| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.2251279135872655, | |
| "eval_steps": 9, | |
| "global_step": 99, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0022740193291642978, | |
| "eval_loss": 2.2858026027679443, | |
| "eval_runtime": 35.4041, | |
| "eval_samples_per_second": 20.93, | |
| "eval_steps_per_second": 2.627, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.006822057987492893, | |
| "grad_norm": 0.60309898853302, | |
| "learning_rate": 3e-05, | |
| "loss": 2.083, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.013644115974985787, | |
| "grad_norm": 0.683789074420929, | |
| "learning_rate": 6e-05, | |
| "loss": 2.3169, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.02046617396247868, | |
| "grad_norm": 0.8379518985748291, | |
| "learning_rate": 9e-05, | |
| "loss": 2.2734, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.02046617396247868, | |
| "eval_loss": 2.204383373260498, | |
| "eval_runtime": 35.8511, | |
| "eval_samples_per_second": 20.669, | |
| "eval_steps_per_second": 2.594, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.027288231949971573, | |
| "grad_norm": 0.6264774799346924, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 2.0739, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.03411028993746447, | |
| "grad_norm": 0.5958771705627441, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 2.0297, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.04093234792495736, | |
| "grad_norm": 1.1767833232879639, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 2.04, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.04093234792495736, | |
| "eval_loss": 1.970629334449768, | |
| "eval_runtime": 35.8984, | |
| "eval_samples_per_second": 20.642, | |
| "eval_steps_per_second": 2.591, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.047754405912450254, | |
| "grad_norm": 0.9643926620483398, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 1.8702, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.054576463899943146, | |
| "grad_norm": 1.1769558191299438, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 1.8235, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.061398521887436046, | |
| "grad_norm": 1.374096393585205, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 1.647, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.061398521887436046, | |
| "eval_loss": 1.761104941368103, | |
| "eval_runtime": 35.916, | |
| "eval_samples_per_second": 20.631, | |
| "eval_steps_per_second": 2.589, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.06822057987492894, | |
| "grad_norm": 0.7258775234222412, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 1.7701, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.07504263786242182, | |
| "grad_norm": 0.5846672654151917, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 1.5202, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.08186469584991472, | |
| "grad_norm": 0.7897035479545593, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 1.6917, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.08186469584991472, | |
| "eval_loss": 1.7188748121261597, | |
| "eval_runtime": 35.9555, | |
| "eval_samples_per_second": 20.609, | |
| "eval_steps_per_second": 2.587, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.08868675383740762, | |
| "grad_norm": 0.4667467176914215, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 1.6435, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.09550881182490051, | |
| "grad_norm": 0.4430086612701416, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 1.6451, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.10233086981239341, | |
| "grad_norm": 0.5567775368690491, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 1.5085, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.10233086981239341, | |
| "eval_loss": 1.696839451789856, | |
| "eval_runtime": 35.9155, | |
| "eval_samples_per_second": 20.632, | |
| "eval_steps_per_second": 2.589, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.10915292779988629, | |
| "grad_norm": 0.4994668662548065, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 1.5962, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.11597498578737919, | |
| "grad_norm": 0.4232550859451294, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 1.7046, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.12279704377487209, | |
| "grad_norm": 0.5421936511993408, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 1.7675, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.12279704377487209, | |
| "eval_loss": 1.6868067979812622, | |
| "eval_runtime": 35.9161, | |
| "eval_samples_per_second": 20.631, | |
| "eval_steps_per_second": 2.589, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.129619101762365, | |
| "grad_norm": 0.4595838785171509, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 1.7213, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.13644115974985788, | |
| "grad_norm": 0.3871839642524719, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 1.6915, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.14326321773735076, | |
| "grad_norm": 0.48633936047554016, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 1.5636, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.14326321773735076, | |
| "eval_loss": 1.6816630363464355, | |
| "eval_runtime": 35.9227, | |
| "eval_samples_per_second": 20.628, | |
| "eval_steps_per_second": 2.589, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.15008527572484365, | |
| "grad_norm": 0.3406151235103607, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 1.7242, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.15690733371233656, | |
| "grad_norm": 0.4090305268764496, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 1.6906, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.16372939169982945, | |
| "grad_norm": 0.39295023679733276, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 1.7433, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.16372939169982945, | |
| "eval_loss": 1.678359866142273, | |
| "eval_runtime": 35.9533, | |
| "eval_samples_per_second": 20.61, | |
| "eval_steps_per_second": 2.587, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.17055144968732233, | |
| "grad_norm": 0.47114595770835876, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 1.7471, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.17737350767481525, | |
| "grad_norm": 0.4847165644168854, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 1.514, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.18419556566230813, | |
| "grad_norm": 0.35954052209854126, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 1.9685, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.18419556566230813, | |
| "eval_loss": 1.6781070232391357, | |
| "eval_runtime": 35.9632, | |
| "eval_samples_per_second": 20.604, | |
| "eval_steps_per_second": 2.586, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.19101762364980102, | |
| "grad_norm": 0.39467528462409973, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 1.7774, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.19783968163729393, | |
| "grad_norm": 0.7397677302360535, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 1.7756, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.20466173962478681, | |
| "grad_norm": 0.33035725355148315, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 1.9051, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.20466173962478681, | |
| "eval_loss": 1.677160620689392, | |
| "eval_runtime": 35.9298, | |
| "eval_samples_per_second": 20.624, | |
| "eval_steps_per_second": 2.588, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.2114837976122797, | |
| "grad_norm": 0.4519857168197632, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 1.394, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.21830585559977259, | |
| "grad_norm": 0.6549511551856995, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 1.6634, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.2251279135872655, | |
| "grad_norm": 0.3595646619796753, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 1.9199, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.2251279135872655, | |
| "eval_loss": 1.6768484115600586, | |
| "eval_runtime": 35.9594, | |
| "eval_samples_per_second": 20.607, | |
| "eval_steps_per_second": 2.586, | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.955421302108979e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |