| { | |
| "best_metric": 1.4534634351730347, | |
| "best_model_checkpoint": "./outputs/checkpoint-2000", | |
| "epoch": 1.4571948998178508, | |
| "eval_steps": 100, | |
| "global_step": 2000, | |
| "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": 2.182, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_loss": 2.0125839710235596, | |
| "eval_runtime": 144.1418, | |
| "eval_samples_per_second": 43.527, | |
| "eval_steps_per_second": 5.446, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0002, | |
| "loss": 1.9626, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 1.9195271730422974, | |
| "eval_runtime": 144.2253, | |
| "eval_samples_per_second": 43.501, | |
| "eval_steps_per_second": 5.443, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8881, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_loss": 1.858821988105774, | |
| "eval_runtime": 144.0313, | |
| "eval_samples_per_second": 43.56, | |
| "eval_steps_per_second": 5.45, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8367, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": 1.8125503063201904, | |
| "eval_runtime": 143.9861, | |
| "eval_samples_per_second": 43.574, | |
| "eval_steps_per_second": 5.452, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0002, | |
| "loss": 1.785, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_loss": 1.7720445394515991, | |
| "eval_runtime": 144.0061, | |
| "eval_samples_per_second": 43.568, | |
| "eval_steps_per_second": 5.451, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0002, | |
| "loss": 1.7562, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_loss": 1.7424511909484863, | |
| "eval_runtime": 144.0245, | |
| "eval_samples_per_second": 43.562, | |
| "eval_steps_per_second": 5.45, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.0002, | |
| "loss": 1.7176, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_loss": 1.7112771272659302, | |
| "eval_runtime": 144.2204, | |
| "eval_samples_per_second": 43.503, | |
| "eval_steps_per_second": 5.443, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.0002, | |
| "loss": 1.6969, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_loss": 1.6823335886001587, | |
| "eval_runtime": 144.0134, | |
| "eval_samples_per_second": 43.565, | |
| "eval_steps_per_second": 5.451, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.0002, | |
| "loss": 1.6619, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_loss": 1.6567240953445435, | |
| "eval_runtime": 144.0655, | |
| "eval_samples_per_second": 43.55, | |
| "eval_steps_per_second": 5.449, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.0002, | |
| "loss": 1.6469, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "eval_loss": 1.6352009773254395, | |
| "eval_runtime": 144.0664, | |
| "eval_samples_per_second": 43.549, | |
| "eval_steps_per_second": 5.449, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0002, | |
| "loss": 1.6461, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 1.6240431070327759, | |
| "eval_runtime": 293.5989, | |
| "eval_samples_per_second": 21.369, | |
| "eval_steps_per_second": 2.674, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5992, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_loss": 1.5974311828613281, | |
| "eval_runtime": 291.7, | |
| "eval_samples_per_second": 21.508, | |
| "eval_steps_per_second": 2.691, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 0.0002, | |
| "loss": 1.6021, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "eval_loss": 1.5751127004623413, | |
| "eval_runtime": 289.9524, | |
| "eval_samples_per_second": 21.638, | |
| "eval_steps_per_second": 2.707, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5538, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_loss": 1.5539450645446777, | |
| "eval_runtime": 287.8748, | |
| "eval_samples_per_second": 21.794, | |
| "eval_steps_per_second": 2.727, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5249, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "eval_loss": 1.5348094701766968, | |
| "eval_runtime": 287.891, | |
| "eval_samples_per_second": 21.793, | |
| "eval_steps_per_second": 2.727, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 0.0002, | |
| "loss": 1.506, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "eval_loss": 1.515953540802002, | |
| "eval_runtime": 289.836, | |
| "eval_samples_per_second": 21.647, | |
| "eval_steps_per_second": 2.708, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5042, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "eval_loss": 1.4988901615142822, | |
| "eval_runtime": 291.5471, | |
| "eval_samples_per_second": 21.52, | |
| "eval_steps_per_second": 2.693, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4762, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "eval_loss": 1.4844294786453247, | |
| "eval_runtime": 293.6668, | |
| "eval_samples_per_second": 21.364, | |
| "eval_steps_per_second": 2.673, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4652, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "eval_loss": 1.4694663286209106, | |
| "eval_runtime": 295.5867, | |
| "eval_samples_per_second": 21.226, | |
| "eval_steps_per_second": 2.656, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4481, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "eval_loss": 1.4534634351730347, | |
| "eval_runtime": 296.7451, | |
| "eval_samples_per_second": 21.143, | |
| "eval_steps_per_second": 2.645, | |
| "step": 2000 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 4116, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 1.192408030183342e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |