| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.090225563909774, | |
| "global_step": 300, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 4.999999999999999e-07, | |
| "loss": 1.8578, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 6.505149978319905e-07, | |
| "loss": 1.719, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 7.38560627359831e-07, | |
| "loss": 1.7023, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 8.01029995663981e-07, | |
| "loss": 1.7111, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 8.494850021680092e-07, | |
| "loss": 1.6419, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 8.890756251918216e-07, | |
| "loss": 1.5888, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 9.225490200071283e-07, | |
| "loss": 1.5672, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 9.515449934959715e-07, | |
| "loss": 1.525, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 9.771212547196622e-07, | |
| "loss": 1.5381, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 9.999999999999997e-07, | |
| "loss": 1.6184, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "eval_oa_private_accuracy": 0.6486486486486487, | |
| "eval_oa_private_loss": 1.4580078125, | |
| "eval_oa_private_runtime": 2.0191, | |
| "eval_oa_private_samples_per_second": 8.42, | |
| "eval_oa_private_steps_per_second": 0.495, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 9.941935483870966e-07, | |
| "loss": 1.4745, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 3.63, | |
| "learning_rate": 9.87741935483871e-07, | |
| "loss": 1.5171, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "learning_rate": 9.812903225806452e-07, | |
| "loss": 1.5434, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 4.24, | |
| "learning_rate": 9.741935483870968e-07, | |
| "loss": 1.5055, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 4.54, | |
| "learning_rate": 9.67741935483871e-07, | |
| "loss": 1.4663, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 4.84, | |
| "learning_rate": 9.612903225806452e-07, | |
| "loss": 1.4662, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 5.15, | |
| "learning_rate": 9.548387096774193e-07, | |
| "loss": 1.4806, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "learning_rate": 9.483870967741935e-07, | |
| "loss": 1.4755, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 5.75, | |
| "learning_rate": 9.419354838709677e-07, | |
| "loss": 1.4745, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "learning_rate": 9.354838709677418e-07, | |
| "loss": 1.4963, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "eval_oa_private_accuracy": 0.6705276705276705, | |
| "eval_oa_private_loss": 1.490234375, | |
| "eval_oa_private_runtime": 0.8407, | |
| "eval_oa_private_samples_per_second": 20.222, | |
| "eval_oa_private_steps_per_second": 1.19, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 6.36, | |
| "learning_rate": 9.290322580645161e-07, | |
| "loss": 1.4592, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 6.66, | |
| "learning_rate": 9.225806451612903e-07, | |
| "loss": 1.447, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 6.96, | |
| "learning_rate": 9.161290322580644e-07, | |
| "loss": 1.4527, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 7.27, | |
| "learning_rate": 9.096774193548387e-07, | |
| "loss": 1.4361, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 7.57, | |
| "learning_rate": 9.032258064516129e-07, | |
| "loss": 1.4065, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 7.87, | |
| "learning_rate": 8.96774193548387e-07, | |
| "loss": 1.4299, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 8.18, | |
| "learning_rate": 8.896774193548387e-07, | |
| "loss": 1.4363, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 8.48, | |
| "learning_rate": 8.832258064516129e-07, | |
| "loss": 1.4161, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 8.78, | |
| "learning_rate": 8.767741935483871e-07, | |
| "loss": 1.4356, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "learning_rate": 8.703225806451613e-07, | |
| "loss": 1.4495, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "eval_oa_private_accuracy": 0.6664378860672615, | |
| "eval_oa_private_loss": 1.3466796875, | |
| "eval_oa_private_runtime": 0.9592, | |
| "eval_oa_private_samples_per_second": 17.722, | |
| "eval_oa_private_steps_per_second": 1.042, | |
| "step": 300 | |
| } | |
| ], | |
| "max_steps": 1650, | |
| "num_train_epochs": 50, | |
| "total_flos": 8.233564448028099e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |