| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "global_step": 280, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.0001942857142857143, | |
| "loss": 2.4106, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 0.00018714285714285716, | |
| "loss": 1.1806, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 0.00018, | |
| "loss": 0.221, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "learning_rate": 0.00017285714285714287, | |
| "loss": 0.0374, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "learning_rate": 0.00016571428571428575, | |
| "loss": 0.0099, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.6155198812484741, | |
| "eval_runtime": 1.2934, | |
| "eval_samples_per_second": 25.515, | |
| "eval_steps_per_second": 3.866, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.29, | |
| "learning_rate": 0.00015857142857142857, | |
| "loss": 0.0047, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 0.00015142857142857143, | |
| "loss": 0.0013, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 5.71, | |
| "learning_rate": 0.00014428571428571428, | |
| "loss": 0.0011, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 6.43, | |
| "learning_rate": 0.00013714285714285716, | |
| "loss": 0.0009, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 7.14, | |
| "learning_rate": 0.00013000000000000002, | |
| "loss": 0.0007, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 7.14, | |
| "eval_accuracy": 0.7575757575757576, | |
| "eval_loss": 0.7440532445907593, | |
| "eval_runtime": 1.2943, | |
| "eval_samples_per_second": 25.497, | |
| "eval_steps_per_second": 3.863, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 7.86, | |
| "learning_rate": 0.00012285714285714287, | |
| "loss": 0.0005, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 8.57, | |
| "learning_rate": 0.00011571428571428574, | |
| "loss": 0.0005, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 9.29, | |
| "learning_rate": 0.00010857142857142856, | |
| "loss": 0.0005, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 0.00010142857142857143, | |
| "loss": 0.0004, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 10.71, | |
| "learning_rate": 9.428571428571429e-05, | |
| "loss": 0.0004, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 10.71, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.6925162076950073, | |
| "eval_runtime": 1.4314, | |
| "eval_samples_per_second": 23.054, | |
| "eval_steps_per_second": 3.493, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 11.43, | |
| "learning_rate": 8.714285714285715e-05, | |
| "loss": 0.0004, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 12.14, | |
| "learning_rate": 8e-05, | |
| "loss": 0.0004, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 12.86, | |
| "learning_rate": 7.285714285714286e-05, | |
| "loss": 0.0004, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 13.57, | |
| "learning_rate": 6.571428571428571e-05, | |
| "loss": 0.0003, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 14.29, | |
| "learning_rate": 5.8571428571428575e-05, | |
| "loss": 0.0003, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 14.29, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.6792979836463928, | |
| "eval_runtime": 1.395, | |
| "eval_samples_per_second": 23.655, | |
| "eval_steps_per_second": 3.584, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 5.142857142857143e-05, | |
| "loss": 0.0003, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 15.71, | |
| "learning_rate": 4.428571428571428e-05, | |
| "loss": 0.0003, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 16.43, | |
| "learning_rate": 3.7142857142857143e-05, | |
| "loss": 0.0003, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 17.14, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0003, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 17.86, | |
| "learning_rate": 2.2857142857142858e-05, | |
| "loss": 0.0003, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 17.86, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.6743510365486145, | |
| "eval_runtime": 1.4417, | |
| "eval_samples_per_second": 22.889, | |
| "eval_steps_per_second": 3.468, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 18.57, | |
| "learning_rate": 1.5714285714285715e-05, | |
| "loss": 0.0003, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 19.29, | |
| "learning_rate": 8.571428571428573e-06, | |
| "loss": 0.0003, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 1.4285714285714286e-06, | |
| "loss": 0.0003, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 280, | |
| "total_flos": 8.52488284741632e+16, | |
| "train_loss": 0.13838704336334817, | |
| "train_runtime": 109.7367, | |
| "train_samples_per_second": 10.024, | |
| "train_steps_per_second": 2.552 | |
| } | |
| ], | |
| "max_steps": 280, | |
| "num_train_epochs": 20, | |
| "total_flos": 8.52488284741632e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |