{ "best_metric": 2.3411779403686523, "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": 2.7403, "step": 100 }, { "epoch": 0.07, "eval_loss": 2.642388343811035, "eval_runtime": 205.2517, "eval_samples_per_second": 30.567, "eval_steps_per_second": 3.825, "step": 100 }, { "epoch": 0.15, "learning_rate": 0.0002, "loss": 2.606, "step": 200 }, { "epoch": 0.15, "eval_loss": 2.5923426151275635, "eval_runtime": 208.3105, "eval_samples_per_second": 30.119, "eval_steps_per_second": 3.768, "step": 200 }, { "epoch": 0.22, "learning_rate": 0.0002, "loss": 2.5632, "step": 300 }, { "epoch": 0.22, "eval_loss": 2.5577170848846436, "eval_runtime": 205.6257, "eval_samples_per_second": 30.512, "eval_steps_per_second": 3.818, "step": 300 }, { "epoch": 0.29, "learning_rate": 0.0002, "loss": 2.5369, "step": 400 }, { "epoch": 0.29, "eval_loss": 2.5247888565063477, "eval_runtime": 205.7237, "eval_samples_per_second": 30.497, "eval_steps_per_second": 3.816, "step": 400 }, { "epoch": 0.36, "learning_rate": 0.0002, "loss": 2.4954, "step": 500 }, { "epoch": 0.36, "eval_loss": 2.498934745788574, "eval_runtime": 207.1055, "eval_samples_per_second": 30.294, "eval_steps_per_second": 3.79, "step": 500 }, { "epoch": 0.44, "learning_rate": 0.0002, "loss": 2.469, "step": 600 }, { "epoch": 0.44, "eval_loss": 2.470606565475464, "eval_runtime": 205.8606, "eval_samples_per_second": 30.477, "eval_steps_per_second": 3.813, "step": 600 }, { "epoch": 0.51, "learning_rate": 0.0002, "loss": 2.4509, "step": 700 }, { "epoch": 0.51, "eval_loss": 2.446812391281128, "eval_runtime": 205.2079, "eval_samples_per_second": 30.574, "eval_steps_per_second": 3.825, "step": 700 }, { "epoch": 0.58, "learning_rate": 0.0002, "loss": 2.4284, "step": 800 }, { "epoch": 0.58, "eval_loss": 2.42557954788208, "eval_runtime": 205.3478, "eval_samples_per_second": 30.553, "eval_steps_per_second": 3.823, "step": 800 }, { "epoch": 0.66, "learning_rate": 0.0002, "loss": 2.3916, "step": 900 }, { "epoch": 0.66, "eval_loss": 2.403062343597412, "eval_runtime": 204.8842, "eval_samples_per_second": 30.622, "eval_steps_per_second": 3.831, "step": 900 }, { "epoch": 0.73, "learning_rate": 0.0002, "loss": 2.3883, "step": 1000 }, { "epoch": 0.73, "eval_loss": 2.384012222290039, "eval_runtime": 204.4165, "eval_samples_per_second": 30.692, "eval_steps_per_second": 3.84, "step": 1000 }, { "epoch": 0.8, "learning_rate": 0.0002, "loss": 2.382, "step": 1100 }, { "epoch": 0.8, "eval_loss": 2.361379384994507, "eval_runtime": 204.6076, "eval_samples_per_second": 30.664, "eval_steps_per_second": 3.837, "step": 1100 }, { "epoch": 0.87, "learning_rate": 0.0002, "loss": 2.3372, "step": 1200 }, { "epoch": 0.87, "eval_loss": 2.3411779403686523, "eval_runtime": 204.8024, "eval_samples_per_second": 30.634, "eval_steps_per_second": 3.833, "step": 1200 } ], "logging_steps": 100, "max_steps": 4116, "num_train_epochs": 3, "save_steps": 100, "total_flos": 3.500881324867584e+16, "trial_name": null, "trial_params": null }