| { | |
| "best_metric": 0.19139504432678223, | |
| "best_model_checkpoint": "pri_docidv2/checkpoint-200", | |
| "epoch": 10.0, | |
| "eval_steps": 100, | |
| "global_step": 310, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.00029032258064516127, | |
| "loss": 0.1105, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.00028064516129032256, | |
| "loss": 0.1204, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 0.00027096774193548386, | |
| "loss": 0.0642, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 0.00026129032258064515, | |
| "loss": 0.0199, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 0.00025161290322580645, | |
| "loss": 0.1239, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 0.00024193548387096771, | |
| "loss": 0.0675, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 0.000232258064516129, | |
| "loss": 0.0115, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 0.0002225806451612903, | |
| "loss": 0.095, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 0.0002129032258064516, | |
| "loss": 0.1924, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 0.00020322580645161287, | |
| "loss": 0.0366, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "eval_accuracy": 0.9285714285714286, | |
| "eval_loss": 0.3684941828250885, | |
| "eval_runtime": 2.4287, | |
| "eval_samples_per_second": 86.467, | |
| "eval_steps_per_second": 11.117, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "learning_rate": 0.00019354838709677416, | |
| "loss": 0.0101, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 0.00018387096774193548, | |
| "loss": 0.0275, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "learning_rate": 0.00017419354838709678, | |
| "loss": 0.0815, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "learning_rate": 0.00016451612903225804, | |
| "loss": 0.0098, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 4.84, | |
| "learning_rate": 0.00015483870967741934, | |
| "loss": 0.0051, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "learning_rate": 0.00014516129032258063, | |
| "loss": 0.1023, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 5.48, | |
| "learning_rate": 0.00013548387096774193, | |
| "loss": 0.0874, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 5.81, | |
| "learning_rate": 0.00012580645161290322, | |
| "loss": 0.0113, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 6.13, | |
| "learning_rate": 0.0001161290322580645, | |
| "loss": 0.0469, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "learning_rate": 0.0001064516129032258, | |
| "loss": 0.006, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "eval_accuracy": 0.9571428571428572, | |
| "eval_loss": 0.19139504432678223, | |
| "eval_runtime": 2.4676, | |
| "eval_samples_per_second": 85.101, | |
| "eval_steps_per_second": 10.942, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 6.77, | |
| "learning_rate": 9.677419354838708e-05, | |
| "loss": 0.0467, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "learning_rate": 8.709677419354839e-05, | |
| "loss": 0.0042, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 7.42, | |
| "learning_rate": 7.741935483870967e-05, | |
| "loss": 0.0036, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 7.74, | |
| "learning_rate": 6.774193548387096e-05, | |
| "loss": 0.0026, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "learning_rate": 5.806451612903225e-05, | |
| "loss": 0.0032, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 8.39, | |
| "learning_rate": 4.838709677419354e-05, | |
| "loss": 0.0028, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 8.71, | |
| "learning_rate": 3.8709677419354835e-05, | |
| "loss": 0.0027, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 2.9032258064516126e-05, | |
| "loss": 0.0028, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 9.35, | |
| "learning_rate": 1.9354838709677417e-05, | |
| "loss": 0.002, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 9.68, | |
| "learning_rate": 9.677419354838709e-06, | |
| "loss": 0.0031, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 9.68, | |
| "eval_accuracy": 0.9571428571428572, | |
| "eval_loss": 0.21088813245296478, | |
| "eval_runtime": 2.491, | |
| "eval_samples_per_second": 84.305, | |
| "eval_steps_per_second": 10.839, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.003, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 310, | |
| "total_flos": 3.7817107732905984e+17, | |
| "train_loss": 0.04214024447625683, | |
| "train_runtime": 135.0852, | |
| "train_samples_per_second": 36.125, | |
| "train_steps_per_second": 2.295 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 310, | |
| "num_train_epochs": 10, | |
| "save_steps": 100, | |
| "total_flos": 3.7817107732905984e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |