| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 13.0, | |
| "global_step": 13273, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 2.90205680705191e-05, | |
| "loss": 0.1464, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 2.80411361410382e-05, | |
| "loss": 0.1261, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9658201932907104, | |
| "eval_loss": 0.10953927785158157, | |
| "eval_runtime": 3.6633, | |
| "eval_samples_per_second": 2547.697, | |
| "eval_steps_per_second": 79.709, | |
| "step": 1021 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 2.70617042115573e-05, | |
| "loss": 0.0789, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 2.6082272282076394e-05, | |
| "loss": 0.0811, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9686059951782227, | |
| "eval_loss": 0.12239421159029007, | |
| "eval_runtime": 3.6281, | |
| "eval_samples_per_second": 2572.455, | |
| "eval_steps_per_second": 80.484, | |
| "step": 2042 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 2.5102840352595495e-05, | |
| "loss": 0.0422, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "learning_rate": 2.4123408423114592e-05, | |
| "loss": 0.0452, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9654987454414368, | |
| "eval_loss": 0.14466284215450287, | |
| "eval_runtime": 3.6125, | |
| "eval_samples_per_second": 2583.514, | |
| "eval_steps_per_second": 80.83, | |
| "step": 3063 | |
| }, | |
| { | |
| "epoch": 3.43, | |
| "learning_rate": 2.3143976493633693e-05, | |
| "loss": 0.022, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "learning_rate": 2.216454456415279e-05, | |
| "loss": 0.0192, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9684988856315613, | |
| "eval_loss": 0.21616922318935394, | |
| "eval_runtime": 3.6831, | |
| "eval_samples_per_second": 2533.973, | |
| "eval_steps_per_second": 79.28, | |
| "step": 4084 | |
| }, | |
| { | |
| "epoch": 4.41, | |
| "learning_rate": 2.118511263467189e-05, | |
| "loss": 0.0078, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "learning_rate": 2.0205680705190993e-05, | |
| "loss": 0.0088, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9695703387260437, | |
| "eval_loss": 0.19787850975990295, | |
| "eval_runtime": 3.624, | |
| "eval_samples_per_second": 2575.335, | |
| "eval_steps_per_second": 80.574, | |
| "step": 5105 | |
| }, | |
| { | |
| "epoch": 5.39, | |
| "learning_rate": 1.922624877571009e-05, | |
| "loss": 0.0085, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 5.88, | |
| "learning_rate": 1.8246816846229187e-05, | |
| "loss": 0.0086, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9717133045196533, | |
| "eval_loss": 0.25486066937446594, | |
| "eval_runtime": 3.6182, | |
| "eval_samples_per_second": 2579.429, | |
| "eval_steps_per_second": 80.702, | |
| "step": 6126 | |
| }, | |
| { | |
| "epoch": 6.37, | |
| "learning_rate": 1.7267384916748285e-05, | |
| "loss": 0.004, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 6.86, | |
| "learning_rate": 1.6287952987267386e-05, | |
| "loss": 0.0054, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9706417918205261, | |
| "eval_loss": 0.2799299657344818, | |
| "eval_runtime": 3.6353, | |
| "eval_samples_per_second": 2567.344, | |
| "eval_steps_per_second": 80.324, | |
| "step": 7147 | |
| }, | |
| { | |
| "epoch": 7.35, | |
| "learning_rate": 1.5308521057786483e-05, | |
| "loss": 0.0011, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 7.84, | |
| "learning_rate": 1.4329089128305584e-05, | |
| "loss": 0.0046, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9684988856315613, | |
| "eval_loss": 0.2791476547718048, | |
| "eval_runtime": 3.6302, | |
| "eval_samples_per_second": 2570.964, | |
| "eval_steps_per_second": 80.437, | |
| "step": 8168 | |
| }, | |
| { | |
| "epoch": 8.33, | |
| "learning_rate": 1.3349657198824682e-05, | |
| "loss": 0.0046, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 8.81, | |
| "learning_rate": 1.237022526934378e-05, | |
| "loss": 0.0029, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9712846875190735, | |
| "eval_loss": 0.27454128861427307, | |
| "eval_runtime": 3.6234, | |
| "eval_samples_per_second": 2575.771, | |
| "eval_steps_per_second": 80.588, | |
| "step": 9189 | |
| }, | |
| { | |
| "epoch": 9.3, | |
| "learning_rate": 1.139079333986288e-05, | |
| "loss": 0.0028, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 9.79, | |
| "learning_rate": 1.041136141038198e-05, | |
| "loss": 0.0033, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9697846174240112, | |
| "eval_loss": 0.2769125699996948, | |
| "eval_runtime": 3.6089, | |
| "eval_samples_per_second": 2586.089, | |
| "eval_steps_per_second": 80.911, | |
| "step": 10210 | |
| }, | |
| { | |
| "epoch": 10.28, | |
| "learning_rate": 9.431929480901077e-06, | |
| "loss": 0.0016, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 10.77, | |
| "learning_rate": 8.452497551420176e-06, | |
| "loss": 0.0016, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.9690346121788025, | |
| "eval_loss": 0.290558785200119, | |
| "eval_runtime": 3.6122, | |
| "eval_samples_per_second": 2583.771, | |
| "eval_steps_per_second": 80.838, | |
| "step": 11231 | |
| }, | |
| { | |
| "epoch": 11.26, | |
| "learning_rate": 7.473065621939276e-06, | |
| "loss": 0.001, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 11.75, | |
| "learning_rate": 6.493633692458374e-06, | |
| "loss": 0.0017, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.9711775183677673, | |
| "eval_loss": 0.2827688753604889, | |
| "eval_runtime": 3.6265, | |
| "eval_samples_per_second": 2573.584, | |
| "eval_steps_per_second": 80.519, | |
| "step": 12252 | |
| }, | |
| { | |
| "epoch": 12.24, | |
| "learning_rate": 5.514201762977473e-06, | |
| "loss": 0.0011, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 12.73, | |
| "learning_rate": 4.534769833496572e-06, | |
| "loss": 0.0012, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.9723561406135559, | |
| "eval_loss": 0.2707096040248871, | |
| "eval_runtime": 3.6138, | |
| "eval_samples_per_second": 2582.613, | |
| "eval_steps_per_second": 80.802, | |
| "step": 13273 | |
| } | |
| ], | |
| "max_steps": 15315, | |
| "num_train_epochs": 15, | |
| "total_flos": 2.7931343434285056e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |