{ "best_metric": 0.738968551158905, "best_model_checkpoint": "./Zeroshot/01-12-23-NousResearch-Nous-Hermes-Llama2-13b_multilang-dataset-3.0.3-portuguese-2_epochs-10_batch_2/checkpoints/checkpoint-13546", "epoch": 4.0, "global_step": 13546, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.15, "learning_rate": 5.8948611931482576e-05, "loss": 1.3183, "step": 500 }, { "epoch": 0.3, "learning_rate": 0.00011801535735380982, "loss": 0.9343, "step": 1000 }, { "epoch": 0.44, "learning_rate": 0.00017708210277613703, "loss": 0.9005, "step": 1500 }, { "epoch": 0.59, "learning_rate": 0.00023614884819846425, "loss": 0.8859, "step": 2000 }, { "epoch": 0.74, "learning_rate": 0.0002952155936207915, "loss": 0.8707, "step": 2500 }, { "epoch": 0.89, "learning_rate": 0.00035416420555227406, "loss": 0.8523, "step": 3000 }, { "epoch": 1.0, "eval_loss": 0.8459745645523071, "eval_runtime": 133.0397, "eval_samples_per_second": 11.32, "eval_steps_per_second": 1.421, "step": 3386 }, { "epoch": 1.03, "learning_rate": 0.00039998666871418663, "loss": 0.8395, "step": 3500 }, { "epoch": 1.18, "learning_rate": 0.00039960207649127196, "loss": 0.8052, "step": 4000 }, { "epoch": 1.33, "learning_rate": 0.0003986872720605752, "loss": 0.8036, "step": 4500 }, { "epoch": 1.48, "learning_rate": 0.00039724468545932536, "loss": 0.7902, "step": 5000 }, { "epoch": 1.62, "learning_rate": 0.0003952781486968895, "loss": 0.7818, "step": 5500 }, { "epoch": 1.77, "learning_rate": 0.00039279288557562877, "loss": 0.7834, "step": 6000 }, { "epoch": 1.92, "learning_rate": 0.0003897954978146717, "loss": 0.7686, "step": 6500 }, { "epoch": 2.0, "eval_loss": 0.7881951332092285, "eval_runtime": 132.7462, "eval_samples_per_second": 11.345, "eval_steps_per_second": 1.424, "step": 6773 }, { "epoch": 2.07, "learning_rate": 0.0003862939475134625, "loss": 0.7371, "step": 7000 }, { "epoch": 2.21, "learning_rate": 0.00038230601585633047, "loss": 0.6893, "step": 7500 }, { "epoch": 2.36, "learning_rate": 0.0003778263160223725, "loss": 0.6936, "step": 8000 }, { "epoch": 2.51, "learning_rate": 0.00037287424793815686, "loss": 0.6923, "step": 8500 }, { "epoch": 2.66, "learning_rate": 0.0003674629660106825, "loss": 0.6862, "step": 9000 }, { "epoch": 2.81, "learning_rate": 0.00036160684447787385, "loss": 0.6869, "step": 9500 }, { "epoch": 2.95, "learning_rate": 0.0003553214392256293, "loss": 0.6856, "step": 10000 }, { "epoch": 3.0, "eval_loss": 0.7562845945358276, "eval_runtime": 132.7178, "eval_samples_per_second": 11.347, "eval_steps_per_second": 1.424, "step": 10159 }, { "epoch": 3.1, "learning_rate": 0.0003486234464660031, "loss": 0.6199, "step": 10500 }, { "epoch": 3.25, "learning_rate": 0.0003415452256149365, "loss": 0.595, "step": 11000 }, { "epoch": 3.4, "learning_rate": 0.0003340772155486902, "loss": 0.5962, "step": 11500 }, { "epoch": 3.54, "learning_rate": 0.0003262690400990177, "loss": 0.6019, "step": 12000 }, { "epoch": 3.69, "learning_rate": 0.00031811015213363493, "loss": 0.6043, "step": 12500 }, { "epoch": 3.84, "learning_rate": 0.0003096375227201415, "loss": 0.5984, "step": 13000 }, { "epoch": 3.99, "learning_rate": 0.0003008736580950757, "loss": 0.6011, "step": 13500 }, { "epoch": 4.0, "eval_loss": 0.738968551158905, "eval_runtime": 132.6823, "eval_samples_per_second": 11.35, "eval_steps_per_second": 1.424, "step": 13546 } ], "max_steps": 33860, "num_train_epochs": 10, "total_flos": 6.037383614590157e+17, "trial_name": null, "trial_params": null }