| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "eval_steps": 66, | |
| "global_step": 1310, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 3e-05, | |
| "loss": 3.7331, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 3e-05, | |
| "loss": 2.345, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 3e-05, | |
| "loss": 2.1491, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 3e-05, | |
| "loss": 2.0739, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_accuracy": 0.6206453178068898, | |
| "eval_loss": 1.8404711484909058, | |
| "eval_runtime": 11.9883, | |
| "eval_samples_per_second": 25.024, | |
| "eval_steps_per_second": 1.585, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_exact_match": 7.666666666666667, | |
| "eval_f1": 10.221428571428572, | |
| "eval_qa_bleu": 1.4042262195131967, | |
| "eval_qa_exact_match": 0.07333333333333333, | |
| "eval_recite_bleu": 8.5956480576491, | |
| "eval_recite_exact_match": 0.0, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 3e-05, | |
| "loss": 1.9722, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 3e-05, | |
| "loss": 2.014, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 3e-05, | |
| "loss": 1.9812, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 3e-05, | |
| "loss": 1.9213, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 3e-05, | |
| "loss": 1.8806, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "eval_accuracy": 0.6365065502183406, | |
| "eval_loss": 1.574863314628601, | |
| "eval_runtime": 9.1867, | |
| "eval_samples_per_second": 32.656, | |
| "eval_steps_per_second": 2.068, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "eval_exact_match": 8.666666666666666, | |
| "eval_f1": 13.8123783922171, | |
| "eval_qa_bleu": 3.0725240037081307, | |
| "eval_qa_exact_match": 0.08666666666666667, | |
| "eval_recite_bleu": 16.386073781113847, | |
| "eval_recite_exact_match": 0.0, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 3e-05, | |
| "loss": 1.6177, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 3e-05, | |
| "loss": 1.4286, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 3e-05, | |
| "loss": 1.4102, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 3e-05, | |
| "loss": 1.4033, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 3e-05, | |
| "loss": 1.3619, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "eval_accuracy": 0.6533139252789908, | |
| "eval_loss": 1.342494010925293, | |
| "eval_runtime": 9.3165, | |
| "eval_samples_per_second": 32.201, | |
| "eval_steps_per_second": 2.039, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "eval_exact_match": 7.0, | |
| "eval_f1": 10.38888888888889, | |
| "eval_qa_bleu": 5.590724094958645, | |
| "eval_qa_exact_match": 0.06666666666666667, | |
| "eval_recite_bleu": 17.070052559724132, | |
| "eval_recite_exact_match": 0.0, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 3e-05, | |
| "loss": 1.3649, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 3e-05, | |
| "loss": 1.2886, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 3e-05, | |
| "loss": 1.2787, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 3e-05, | |
| "loss": 1.283, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "eval_accuracy": 0.6685152838427948, | |
| "eval_loss": 1.1253471374511719, | |
| "eval_runtime": 9.3443, | |
| "eval_samples_per_second": 32.105, | |
| "eval_steps_per_second": 2.033, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "eval_exact_match": 9.0, | |
| "eval_f1": 13.042195767195768, | |
| "eval_qa_bleu": 5.743589453455261, | |
| "eval_qa_exact_match": 0.08666666666666667, | |
| "eval_recite_bleu": 15.4059437725803, | |
| "eval_recite_exact_match": 0.0, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 3e-05, | |
| "loss": 1.112, | |
| "step": 266 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 3e-05, | |
| "loss": 0.8377, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 3e-05, | |
| "loss": 0.8533, | |
| "step": 294 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 3e-05, | |
| "loss": 0.8407, | |
| "step": 308 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 3e-05, | |
| "loss": 0.8433, | |
| "step": 322 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "eval_accuracy": 0.6825278990781174, | |
| "eval_loss": 0.9735248684883118, | |
| "eval_runtime": 9.3401, | |
| "eval_samples_per_second": 32.12, | |
| "eval_steps_per_second": 2.034, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "eval_exact_match": 11.666666666666666, | |
| "eval_f1": 15.95859788359788, | |
| "eval_qa_bleu": 6.31601664322449, | |
| "eval_qa_exact_match": 0.11, | |
| "eval_recite_bleu": 21.285839887136575, | |
| "eval_recite_exact_match": 0.0, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 3e-05, | |
| "loss": 0.8632, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 3e-05, | |
| "loss": 0.8308, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 3e-05, | |
| "loss": 0.8475, | |
| "step": 364 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 3e-05, | |
| "loss": 0.7829, | |
| "step": 378 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 3e-05, | |
| "loss": 0.7629, | |
| "step": 392 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "eval_accuracy": 0.6982241630276564, | |
| "eval_loss": 0.7873561382293701, | |
| "eval_runtime": 9.0856, | |
| "eval_samples_per_second": 33.019, | |
| "eval_steps_per_second": 2.091, | |
| "step": 396 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "eval_exact_match": 12.333333333333334, | |
| "eval_f1": 16.93478835978836, | |
| "eval_qa_bleu": 7.7589360856745815, | |
| "eval_qa_exact_match": 0.12, | |
| "eval_recite_bleu": 22.16978951972916, | |
| "eval_recite_exact_match": 0.0, | |
| "step": 396 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 3e-05, | |
| "loss": 0.5537, | |
| "step": 406 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "learning_rate": 3e-05, | |
| "loss": 0.5111, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "learning_rate": 3e-05, | |
| "loss": 0.479, | |
| "step": 434 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4904, | |
| "step": 448 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "learning_rate": 3e-05, | |
| "loss": 0.5058, | |
| "step": 462 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "eval_accuracy": 0.7086026200873362, | |
| "eval_loss": 0.6920613646507263, | |
| "eval_runtime": 9.3473, | |
| "eval_samples_per_second": 32.095, | |
| "eval_steps_per_second": 2.033, | |
| "step": 462 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "eval_exact_match": 15.333333333333334, | |
| "eval_f1": 20.8812384782973, | |
| "eval_qa_bleu": 11.001436952878551, | |
| "eval_qa_exact_match": 0.14, | |
| "eval_recite_bleu": 25.319328021621963, | |
| "eval_recite_exact_match": 0.0, | |
| "step": 462 | |
| }, | |
| { | |
| "epoch": 3.63, | |
| "learning_rate": 3e-05, | |
| "loss": 0.5456, | |
| "step": 476 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4998, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "learning_rate": 3e-05, | |
| "loss": 0.5075, | |
| "step": 504 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4593, | |
| "step": 518 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "eval_accuracy": 0.7196943231441048, | |
| "eval_loss": 0.564062237739563, | |
| "eval_runtime": 9.2784, | |
| "eval_samples_per_second": 32.333, | |
| "eval_steps_per_second": 2.048, | |
| "step": 528 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "eval_exact_match": 13.333333333333334, | |
| "eval_f1": 18.11966301672184, | |
| "eval_qa_bleu": 4.47266646700781, | |
| "eval_qa_exact_match": 0.12666666666666668, | |
| "eval_recite_bleu": 26.136161725029392, | |
| "eval_recite_exact_match": 0.0033333333333333335, | |
| "step": 528 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3952, | |
| "step": 532 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2996, | |
| "step": 546 | |
| }, | |
| { | |
| "epoch": 4.27, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3354, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 4.38, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3179, | |
| "step": 574 | |
| }, | |
| { | |
| "epoch": 4.49, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3064, | |
| "step": 588 | |
| }, | |
| { | |
| "epoch": 4.53, | |
| "eval_accuracy": 0.7245269286754003, | |
| "eval_loss": 0.5348330736160278, | |
| "eval_runtime": 9.5519, | |
| "eval_samples_per_second": 31.407, | |
| "eval_steps_per_second": 1.989, | |
| "step": 594 | |
| }, | |
| { | |
| "epoch": 4.53, | |
| "eval_exact_match": 13.666666666666666, | |
| "eval_f1": 19.689682539682536, | |
| "eval_qa_bleu": 4.45260487005976, | |
| "eval_qa_exact_match": 0.12666666666666668, | |
| "eval_recite_bleu": 28.855187002245795, | |
| "eval_recite_exact_match": 0.0, | |
| "step": 594 | |
| }, | |
| { | |
| "epoch": 4.6, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3258, | |
| "step": 602 | |
| }, | |
| { | |
| "epoch": 4.7, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3384, | |
| "step": 616 | |
| }, | |
| { | |
| "epoch": 4.81, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3258, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 4.92, | |
| "learning_rate": 3e-05, | |
| "loss": 0.312, | |
| "step": 644 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2967, | |
| "step": 658 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "eval_accuracy": 0.7303784570596797, | |
| "eval_loss": 0.47703343629837036, | |
| "eval_runtime": 9.4774, | |
| "eval_samples_per_second": 31.654, | |
| "eval_steps_per_second": 2.005, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "eval_exact_match": 12.333333333333334, | |
| "eval_f1": 18.61798941798941, | |
| "eval_qa_bleu": 6.60839076276961, | |
| "eval_qa_exact_match": 0.12, | |
| "eval_recite_bleu": 29.5293953590396, | |
| "eval_recite_exact_match": 0.0, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 5.13, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2148, | |
| "step": 672 | |
| }, | |
| { | |
| "epoch": 5.24, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2275, | |
| "step": 686 | |
| }, | |
| { | |
| "epoch": 5.34, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2158, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2167, | |
| "step": 714 | |
| }, | |
| { | |
| "epoch": 5.54, | |
| "eval_accuracy": 0.732372634643377, | |
| "eval_loss": 0.458192378282547, | |
| "eval_runtime": 9.1255, | |
| "eval_samples_per_second": 32.875, | |
| "eval_steps_per_second": 2.082, | |
| "step": 726 | |
| }, | |
| { | |
| "epoch": 5.54, | |
| "eval_exact_match": 14.333333333333334, | |
| "eval_f1": 19.81243386243386, | |
| "eval_qa_bleu": 8.961623752889384, | |
| "eval_qa_exact_match": 0.13666666666666666, | |
| "eval_recite_bleu": 33.41904095334099, | |
| "eval_recite_exact_match": 0.0, | |
| "step": 726 | |
| }, | |
| { | |
| "epoch": 5.56, | |
| "learning_rate": 3e-05, | |
| "loss": 0.229, | |
| "step": 728 | |
| }, | |
| { | |
| "epoch": 5.66, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2275, | |
| "step": 742 | |
| }, | |
| { | |
| "epoch": 5.77, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2211, | |
| "step": 756 | |
| }, | |
| { | |
| "epoch": 5.88, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2231, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 5.98, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2157, | |
| "step": 784 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "eval_accuracy": 0.7358369723435225, | |
| "eval_loss": 0.4307834804058075, | |
| "eval_runtime": 9.1847, | |
| "eval_samples_per_second": 32.663, | |
| "eval_steps_per_second": 2.069, | |
| "step": 792 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "eval_exact_match": 16.666666666666668, | |
| "eval_f1": 21.786772486772488, | |
| "eval_qa_bleu": 7.511563755726586, | |
| "eval_qa_exact_match": 0.16, | |
| "eval_recite_bleu": 37.63384023220464, | |
| "eval_recite_exact_match": 0.013333333333333334, | |
| "step": 792 | |
| }, | |
| { | |
| "epoch": 6.09, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1669, | |
| "step": 798 | |
| }, | |
| { | |
| "epoch": 6.2, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1712, | |
| "step": 812 | |
| }, | |
| { | |
| "epoch": 6.31, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1601, | |
| "step": 826 | |
| }, | |
| { | |
| "epoch": 6.41, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1608, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 6.52, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1597, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 6.55, | |
| "eval_accuracy": 0.7373410965550704, | |
| "eval_loss": 0.4301389157772064, | |
| "eval_runtime": 9.7372, | |
| "eval_samples_per_second": 30.81, | |
| "eval_steps_per_second": 1.951, | |
| "step": 858 | |
| }, | |
| { | |
| "epoch": 6.55, | |
| "eval_exact_match": 15.666666666666666, | |
| "eval_f1": 21.056661856661847, | |
| "eval_qa_bleu": 12.649140852831426, | |
| "eval_qa_exact_match": 0.14666666666666667, | |
| "eval_recite_bleu": 42.22665248887737, | |
| "eval_recite_exact_match": 0.013333333333333334, | |
| "step": 858 | |
| }, | |
| { | |
| "epoch": 6.63, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1623, | |
| "step": 868 | |
| }, | |
| { | |
| "epoch": 6.73, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1668, | |
| "step": 882 | |
| }, | |
| { | |
| "epoch": 6.84, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1624, | |
| "step": 896 | |
| }, | |
| { | |
| "epoch": 6.95, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1648, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1481, | |
| "step": 924 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "eval_accuracy": 0.7385298398835517, | |
| "eval_loss": 0.42236796021461487, | |
| "eval_runtime": 9.3603, | |
| "eval_samples_per_second": 32.05, | |
| "eval_steps_per_second": 2.03, | |
| "step": 924 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "eval_exact_match": 18.666666666666668, | |
| "eval_f1": 25.187830687830694, | |
| "eval_qa_bleu": 8.014835952265651, | |
| "eval_qa_exact_match": 0.18, | |
| "eval_recite_bleu": 42.38987173856079, | |
| "eval_recite_exact_match": 0.016666666666666666, | |
| "step": 924 | |
| }, | |
| { | |
| "epoch": 7.16, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1227, | |
| "step": 938 | |
| }, | |
| { | |
| "epoch": 7.27, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1272, | |
| "step": 952 | |
| }, | |
| { | |
| "epoch": 7.37, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1312, | |
| "step": 966 | |
| }, | |
| { | |
| "epoch": 7.48, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1293, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 7.56, | |
| "eval_accuracy": 0.739422610383309, | |
| "eval_loss": 0.41248488426208496, | |
| "eval_runtime": 9.7486, | |
| "eval_samples_per_second": 30.774, | |
| "eval_steps_per_second": 1.949, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 7.56, | |
| "eval_exact_match": 15.666666666666666, | |
| "eval_f1": 22.348196248196246, | |
| "eval_qa_bleu": 6.260683683577316, | |
| "eval_qa_exact_match": 0.14666666666666667, | |
| "eval_recite_bleu": 45.04221504063147, | |
| "eval_recite_exact_match": 0.023333333333333334, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 7.59, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1345, | |
| "step": 994 | |
| }, | |
| { | |
| "epoch": 7.69, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1325, | |
| "step": 1008 | |
| }, | |
| { | |
| "epoch": 7.8, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1273, | |
| "step": 1022 | |
| }, | |
| { | |
| "epoch": 7.91, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1362, | |
| "step": 1036 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "learning_rate": 3e-05, | |
| "loss": 0.125, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "eval_accuracy": 0.7399902959728287, | |
| "eval_loss": 0.41223272681236267, | |
| "eval_runtime": 9.2562, | |
| "eval_samples_per_second": 32.411, | |
| "eval_steps_per_second": 2.053, | |
| "step": 1056 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "eval_exact_match": 18.0, | |
| "eval_f1": 25.051058201058197, | |
| "eval_qa_bleu": 10.352483602423003, | |
| "eval_qa_exact_match": 0.17, | |
| "eval_recite_bleu": 46.26228887496748, | |
| "eval_recite_exact_match": 0.04666666666666667, | |
| "step": 1056 | |
| }, | |
| { | |
| "epoch": 8.12, | |
| "learning_rate": 3e-05, | |
| "loss": 0.11, | |
| "step": 1064 | |
| }, | |
| { | |
| "epoch": 8.23, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1045, | |
| "step": 1078 | |
| }, | |
| { | |
| "epoch": 8.34, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1179, | |
| "step": 1092 | |
| }, | |
| { | |
| "epoch": 8.44, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1083, | |
| "step": 1106 | |
| }, | |
| { | |
| "epoch": 8.55, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1139, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 8.56, | |
| "eval_accuracy": 0.7406501698204755, | |
| "eval_loss": 0.40691348910331726, | |
| "eval_runtime": 9.0014, | |
| "eval_samples_per_second": 33.328, | |
| "eval_steps_per_second": 2.111, | |
| "step": 1122 | |
| }, | |
| { | |
| "epoch": 8.56, | |
| "eval_exact_match": 16.333333333333332, | |
| "eval_f1": 23.766955266955264, | |
| "eval_qa_bleu": 11.966740392922118, | |
| "eval_qa_exact_match": 0.15333333333333332, | |
| "eval_recite_bleu": 45.74271226096357, | |
| "eval_recite_exact_match": 0.03333333333333333, | |
| "step": 1122 | |
| }, | |
| { | |
| "epoch": 8.66, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1108, | |
| "step": 1134 | |
| }, | |
| { | |
| "epoch": 8.76, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1168, | |
| "step": 1148 | |
| }, | |
| { | |
| "epoch": 8.87, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1132, | |
| "step": 1162 | |
| }, | |
| { | |
| "epoch": 8.98, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1141, | |
| "step": 1176 | |
| }, | |
| { | |
| "epoch": 9.07, | |
| "eval_accuracy": 0.7409364386220282, | |
| "eval_loss": 0.40822312235832214, | |
| "eval_runtime": 9.119, | |
| "eval_samples_per_second": 32.898, | |
| "eval_steps_per_second": 2.084, | |
| "step": 1188 | |
| }, | |
| { | |
| "epoch": 9.07, | |
| "eval_exact_match": 17.333333333333332, | |
| "eval_f1": 23.91428571428571, | |
| "eval_qa_bleu": 11.956602534707741, | |
| "eval_qa_exact_match": 0.16666666666666666, | |
| "eval_recite_bleu": 49.23892010841006, | |
| "eval_recite_exact_match": 0.056666666666666664, | |
| "step": 1188 | |
| }, | |
| { | |
| "epoch": 9.08, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0925, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 9.19, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0918, | |
| "step": 1204 | |
| }, | |
| { | |
| "epoch": 9.3, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0975, | |
| "step": 1218 | |
| }, | |
| { | |
| "epoch": 9.4, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0998, | |
| "step": 1232 | |
| }, | |
| { | |
| "epoch": 9.51, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0994, | |
| "step": 1246 | |
| }, | |
| { | |
| "epoch": 9.57, | |
| "eval_accuracy": 0.7412130033964095, | |
| "eval_loss": 0.40646126866340637, | |
| "eval_runtime": 9.2467, | |
| "eval_samples_per_second": 32.444, | |
| "eval_steps_per_second": 2.055, | |
| "step": 1254 | |
| }, | |
| { | |
| "epoch": 9.57, | |
| "eval_exact_match": 18.0, | |
| "eval_f1": 25.15901875901876, | |
| "eval_qa_bleu": 9.885421847181547, | |
| "eval_qa_exact_match": 0.16666666666666666, | |
| "eval_recite_bleu": 48.32004544781462, | |
| "eval_recite_exact_match": 0.06, | |
| "step": 1254 | |
| }, | |
| { | |
| "epoch": 9.62, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0972, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 9.73, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1074, | |
| "step": 1274 | |
| }, | |
| { | |
| "epoch": 9.83, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0998, | |
| "step": 1288 | |
| }, | |
| { | |
| "epoch": 9.94, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1094, | |
| "step": 1302 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 1310, | |
| "total_flos": 5.95362899625984e+16, | |
| "train_loss": 0.5945830505312855, | |
| "train_runtime": 4268.7073, | |
| "train_samples_per_second": 4.905, | |
| "train_steps_per_second": 0.307 | |
| } | |
| ], | |
| "logging_steps": 14, | |
| "max_steps": 1310, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 5.95362899625984e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |