| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.3325020781379884, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0033250207813798837, | |
| "eval_loss": 2.4745399951934814, | |
| "eval_runtime": 35.0849, | |
| "eval_samples_per_second": 14.451, | |
| "eval_steps_per_second": 1.824, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00997506234413965, | |
| "grad_norm": 1.8293392658233643, | |
| "learning_rate": 1.5e-05, | |
| "loss": 2.4184, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0199501246882793, | |
| "grad_norm": 2.3942174911499023, | |
| "learning_rate": 3e-05, | |
| "loss": 2.4593, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.029925187032418952, | |
| "grad_norm": 1.915959358215332, | |
| "learning_rate": 4.5e-05, | |
| "loss": 2.2337, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.029925187032418952, | |
| "eval_loss": 2.385331153869629, | |
| "eval_runtime": 35.4941, | |
| "eval_samples_per_second": 14.284, | |
| "eval_steps_per_second": 1.803, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0399002493765586, | |
| "grad_norm": 2.4679598808288574, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 2.4138, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.04987531172069826, | |
| "grad_norm": 2.2216885089874268, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 2.1899, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.059850374064837904, | |
| "grad_norm": 2.1264395713806152, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 1.9607, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.059850374064837904, | |
| "eval_loss": 1.7929750680923462, | |
| "eval_runtime": 35.5582, | |
| "eval_samples_per_second": 14.258, | |
| "eval_steps_per_second": 1.8, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.06982543640897755, | |
| "grad_norm": 1.875053882598877, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 1.7054, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0798004987531172, | |
| "grad_norm": 1.6136735677719116, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 1.4407, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.08977556109725686, | |
| "grad_norm": 1.430875301361084, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 1.4597, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.08977556109725686, | |
| "eval_loss": 1.4339165687561035, | |
| "eval_runtime": 35.58, | |
| "eval_samples_per_second": 14.25, | |
| "eval_steps_per_second": 1.799, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.09975062344139651, | |
| "grad_norm": 1.4442172050476074, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 1.4351, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.10972568578553615, | |
| "grad_norm": 1.886985421180725, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 1.4478, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.11970074812967581, | |
| "grad_norm": 1.4151262044906616, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 1.2824, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.11970074812967581, | |
| "eval_loss": 1.3912534713745117, | |
| "eval_runtime": 35.581, | |
| "eval_samples_per_second": 14.249, | |
| "eval_steps_per_second": 1.799, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.12967581047381546, | |
| "grad_norm": 1.5250979661941528, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 1.3491, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.1396508728179551, | |
| "grad_norm": 1.3472334146499634, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 1.4616, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.14962593516209477, | |
| "grad_norm": 1.2380613088607788, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 1.2844, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.14962593516209477, | |
| "eval_loss": 1.3702603578567505, | |
| "eval_runtime": 35.6209, | |
| "eval_samples_per_second": 14.233, | |
| "eval_steps_per_second": 1.797, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.1596009975062344, | |
| "grad_norm": 1.376523494720459, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 1.3452, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.16957605985037408, | |
| "grad_norm": 1.2102608680725098, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 1.3209, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.17955112219451372, | |
| "grad_norm": 1.7671351432800293, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 1.4073, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.17955112219451372, | |
| "eval_loss": 1.3600367307662964, | |
| "eval_runtime": 35.6092, | |
| "eval_samples_per_second": 14.238, | |
| "eval_steps_per_second": 1.797, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.18952618453865336, | |
| "grad_norm": 1.6414278745651245, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 1.4529, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.19950124688279303, | |
| "grad_norm": 1.6693403720855713, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 1.263, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.20947630922693267, | |
| "grad_norm": 1.2293144464492798, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 1.374, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.20947630922693267, | |
| "eval_loss": 1.3520565032958984, | |
| "eval_runtime": 35.64, | |
| "eval_samples_per_second": 14.226, | |
| "eval_steps_per_second": 1.796, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.2194513715710723, | |
| "grad_norm": 1.280704379081726, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 1.2475, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.22942643391521197, | |
| "grad_norm": 1.503631591796875, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 1.3387, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.23940149625935161, | |
| "grad_norm": 1.318060278892517, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 1.2855, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.23940149625935161, | |
| "eval_loss": 1.346817970275879, | |
| "eval_runtime": 35.6, | |
| "eval_samples_per_second": 14.242, | |
| "eval_steps_per_second": 1.798, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.24937655860349128, | |
| "grad_norm": 1.3991254568099976, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 1.2667, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.2593516209476309, | |
| "grad_norm": 1.3369468450546265, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 1.2339, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.26932668329177056, | |
| "grad_norm": 1.3982964754104614, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 1.3598, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.26932668329177056, | |
| "eval_loss": 1.3438166379928589, | |
| "eval_runtime": 35.6426, | |
| "eval_samples_per_second": 14.225, | |
| "eval_steps_per_second": 1.796, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.2793017456359102, | |
| "grad_norm": 1.6268279552459717, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 1.258, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.2892768079800499, | |
| "grad_norm": 1.3860325813293457, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 1.3495, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.29925187032418954, | |
| "grad_norm": 1.7704390287399292, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 1.312, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.29925187032418954, | |
| "eval_loss": 1.3430734872817993, | |
| "eval_runtime": 35.5853, | |
| "eval_samples_per_second": 14.247, | |
| "eval_steps_per_second": 1.798, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.3092269326683292, | |
| "grad_norm": 1.516430377960205, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 1.2602, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.3192019950124688, | |
| "grad_norm": 1.294288158416748, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 1.3084, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.32917705735660846, | |
| "grad_norm": 1.539898157119751, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 1.4346, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.32917705735660846, | |
| "eval_loss": 1.3427472114562988, | |
| "eval_runtime": 35.5324, | |
| "eval_samples_per_second": 14.269, | |
| "eval_steps_per_second": 1.801, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.483774567120896e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |