{ "best_global_step": null, "best_metric": null, "best_model_checkpoint": null, "epoch": 2.0, "eval_steps": 500, "global_step": 420, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.0477326968973747, "grad_norm": 3.792598396402823, "learning_rate": 6.923076923076923e-06, "loss": 1.7804, "step": 10 }, { "epoch": 0.0954653937947494, "grad_norm": 3.133312331753866, "learning_rate": 9.994638629661157e-06, "loss": 1.0888, "step": 20 }, { "epoch": 0.1431980906921241, "grad_norm": 2.5738359920998066, "learning_rate": 9.96191632837388e-06, "loss": 0.9481, "step": 30 }, { "epoch": 0.1909307875894988, "grad_norm": 2.4761984586954533, "learning_rate": 9.899644895607362e-06, "loss": 0.8732, "step": 40 }, { "epoch": 0.2386634844868735, "grad_norm": 2.23923340872218, "learning_rate": 9.808195169286402e-06, "loss": 0.8408, "step": 50 }, { "epoch": 0.2863961813842482, "grad_norm": 2.2576561797308656, "learning_rate": 9.688111749482416e-06, "loss": 0.7931, "step": 60 }, { "epoch": 0.3341288782816229, "grad_norm": 2.394086860913576, "learning_rate": 9.540109755219068e-06, "loss": 0.8051, "step": 70 }, { "epoch": 0.3818615751789976, "grad_norm": 2.0509490805698016, "learning_rate": 9.365070565805941e-06, "loss": 0.7735, "step": 80 }, { "epoch": 0.4295942720763723, "grad_norm": 2.306102249452906, "learning_rate": 9.164036572061354e-06, "loss": 0.7734, "step": 90 }, { "epoch": 0.477326968973747, "grad_norm": 2.1901447177925273, "learning_rate": 8.938204968681838e-06, "loss": 0.7752, "step": 100 }, { "epoch": 0.5250596658711217, "grad_norm": 2.156754909638999, "learning_rate": 8.688920624725798e-06, "loss": 0.7687, "step": 110 }, { "epoch": 0.5727923627684964, "grad_norm": 2.0929605770239617, "learning_rate": 8.417668074668988e-06, "loss": 0.7602, "step": 120 }, { "epoch": 0.6205250596658711, "grad_norm": 2.234313037071736, "learning_rate": 8.126062677726529e-06, "loss": 0.7578, "step": 130 }, { "epoch": 0.6682577565632458, "grad_norm": 2.1626950039794712, "learning_rate": 7.815840998089368e-06, "loss": 0.7383, "step": 140 }, { "epoch": 0.7159904534606205, "grad_norm": 1.7928525458645652, "learning_rate": 7.488850463362616e-06, "loss": 0.7113, "step": 150 }, { "epoch": 0.7637231503579952, "grad_norm": 2.082018682407666, "learning_rate": 7.147038362791731e-06, "loss": 0.7276, "step": 160 }, { "epoch": 0.8114558472553699, "grad_norm": 2.1101539500309014, "learning_rate": 6.7924402507941125e-06, "loss": 0.7243, "step": 170 }, { "epoch": 0.8591885441527446, "grad_norm": 1.8276935874791898, "learning_rate": 6.427167824855218e-06, "loss": 0.7276, "step": 180 }, { "epoch": 0.9069212410501193, "grad_norm": 1.9723562454703483, "learning_rate": 6.053396349978632e-06, "loss": 0.7373, "step": 190 }, { "epoch": 0.954653937947494, "grad_norm": 1.9619509428594932, "learning_rate": 5.673351704579747e-06, "loss": 0.7104, "step": 200 }, { "epoch": 1.0, "grad_norm": 1.9508521881974266, "learning_rate": 5.289297124967216e-06, "loss": 0.6939, "step": 210 }, { "epoch": 1.0477326968973748, "grad_norm": 2.165080867978587, "learning_rate": 4.903519727351218e-06, "loss": 0.5156, "step": 220 }, { "epoch": 1.0954653937947494, "grad_norm": 2.155640295025137, "learning_rate": 4.518316887642484e-06, "loss": 0.4936, "step": 230 }, { "epoch": 1.1431980906921242, "grad_norm": 2.2448096170945413, "learning_rate": 4.135982560152907e-06, "loss": 0.4903, "step": 240 }, { "epoch": 1.1909307875894988, "grad_norm": 2.335307090429571, "learning_rate": 3.75879361667241e-06, "loss": 0.5153, "step": 250 }, { "epoch": 1.2386634844868736, "grad_norm": 1.9880728910364924, "learning_rate": 3.3889962872753834e-06, "loss": 0.4993, "step": 260 }, { "epoch": 1.2863961813842482, "grad_norm": 1.9991557534238604, "learning_rate": 3.028792783604242e-06, "loss": 0.4989, "step": 270 }, { "epoch": 1.334128878281623, "grad_norm": 2.256149746260398, "learning_rate": 2.6803281842908823e-06, "loss": 0.5111, "step": 280 }, { "epoch": 1.3818615751789975, "grad_norm": 1.8941648223063126, "learning_rate": 2.3456776606158227e-06, "loss": 0.5062, "step": 290 }, { "epoch": 1.4295942720763724, "grad_norm": 2.0336663985853023, "learning_rate": 2.0268341184785674e-06, "loss": 0.4831, "step": 300 }, { "epoch": 1.477326968973747, "grad_norm": 1.8511580250790538, "learning_rate": 1.7256963302735752e-06, "loss": 0.4993, "step": 310 }, { "epoch": 1.5250596658711217, "grad_norm": 1.9762983912287935, "learning_rate": 1.4440576273486856e-06, "loss": 0.5162, "step": 320 }, { "epoch": 1.5727923627684963, "grad_norm": 1.9999117959274775, "learning_rate": 1.1835952203845352e-06, "loss": 0.4896, "step": 330 }, { "epoch": 1.6205250596658711, "grad_norm": 2.103694501545566, "learning_rate": 9.458602112940618e-07, "loss": 0.4966, "step": 340 }, { "epoch": 1.668257756563246, "grad_norm": 2.0382561690750802, "learning_rate": 7.322683561231025e-07, "loss": 0.4852, "step": 350 }, { "epoch": 1.7159904534606205, "grad_norm": 2.319579256350246, "learning_rate": 5.440916339607138e-07, "loss": 0.4738, "step": 360 }, { "epoch": 1.763723150357995, "grad_norm": 2.0499623729236314, "learning_rate": 3.824506720679233e-07, "loss": 0.5048, "step": 370 }, { "epoch": 1.81145584725537, "grad_norm": 2.0511457934626196, "learning_rate": 2.4830807233466645e-07, "loss": 0.4855, "step": 380 }, { "epoch": 1.8591885441527447, "grad_norm": 2.1373753031510665, "learning_rate": 1.4246267880710817e-07, "loss": 0.4815, "step": 390 }, { "epoch": 1.9069212410501193, "grad_norm": 2.0720761339135167, "learning_rate": 6.554482042327004e-08, "loss": 0.4956, "step": 400 }, { "epoch": 1.9546539379474939, "grad_norm": 2.159348679673841, "learning_rate": 1.8012557287367394e-08, "loss": 0.4973, "step": 410 }, { "epoch": 2.0, "grad_norm": 2.9991141585377976, "learning_rate": 1.4895283698412867e-10, "loss": 0.4761, "step": 420 }, { "epoch": 2.0, "step": 420, "total_flos": 13232307830784.0, "train_loss": 0.6648564997173491, "train_runtime": 594.9028, "train_samples_per_second": 11.259, "train_steps_per_second": 0.706 } ], "logging_steps": 10, "max_steps": 420, "num_input_tokens_seen": 0, "num_train_epochs": 2, "save_steps": 500, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 13232307830784.0, "train_batch_size": 2, "trial_name": null, "trial_params": null }