| { | |
| "best_metric": 0.7070142891290766, | |
| "best_model_checkpoint": "/home/felix/kubhist2-testing/SuperLim-2-Testing/results/trainer_output/run-5d9be_00002/checkpoint-8896", | |
| "epoch": 4.0, | |
| "eval_steps": 500, | |
| "global_step": 8896, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 1.1146067415730336e-05, | |
| "loss": 0.5378, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 2.2382022471910112e-05, | |
| "loss": 0.3896, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 2.976895479550347e-05, | |
| "loss": 0.3679, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 2.905142310452045e-05, | |
| "loss": 0.3735, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8432581880051042, | |
| "eval_krippendorff_nominal": 0.6854738483266531, | |
| "eval_loss": 0.3858277499675751, | |
| "eval_runtime": 3.9653, | |
| "eval_samples_per_second": 1185.781, | |
| "eval_steps_per_second": 18.662, | |
| "learning_rate": 2.8729968906960057e-05, | |
| "step": 2224 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 2.833389141353743e-05, | |
| "loss": 0.3276, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 2.7616359722554415e-05, | |
| "loss": 0.3014, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 2.6898828031571395e-05, | |
| "loss": 0.3091, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 2.6181296340588376e-05, | |
| "loss": 0.3179, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8441088898341131, | |
| "eval_krippendorff_nominal": 0.6860352395475762, | |
| "eval_loss": 0.3822194039821625, | |
| "eval_runtime": 3.9064, | |
| "eval_samples_per_second": 1203.675, | |
| "eval_steps_per_second": 18.943, | |
| "learning_rate": 2.5539823008849558e-05, | |
| "step": 4448 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 2.5465199712987326e-05, | |
| "loss": 0.3132, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 2.4749103085386272e-05, | |
| "loss": 0.2406, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 2.4031571394403256e-05, | |
| "loss": 0.2518, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 2.3314039703420236e-05, | |
| "loss": 0.2558, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 2.2596508012437217e-05, | |
| "loss": 0.2739, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.839430029774564, | |
| "eval_krippendorff_nominal": 0.6751470306348389, | |
| "eval_loss": 0.5222128629684448, | |
| "eval_runtime": 3.8961, | |
| "eval_samples_per_second": 1206.858, | |
| "eval_steps_per_second": 18.994, | |
| "learning_rate": 2.234967711073906e-05, | |
| "step": 6672 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "learning_rate": 2.1878976321454197e-05, | |
| "loss": 0.2358, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "learning_rate": 2.116144463047118e-05, | |
| "loss": 0.2057, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 2.044391293948816e-05, | |
| "loss": 0.2157, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.82, | |
| "learning_rate": 1.972638124850514e-05, | |
| "loss": 0.2295, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8543173117822204, | |
| "eval_krippendorff_nominal": 0.7070142891290766, | |
| "eval_loss": 0.44575992226600647, | |
| "eval_runtime": 3.9739, | |
| "eval_samples_per_second": 1183.225, | |
| "eval_steps_per_second": 18.622, | |
| "learning_rate": 1.9158096149246593e-05, | |
| "step": 8896 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 22240, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 5898489416824320.0, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 3e-05, | |
| "per_device_train_batch_size": 16 | |
| } | |
| } | |