| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.03753554502369668, | |
| "eval_steps": 9, | |
| "global_step": 99, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0003791469194312796, | |
| "eval_loss": 1.3354411125183105, | |
| "eval_runtime": 561.0446, | |
| "eval_samples_per_second": 7.917, | |
| "eval_steps_per_second": 0.991, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.001137440758293839, | |
| "grad_norm": 8.923999786376953, | |
| "learning_rate": 3e-05, | |
| "loss": 5.2252, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.002274881516587678, | |
| "grad_norm": 9.253809928894043, | |
| "learning_rate": 6e-05, | |
| "loss": 5.0773, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0034123222748815166, | |
| "grad_norm": 7.367655277252197, | |
| "learning_rate": 9e-05, | |
| "loss": 4.104, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0034123222748815166, | |
| "eval_loss": 0.9018532037734985, | |
| "eval_runtime": 565.3238, | |
| "eval_samples_per_second": 7.857, | |
| "eval_steps_per_second": 0.984, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.004549763033175356, | |
| "grad_norm": 7.7459797859191895, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 3.5208, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.005687203791469194, | |
| "grad_norm": 20.93357276916504, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 2.6521, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.006824644549763033, | |
| "grad_norm": 6.177847862243652, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 2.6246, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.006824644549763033, | |
| "eval_loss": 0.6228094100952148, | |
| "eval_runtime": 565.3914, | |
| "eval_samples_per_second": 7.857, | |
| "eval_steps_per_second": 0.983, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.007962085308056872, | |
| "grad_norm": 5.295720100402832, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 2.5272, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.009099526066350712, | |
| "grad_norm": 4.090085983276367, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 2.3416, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.010236966824644549, | |
| "grad_norm": 4.162299156188965, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 2.0675, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.010236966824644549, | |
| "eval_loss": 0.562384307384491, | |
| "eval_runtime": 565.6143, | |
| "eval_samples_per_second": 7.853, | |
| "eval_steps_per_second": 0.983, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.011374407582938388, | |
| "grad_norm": 4.30471658706665, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 2.0407, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.012511848341232227, | |
| "grad_norm": 4.804416179656982, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 2.0456, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.013649289099526066, | |
| "grad_norm": 3.96138596534729, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 2.0244, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.013649289099526066, | |
| "eval_loss": 0.5297911167144775, | |
| "eval_runtime": 565.4741, | |
| "eval_samples_per_second": 7.855, | |
| "eval_steps_per_second": 0.983, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.014786729857819906, | |
| "grad_norm": 4.675422191619873, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 2.015, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.015924170616113745, | |
| "grad_norm": 4.122074604034424, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 2.0857, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.017061611374407582, | |
| "grad_norm": 4.02851676940918, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 2.0372, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.017061611374407582, | |
| "eval_loss": 0.5146540999412537, | |
| "eval_runtime": 565.5761, | |
| "eval_samples_per_second": 7.854, | |
| "eval_steps_per_second": 0.983, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.018199052132701423, | |
| "grad_norm": 4.6452507972717285, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 1.932, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.01933649289099526, | |
| "grad_norm": 4.3586039543151855, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 2.1188, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.020473933649289098, | |
| "grad_norm": 4.268399715423584, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 1.9989, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.020473933649289098, | |
| "eval_loss": 0.5036410689353943, | |
| "eval_runtime": 565.5702, | |
| "eval_samples_per_second": 7.854, | |
| "eval_steps_per_second": 0.983, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.02161137440758294, | |
| "grad_norm": 3.966719627380371, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 2.1469, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.022748815165876776, | |
| "grad_norm": 4.320970058441162, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 2.1051, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.023886255924170617, | |
| "grad_norm": 4.397201061248779, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 1.864, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.023886255924170617, | |
| "eval_loss": 0.4941074252128601, | |
| "eval_runtime": 565.4444, | |
| "eval_samples_per_second": 7.856, | |
| "eval_steps_per_second": 0.983, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.025023696682464455, | |
| "grad_norm": 4.262881278991699, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 1.9055, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.026161137440758295, | |
| "grad_norm": 4.537021636962891, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 2.0358, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.027298578199052133, | |
| "grad_norm": 3.677988052368164, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 1.6868, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.027298578199052133, | |
| "eval_loss": 0.48585864901542664, | |
| "eval_runtime": 565.2465, | |
| "eval_samples_per_second": 7.859, | |
| "eval_steps_per_second": 0.984, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.02843601895734597, | |
| "grad_norm": 4.369206428527832, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 2.1532, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.02957345971563981, | |
| "grad_norm": 4.181845664978027, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 2.0465, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.03071090047393365, | |
| "grad_norm": 4.068225383758545, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 1.9805, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.03071090047393365, | |
| "eval_loss": 0.48237505555152893, | |
| "eval_runtime": 565.3076, | |
| "eval_samples_per_second": 7.858, | |
| "eval_steps_per_second": 0.984, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.03184834123222749, | |
| "grad_norm": 4.095105171203613, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 2.1349, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.03298578199052133, | |
| "grad_norm": 4.034782886505127, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 1.989, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.034123222748815164, | |
| "grad_norm": 4.185176372528076, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 1.9752, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.034123222748815164, | |
| "eval_loss": 0.48054438829421997, | |
| "eval_runtime": 565.9474, | |
| "eval_samples_per_second": 7.849, | |
| "eval_steps_per_second": 0.982, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.035260663507109005, | |
| "grad_norm": 3.7425358295440674, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 1.9, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.036398104265402846, | |
| "grad_norm": 3.9407403469085693, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 1.9243, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.03753554502369668, | |
| "grad_norm": 3.88885760307312, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 1.9794, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.03753554502369668, | |
| "eval_loss": 0.4799988865852356, | |
| "eval_runtime": 565.6344, | |
| "eval_samples_per_second": 7.853, | |
| "eval_steps_per_second": 0.983, | |
| "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": 2.6599302131141837e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |