuracy": null, "eval_bertscore_f1": 0.933153417701039, "eval_bleu": 0.4002469468404885, "eval_f1": null, "eval_loss": 1.7909314632415771, "eval_rouge1": 0.5975097321291585, "eval_rouge2": 0.3847152683676083, "eval_rougeL": 0.5420448625219438, "eval_rougeLsum": 0.5439086899907004, "eval_runtime": 163.5422, "eval_samples_per_second": 60.18, "eval_steps_per_second": 0.942, "step": 4000 }, { "epoch": 0.7, "learning_rate": 0.0009790552803256977, "loss": 1.3804, "step": 6000 }, { "epoch": 0.7, "eval_accuracy": null, "eval_bertscore_f1": 0.9348857560079773, "eval_bleu": 0.41900709929386093, "eval_f1": null, "eval_loss": 1.7246909141540527, "eval_rouge1": 0.616316027086431, "eval_rouge2": 0.4107303410412238, "eval_rougeL": 0.5610382702035865, "eval_rougeLsum": 0.5630504299315114, "eval_runtime": 164.6404, "eval_samples_per_second": 59.779, "eval_steps_per_second": 0.935, "step": 6000 }, { "epoch": 0.93, "learning_rate": 0.0009545298474518076, "loss": 1.3396, "step": 8000 }, { "epoch": 0.93, "eval_accuracy": null, "eval_bertscore_f1": 0.9358202236390555, "eval_bleu": 0.41886081162261424, "eval_f1": null, "eval_loss": 1.7622839212417603, "eval_rouge1": 0.6176280342703749, "eval_rouge2": 0.4078573926320103, "eval_rougeL": 0.5616472881864998, "eval_rougeLsum": 0.563450597865839, "eval_runtime": 154.9703, "eval_samples_per_second": 63.509, "eval_steps_per_second": 0.994, "step": 8000 }, { "epoch": 1.16, "learning_rate": 0.0009300044145779173, "loss": 1.2871, "step": 10000 }, { "epoch": 1.16, "eval_accuracy": null, "eval_bertscore_f1": 0.9358213962977603, "eval_bleu": 0.4193950275310417, "eval_f1": null, "eval_loss": 1.7429882287979126, "eval_rouge1": 0.6213453720058478, "eval_rouge2": 0.4161815104047649, "eval_rougeL": 0.5663236030411948, "eval_rougeLsum": 0.5683627905966635, "eval_runtime": 166.3727, "eval_samples_per_second": 59.156, "eval_steps_per_second": 0.926, "step": 10000 }, { "epoch": 1.4, "learning_rate": 0.0009054789817040271, "loss": 1.2598, "step": 12000 }, { "epoch": 1.4, "eval_accuracy": null, "eval_bertscore_f1": 0.9357474451383859, "eval_bleu": 0.4237575236205067, "eval_f1": null, "eval_loss": 1.7180376052856445, "eval_rouge1": 0.6178787224462856, "eval_rouge2": 0.41096129012621224, "eval_rougeL": 0.5625046403576894, "eval_rougeLsum": 0.5643579355959574, "eval_runtime": 161.2048, "eval_samples_per_second": 61.053, "eval_steps_per_second": 0.955, "step": 12000 }, { "epoch": 1.63, "learning_rate": 0.0008809535488301369, "loss": 1.2527, "step": 14000 }, { "epoch": 1.63, "eval_accuracy": null, "eval_bertscore_f1": 0.9347525270301088, "eval_bleu": 0.4103575347376984, "eval_f1": null, "eval_loss": 1.7438157796859741, "eval_rouge1": 0.6163196253092696, "eval_rouge2": 0.40951276436422024, "eval_rougeL": 0.561512744919562, "eval_rougeLsum": 0.5636171995683823, "eval_runtime": 163.0281, "eval_samples_per_second": 60.37, "eval_steps_per_second": 0.945, "step": 14000 }, { "epoch": 1.86, "learning_rate": 0.0008564281159562467, "loss": 1.2367, "step": 16000 }, { "epoch": 1.86, "eval_accuracy": null, "eval_bertscore_f1": 0.9367065161642227, "eval_bleu": 0.4269002075734853, "eval_f1": null, "eval_loss": 1.692925214767456, "eval_rouge1": 0.6232036846315818, "eval_rouge2": 0.4180000668098338, "eval_rougeL": 0.5680648753273818, "eval_rougeLsum": 0.570113733783007, "eval_runtime": 165.5988, "eval_samples_per_second": 59.433, "eval_steps_per_second": 0.93, "step": 16000 }, { "epoch": 2.1, "learning_rate": 0.0008319026830823564, "loss": 1.2084, "step": 18000 }, { "epoch": 2.1, "eval_accuracy": null, "eval_bertscore_f1": 0.9346273795566624, "eval_bleu": 0.4082124628614291, "eval_f1": null, "eval_loss": 1.7380170822143555, "eval_rouge1": 0.6207188358704087, "eval_rouge2": 0.4177034932099197, "eval_rougeL": 0.5656127494093233, "eval_rougeLsum": 0.5677652123372579, "eval_runtime": 171.9389, "eval_samples_per_second": 57.241, "eval_steps_per_second": 0.896, "step": 18000 }, { "epoch": 2.33, "learning_rate": 0.0008073772502084662, "loss": 1.1764, "step": 20000 }, { "epoch": 2.33, "eval_accuracy": null, "eval_bertscore_f1": 0.9354710851566406, "eval_bleu": 0.40988266787912037, "eval_f1": null, "eval_loss": 1.693483591079712, "eval_rouge1": 0.6180761782666475, "eval_rouge2": 0.41114120287967504, "eval_rougeL": 0.5614112736943433, "eval_rougeLsum": 0.5634652483896412, "eval_runtime": 166.4526, "eval_samples_per_second": 59.128, "eval_steps_per_second": 0.925, "step": 20000 }, { "epoch": 2.56, "learning_rate": 0.000782851817334576, "loss": 1.1755, "step": 22000 }, { "epoch": 2.56, "eval_accuracy": null, "eval_bertscore_f1": 0.9346324508266705, "eval_bleu": 0.40634655287582505, "eval_f1": null, "eval_loss": 1.7524782419204712, "eval_rouge1": 0.6135893786369074, "eval_rouge2": 0.40424644238900054, "eval_rougeL": 0.5569845265395319, "eval_rougeLsum": 0.5589737599230747, "eval_runtime": 160.1533, "eval_samples_per_second": 61.454, "eval_steps_per_second": 0.962, "step": 22000 }, { "epoch": 2.8, "learning_rate": 0.0007583263844606856, "loss": 1.177, "step": 24000 }, { "epoch": 2.8, "eval_accuracy": null, "eval_bertscore_f1": 0.9357648674997463, "eval_bleu": 0.4166788473251748, "eval_f1": null, "eval_loss": 1.6996724605560303, "eval_rouge1": 0.6259385124974665, "eval_rouge2": 0.4221804267801674, "eval_rougeL": 0.5702138311702792, "eval_rougeLsum": 0.5724783022726267, "eval_runtime": 168.0704, "eval_samples_per_second": 58.559, "eval_steps_per_second": 0.916, "step": 24000 }, { "epoch": 3.03, "learning_rate": 0.0007338009515867955, "loss": 1.1627, "step": 26000 }, { "epoch": 3.03, "eval_accuracy": null, "eval_bertscore_f1": 0.9365227118034573, "eval_bleu": 0.42188586391601435, "eval_f1": null, "eval_loss": 1.7493563890457153, "eval_rouge1": 0.6268137047490179, "eval_rouge2": 0.42133389000894483, "eval_rougeL": 0.5704007507856375, "eval_rougeLsum": 0.5723283103728103, "eval_runtime": 165.0375, "eval_samples_per_second": 59.635, "eval_steps_per_second": 0.933, "step": 26000 }, { "epoch": 3.26, "learning_rate": 0.0007092755187129053, "loss": 1.1213, "step": 28000 }, { "epoch": 3.26, "eval_accuracy": null, "eval_bertscore_f1": 0.9357226263202562, "eval_bleu": 0.4211338041409779, "eval_f1": null, "eval_loss": 1.6879340410232544, "eval_rouge1": 0.6241785396960564, "eval_rouge2": 0.4172784618159209, "eval_rougeL": 0.5676477480665587, "eval_rougeLsum": 0.5698542368429731, "eval_runtime": 161.3196, "eval_samples_per_second": 61.009, "eval_steps_per_second": 0.955, "step": 28000 }, { "epoch": 3.49, "learning_rate": 0.0006847500858390151, "loss": 1.121, "step": 30000 }, { "epoch": 3.49, "eval_accuracy": null, "eval_bertscore_f1": 0.9367988150164839, "eval_bleu": 0.42462787525751705, "eval_f1": null, "eval_loss": 1.7076311111450195, "eval_rouge1": 0.6318247622031307, "eval_rouge2": 0.42560950906924944, "eval_rougeL": 0.5726212713576446, "eval_rougeLsum": 0.5746959557098454, "eval_runtime": 159.1906, "eval_samples_per_second": 61.825, "eval_steps_per_second": 0.967, "step": 30000 }, { "epoch": 3.73, "learning_rate": 0.0006602246529651248, "loss": 1.1252, "step": 32000 }, { "epoch": 3.73, "eval_accuracy": null, "eval_bertscore_f1": 0.9359545163678532, "eval_bleu": 0.42344383498507465, "eval_f1": null, "eval_loss": 1.709180474281311, "eval_rouge1": 0.6240469583047095, "eval_rouge2": 0.4172449697108748, "eval_rougeL": 0.5686400389756898, "eval_rougeLsum": 0.5707285873482231, "eval_runtime": 166.6453, "eval_samples_per_second": 59.06, "eval_steps_per_second": 0.924, "step": 32000 }, { "epoch": 3.96, "learning_rate": 0.0006356992200912346, "loss": 1.1249, "step": 34000 }, { "epoch": 3.96, "eval_accuracy": null, "eval_bertscore_f1": 0.935824385111566, "eval_bleu": 0.42158432020818837, "eval_f1": null, "eval_loss": 1.717002511024475, "eval_rouge1": 0.626629100727167, "eval_rouge2": 0.41890429908455223, "eval_rougeL": 0.5695943813788329, "eval_rougeLsum": 0.5715380718252487, "eval_runtime": 163.4292, "eval_samples_per_second": 60.222, "eval_steps_per_second": 0.942, "step": 34000 }, { "epoch": 4.19, "learning_rate": 0.0006111737872173444, "loss": 1.081, "step": 36000 }, { "epoch": 4.19, "eval_accuracy": null, "eval_bertscore_f1": 0.9366190056446194, "eval_bleu": 0.4235289371475263, "eval_f1": null, "eval_loss": 1.7053654193878174, "eval_rouge1": 0.6275525594646458, "eval_rouge2": 0.4228795682628885, "eval_rougeL": 0.5724176993561514, "eval_rougeLsum": 0.5744128212816374, "eval_runtime": 161.6199, "eval_samples_per_second": 60.896, "eval_steps_per_second": 0.953, "step": 36000 }, { "epoch": 4.43, "learning_rate": 0.0005866483543434542, "loss": 1.0775, "step": 38000 }, { "epoch": 4.43, "eval_accuracy": null, "eval_bertscore_f1": 0.9358997231175613, "eval_bleu": 0.42469208733709635, "eval_f1": null, "eval_loss": 1.7025026082992554, "eval_rouge1": 0.624191647944516, "eval_rouge2": 0.4156753967572323, "eval_rougeL": 0.5654439558669471, "eval_rougeLsum": 0.5676313065358018, "eval_runtime": 161.1453, "eval_samples_per_second": 61.075, "eval_steps_per_second": 0.956, "step": 38000 }, { "epoch": 4.66, "learning_rate": 0.0005621229214695639, "loss": 1.0817, "step": 40000 }, { "epoch": 4.66, "eval_accuracy": null, "eval_bertscore_f1": 0.9371525604679827, "eval_bleu": 0.4290378828512276, "eval_f1": null, "eval_loss": 1.759724736213684, "eval_rouge1": 0.6329216329529901, "eval_rouge2": 0.428712988798996, "eval_rougeL": 0.5766896544496578, "eval_rougeLsum": 0.5786424518723892, "eval_runtime": 164.8993, "eval_samples_per_second": 59.685, "eval_steps_per_second": 0.934, "step": 40000 }, { "epoch": 4.89, "learning_rate": 0.0005375974885956737, "loss": 1.0858, "step": 42000 }, { "epoch": 4.89, "eval_accuracy": null, "eval_bertscore_f1": 0.9373510686986889, "eval_bleu": 0.432049715666974, "eval_f1": null, "eval_loss": 1.7263308763504028, "eval_rouge1": 0.6324147183021485, "eval_rouge2": 0.42847933630650004, "eval_rougeL": 0.5752177231277784, "eval_rougeLsum": 0.5770958516370777, "eval_runtime": 161.8865, "eval_samples_per_second": 60.796, "eval_steps_per_second": 0.951, "step": 42000 }, { "epoch": 5.13, "learning_rate": 0.0005130720557217835, "loss": 1.0571, "step": 44000 }, { "epoch": 5.13, "eval_accuracy": null, "eval_bertscore_f1": 0.9361940296387629, "eval_bleu": 0.42251404411209087, "eval_f1": null, "eval_loss": 1.7361332178115845, "eval_rouge1": 0.623944320741679, "eval_rouge2": 0.41785433009247724, "eval_rougeL": 0.5665420760123123, "eval_rougeLsum": 0.5687571348191405, "eval_runtime": 164.5399, "eval_samples_per_second": 59.815, "eval_steps_per_second": 0.936, "step": 44000 } ], "max_steps": 85840, "num_train_epochs": 10, "total_flos": 6.649203110355149e+16, "trial_name": null, "trial_params": null }