| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 19.99874843554443, | |
| "global_step": 7980, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 2.495e-05, | |
| "loss": 2.2002, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "eval_accuracy": 0.3199999928474426, | |
| "eval_loss": 1.9293218851089478, | |
| "eval_runtime": 130.8072, | |
| "eval_samples_per_second": 1.529, | |
| "eval_steps_per_second": 1.529, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 4.99e-05, | |
| "loss": 1.7914, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "eval_accuracy": 0.4699999988079071, | |
| "eval_loss": 1.528580904006958, | |
| "eval_runtime": 135.4283, | |
| "eval_samples_per_second": 1.477, | |
| "eval_steps_per_second": 1.477, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "learning_rate": 7.475000000000001e-05, | |
| "loss": 1.6111, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "eval_accuracy": 0.3050000071525574, | |
| "eval_loss": 2.1590416431427, | |
| "eval_runtime": 133.2351, | |
| "eval_samples_per_second": 1.501, | |
| "eval_steps_per_second": 1.501, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "learning_rate": 9.975000000000001e-05, | |
| "loss": 1.4597, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "eval_accuracy": 0.46000000834465027, | |
| "eval_loss": 1.557754635810852, | |
| "eval_runtime": 131.9409, | |
| "eval_samples_per_second": 1.516, | |
| "eval_steps_per_second": 1.516, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.27, | |
| "learning_rate": 3.7578814627994955e-05, | |
| "loss": 1.284, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 6.27, | |
| "eval_accuracy": 0.6650000214576721, | |
| "eval_loss": 1.078925609588623, | |
| "eval_runtime": 130.6559, | |
| "eval_samples_per_second": 1.531, | |
| "eval_steps_per_second": 1.531, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 7.52, | |
| "learning_rate": 8.336120401337793e-05, | |
| "loss": 1.1617, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 7.52, | |
| "eval_accuracy": 0.6050000190734863, | |
| "eval_loss": 1.660351276397705, | |
| "eval_runtime": 133.1101, | |
| "eval_samples_per_second": 1.503, | |
| "eval_steps_per_second": 1.503, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 8.77, | |
| "learning_rate": 7.501672240802676e-05, | |
| "loss": 1.0521, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 8.77, | |
| "eval_accuracy": 0.699999988079071, | |
| "eval_loss": 1.3572155237197876, | |
| "eval_runtime": 132.8422, | |
| "eval_samples_per_second": 1.506, | |
| "eval_steps_per_second": 1.506, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 10.03, | |
| "learning_rate": 6.665551839464883e-05, | |
| "loss": 0.928, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 10.03, | |
| "eval_accuracy": 0.7200000286102295, | |
| "eval_loss": 1.5309524536132812, | |
| "eval_runtime": 133.7214, | |
| "eval_samples_per_second": 1.496, | |
| "eval_steps_per_second": 1.496, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 11.28, | |
| "learning_rate": 5.829431438127091e-05, | |
| "loss": 0.7632, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 11.28, | |
| "eval_accuracy": 0.7749999761581421, | |
| "eval_loss": 1.2670022249221802, | |
| "eval_runtime": 132.539, | |
| "eval_samples_per_second": 1.509, | |
| "eval_steps_per_second": 1.509, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 12.53, | |
| "learning_rate": 4.993311036789298e-05, | |
| "loss": 0.6718, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 12.53, | |
| "eval_accuracy": 0.7900000214576721, | |
| "eval_loss": 1.2351398468017578, | |
| "eval_runtime": 132.8123, | |
| "eval_samples_per_second": 1.506, | |
| "eval_steps_per_second": 1.506, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 13.78, | |
| "learning_rate": 4.157190635451505e-05, | |
| "loss": 0.4634, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 13.78, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 1.7186837196350098, | |
| "eval_runtime": 131.5974, | |
| "eval_samples_per_second": 1.52, | |
| "eval_steps_per_second": 1.52, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 15.04, | |
| "learning_rate": 3.322742474916388e-05, | |
| "loss": 0.373, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 15.04, | |
| "eval_accuracy": 0.7400000095367432, | |
| "eval_loss": 1.6782505512237549, | |
| "eval_runtime": 130.9999, | |
| "eval_samples_per_second": 1.527, | |
| "eval_steps_per_second": 1.527, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 16.29, | |
| "learning_rate": 2.4866220735785955e-05, | |
| "loss": 0.349, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 16.29, | |
| "eval_accuracy": 0.7900000214576721, | |
| "eval_loss": 1.2969427108764648, | |
| "eval_runtime": 131.4011, | |
| "eval_samples_per_second": 1.522, | |
| "eval_steps_per_second": 1.522, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 17.54, | |
| "learning_rate": 1.650501672240803e-05, | |
| "loss": 0.2225, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 17.54, | |
| "eval_accuracy": 0.7799999713897705, | |
| "eval_loss": 1.6187431812286377, | |
| "eval_runtime": 130.156, | |
| "eval_samples_per_second": 1.537, | |
| "eval_steps_per_second": 1.537, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 18.8, | |
| "learning_rate": 8.1438127090301e-06, | |
| "loss": 0.2156, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 18.8, | |
| "eval_accuracy": 0.7850000262260437, | |
| "eval_loss": 1.5765472650527954, | |
| "eval_runtime": 130.2954, | |
| "eval_samples_per_second": 1.535, | |
| "eval_steps_per_second": 1.535, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 7980, | |
| "total_flos": 4.373813712366772e+18, | |
| "train_runtime": 7059.2772, | |
| "train_samples_per_second": 2.264, | |
| "train_steps_per_second": 1.13 | |
| } | |
| ], | |
| "max_steps": 7980, | |
| "num_train_epochs": 20, | |
| "total_flos": 4.373813712366772e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |