{ "all_params": 2211058176, "memory_footprint": 4426269008, "total_flos": 2.136820048293888e+16, "train_loss": 0.2946822406862069, "train_runtime": 4276.0328, "train_samples_per_second": 0.745, "train_steps_per_second": 0.093, "trainable_params": 2072576, "trainable_params_percent": 0.09373683707180756 }