| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.01365280906546522, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0001365280906546522, | |
| "eval_loss": 3.0241544246673584, | |
| "eval_runtime": 335.8469, | |
| "eval_samples_per_second": 36.731, | |
| "eval_steps_per_second": 4.591, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00040958427196395656, | |
| "grad_norm": 0.23248405754566193, | |
| "learning_rate": 3e-05, | |
| "loss": 2.9839, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0008191685439279131, | |
| "grad_norm": 0.17891249060630798, | |
| "learning_rate": 6e-05, | |
| "loss": 3.0261, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0012287528158918697, | |
| "grad_norm": 0.20870865881443024, | |
| "learning_rate": 9e-05, | |
| "loss": 2.9711, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0012287528158918697, | |
| "eval_loss": 3.003326416015625, | |
| "eval_runtime": 335.1248, | |
| "eval_samples_per_second": 36.81, | |
| "eval_steps_per_second": 4.601, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0016383370878558263, | |
| "grad_norm": 0.23463056981563568, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 3.0762, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0020479213598197828, | |
| "grad_norm": 0.2302045226097107, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 2.8429, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0024575056317837395, | |
| "grad_norm": 0.22244347631931305, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 2.7983, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0024575056317837395, | |
| "eval_loss": 2.8648135662078857, | |
| "eval_runtime": 336.04, | |
| "eval_samples_per_second": 36.71, | |
| "eval_steps_per_second": 4.589, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0028670899037476962, | |
| "grad_norm": 0.17184042930603027, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 2.818, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0032766741757116525, | |
| "grad_norm": 0.20670373737812042, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 2.8476, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0036862584476756092, | |
| "grad_norm": 0.21791444718837738, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 2.7224, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0036862584476756092, | |
| "eval_loss": 2.7546896934509277, | |
| "eval_runtime": 336.2625, | |
| "eval_samples_per_second": 36.686, | |
| "eval_steps_per_second": 4.586, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0040958427196395655, | |
| "grad_norm": 0.23728761076927185, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 2.7488, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.004505426991603522, | |
| "grad_norm": 0.19305483996868134, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 2.6667, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.004915011263567479, | |
| "grad_norm": 0.19373920559883118, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 2.6568, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.004915011263567479, | |
| "eval_loss": 2.697031259536743, | |
| "eval_runtime": 335.7735, | |
| "eval_samples_per_second": 36.739, | |
| "eval_steps_per_second": 4.592, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.005324595535531436, | |
| "grad_norm": 0.1955287754535675, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 2.6984, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0057341798074953924, | |
| "grad_norm": 0.18126782774925232, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 2.718, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.006143764079459349, | |
| "grad_norm": 0.19327886402606964, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 2.6685, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.006143764079459349, | |
| "eval_loss": 2.66717267036438, | |
| "eval_runtime": 335.9741, | |
| "eval_samples_per_second": 36.717, | |
| "eval_steps_per_second": 4.59, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.006553348351423305, | |
| "grad_norm": 0.18886631727218628, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 2.6382, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.006962932623387262, | |
| "grad_norm": 0.23726007342338562, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 2.6984, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.0073725168953512185, | |
| "grad_norm": 0.20413297414779663, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 2.5525, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.0073725168953512185, | |
| "eval_loss": 2.6474061012268066, | |
| "eval_runtime": 335.8668, | |
| "eval_samples_per_second": 36.729, | |
| "eval_steps_per_second": 4.591, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.007782101167315175, | |
| "grad_norm": 0.182687446475029, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 2.5935, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.008191685439279131, | |
| "grad_norm": 0.20747148990631104, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 2.671, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.008601269711243089, | |
| "grad_norm": 0.17653383314609528, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 2.6424, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.008601269711243089, | |
| "eval_loss": 2.635723114013672, | |
| "eval_runtime": 335.758, | |
| "eval_samples_per_second": 36.741, | |
| "eval_steps_per_second": 4.593, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.009010853983207045, | |
| "grad_norm": 0.2043198049068451, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 2.6353, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.009420438255171002, | |
| "grad_norm": 0.1920263171195984, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 2.6312, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.009830022527134958, | |
| "grad_norm": 0.2099771350622177, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 2.7014, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.009830022527134958, | |
| "eval_loss": 2.62795352935791, | |
| "eval_runtime": 335.8487, | |
| "eval_samples_per_second": 36.731, | |
| "eval_steps_per_second": 4.591, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.010239606799098914, | |
| "grad_norm": 0.17937679588794708, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 2.5686, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.010649191071062871, | |
| "grad_norm": 0.17208440601825714, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 2.6691, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.011058775343026827, | |
| "grad_norm": 0.1848759800195694, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 2.5746, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.011058775343026827, | |
| "eval_loss": 2.6240718364715576, | |
| "eval_runtime": 335.5775, | |
| "eval_samples_per_second": 36.761, | |
| "eval_steps_per_second": 4.595, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.011468359614990785, | |
| "grad_norm": 0.18467338383197784, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 2.6069, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.01187794388695474, | |
| "grad_norm": 0.19607900083065033, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 2.4989, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.012287528158918698, | |
| "grad_norm": 0.18552134931087494, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 2.675, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.012287528158918698, | |
| "eval_loss": 2.622399091720581, | |
| "eval_runtime": 335.3275, | |
| "eval_samples_per_second": 36.788, | |
| "eval_steps_per_second": 4.598, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.012697112430882654, | |
| "grad_norm": 0.18758685886859894, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 2.5761, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.01310669670284661, | |
| "grad_norm": 0.19522517919540405, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 2.6464, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.013516280974810568, | |
| "grad_norm": 0.20688404142856598, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 2.658, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.013516280974810568, | |
| "eval_loss": 2.622002124786377, | |
| "eval_runtime": 335.7295, | |
| "eval_samples_per_second": 36.744, | |
| "eval_steps_per_second": 4.593, | |
| "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": 3.234376648438579e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |