| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.7584, | |
| "global_step": 1896, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1.9624e-05, | |
| "loss": 0.9388, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 1.9232e-05, | |
| "loss": 0.9522, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 1.8832000000000002e-05, | |
| "loss": 0.8005, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 1.8432000000000002e-05, | |
| "loss": 0.6908, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 1.8032e-05, | |
| "loss": 0.7927, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_accuracy": 0.6659167604049494, | |
| "eval_loss": 0.7351371645927429, | |
| "eval_runtime": 434.6411, | |
| "eval_samples_per_second": 2.045, | |
| "eval_steps_per_second": 2.045, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 1.764e-05, | |
| "loss": 0.5962, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 1.724e-05, | |
| "loss": 0.6989, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 1.684e-05, | |
| "loss": 0.7051, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 1.6440000000000002e-05, | |
| "loss": 0.6664, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 1.6040000000000002e-05, | |
| "loss": 0.6547, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_accuracy": 0.7041619797525309, | |
| "eval_loss": 0.6934666037559509, | |
| "eval_runtime": 436.2176, | |
| "eval_samples_per_second": 2.038, | |
| "eval_steps_per_second": 2.038, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 1.5640000000000003e-05, | |
| "loss": 0.665, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 1.5240000000000001e-05, | |
| "loss": 0.654, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 1.4840000000000002e-05, | |
| "loss": 0.6714, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 1.444e-05, | |
| "loss": 0.7395, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 1.4040000000000001e-05, | |
| "loss": 0.5393, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_accuracy": 0.7142857142857143, | |
| "eval_loss": 0.621578574180603, | |
| "eval_runtime": 436.3187, | |
| "eval_samples_per_second": 2.038, | |
| "eval_steps_per_second": 2.038, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 1.3640000000000002e-05, | |
| "loss": 0.5185, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 1.3240000000000002e-05, | |
| "loss": 0.6009, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 1.284e-05, | |
| "loss": 0.6588, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 1.2440000000000001e-05, | |
| "loss": 0.6022, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 1.204e-05, | |
| "loss": 0.7316, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_accuracy": 0.734533183352081, | |
| "eval_loss": 0.5916205644607544, | |
| "eval_runtime": 436.6514, | |
| "eval_samples_per_second": 2.036, | |
| "eval_steps_per_second": 2.036, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 1.164e-05, | |
| "loss": 0.6086, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 1.1240000000000002e-05, | |
| "loss": 0.5806, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 1.0840000000000001e-05, | |
| "loss": 0.5992, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 1.0440000000000002e-05, | |
| "loss": 0.5807, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 1.004e-05, | |
| "loss": 0.5667, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_accuracy": 0.734533183352081, | |
| "eval_loss": 0.5785398483276367, | |
| "eval_runtime": 436.2096, | |
| "eval_samples_per_second": 2.038, | |
| "eval_steps_per_second": 2.038, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 9.640000000000001e-06, | |
| "loss": 0.4989, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 9.240000000000001e-06, | |
| "loss": 0.6015, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 8.848e-06, | |
| "loss": 0.5728, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 8.448000000000001e-06, | |
| "loss": 0.6285, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 8.048e-06, | |
| "loss": 0.498, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_accuracy": 0.7435320584926884, | |
| "eval_loss": 0.5632913708686829, | |
| "eval_runtime": 436.3374, | |
| "eval_samples_per_second": 2.037, | |
| "eval_steps_per_second": 2.037, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 7.648e-06, | |
| "loss": 0.5134, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 7.248000000000001e-06, | |
| "loss": 0.4582, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 6.848e-06, | |
| "loss": 0.534, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 6.448000000000001e-06, | |
| "loss": 0.6765, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 6.048e-06, | |
| "loss": 0.6598, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "eval_accuracy": 0.7457817772778402, | |
| "eval_loss": 0.565944254398346, | |
| "eval_runtime": 435.7582, | |
| "eval_samples_per_second": 2.04, | |
| "eval_steps_per_second": 2.04, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 5.648e-06, | |
| "loss": 0.5006, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 5.248000000000001e-06, | |
| "loss": 0.579, | |
| "step": 1850 | |
| } | |
| ], | |
| "max_steps": 2500, | |
| "num_train_epochs": 1, | |
| "total_flos": 0.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |