| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 1452, | |
| "global_step": 29040, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 3e-05, | |
| "loss": 2.2721, | |
| "step": 291 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 3e-05, | |
| "loss": 2.0539, | |
| "step": 582 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 3e-05, | |
| "loss": 2.0079, | |
| "step": 873 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 3e-05, | |
| "loss": 1.9686, | |
| "step": 1164 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.668149394347241, | |
| "eval_loss": 1.5662202835083008, | |
| "eval_runtime": 137.7572, | |
| "eval_samples_per_second": 36.296, | |
| "eval_steps_per_second": 2.272, | |
| "step": 1452 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_exact_match": 9.04, | |
| "eval_f1": 13.087046780017364, | |
| "eval_qa_bleu": 5.297766450359973, | |
| "eval_qa_exact_match": 0.085, | |
| "eval_recite_bleu": 14.716037077738477, | |
| "eval_recite_exact_match": 0.0, | |
| "step": 1452 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 3e-05, | |
| "loss": 1.9076, | |
| "step": 1455 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 3e-05, | |
| "loss": 1.5211, | |
| "step": 1746 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 3e-05, | |
| "loss": 1.5039, | |
| "step": 2037 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 3e-05, | |
| "loss": 1.4834, | |
| "step": 2328 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 3e-05, | |
| "loss": 1.4604, | |
| "step": 2619 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6914514131897712, | |
| "eval_loss": 1.206884741783142, | |
| "eval_runtime": 139.7305, | |
| "eval_samples_per_second": 35.783, | |
| "eval_steps_per_second": 2.24, | |
| "step": 2904 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_exact_match": 11.64, | |
| "eval_f1": 16.917593370997537, | |
| "eval_qa_bleu": 7.627957176996819, | |
| "eval_qa_exact_match": 0.1106, | |
| "eval_recite_bleu": 18.356743056908567, | |
| "eval_recite_exact_match": 0.0, | |
| "step": 2904 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 3e-05, | |
| "loss": 1.4368, | |
| "step": 2910 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 3e-05, | |
| "loss": 1.0705, | |
| "step": 3201 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 3e-05, | |
| "loss": 1.0685, | |
| "step": 3492 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 3e-05, | |
| "loss": 1.0554, | |
| "step": 3783 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 3e-05, | |
| "loss": 1.0522, | |
| "step": 4074 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7148896366083446, | |
| "eval_loss": 0.9045500159263611, | |
| "eval_runtime": 141.6455, | |
| "eval_samples_per_second": 35.299, | |
| "eval_steps_per_second": 2.21, | |
| "step": 4356 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_exact_match": 12.84, | |
| "eval_f1": 18.555356874635468, | |
| "eval_qa_bleu": 8.395879921702353, | |
| "eval_qa_exact_match": 0.1188, | |
| "eval_recite_bleu": 20.824221134285214, | |
| "eval_recite_exact_match": 0.0, | |
| "step": 4356 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 3e-05, | |
| "loss": 1.0375, | |
| "step": 4365 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "learning_rate": 3e-05, | |
| "loss": 0.7373, | |
| "step": 4656 | |
| }, | |
| { | |
| "epoch": 3.41, | |
| "learning_rate": 3e-05, | |
| "loss": 0.7497, | |
| "step": 4947 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "learning_rate": 3e-05, | |
| "loss": 0.7479, | |
| "step": 5238 | |
| }, | |
| { | |
| "epoch": 3.81, | |
| "learning_rate": 3e-05, | |
| "loss": 0.7455, | |
| "step": 5529 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7340508748317631, | |
| "eval_loss": 0.6843858957290649, | |
| "eval_runtime": 143.5169, | |
| "eval_samples_per_second": 34.839, | |
| "eval_steps_per_second": 2.181, | |
| "step": 5808 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_exact_match": 13.44, | |
| "eval_f1": 19.418847592695993, | |
| "eval_qa_bleu": 8.731752840499793, | |
| "eval_qa_exact_match": 0.1256, | |
| "eval_recite_bleu": 24.11683390270685, | |
| "eval_recite_exact_match": 0.0002, | |
| "step": 5808 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 3e-05, | |
| "loss": 0.7226, | |
| "step": 5820 | |
| }, | |
| { | |
| "epoch": 4.21, | |
| "learning_rate": 3e-05, | |
| "loss": 0.5117, | |
| "step": 6111 | |
| }, | |
| { | |
| "epoch": 4.41, | |
| "learning_rate": 3e-05, | |
| "loss": 0.5208, | |
| "step": 6402 | |
| }, | |
| { | |
| "epoch": 4.61, | |
| "learning_rate": 3e-05, | |
| "loss": 0.5271, | |
| "step": 6693 | |
| }, | |
| { | |
| "epoch": 4.81, | |
| "learning_rate": 3e-05, | |
| "loss": 0.5307, | |
| "step": 6984 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7474745625841185, | |
| "eval_loss": 0.5420497059822083, | |
| "eval_runtime": 144.0475, | |
| "eval_samples_per_second": 34.711, | |
| "eval_steps_per_second": 2.173, | |
| "step": 7260 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_exact_match": 15.08, | |
| "eval_f1": 21.4491845879152, | |
| "eval_qa_bleu": 10.89471390420015, | |
| "eval_qa_exact_match": 0.1434, | |
| "eval_recite_bleu": 27.62075386163555, | |
| "eval_recite_exact_match": 0.0014, | |
| "step": 7260 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "learning_rate": 3e-05, | |
| "loss": 0.5165, | |
| "step": 7275 | |
| }, | |
| { | |
| "epoch": 5.21, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3624, | |
| "step": 7566 | |
| }, | |
| { | |
| "epoch": 5.41, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3748, | |
| "step": 7857 | |
| }, | |
| { | |
| "epoch": 5.61, | |
| "learning_rate": 3e-05, | |
| "loss": 0.383, | |
| "step": 8148 | |
| }, | |
| { | |
| "epoch": 5.81, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3796, | |
| "step": 8439 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7560010767160161, | |
| "eval_loss": 0.4608670473098755, | |
| "eval_runtime": 144.2314, | |
| "eval_samples_per_second": 34.667, | |
| "eval_steps_per_second": 2.17, | |
| "step": 8712 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_exact_match": 16.22, | |
| "eval_f1": 22.5212141898349, | |
| "eval_qa_bleu": 11.517856911949949, | |
| "eval_qa_exact_match": 0.1524, | |
| "eval_recite_bleu": 32.1802083615024, | |
| "eval_recite_exact_match": 0.0044, | |
| "step": 8712 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3803, | |
| "step": 8730 | |
| }, | |
| { | |
| "epoch": 6.21, | |
| "learning_rate": 3e-05, | |
| "loss": 0.27, | |
| "step": 9021 | |
| }, | |
| { | |
| "epoch": 6.41, | |
| "learning_rate": 3e-05, | |
| "loss": 0.28, | |
| "step": 9312 | |
| }, | |
| { | |
| "epoch": 6.61, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2889, | |
| "step": 9603 | |
| }, | |
| { | |
| "epoch": 6.81, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2912, | |
| "step": 9894 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.7603886944818304, | |
| "eval_loss": 0.43113213777542114, | |
| "eval_runtime": 144.1381, | |
| "eval_samples_per_second": 34.689, | |
| "eval_steps_per_second": 2.172, | |
| "step": 10164 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_exact_match": 17.64, | |
| "eval_f1": 24.23915265275915, | |
| "eval_qa_bleu": 12.404229060978194, | |
| "eval_qa_exact_match": 0.1648, | |
| "eval_recite_bleu": 37.23348279932964, | |
| "eval_recite_exact_match": 0.0102, | |
| "step": 10164 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2835, | |
| "step": 10185 | |
| }, | |
| { | |
| "epoch": 7.21, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2131, | |
| "step": 10476 | |
| }, | |
| { | |
| "epoch": 7.42, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2213, | |
| "step": 10767 | |
| }, | |
| { | |
| "epoch": 7.62, | |
| "learning_rate": 3e-05, | |
| "loss": 0.226, | |
| "step": 11058 | |
| }, | |
| { | |
| "epoch": 7.82, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2282, | |
| "step": 11349 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7626543741588157, | |
| "eval_loss": 0.41835975646972656, | |
| "eval_runtime": 144.0991, | |
| "eval_samples_per_second": 34.698, | |
| "eval_steps_per_second": 2.172, | |
| "step": 11616 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_exact_match": 17.78, | |
| "eval_f1": 25.32648324184451, | |
| "eval_qa_bleu": 12.460459536013193, | |
| "eval_qa_exact_match": 0.1676, | |
| "eval_recite_bleu": 40.14518456576004, | |
| "eval_recite_exact_match": 0.0168, | |
| "step": 11616 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1645, | |
| "step": 11640 | |
| }, | |
| { | |
| "epoch": 8.22, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1768, | |
| "step": 11931 | |
| }, | |
| { | |
| "epoch": 8.42, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1851, | |
| "step": 12222 | |
| }, | |
| { | |
| "epoch": 8.62, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1895, | |
| "step": 12513 | |
| }, | |
| { | |
| "epoch": 8.82, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1905, | |
| "step": 12804 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.7639776581426648, | |
| "eval_loss": 0.4136393666267395, | |
| "eval_runtime": 146.8079, | |
| "eval_samples_per_second": 34.058, | |
| "eval_steps_per_second": 2.132, | |
| "step": 13068 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_exact_match": 18.58, | |
| "eval_f1": 26.144983963782675, | |
| "eval_qa_bleu": 12.723130940050861, | |
| "eval_qa_exact_match": 0.1766, | |
| "eval_recite_bleu": 44.36820281112791, | |
| "eval_recite_exact_match": 0.0212, | |
| "step": 13068 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1919, | |
| "step": 13095 | |
| }, | |
| { | |
| "epoch": 9.22, | |
| "learning_rate": 3e-05, | |
| "loss": 0.156, | |
| "step": 13386 | |
| }, | |
| { | |
| "epoch": 9.42, | |
| "learning_rate": 3e-05, | |
| "loss": 0.158, | |
| "step": 13677 | |
| }, | |
| { | |
| "epoch": 9.62, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1656, | |
| "step": 13968 | |
| }, | |
| { | |
| "epoch": 9.82, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1687, | |
| "step": 14259 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.7647776581426649, | |
| "eval_loss": 0.41753506660461426, | |
| "eval_runtime": 149.0935, | |
| "eval_samples_per_second": 33.536, | |
| "eval_steps_per_second": 2.099, | |
| "step": 14520 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_exact_match": 19.8, | |
| "eval_f1": 27.235079497903637, | |
| "eval_qa_bleu": 13.874335090146856, | |
| "eval_qa_exact_match": 0.1886, | |
| "eval_recite_bleu": 46.53984717815193, | |
| "eval_recite_exact_match": 0.0276, | |
| "step": 14520 | |
| }, | |
| { | |
| "epoch": 10.02, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1684, | |
| "step": 14550 | |
| }, | |
| { | |
| "epoch": 10.22, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1396, | |
| "step": 14841 | |
| }, | |
| { | |
| "epoch": 10.42, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1457, | |
| "step": 15132 | |
| }, | |
| { | |
| "epoch": 10.62, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1542, | |
| "step": 15423 | |
| }, | |
| { | |
| "epoch": 10.82, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1553, | |
| "step": 15714 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.7651154777927321, | |
| "eval_loss": 0.4211733937263489, | |
| "eval_runtime": 150.5664, | |
| "eval_samples_per_second": 33.208, | |
| "eval_steps_per_second": 2.079, | |
| "step": 15972 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_exact_match": 19.1, | |
| "eval_f1": 26.696266939458862, | |
| "eval_qa_bleu": 14.962573389382232, | |
| "eval_qa_exact_match": 0.1814, | |
| "eval_recite_bleu": 46.889182609591046, | |
| "eval_recite_exact_match": 0.032, | |
| "step": 15972 | |
| }, | |
| { | |
| "epoch": 11.02, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1549, | |
| "step": 16005 | |
| }, | |
| { | |
| "epoch": 11.22, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1335, | |
| "step": 16296 | |
| }, | |
| { | |
| "epoch": 11.42, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1381, | |
| "step": 16587 | |
| }, | |
| { | |
| "epoch": 11.62, | |
| "learning_rate": 3e-05, | |
| "loss": 0.143, | |
| "step": 16878 | |
| }, | |
| { | |
| "epoch": 11.82, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1447, | |
| "step": 17169 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.7653222072678331, | |
| "eval_loss": 0.4283375144004822, | |
| "eval_runtime": 152.1173, | |
| "eval_samples_per_second": 32.869, | |
| "eval_steps_per_second": 2.058, | |
| "step": 17424 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_exact_match": 19.7, | |
| "eval_f1": 27.086380949947664, | |
| "eval_qa_bleu": 13.161442853987554, | |
| "eval_qa_exact_match": 0.1884, | |
| "eval_recite_bleu": 48.94647260503517, | |
| "eval_recite_exact_match": 0.0318, | |
| "step": 17424 | |
| }, | |
| { | |
| "epoch": 12.02, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1449, | |
| "step": 17460 | |
| }, | |
| { | |
| "epoch": 12.23, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1267, | |
| "step": 17751 | |
| }, | |
| { | |
| "epoch": 12.43, | |
| "learning_rate": 3e-05, | |
| "loss": 0.131, | |
| "step": 18042 | |
| }, | |
| { | |
| "epoch": 12.63, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1366, | |
| "step": 18333 | |
| }, | |
| { | |
| "epoch": 12.83, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1388, | |
| "step": 18624 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.7655897711978465, | |
| "eval_loss": 0.4287155270576477, | |
| "eval_runtime": 152.6887, | |
| "eval_samples_per_second": 32.746, | |
| "eval_steps_per_second": 2.05, | |
| "step": 18876 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_exact_match": 20.12, | |
| "eval_f1": 27.673234730738653, | |
| "eval_qa_bleu": 14.175044837001105, | |
| "eval_qa_exact_match": 0.191, | |
| "eval_recite_bleu": 49.79523539787241, | |
| "eval_recite_exact_match": 0.0446, | |
| "step": 18876 | |
| }, | |
| { | |
| "epoch": 13.03, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1393, | |
| "step": 18915 | |
| }, | |
| { | |
| "epoch": 13.23, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1243, | |
| "step": 19206 | |
| }, | |
| { | |
| "epoch": 13.43, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1277, | |
| "step": 19497 | |
| }, | |
| { | |
| "epoch": 13.63, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1324, | |
| "step": 19788 | |
| }, | |
| { | |
| "epoch": 13.83, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1329, | |
| "step": 20079 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.7656621803499327, | |
| "eval_loss": 0.4349338412284851, | |
| "eval_runtime": 153.0165, | |
| "eval_samples_per_second": 32.676, | |
| "eval_steps_per_second": 2.046, | |
| "step": 20328 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_exact_match": 19.74, | |
| "eval_f1": 27.539494437309155, | |
| "eval_qa_bleu": 14.653456371114471, | |
| "eval_qa_exact_match": 0.1874, | |
| "eval_recite_bleu": 49.667204132453634, | |
| "eval_recite_exact_match": 0.0442, | |
| "step": 20328 | |
| }, | |
| { | |
| "epoch": 14.03, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1324, | |
| "step": 20370 | |
| }, | |
| { | |
| "epoch": 14.23, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1191, | |
| "step": 20661 | |
| }, | |
| { | |
| "epoch": 14.43, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1228, | |
| "step": 20952 | |
| }, | |
| { | |
| "epoch": 14.63, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1265, | |
| "step": 21243 | |
| }, | |
| { | |
| "epoch": 14.83, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1292, | |
| "step": 21534 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.7657300134589502, | |
| "eval_loss": 0.43529966473579407, | |
| "eval_runtime": 152.7159, | |
| "eval_samples_per_second": 32.741, | |
| "eval_steps_per_second": 2.05, | |
| "step": 21780 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_exact_match": 20.14, | |
| "eval_f1": 28.059788808027182, | |
| "eval_qa_bleu": 14.397818558683056, | |
| "eval_qa_exact_match": 0.1928, | |
| "eval_recite_bleu": 49.57947130892122, | |
| "eval_recite_exact_match": 0.0444, | |
| "step": 21780 | |
| }, | |
| { | |
| "epoch": 15.03, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1313, | |
| "step": 21825 | |
| }, | |
| { | |
| "epoch": 15.23, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1185, | |
| "step": 22116 | |
| }, | |
| { | |
| "epoch": 15.43, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1199, | |
| "step": 22407 | |
| }, | |
| { | |
| "epoch": 15.63, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1226, | |
| "step": 22698 | |
| }, | |
| { | |
| "epoch": 15.83, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1267, | |
| "step": 22989 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.7659200538358009, | |
| "eval_loss": 0.438266783952713, | |
| "eval_runtime": 153.2915, | |
| "eval_samples_per_second": 32.618, | |
| "eval_steps_per_second": 2.042, | |
| "step": 23232 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_exact_match": 20.64, | |
| "eval_f1": 28.191361890575656, | |
| "eval_qa_bleu": 14.695055084130074, | |
| "eval_qa_exact_match": 0.1918, | |
| "eval_recite_bleu": 50.268257039645256, | |
| "eval_recite_exact_match": 0.0478, | |
| "step": 23232 | |
| }, | |
| { | |
| "epoch": 16.03, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1133, | |
| "step": 23280 | |
| }, | |
| { | |
| "epoch": 16.23, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1145, | |
| "step": 23571 | |
| }, | |
| { | |
| "epoch": 16.43, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1174, | |
| "step": 23862 | |
| }, | |
| { | |
| "epoch": 16.63, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1218, | |
| "step": 24153 | |
| }, | |
| { | |
| "epoch": 16.83, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1251, | |
| "step": 24444 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.7660632570659489, | |
| "eval_loss": 0.4416392147541046, | |
| "eval_runtime": 148.1371, | |
| "eval_samples_per_second": 33.753, | |
| "eval_steps_per_second": 2.113, | |
| "step": 24684 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_exact_match": 20.62, | |
| "eval_f1": 28.101175315458292, | |
| "eval_qa_bleu": 14.458032503572724, | |
| "eval_qa_exact_match": 0.1966, | |
| "eval_recite_bleu": 51.2644528074332, | |
| "eval_recite_exact_match": 0.0496, | |
| "step": 24684 | |
| }, | |
| { | |
| "epoch": 17.04, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1227, | |
| "step": 24735 | |
| }, | |
| { | |
| "epoch": 17.24, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1117, | |
| "step": 25026 | |
| }, | |
| { | |
| "epoch": 17.44, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1163, | |
| "step": 25317 | |
| }, | |
| { | |
| "epoch": 17.64, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1209, | |
| "step": 25608 | |
| }, | |
| { | |
| "epoch": 17.84, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1201, | |
| "step": 25899 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.7658912516823688, | |
| "eval_loss": 0.4466552734375, | |
| "eval_runtime": 149.6031, | |
| "eval_samples_per_second": 33.422, | |
| "eval_steps_per_second": 2.092, | |
| "step": 26136 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_exact_match": 19.72, | |
| "eval_f1": 27.705882516354148, | |
| "eval_qa_bleu": 13.41419171329321, | |
| "eval_qa_exact_match": 0.189, | |
| "eval_recite_bleu": 49.4786180883858, | |
| "eval_recite_exact_match": 0.0434, | |
| "step": 26136 | |
| }, | |
| { | |
| "epoch": 18.04, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1215, | |
| "step": 26190 | |
| }, | |
| { | |
| "epoch": 18.24, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1111, | |
| "step": 26481 | |
| }, | |
| { | |
| "epoch": 18.44, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1164, | |
| "step": 26772 | |
| }, | |
| { | |
| "epoch": 18.64, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1162, | |
| "step": 27063 | |
| }, | |
| { | |
| "epoch": 18.84, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1186, | |
| "step": 27354 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.7659983849259758, | |
| "eval_loss": 0.4507902264595032, | |
| "eval_runtime": 151.1093, | |
| "eval_samples_per_second": 33.089, | |
| "eval_steps_per_second": 2.071, | |
| "step": 27588 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_exact_match": 20.78, | |
| "eval_f1": 28.407049206780048, | |
| "eval_qa_bleu": 14.348972564948328, | |
| "eval_qa_exact_match": 0.195, | |
| "eval_recite_bleu": 50.89618814085128, | |
| "eval_recite_exact_match": 0.0544, | |
| "step": 27588 | |
| }, | |
| { | |
| "epoch": 19.04, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1198, | |
| "step": 27645 | |
| }, | |
| { | |
| "epoch": 19.24, | |
| "learning_rate": 3e-05, | |
| "loss": 0.109, | |
| "step": 27936 | |
| }, | |
| { | |
| "epoch": 19.44, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1136, | |
| "step": 28227 | |
| }, | |
| { | |
| "epoch": 19.64, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1153, | |
| "step": 28518 | |
| }, | |
| { | |
| "epoch": 19.84, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1176, | |
| "step": 28809 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.7660672947510094, | |
| "eval_loss": 0.4521976709365845, | |
| "eval_runtime": 152.6418, | |
| "eval_samples_per_second": 32.756, | |
| "eval_steps_per_second": 2.051, | |
| "step": 29040 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_exact_match": 20.7, | |
| "eval_f1": 28.598247849322437, | |
| "eval_qa_bleu": 14.395309016288971, | |
| "eval_qa_exact_match": 0.1954, | |
| "eval_recite_bleu": 51.41424385892475, | |
| "eval_recite_exact_match": 0.0504, | |
| "step": 29040 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 29040, | |
| "total_flos": 1.326010952896512e+18, | |
| "train_loss": 0.023505237591824914, | |
| "train_runtime": 14139.9896, | |
| "train_samples_per_second": 32.849, | |
| "train_steps_per_second": 2.054 | |
| } | |
| ], | |
| "logging_steps": 291, | |
| "max_steps": 29040, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 1.326010952896512e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |