| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.1154068090017311, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.001154068090017311, | |
| "eval_loss": 2.0465598106384277, | |
| "eval_runtime": 103.0211, | |
| "eval_samples_per_second": 14.172, | |
| "eval_steps_per_second": 1.776, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.003462204270051933, | |
| "grad_norm": 8.097893714904785, | |
| "learning_rate": 3e-05, | |
| "loss": 8.0719, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.006924408540103866, | |
| "grad_norm": 5.514200210571289, | |
| "learning_rate": 6e-05, | |
| "loss": 8.0211, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0103866128101558, | |
| "grad_norm": 5.214906692504883, | |
| "learning_rate": 9e-05, | |
| "loss": 8.0571, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0103866128101558, | |
| "eval_loss": 1.9573053121566772, | |
| "eval_runtime": 104.5011, | |
| "eval_samples_per_second": 13.971, | |
| "eval_steps_per_second": 1.751, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.013848817080207732, | |
| "grad_norm": 4.61670446395874, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 7.7237, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.017311021350259664, | |
| "grad_norm": 4.776439666748047, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 7.5723, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0207732256203116, | |
| "grad_norm": 3.863844394683838, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 7.8625, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0207732256203116, | |
| "eval_loss": 1.938367247581482, | |
| "eval_runtime": 104.6138, | |
| "eval_samples_per_second": 13.956, | |
| "eval_steps_per_second": 1.749, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.024235429890363532, | |
| "grad_norm": 3.75036358833313, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 7.5972, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.027697634160415464, | |
| "grad_norm": 3.694056510925293, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 7.325, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.031159838430467397, | |
| "grad_norm": 3.767008066177368, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 7.8502, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.031159838430467397, | |
| "eval_loss": 1.9282969236373901, | |
| "eval_runtime": 104.5512, | |
| "eval_samples_per_second": 13.964, | |
| "eval_steps_per_second": 1.75, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.03462204270051933, | |
| "grad_norm": 3.892972230911255, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 7.737, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.038084246970571264, | |
| "grad_norm": 4.205379962921143, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 7.7768, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.0415464512406232, | |
| "grad_norm": 4.217511177062988, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 7.7462, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0415464512406232, | |
| "eval_loss": 1.9257190227508545, | |
| "eval_runtime": 104.5925, | |
| "eval_samples_per_second": 13.959, | |
| "eval_steps_per_second": 1.75, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.04500865551067513, | |
| "grad_norm": 4.240529537200928, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 7.9651, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.048470859780727064, | |
| "grad_norm": 4.575836658477783, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 7.7481, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.05193306405077899, | |
| "grad_norm": 4.351076126098633, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 7.9443, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.05193306405077899, | |
| "eval_loss": 1.9247784614562988, | |
| "eval_runtime": 104.5498, | |
| "eval_samples_per_second": 13.965, | |
| "eval_steps_per_second": 1.75, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.05539526832083093, | |
| "grad_norm": 4.322741508483887, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 7.5892, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.058857472590882864, | |
| "grad_norm": 3.949061870574951, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 7.5119, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.06231967686093479, | |
| "grad_norm": 4.710406303405762, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 7.8128, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.06231967686093479, | |
| "eval_loss": 1.9215248823165894, | |
| "eval_runtime": 104.5083, | |
| "eval_samples_per_second": 13.97, | |
| "eval_steps_per_second": 1.751, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.06578188113098672, | |
| "grad_norm": 3.8922553062438965, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 7.6261, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.06924408540103866, | |
| "grad_norm": 3.532529830932617, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 7.799, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.0727062896710906, | |
| "grad_norm": 3.5513439178466797, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 7.5004, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.0727062896710906, | |
| "eval_loss": 1.9200663566589355, | |
| "eval_runtime": 104.4709, | |
| "eval_samples_per_second": 13.975, | |
| "eval_steps_per_second": 1.752, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.07616849394114253, | |
| "grad_norm": 3.730220317840576, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 7.9821, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.07963069821119446, | |
| "grad_norm": 3.3771941661834717, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 7.2337, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.0830929024812464, | |
| "grad_norm": 3.681811809539795, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 7.4931, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.0830929024812464, | |
| "eval_loss": 1.9186850786209106, | |
| "eval_runtime": 104.5007, | |
| "eval_samples_per_second": 13.971, | |
| "eval_steps_per_second": 1.751, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.08655510675129832, | |
| "grad_norm": 3.645982027053833, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 7.5561, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.09001731102135026, | |
| "grad_norm": 4.192688941955566, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 7.7532, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.0934795152914022, | |
| "grad_norm": 4.39183235168457, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 7.7928, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.0934795152914022, | |
| "eval_loss": 1.9180028438568115, | |
| "eval_runtime": 104.5172, | |
| "eval_samples_per_second": 13.969, | |
| "eval_steps_per_second": 1.751, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.09694171956145413, | |
| "grad_norm": 3.5582807064056396, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 7.6637, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.10040392383150606, | |
| "grad_norm": 4.048159122467041, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 7.3083, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.10386612810155799, | |
| "grad_norm": 3.8467483520507812, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 7.7565, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.10386612810155799, | |
| "eval_loss": 1.91769278049469, | |
| "eval_runtime": 104.4893, | |
| "eval_samples_per_second": 13.973, | |
| "eval_steps_per_second": 1.751, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.10732833237160992, | |
| "grad_norm": 3.6378891468048096, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 7.2697, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.11079053664166186, | |
| "grad_norm": 3.763043165206909, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 7.5429, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.1142527409117138, | |
| "grad_norm": 3.8854165077209473, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 7.6568, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.1142527409117138, | |
| "eval_loss": 1.9176751375198364, | |
| "eval_runtime": 104.5578, | |
| "eval_samples_per_second": 13.964, | |
| "eval_steps_per_second": 1.75, | |
| "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": 1.402137439371264e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |