| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "global_step": 2500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1.9632e-05, | |
| "loss": 0.9899, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 1.9232e-05, | |
| "loss": 0.9949, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 1.8832000000000002e-05, | |
| "loss": 0.828, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 1.8432000000000002e-05, | |
| "loss": 0.8372, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 1.8032e-05, | |
| "loss": 0.8409, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_accuracy": 0.6220472440944882, | |
| "eval_loss": 0.8243474364280701, | |
| "eval_runtime": 435.8774, | |
| "eval_samples_per_second": 2.04, | |
| "eval_steps_per_second": 2.04, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 1.7632000000000003e-05, | |
| "loss": 0.694, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 1.7232000000000004e-05, | |
| "loss": 0.6918, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 1.6832e-05, | |
| "loss": 0.6794, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 1.6432e-05, | |
| "loss": 0.6338, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 1.6032e-05, | |
| "loss": 0.6288, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_accuracy": 0.671541057367829, | |
| "eval_loss": 0.7539446949958801, | |
| "eval_runtime": 434.4073, | |
| "eval_samples_per_second": 2.046, | |
| "eval_steps_per_second": 2.046, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 1.5632000000000002e-05, | |
| "loss": 0.6622, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 1.5232000000000003e-05, | |
| "loss": 0.773, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 1.4832000000000001e-05, | |
| "loss": 0.6051, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 1.4432000000000002e-05, | |
| "loss": 0.7805, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 1.4032e-05, | |
| "loss": 0.5882, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_accuracy": 0.7075365579302587, | |
| "eval_loss": 0.6791747808456421, | |
| "eval_runtime": 433.8268, | |
| "eval_samples_per_second": 2.049, | |
| "eval_steps_per_second": 2.049, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 1.3632000000000001e-05, | |
| "loss": 0.5672, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 1.3232e-05, | |
| "loss": 0.6807, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 1.2832e-05, | |
| "loss": 0.6796, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 1.2432000000000002e-05, | |
| "loss": 0.6922, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 1.2032000000000001e-05, | |
| "loss": 0.7671, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_accuracy": 0.7334083239595051, | |
| "eval_loss": 0.6129724383354187, | |
| "eval_runtime": 433.287, | |
| "eval_samples_per_second": 2.052, | |
| "eval_steps_per_second": 2.052, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 1.1632000000000001e-05, | |
| "loss": 0.645, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 1.1232e-05, | |
| "loss": 0.5891, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 1.0832e-05, | |
| "loss": 0.6426, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 1.0432e-05, | |
| "loss": 0.567, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 1.0032000000000002e-05, | |
| "loss": 0.5782, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_accuracy": 0.7255343082114736, | |
| "eval_loss": 0.6114887595176697, | |
| "eval_runtime": 433.3273, | |
| "eval_samples_per_second": 2.052, | |
| "eval_steps_per_second": 2.052, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 9.632e-06, | |
| "loss": 0.5736, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 9.232e-06, | |
| "loss": 0.6849, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 8.832000000000001e-06, | |
| "loss": 0.5305, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 8.432e-06, | |
| "loss": 0.7265, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 8.032e-06, | |
| "loss": 0.5691, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_accuracy": 0.7412823397075365, | |
| "eval_loss": 0.5794617533683777, | |
| "eval_runtime": 433.4136, | |
| "eval_samples_per_second": 2.051, | |
| "eval_steps_per_second": 2.051, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 7.632e-06, | |
| "loss": 0.519, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 7.232e-06, | |
| "loss": 0.5378, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 6.832000000000001e-06, | |
| "loss": 0.5982, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 6.432e-06, | |
| "loss": 0.7027, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 6.032e-06, | |
| "loss": 0.6579, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "eval_accuracy": 0.7469066366704162, | |
| "eval_loss": 0.5774183869361877, | |
| "eval_runtime": 433.4068, | |
| "eval_samples_per_second": 2.051, | |
| "eval_steps_per_second": 2.051, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 5.6320000000000005e-06, | |
| "loss": 0.5044, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 5.232e-06, | |
| "loss": 0.6482, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 4.8320000000000005e-06, | |
| "loss": 0.5406, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 4.432e-06, | |
| "loss": 0.5372, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 4.0320000000000005e-06, | |
| "loss": 0.6107, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_accuracy": 0.7401574803149606, | |
| "eval_loss": 0.5690832734107971, | |
| "eval_runtime": 433.3967, | |
| "eval_samples_per_second": 2.051, | |
| "eval_steps_per_second": 2.051, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 3.6320000000000005e-06, | |
| "loss": 0.4043, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 3.2400000000000003e-06, | |
| "loss": 0.5344, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 2.84e-06, | |
| "loss": 0.7056, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 2.4400000000000004e-06, | |
| "loss": 0.5719, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 2.04e-06, | |
| "loss": 0.6255, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "eval_accuracy": 0.7435320584926884, | |
| "eval_loss": 0.570974588394165, | |
| "eval_runtime": 433.4106, | |
| "eval_samples_per_second": 2.051, | |
| "eval_steps_per_second": 2.051, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 1.6400000000000002e-06, | |
| "loss": 0.5958, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 1.2400000000000002e-06, | |
| "loss": 0.5984, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 8.480000000000001e-07, | |
| "loss": 0.6103, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 4.4800000000000004e-07, | |
| "loss": 0.612, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.8e-08, | |
| "loss": 0.7034, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7435320584926884, | |
| "eval_loss": 0.5713425874710083, | |
| "eval_runtime": 434.3078, | |
| "eval_samples_per_second": 2.047, | |
| "eval_steps_per_second": 2.047, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 2500, | |
| "total_flos": 0.0, | |
| "train_loss": 0.6507886672973633, | |
| "train_runtime": 17457.236, | |
| "train_samples_per_second": 0.573, | |
| "train_steps_per_second": 0.143 | |
| } | |
| ], | |
| "max_steps": 2500, | |
| "num_train_epochs": 1, | |
| "total_flos": 0.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |