{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.993006993006993, "eval_steps": 36, "global_step": 71, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.013986013986013986, "eval_loss": 0.8843945860862732, "eval_runtime": 7.527, "eval_samples_per_second": 3.986, "eval_steps_per_second": 1.063, "step": 1 }, { "epoch": 0.13986013986013987, "grad_norm": 0.30088093876838684, "learning_rate": 0.0002, "loss": 0.78, "step": 10 }, { "epoch": 0.27972027972027974, "grad_norm": 0.16369399428367615, "learning_rate": 0.00019718115683235417, "loss": 0.1395, "step": 20 }, { "epoch": 0.4195804195804196, "grad_norm": 0.08310074359178543, "learning_rate": 0.00018888354486549237, "loss": 0.0605, "step": 30 }, { "epoch": 0.5034965034965035, "eval_loss": 0.02842050977051258, "eval_runtime": 7.5377, "eval_samples_per_second": 3.98, "eval_steps_per_second": 1.061, "step": 36 }, { "epoch": 0.5594405594405595, "grad_norm": 0.12133278697729111, "learning_rate": 0.00017557495743542585, "loss": 0.0527, "step": 40 }, { "epoch": 0.6993006993006993, "grad_norm": 0.11081533133983612, "learning_rate": 0.00015800569095711982, "loss": 0.0447, "step": 50 }, { "epoch": 0.8391608391608392, "grad_norm": 0.07465693354606628, "learning_rate": 0.00013716624556603274, "loss": 0.0408, "step": 60 }, { "epoch": 0.9790209790209791, "grad_norm": 0.15174183249473572, "learning_rate": 0.00011423148382732853, "loss": 0.0526, "step": 70 } ], "logging_steps": 10, "max_steps": 142, "num_input_tokens_seen": 0, "num_train_epochs": 2, "save_steps": 71, "total_flos": 4.667501229952205e+16, "train_batch_size": 4, "trial_name": null, "trial_params": null }