| { | |
| "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": 3.2618825722273996, | |
| "global_step": 28000, | |
| "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 | |
| } | |
| ], | |
| "max_steps": 85840, | |
| "num_train_epochs": 10, | |
| "total_flos": 4.230446910422016e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |