| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 93, | |
| "global_step": 1860, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0001, | |
| "loss": 3.2142, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.0001, | |
| "loss": 2.3949, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 0.0001, | |
| "loss": 2.1979, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 0.0001, | |
| "loss": 2.0096, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6793333333333333, | |
| "eval_loss": 1.380240559577942, | |
| "eval_runtime": 7.9893, | |
| "eval_samples_per_second": 37.55, | |
| "eval_steps_per_second": 2.378, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_exact_match": 3.3333333333333335, | |
| "eval_f1": 5.83015873015873, | |
| "eval_qa_bleu": 1.6970908056732688, | |
| "eval_qa_exact_match": 0.023333333333333334, | |
| "eval_recite_bleu": 8.149431756948522, | |
| "eval_recite_exact_match": 0.0, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.0001, | |
| "loss": 1.7631, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 0.0001, | |
| "loss": 0.9463, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 0.0001, | |
| "loss": 0.8797, | |
| "step": 133 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 0.0001, | |
| "loss": 0.8213, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 0.0001, | |
| "loss": 0.7666, | |
| "step": 171 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7848627450980392, | |
| "eval_loss": 0.6686084270477295, | |
| "eval_runtime": 8.4979, | |
| "eval_samples_per_second": 35.303, | |
| "eval_steps_per_second": 2.236, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_exact_match": 4.666666666666667, | |
| "eval_f1": 7.87063492063492, | |
| "eval_qa_bleu": 4.158944460425248, | |
| "eval_qa_exact_match": 0.03333333333333333, | |
| "eval_recite_bleu": 12.887123438782695, | |
| "eval_recite_exact_match": 0.0033333333333333335, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 0.0001, | |
| "loss": 0.5917, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 0.0001, | |
| "loss": 0.3191, | |
| "step": 209 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 0.0001, | |
| "loss": 0.3119, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "learning_rate": 0.0001, | |
| "loss": 0.3058, | |
| "step": 247 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "learning_rate": 0.0001, | |
| "loss": 0.31, | |
| "step": 266 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8184183006535948, | |
| "eval_loss": 0.4719126522541046, | |
| "eval_runtime": 8.2778, | |
| "eval_samples_per_second": 36.242, | |
| "eval_steps_per_second": 2.295, | |
| "step": 279 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_exact_match": 8.0, | |
| "eval_f1": 11.731216931216931, | |
| "eval_qa_bleu": 3.2525404560535587, | |
| "eval_qa_exact_match": 0.06666666666666667, | |
| "eval_recite_bleu": 17.151706969674898, | |
| "eval_recite_exact_match": 0.006666666666666667, | |
| "step": 279 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 0.0001, | |
| "loss": 0.2319, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1567, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 3.47, | |
| "learning_rate": 0.0001, | |
| "loss": 0.163, | |
| "step": 323 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1494, | |
| "step": 342 | |
| }, | |
| { | |
| "epoch": 3.88, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1608, | |
| "step": 361 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8310980392156863, | |
| "eval_loss": 0.4038252830505371, | |
| "eval_runtime": 8.4283, | |
| "eval_samples_per_second": 35.594, | |
| "eval_steps_per_second": 2.254, | |
| "step": 372 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_exact_match": 9.666666666666666, | |
| "eval_f1": 13.638095238095238, | |
| "eval_qa_bleu": 11.09109473987401, | |
| "eval_qa_exact_match": 0.07, | |
| "eval_recite_bleu": 22.969082796041224, | |
| "eval_recite_exact_match": 0.02666666666666667, | |
| "step": 372 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "learning_rate": 0.0001, | |
| "loss": 0.139, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 4.29, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1032, | |
| "step": 399 | |
| }, | |
| { | |
| "epoch": 4.49, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1056, | |
| "step": 418 | |
| }, | |
| { | |
| "epoch": 4.7, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1006, | |
| "step": 437 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1101, | |
| "step": 456 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8371764705882353, | |
| "eval_loss": 0.37421324849128723, | |
| "eval_runtime": 8.4581, | |
| "eval_samples_per_second": 35.469, | |
| "eval_steps_per_second": 2.246, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_exact_match": 9.0, | |
| "eval_f1": 13.622222222222222, | |
| "eval_qa_bleu": 10.209463809998487, | |
| "eval_qa_exact_match": 0.06, | |
| "eval_recite_bleu": 25.315439762957883, | |
| "eval_recite_exact_match": 0.07, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "learning_rate": 0.0001, | |
| "loss": 0.097, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 5.31, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0835, | |
| "step": 494 | |
| }, | |
| { | |
| "epoch": 5.52, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0924, | |
| "step": 513 | |
| }, | |
| { | |
| "epoch": 5.72, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0816, | |
| "step": 532 | |
| }, | |
| { | |
| "epoch": 5.92, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0839, | |
| "step": 551 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8392679738562091, | |
| "eval_loss": 0.3733816146850586, | |
| "eval_runtime": 8.3329, | |
| "eval_samples_per_second": 36.002, | |
| "eval_steps_per_second": 2.28, | |
| "step": 558 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_exact_match": 10.333333333333334, | |
| "eval_f1": 13.493121693121692, | |
| "eval_qa_bleu": 10.482710761607542, | |
| "eval_qa_exact_match": 0.06, | |
| "eval_recite_bleu": 26.495515647579698, | |
| "eval_recite_exact_match": 0.09, | |
| "step": 558 | |
| }, | |
| { | |
| "epoch": 6.13, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0844, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 6.33, | |
| "learning_rate": 0.0001, | |
| "loss": 0.077, | |
| "step": 589 | |
| }, | |
| { | |
| "epoch": 6.54, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0759, | |
| "step": 608 | |
| }, | |
| { | |
| "epoch": 6.74, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0746, | |
| "step": 627 | |
| }, | |
| { | |
| "epoch": 6.95, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0743, | |
| "step": 646 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8404183006535948, | |
| "eval_loss": 0.3625450134277344, | |
| "eval_runtime": 7.971, | |
| "eval_samples_per_second": 37.637, | |
| "eval_steps_per_second": 2.384, | |
| "step": 651 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_exact_match": 12.333333333333334, | |
| "eval_f1": 16.304761904761904, | |
| "eval_qa_bleu": 11.931512542100624, | |
| "eval_qa_exact_match": 0.09, | |
| "eval_recite_bleu": 28.955751208977897, | |
| "eval_recite_exact_match": 0.09666666666666666, | |
| "step": 651 | |
| }, | |
| { | |
| "epoch": 7.15, | |
| "learning_rate": 0.0001, | |
| "loss": 0.069, | |
| "step": 665 | |
| }, | |
| { | |
| "epoch": 7.35, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0653, | |
| "step": 684 | |
| }, | |
| { | |
| "epoch": 7.56, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0678, | |
| "step": 703 | |
| }, | |
| { | |
| "epoch": 7.76, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0711, | |
| "step": 722 | |
| }, | |
| { | |
| "epoch": 7.97, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0756, | |
| "step": 741 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8399215686274509, | |
| "eval_loss": 0.36541667580604553, | |
| "eval_runtime": 8.1637, | |
| "eval_samples_per_second": 36.748, | |
| "eval_steps_per_second": 2.327, | |
| "step": 744 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_exact_match": 11.333333333333334, | |
| "eval_f1": 15.38253968253968, | |
| "eval_qa_bleu": 8.755756508741644, | |
| "eval_qa_exact_match": 0.09, | |
| "eval_recite_bleu": 28.742438290087694, | |
| "eval_recite_exact_match": 0.08666666666666667, | |
| "step": 744 | |
| }, | |
| { | |
| "epoch": 8.17, | |
| "learning_rate": 0.0001, | |
| "loss": 0.068, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 8.38, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0665, | |
| "step": 779 | |
| }, | |
| { | |
| "epoch": 8.58, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0679, | |
| "step": 798 | |
| }, | |
| { | |
| "epoch": 8.78, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0703, | |
| "step": 817 | |
| }, | |
| { | |
| "epoch": 8.99, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0694, | |
| "step": 836 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8399869281045752, | |
| "eval_loss": 0.37417492270469666, | |
| "eval_runtime": 8.3008, | |
| "eval_samples_per_second": 36.141, | |
| "eval_steps_per_second": 2.289, | |
| "step": 837 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_exact_match": 11.0, | |
| "eval_f1": 14.560317460317458, | |
| "eval_qa_bleu": 11.653366914054544, | |
| "eval_qa_exact_match": 0.07333333333333333, | |
| "eval_recite_bleu": 30.635594644467947, | |
| "eval_recite_exact_match": 0.09, | |
| "step": 837 | |
| }, | |
| { | |
| "epoch": 9.19, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0652, | |
| "step": 855 | |
| }, | |
| { | |
| "epoch": 9.4, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0643, | |
| "step": 874 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0662, | |
| "step": 893 | |
| }, | |
| { | |
| "epoch": 9.81, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0669, | |
| "step": 912 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8402875816993464, | |
| "eval_loss": 0.37116506695747375, | |
| "eval_runtime": 8.3018, | |
| "eval_samples_per_second": 36.137, | |
| "eval_steps_per_second": 2.289, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_exact_match": 11.0, | |
| "eval_f1": 14.528571428571425, | |
| "eval_qa_bleu": 10.001071696850747, | |
| "eval_qa_exact_match": 0.07666666666666666, | |
| "eval_recite_bleu": 25.0280074067495, | |
| "eval_recite_exact_match": 0.07333333333333333, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 10.01, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0687, | |
| "step": 931 | |
| }, | |
| { | |
| "epoch": 10.22, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0619, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 10.42, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0667, | |
| "step": 969 | |
| }, | |
| { | |
| "epoch": 10.62, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0688, | |
| "step": 988 | |
| }, | |
| { | |
| "epoch": 10.83, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0692, | |
| "step": 1007 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8396993464052288, | |
| "eval_loss": 0.3811953365802765, | |
| "eval_runtime": 8.6273, | |
| "eval_samples_per_second": 34.773, | |
| "eval_steps_per_second": 2.202, | |
| "step": 1023 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_exact_match": 8.0, | |
| "eval_f1": 12.665512265512264, | |
| "eval_qa_bleu": 9.029340244302077, | |
| "eval_qa_exact_match": 0.056666666666666664, | |
| "eval_recite_bleu": 27.692170109150023, | |
| "eval_recite_exact_match": 0.09, | |
| "step": 1023 | |
| }, | |
| { | |
| "epoch": 11.03, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0721, | |
| "step": 1026 | |
| }, | |
| { | |
| "epoch": 11.24, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0653, | |
| "step": 1045 | |
| }, | |
| { | |
| "epoch": 11.44, | |
| "learning_rate": 0.0001, | |
| "loss": 0.069, | |
| "step": 1064 | |
| }, | |
| { | |
| "epoch": 11.65, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0749, | |
| "step": 1083 | |
| }, | |
| { | |
| "epoch": 11.85, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0717, | |
| "step": 1102 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8394901960784313, | |
| "eval_loss": 0.3796521723270416, | |
| "eval_runtime": 8.4545, | |
| "eval_samples_per_second": 35.484, | |
| "eval_steps_per_second": 2.247, | |
| "step": 1116 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_exact_match": 11.666666666666666, | |
| "eval_f1": 15.368253968253963, | |
| "eval_qa_bleu": 14.544634955750682, | |
| "eval_qa_exact_match": 0.08666666666666667, | |
| "eval_recite_bleu": 28.00469452055886, | |
| "eval_recite_exact_match": 0.09666666666666666, | |
| "step": 1116 | |
| }, | |
| { | |
| "epoch": 12.05, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0747, | |
| "step": 1121 | |
| }, | |
| { | |
| "epoch": 12.26, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0671, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 12.46, | |
| "learning_rate": 0.0001, | |
| "loss": 0.076, | |
| "step": 1159 | |
| }, | |
| { | |
| "epoch": 12.67, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0793, | |
| "step": 1178 | |
| }, | |
| { | |
| "epoch": 12.87, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0762, | |
| "step": 1197 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8392679738562091, | |
| "eval_loss": 0.3891879618167877, | |
| "eval_runtime": 8.3465, | |
| "eval_samples_per_second": 35.943, | |
| "eval_steps_per_second": 2.276, | |
| "step": 1209 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_exact_match": 11.666666666666666, | |
| "eval_f1": 16.049206349206347, | |
| "eval_qa_bleu": 9.975889576329717, | |
| "eval_qa_exact_match": 0.09, | |
| "eval_recite_bleu": 28.604552918150617, | |
| "eval_recite_exact_match": 0.1, | |
| "step": 1209 | |
| }, | |
| { | |
| "epoch": 13.08, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0721, | |
| "step": 1216 | |
| }, | |
| { | |
| "epoch": 13.28, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0795, | |
| "step": 1235 | |
| }, | |
| { | |
| "epoch": 13.48, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0761, | |
| "step": 1254 | |
| }, | |
| { | |
| "epoch": 13.69, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0823, | |
| "step": 1273 | |
| }, | |
| { | |
| "epoch": 13.89, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0823, | |
| "step": 1292 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8383790849673203, | |
| "eval_loss": 0.3992992043495178, | |
| "eval_runtime": 8.5283, | |
| "eval_samples_per_second": 35.177, | |
| "eval_steps_per_second": 2.228, | |
| "step": 1302 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_exact_match": 9.666666666666666, | |
| "eval_f1": 12.499470899470898, | |
| "eval_qa_bleu": 8.034011406889757, | |
| "eval_qa_exact_match": 0.08333333333333333, | |
| "eval_recite_bleu": 27.04820432137949, | |
| "eval_recite_exact_match": 0.06333333333333334, | |
| "step": 1302 | |
| }, | |
| { | |
| "epoch": 14.1, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0756, | |
| "step": 1311 | |
| }, | |
| { | |
| "epoch": 14.3, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0731, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 14.51, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0795, | |
| "step": 1349 | |
| }, | |
| { | |
| "epoch": 14.71, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0797, | |
| "step": 1368 | |
| }, | |
| { | |
| "epoch": 14.91, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0789, | |
| "step": 1387 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8389281045751634, | |
| "eval_loss": 0.3946473002433777, | |
| "eval_runtime": 8.3483, | |
| "eval_samples_per_second": 35.935, | |
| "eval_steps_per_second": 2.276, | |
| "step": 1395 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_exact_match": 10.333333333333334, | |
| "eval_f1": 13.693650793650791, | |
| "eval_qa_bleu": 10.968521104903296, | |
| "eval_qa_exact_match": 0.08666666666666667, | |
| "eval_recite_bleu": 25.816615313359897, | |
| "eval_recite_exact_match": 0.08, | |
| "step": 1395 | |
| }, | |
| { | |
| "epoch": 15.12, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0728, | |
| "step": 1406 | |
| }, | |
| { | |
| "epoch": 15.32, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0695, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 15.53, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0699, | |
| "step": 1444 | |
| }, | |
| { | |
| "epoch": 15.73, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0771, | |
| "step": 1463 | |
| }, | |
| { | |
| "epoch": 15.94, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0737, | |
| "step": 1482 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8392941176470589, | |
| "eval_loss": 0.39267826080322266, | |
| "eval_runtime": 7.9822, | |
| "eval_samples_per_second": 37.584, | |
| "eval_steps_per_second": 2.38, | |
| "step": 1488 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_exact_match": 8.0, | |
| "eval_f1": 10.6, | |
| "eval_qa_bleu": 7.290561593538973, | |
| "eval_qa_exact_match": 0.06333333333333334, | |
| "eval_recite_bleu": 23.399757672969518, | |
| "eval_recite_exact_match": 0.07333333333333333, | |
| "step": 1488 | |
| }, | |
| { | |
| "epoch": 16.14, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0672, | |
| "step": 1501 | |
| }, | |
| { | |
| "epoch": 16.34, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0686, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 16.55, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0719, | |
| "step": 1539 | |
| }, | |
| { | |
| "epoch": 16.75, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0752, | |
| "step": 1558 | |
| }, | |
| { | |
| "epoch": 16.96, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0739, | |
| "step": 1577 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8381045751633986, | |
| "eval_loss": 0.3977350890636444, | |
| "eval_runtime": 8.3772, | |
| "eval_samples_per_second": 35.811, | |
| "eval_steps_per_second": 2.268, | |
| "step": 1581 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_exact_match": 11.0, | |
| "eval_f1": 14.11269841269841, | |
| "eval_qa_bleu": 11.007855081705829, | |
| "eval_qa_exact_match": 0.07333333333333333, | |
| "eval_recite_bleu": 29.431067552519558, | |
| "eval_recite_exact_match": 0.08, | |
| "step": 1581 | |
| }, | |
| { | |
| "epoch": 17.16, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0703, | |
| "step": 1596 | |
| }, | |
| { | |
| "epoch": 17.37, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0737, | |
| "step": 1615 | |
| }, | |
| { | |
| "epoch": 17.57, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0784, | |
| "step": 1634 | |
| }, | |
| { | |
| "epoch": 17.77, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0729, | |
| "step": 1653 | |
| }, | |
| { | |
| "epoch": 17.98, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0741, | |
| "step": 1672 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8378562091503268, | |
| "eval_loss": 0.40596938133239746, | |
| "eval_runtime": 8.2411, | |
| "eval_samples_per_second": 36.403, | |
| "eval_steps_per_second": 2.306, | |
| "step": 1674 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_exact_match": 8.333333333333334, | |
| "eval_f1": 11.382539682539681, | |
| "eval_qa_bleu": 7.279783117269195, | |
| "eval_qa_exact_match": 0.056666666666666664, | |
| "eval_recite_bleu": 23.09547487104553, | |
| "eval_recite_exact_match": 0.07333333333333333, | |
| "step": 1674 | |
| }, | |
| { | |
| "epoch": 18.18, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0737, | |
| "step": 1691 | |
| }, | |
| { | |
| "epoch": 18.39, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0666, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 18.59, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0764, | |
| "step": 1729 | |
| }, | |
| { | |
| "epoch": 18.8, | |
| "learning_rate": 0.0001, | |
| "loss": 0.072, | |
| "step": 1748 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0741, | |
| "step": 1767 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8388888888888889, | |
| "eval_loss": 0.40468740463256836, | |
| "eval_runtime": 7.9826, | |
| "eval_samples_per_second": 37.582, | |
| "eval_steps_per_second": 2.38, | |
| "step": 1767 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_exact_match": 12.0, | |
| "eval_f1": 15.778571428571425, | |
| "eval_qa_bleu": 12.003446246092967, | |
| "eval_qa_exact_match": 0.08666666666666667, | |
| "eval_recite_bleu": 31.433479842615494, | |
| "eval_recite_exact_match": 0.11666666666666667, | |
| "step": 1767 | |
| }, | |
| { | |
| "epoch": 19.2, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0622, | |
| "step": 1786 | |
| }, | |
| { | |
| "epoch": 19.41, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0684, | |
| "step": 1805 | |
| }, | |
| { | |
| "epoch": 19.61, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0721, | |
| "step": 1824 | |
| }, | |
| { | |
| "epoch": 19.82, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0715, | |
| "step": 1843 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8390196078431372, | |
| "eval_loss": 0.39820176362991333, | |
| "eval_runtime": 8.1892, | |
| "eval_samples_per_second": 36.634, | |
| "eval_steps_per_second": 2.32, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_exact_match": 11.333333333333334, | |
| "eval_f1": 14.327777777777778, | |
| "eval_qa_bleu": 10.816198494730774, | |
| "eval_qa_exact_match": 0.08, | |
| "eval_recite_bleu": 26.73749978175143, | |
| "eval_recite_exact_match": 0.1, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 1860, | |
| "total_flos": 4.95263869850112e+16, | |
| "train_loss": 0.24169133318367825, | |
| "train_runtime": 4163.8318, | |
| "train_samples_per_second": 7.123, | |
| "train_steps_per_second": 0.447 | |
| } | |
| ], | |
| "logging_steps": 19, | |
| "max_steps": 1860, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 4.95263869850112e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |