{ "best_global_step": null, "best_metric": null, "best_model_checkpoint": null, "epoch": 2.0409381663113004, "eval_steps": 500, "global_step": 600, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.08528784648187633, "grad_norm": 0.778593122959137, "learning_rate": 8.999999999999999e-05, "loss": 5.398508911132812, "step": 25 }, { "epoch": 0.17057569296375266, "grad_norm": 1.1395962238311768, "learning_rate": 0.0001475892857142857, "loss": 5.228872680664063, "step": 50 }, { "epoch": 0.255863539445629, "grad_norm": 0.5455451607704163, "learning_rate": 0.00014089285714285712, "loss": 4.985019226074218, "step": 75 }, { "epoch": 0.3411513859275053, "grad_norm": 0.6568529605865479, "learning_rate": 0.00013419642857142856, "loss": 4.8717037963867185, "step": 100 }, { "epoch": 0.42643923240938164, "grad_norm": 0.6314603686332703, "learning_rate": 0.00012749999999999998, "loss": 4.727347412109375, "step": 125 }, { "epoch": 0.511727078891258, "grad_norm": 0.7707880139350891, "learning_rate": 0.00012080357142857143, "loss": 4.644950256347657, "step": 150 }, { "epoch": 0.5970149253731343, "grad_norm": 0.536558985710144, "learning_rate": 0.00011410714285714285, "loss": 4.557723388671875, "step": 175 }, { "epoch": 0.6823027718550106, "grad_norm": 0.5920282602310181, "learning_rate": 0.00010741071428571428, "loss": 4.463509521484375, "step": 200 }, { "epoch": 0.767590618336887, "grad_norm": 0.533022403717041, "learning_rate": 0.0001007142857142857, "loss": 4.413720703125, "step": 225 }, { "epoch": 0.8528784648187633, "grad_norm": 0.5817797780036926, "learning_rate": 9.401785714285713e-05, "loss": 4.353083801269531, "step": 250 }, { "epoch": 0.9381663113006397, "grad_norm": 0.5374537110328674, "learning_rate": 8.732142857142858e-05, "loss": 4.331479797363281, "step": 275 }, { "epoch": 1.0204690831556502, "grad_norm": 0.5683950781822205, "learning_rate": 8.0625e-05, "loss": 4.214647827148437, "step": 300 }, { "epoch": 1.1057569296375267, "grad_norm": 0.5244670510292053, "learning_rate": 7.392857142857143e-05, "loss": 4.057185668945312, "step": 325 }, { "epoch": 1.191044776119403, "grad_norm": 0.5471984148025513, "learning_rate": 6.723214285714286e-05, "loss": 4.005531921386718, "step": 350 }, { "epoch": 1.2763326226012794, "grad_norm": 0.5038277506828308, "learning_rate": 6.0535714285714285e-05, "loss": 4.033403015136718, "step": 375 }, { "epoch": 1.3616204690831557, "grad_norm": 0.486166387796402, "learning_rate": 5.383928571428571e-05, "loss": 3.9649624633789062, "step": 400 }, { "epoch": 1.446908315565032, "grad_norm": 0.45619258284568787, "learning_rate": 4.7142857142857136e-05, "loss": 3.946517028808594, "step": 425 }, { "epoch": 1.5321961620469082, "grad_norm": 0.45738720893859863, "learning_rate": 4.044642857142857e-05, "loss": 3.9221026611328127, "step": 450 }, { "epoch": 1.6174840085287845, "grad_norm": 0.4545121192932129, "learning_rate": 3.375e-05, "loss": 3.907404479980469, "step": 475 }, { "epoch": 1.702771855010661, "grad_norm": 0.458916574716568, "learning_rate": 2.7053571428571426e-05, "loss": 3.874950256347656, "step": 500 }, { "epoch": 1.7880597014925375, "grad_norm": 0.48818081617355347, "learning_rate": 2.0357142857142855e-05, "loss": 3.8785760498046873, "step": 525 }, { "epoch": 1.8733475479744137, "grad_norm": 0.4515751898288727, "learning_rate": 1.3660714285714284e-05, "loss": 3.8657327270507813, "step": 550 }, { "epoch": 1.95863539445629, "grad_norm": 0.45125558972358704, "learning_rate": 6.9642857142857136e-06, "loss": 3.8602328491210938, "step": 575 }, { "epoch": 2.0409381663113004, "grad_norm": 0.4686155915260315, "learning_rate": 2.6785714285714284e-07, "loss": 3.7586907958984375, "step": 600 } ], "logging_steps": 25, "max_steps": 600, "num_input_tokens_seen": 0, "num_train_epochs": 3, "save_steps": 200, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 7.896972537470362e+16, "train_batch_size": 4, "trial_name": null, "trial_params": null }