{ "best_global_step": null, "best_metric": null, "best_model_checkpoint": null, "epoch": 3.0, "eval_steps": 500, "global_step": 1137, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.052857614800132145, "grad_norm": 2.238008737564087, "learning_rate": 1.9e-05, "loss": 8.4999755859375, "step": 20 }, { "epoch": 0.10571522960026429, "grad_norm": 1.7257620096206665, "learning_rate": 3.9e-05, "loss": 7.775839233398438, "step": 40 }, { "epoch": 0.15857284440039643, "grad_norm": 1.491312026977539, "learning_rate": 5.899999999999999e-05, "loss": 7.235868835449219, "step": 60 }, { "epoch": 0.21143045920052858, "grad_norm": 0.9560289978981018, "learning_rate": 7.899999999999998e-05, "loss": 6.741374206542969, "step": 80 }, { "epoch": 0.26428807400066073, "grad_norm": 0.9636826515197754, "learning_rate": 9.9e-05, "loss": 6.289650726318359, "step": 100 }, { "epoch": 0.31714568880079286, "grad_norm": 1.0701491832733154, "learning_rate": 0.00011899999999999999, "loss": 6.034837341308593, "step": 120 }, { "epoch": 0.370003303600925, "grad_norm": 1.4025605916976929, "learning_rate": 0.000139, "loss": 5.823862075805664, "step": 140 }, { "epoch": 0.42286091840105716, "grad_norm": 0.7738357186317444, "learning_rate": 0.000159, "loss": 5.675333404541016, "step": 160 }, { "epoch": 0.4757185332011893, "grad_norm": 0.8328272700309753, "learning_rate": 0.000179, "loss": 5.511610412597657, "step": 180 }, { "epoch": 0.5285761480013215, "grad_norm": 0.8638654947280884, "learning_rate": 0.00019899999999999999, "loss": 5.4444831848144535, "step": 200 }, { "epoch": 0.5814337628014535, "grad_norm": 0.8530329465866089, "learning_rate": 0.00021899999999999998, "loss": 5.359589385986328, "step": 220 }, { "epoch": 0.6342913776015857, "grad_norm": 0.7593206167221069, "learning_rate": 0.00023899999999999998, "loss": 5.265158843994141, "step": 240 }, { "epoch": 0.6871489924017179, "grad_norm": 0.8268908262252808, "learning_rate": 0.00025899999999999995, "loss": 5.213626861572266, "step": 260 }, { "epoch": 0.74000660720185, "grad_norm": 0.9356043338775635, "learning_rate": 0.000279, "loss": 5.14818115234375, "step": 280 }, { "epoch": 0.7928642220019821, "grad_norm": 1.2305967807769775, "learning_rate": 0.000299, "loss": 5.070086669921875, "step": 300 }, { "epoch": 0.8457218368021143, "grad_norm": 0.8263030052185059, "learning_rate": 0.0002996187294797388, "loss": 5.018085861206055, "step": 320 }, { "epoch": 0.8985794516022465, "grad_norm": 1.0647673606872559, "learning_rate": 0.00029839578094886874, "loss": 4.944237518310547, "step": 340 }, { "epoch": 0.9514370664023786, "grad_norm": 1.057557463645935, "learning_rate": 0.00029633698612563077, "loss": 4.900520324707031, "step": 360 }, { "epoch": 1.0026428807400065, "grad_norm": 0.9045482277870178, "learning_rate": 0.00029345394126976173, "loss": 4.840346908569336, "step": 380 }, { "epoch": 1.0555004955401388, "grad_norm": 1.0554182529449463, "learning_rate": 0.0002897628852685002, "loss": 4.7415111541748045, "step": 400 }, { "epoch": 1.1083581103402709, "grad_norm": 0.8702770471572876, "learning_rate": 0.00028528460817028195, "loss": 4.725639343261719, "step": 420 }, { "epoch": 1.161215725140403, "grad_norm": 0.9115422964096069, "learning_rate": 0.0002800443340838033, "loss": 4.660248947143555, "step": 440 }, { "epoch": 1.2140733399405352, "grad_norm": 1.0200954675674438, "learning_rate": 0.00027407157910202976, "loss": 4.621082305908203, "step": 460 }, { "epoch": 1.2669309547406673, "grad_norm": 0.972144603729248, "learning_rate": 0.0002673999850513988, "loss": 4.590683364868164, "step": 480 }, { "epoch": 1.3197885695407994, "grad_norm": 0.8780707120895386, "learning_rate": 0.0002600671300026329, "loss": 4.533613204956055, "step": 500 }, { "epoch": 1.3726461843409317, "grad_norm": 0.8141133785247803, "learning_rate": 0.0002521143166104692, "loss": 4.505572891235351, "step": 520 }, { "epoch": 1.4255037991410637, "grad_norm": 0.8241302371025085, "learning_rate": 0.00024358633947449282, "loss": 4.485667419433594, "step": 540 }, { "epoch": 1.478361413941196, "grad_norm": 0.9647106528282166, "learning_rate": 0.00023453123283142488, "loss": 4.4454345703125, "step": 560 }, { "epoch": 1.531219028741328, "grad_norm": 0.9234511256217957, "learning_rate": 0.000225, "loss": 4.416561889648437, "step": 580 }, { "epoch": 1.5840766435414602, "grad_norm": 0.9763705134391785, "learning_rate": 0.00021504632610234826, "loss": 4.387359237670898, "step": 600 }, { "epoch": 1.6369342583415922, "grad_norm": 0.8934098482131958, "learning_rate": 0.00020472627567999242, "loss": 4.348335266113281, "step": 620 }, { "epoch": 1.6897918731417245, "grad_norm": 0.9640178680419922, "learning_rate": 0.00019409797690765154, "loss": 4.313744735717774, "step": 640 }, { "epoch": 1.7426494879418566, "grad_norm": 0.9510703682899475, "learning_rate": 0.0001832212941835331, "loss": 4.288594818115234, "step": 660 }, { "epoch": 1.7955071027419889, "grad_norm": 0.8832435011863708, "learning_rate": 0.00017215749094026366, "loss": 4.275389862060547, "step": 680 }, { "epoch": 1.848364717542121, "grad_norm": 0.9560243487358093, "learning_rate": 0.0001609688845756914, "loss": 4.2810821533203125, "step": 700 }, { "epoch": 1.901222332342253, "grad_norm": 1.0186476707458496, "learning_rate": 0.00014971849544717942, "loss": 4.235449600219726, "step": 720 }, { "epoch": 1.954079947142385, "grad_norm": 1.1411651372909546, "learning_rate": 0.0001384696919064467, "loss": 4.205095291137695, "step": 740 }, { "epoch": 2.005285761480013, "grad_norm": 1.0430322885513306, "learning_rate": 0.00012728583337431353, "loss": 4.177823257446289, "step": 760 }, { "epoch": 2.0581433762801455, "grad_norm": 1.0838340520858765, "learning_rate": 0.00011622991346575104, "loss": 4.1247001647949215, "step": 780 }, { "epoch": 2.1110009910802776, "grad_norm": 0.9293513894081116, "learning_rate": 0.00010536420517534995, "loss": 4.097352600097656, "step": 800 }, { "epoch": 2.1638586058804097, "grad_norm": 1.2049766778945923, "learning_rate": 9.474991012171693e-05, "loss": 4.0907234191894535, "step": 820 }, { "epoch": 2.2167162206805417, "grad_norm": 0.975261926651001, "learning_rate": 8.444681382644803e-05, "loss": 4.077187347412109, "step": 840 }, { "epoch": 2.269573835480674, "grad_norm": 0.9978065490722656, "learning_rate": 7.451294896933534e-05, "loss": 4.0670654296875, "step": 860 }, { "epoch": 2.322431450280806, "grad_norm": 1.1385101079940796, "learning_rate": 6.500426851653885e-05, "loss": 4.044752502441407, "step": 880 }, { "epoch": 2.3752890650809384, "grad_norm": 1.1333826780319214, "learning_rate": 5.59743305628436e-05, "loss": 4.038077545166016, "step": 900 }, { "epoch": 2.4281466798810705, "grad_norm": 0.99233478307724, "learning_rate": 4.747399666314473e-05, "loss": 4.00207290649414, "step": 920 }, { "epoch": 2.4810042946812025, "grad_norm": 1.0230666399002075, "learning_rate": 3.9551145352320425e-05, "loss": 4.000417327880859, "step": 940 }, { "epoch": 2.5338619094813346, "grad_norm": 0.9381811618804932, "learning_rate": 3.225040246710802e-05, "loss": 4.000282669067383, "step": 960 }, { "epoch": 2.5867195242814667, "grad_norm": 1.0343655347824097, "learning_rate": 2.5612889788956553e-05, "loss": 4.005098342895508, "step": 980 }, { "epoch": 2.6395771390815987, "grad_norm": 1.0831797122955322, "learning_rate": 1.9675993423637447e-05, "loss": 4.0017749786376955, "step": 1000 }, { "epoch": 2.692434753881731, "grad_norm": 0.9464903473854065, "learning_rate": 1.4473153222225864e-05, "loss": 4.004263687133789, "step": 1020 }, { "epoch": 2.7452923686818633, "grad_norm": 0.9014742374420166, "learning_rate": 1.0033674429549738e-05, "loss": 3.969345474243164, "step": 1040 }, { "epoch": 2.7981499834819954, "grad_norm": 0.8502050042152405, "learning_rate": 6.382562621004678e-06, "loss": 3.9854698181152344, "step": 1060 }, { "epoch": 2.8510075982821275, "grad_norm": 0.8504449129104614, "learning_rate": 3.54038285746151e-06, "loss": 3.9741233825683593, "step": 1080 }, { "epoch": 2.9038652130822595, "grad_norm": 0.8589563965797424, "learning_rate": 1.5231438515831807e-06, "loss": 3.9875431060791016, "step": 1100 }, { "epoch": 2.956722827882392, "grad_norm": 0.8136894702911377, "learning_rate": 3.422077979893911e-07, "loss": 3.9866878509521486, "step": 1120 }, { "epoch": 3.0, "step": 1137, "total_flos": 2813778267734016.0, "train_loss": 4.764855538320416, "train_runtime": 782.9505, "train_samples_per_second": 46.386, "train_steps_per_second": 1.452 } ], "logging_steps": 20, "max_steps": 1137, "num_input_tokens_seen": 0, "num_train_epochs": 3, "save_steps": 200, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 2813778267734016.0, "train_batch_size": 4, "trial_name": null, "trial_params": null }