{ "best_metric": 2.42663836479187, "best_model_checkpoint": "./outputs/checkpoint-1200", "epoch": 0.8743169398907104, "eval_steps": 100, "global_step": 1200, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.07, "learning_rate": 0.0002, "loss": 2.8181, "step": 100 }, { "epoch": 0.07, "eval_loss": 2.7240540981292725, "eval_runtime": 204.8489, "eval_samples_per_second": 30.627, "eval_steps_per_second": 3.832, "step": 100 }, { "epoch": 0.15, "learning_rate": 0.0002, "loss": 2.6957, "step": 200 }, { "epoch": 0.15, "eval_loss": 2.6725997924804688, "eval_runtime": 204.875, "eval_samples_per_second": 30.624, "eval_steps_per_second": 3.832, "step": 200 }, { "epoch": 0.22, "learning_rate": 0.0002, "loss": 2.6509, "step": 300 }, { "epoch": 0.22, "eval_loss": 2.6377696990966797, "eval_runtime": 204.8779, "eval_samples_per_second": 30.623, "eval_steps_per_second": 3.832, "step": 300 }, { "epoch": 0.29, "learning_rate": 0.0002, "loss": 2.6248, "step": 400 }, { "epoch": 0.29, "eval_loss": 2.6048855781555176, "eval_runtime": 204.7551, "eval_samples_per_second": 30.641, "eval_steps_per_second": 3.834, "step": 400 }, { "epoch": 0.36, "learning_rate": 0.0002, "loss": 2.5822, "step": 500 }, { "epoch": 0.36, "eval_loss": 2.5791709423065186, "eval_runtime": 204.976, "eval_samples_per_second": 30.608, "eval_steps_per_second": 3.83, "step": 500 }, { "epoch": 0.44, "learning_rate": 0.0002, "loss": 2.5557, "step": 600 }, { "epoch": 0.44, "eval_loss": 2.5526647567749023, "eval_runtime": 208.5751, "eval_samples_per_second": 30.08, "eval_steps_per_second": 3.764, "step": 600 }, { "epoch": 0.51, "learning_rate": 0.0002, "loss": 2.5397, "step": 700 }, { "epoch": 0.51, "eval_loss": 2.5280492305755615, "eval_runtime": 204.8306, "eval_samples_per_second": 30.63, "eval_steps_per_second": 3.832, "step": 700 }, { "epoch": 0.58, "learning_rate": 0.0002, "loss": 2.5165, "step": 800 }, { "epoch": 0.58, "eval_loss": 2.507997512817383, "eval_runtime": 204.6726, "eval_samples_per_second": 30.654, "eval_steps_per_second": 3.835, "step": 800 }, { "epoch": 0.66, "learning_rate": 0.0002, "loss": 2.478, "step": 900 }, { "epoch": 0.66, "eval_loss": 2.485630512237549, "eval_runtime": 204.5609, "eval_samples_per_second": 30.671, "eval_steps_per_second": 3.837, "step": 900 }, { "epoch": 0.73, "learning_rate": 0.0002, "loss": 2.4748, "step": 1000 }, { "epoch": 0.73, "eval_loss": 2.4663589000701904, "eval_runtime": 204.9278, "eval_samples_per_second": 30.616, "eval_steps_per_second": 3.831, "step": 1000 }, { "epoch": 0.8, "learning_rate": 0.0002, "loss": 2.4692, "step": 1100 }, { "epoch": 0.8, "eval_loss": 2.4438841342926025, "eval_runtime": 204.9158, "eval_samples_per_second": 30.617, "eval_steps_per_second": 3.831, "step": 1100 }, { "epoch": 0.87, "learning_rate": 0.0002, "loss": 2.4221, "step": 1200 }, { "epoch": 0.87, "eval_loss": 2.42663836479187, "eval_runtime": 204.7233, "eval_samples_per_second": 30.646, "eval_steps_per_second": 3.834, "step": 1200 } ], "logging_steps": 100, "max_steps": 4116, "num_train_epochs": 3, "save_steps": 100, "total_flos": 3.41664786505728e+16, "trial_name": null, "trial_params": null }