| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 3077, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.032499187520311994, | |
| "grad_norm": 0.38231489062309265, | |
| "learning_rate": 4.8375040623984405e-05, | |
| "loss": 0.1722, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.06499837504062399, | |
| "grad_norm": 0.29991772770881653, | |
| "learning_rate": 4.67500812479688e-05, | |
| "loss": 0.1983, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.09749756256093597, | |
| "grad_norm": 0.3033580780029297, | |
| "learning_rate": 4.51251218719532e-05, | |
| "loss": 0.2034, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.12999675008124797, | |
| "grad_norm": 0.31061500310897827, | |
| "learning_rate": 4.3500162495937604e-05, | |
| "loss": 0.2022, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.16249593760155995, | |
| "grad_norm": 0.27624836564064026, | |
| "learning_rate": 4.1875203119922e-05, | |
| "loss": 0.1975, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.16249593760155995, | |
| "eval_gen_len": 19.964290067032298, | |
| "eval_loss": 0.18788783252239227, | |
| "eval_rouge1": 0.0068, | |
| "eval_rouge2": 0.0, | |
| "eval_rougeL": 0.0068, | |
| "eval_rougeLsum": 0.0068, | |
| "eval_runtime": 2174.226, | |
| "eval_samples_per_second": 11.321, | |
| "eval_steps_per_second": 0.354, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.19499512512187195, | |
| "grad_norm": 0.2493065893650055, | |
| "learning_rate": 4.02502437439064e-05, | |
| "loss": 0.1943, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.22749431264218395, | |
| "grad_norm": 0.2768040597438812, | |
| "learning_rate": 3.8625284367890804e-05, | |
| "loss": 0.1919, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.25999350016249595, | |
| "grad_norm": 0.2641097605228424, | |
| "learning_rate": 3.7000324991875206e-05, | |
| "loss": 0.1883, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.29249268768280795, | |
| "grad_norm": 0.266985148191452, | |
| "learning_rate": 3.537536561585961e-05, | |
| "loss": 0.1866, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.3249918752031199, | |
| "grad_norm": 0.2505531311035156, | |
| "learning_rate": 3.3750406239844004e-05, | |
| "loss": 0.186, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.3249918752031199, | |
| "eval_gen_len": 19.9981718464351, | |
| "eval_loss": 0.1798761785030365, | |
| "eval_rouge1": 0.0108, | |
| "eval_rouge2": 0.0041, | |
| "eval_rougeL": 0.0108, | |
| "eval_rougeLsum": 0.0108, | |
| "eval_runtime": 2176.3057, | |
| "eval_samples_per_second": 11.31, | |
| "eval_steps_per_second": 0.354, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.3574910627234319, | |
| "grad_norm": 0.2498556226491928, | |
| "learning_rate": 3.2125446863828406e-05, | |
| "loss": 0.1839, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.3899902502437439, | |
| "grad_norm": 0.2563565969467163, | |
| "learning_rate": 3.0500487487812808e-05, | |
| "loss": 0.1835, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.4224894377640559, | |
| "grad_norm": 0.25993165373802185, | |
| "learning_rate": 2.8875528111797207e-05, | |
| "loss": 0.1837, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.4549886252843679, | |
| "grad_norm": 0.2639683485031128, | |
| "learning_rate": 2.7250568735781606e-05, | |
| "loss": 0.1814, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.4874878128046799, | |
| "grad_norm": 0.25668901205062866, | |
| "learning_rate": 2.5625609359766008e-05, | |
| "loss": 0.1762, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.4874878128046799, | |
| "eval_gen_len": 19.983384115376804, | |
| "eval_loss": 0.17473776638507843, | |
| "eval_rouge1": 0.1836, | |
| "eval_rouge2": 0.0, | |
| "eval_rougeL": 0.1828, | |
| "eval_rougeLsum": 0.1851, | |
| "eval_runtime": 2167.8807, | |
| "eval_samples_per_second": 11.354, | |
| "eval_steps_per_second": 0.355, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.5199870003249919, | |
| "grad_norm": 0.2645615041255951, | |
| "learning_rate": 2.4000649983750407e-05, | |
| "loss": 0.1794, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.5524861878453039, | |
| "grad_norm": 0.24295924603939056, | |
| "learning_rate": 2.237569060773481e-05, | |
| "loss": 0.1769, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.5849853753656159, | |
| "grad_norm": 0.25981849431991577, | |
| "learning_rate": 2.0750731231719208e-05, | |
| "loss": 0.1785, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.6174845628859279, | |
| "grad_norm": 0.24553097784519196, | |
| "learning_rate": 1.912577185570361e-05, | |
| "loss": 0.173, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.6499837504062398, | |
| "grad_norm": 0.24243266880512238, | |
| "learning_rate": 1.750081247968801e-05, | |
| "loss": 0.1764, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.6499837504062398, | |
| "eval_gen_len": 19.99471866747918, | |
| "eval_loss": 0.17154192924499512, | |
| "eval_rouge1": 0.1151, | |
| "eval_rouge2": 0.0, | |
| "eval_rougeL": 0.1151, | |
| "eval_rougeLsum": 0.1165, | |
| "eval_runtime": 2157.4938, | |
| "eval_samples_per_second": 11.409, | |
| "eval_steps_per_second": 0.357, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.6824829379265518, | |
| "grad_norm": 0.26906952261924744, | |
| "learning_rate": 1.5875853103672408e-05, | |
| "loss": 0.1789, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.7149821254468638, | |
| "grad_norm": 0.2654452919960022, | |
| "learning_rate": 1.4250893727656808e-05, | |
| "loss": 0.1727, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.7474813129671758, | |
| "grad_norm": 0.24443137645721436, | |
| "learning_rate": 1.2625934351641208e-05, | |
| "loss": 0.1767, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.7799805004874878, | |
| "grad_norm": 0.24741144478321075, | |
| "learning_rate": 1.1000974975625609e-05, | |
| "loss": 0.1745, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.8124796880077998, | |
| "grad_norm": 0.2507327198982239, | |
| "learning_rate": 9.376015599610011e-06, | |
| "loss": 0.1756, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.8124796880077998, | |
| "eval_gen_len": 19.99869997968718, | |
| "eval_loss": 0.1691381335258484, | |
| "eval_rouge1": 0.0027, | |
| "eval_rouge2": 0.0, | |
| "eval_rougeL": 0.0027, | |
| "eval_rougeLsum": 0.0027, | |
| "eval_runtime": 2163.6252, | |
| "eval_samples_per_second": 11.377, | |
| "eval_steps_per_second": 0.356, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.8449788755281118, | |
| "grad_norm": 0.23773913085460663, | |
| "learning_rate": 7.75105622359441e-06, | |
| "loss": 0.1733, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.8774780630484238, | |
| "grad_norm": 0.24695108830928802, | |
| "learning_rate": 6.1260968475788104e-06, | |
| "loss": 0.1744, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.9099772505687358, | |
| "grad_norm": 0.23871463537216187, | |
| "learning_rate": 4.501137471563211e-06, | |
| "loss": 0.1741, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.9424764380890478, | |
| "grad_norm": 0.24402374029159546, | |
| "learning_rate": 2.8761780955476114e-06, | |
| "loss": 0.1703, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.9749756256093598, | |
| "grad_norm": 0.24215468764305115, | |
| "learning_rate": 1.2512187195320117e-06, | |
| "loss": 0.1724, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.9749756256093598, | |
| "eval_gen_len": 19.99163111923624, | |
| "eval_loss": 0.1676524430513382, | |
| "eval_rouge1": 0.0257, | |
| "eval_rouge2": 0.0, | |
| "eval_rougeL": 0.0244, | |
| "eval_rougeLsum": 0.0257, | |
| "eval_runtime": 2148.352, | |
| "eval_samples_per_second": 11.458, | |
| "eval_steps_per_second": 0.358, | |
| "step": 3000 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 3077, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.7819450187495506e+19, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |