| { | |
| "best_metric": 2.273698555654846e-06, | |
| "best_model_checkpoint": "data/further_pretraining/roberta-base/checkpoint-11695", | |
| "epoch": 6.0, | |
| "global_step": 14034, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 1.9572466866182133e-05, | |
| "loss": 0.0508, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 1.914493373236426e-05, | |
| "loss": 0.0326, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 1.8717400598546387e-05, | |
| "loss": 0.0135, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 1.828986746472852e-05, | |
| "loss": 0.0023, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.0002380974474363029, | |
| "eval_runtime": 225.964, | |
| "eval_samples_per_second": 331.212, | |
| "eval_steps_per_second": 10.351, | |
| "step": 2339 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 1.7862334330910646e-05, | |
| "loss": 0.0009, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 1.7434801197092777e-05, | |
| "loss": 0.0004, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 1.7007268063274904e-05, | |
| "loss": 0.0004, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 1.6579734929457035e-05, | |
| "loss": 0.0004, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 1.6152201795639163e-05, | |
| "loss": 0.0003, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 2.140663673344534e-05, | |
| "eval_runtime": 228.247, | |
| "eval_samples_per_second": 327.899, | |
| "eval_steps_per_second": 10.248, | |
| "step": 4678 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 1.5724668661821293e-05, | |
| "loss": 0.0001, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 1.529713552800342e-05, | |
| "loss": 0.0001, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 1.4869602394185552e-05, | |
| "loss": 0.0002, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 1.444206926036768e-05, | |
| "loss": 0.0001, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 1.4014536126549808e-05, | |
| "loss": 0.0001, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 5.5228130804607645e-05, | |
| "eval_runtime": 228.2524, | |
| "eval_samples_per_second": 327.891, | |
| "eval_steps_per_second": 10.247, | |
| "step": 7017 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "learning_rate": 1.3587002992731938e-05, | |
| "loss": 0.0001, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "learning_rate": 1.3159469858914067e-05, | |
| "loss": 0.0001, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.63, | |
| "learning_rate": 1.2731936725096196e-05, | |
| "loss": 0.0001, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "learning_rate": 1.2304403591278325e-05, | |
| "loss": 0.0004, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 6.755703907401767e-06, | |
| "eval_runtime": 228.2467, | |
| "eval_samples_per_second": 327.9, | |
| "eval_steps_per_second": 10.248, | |
| "step": 9356 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "learning_rate": 1.1876870457460454e-05, | |
| "loss": 0.0001, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "learning_rate": 1.1449337323642584e-05, | |
| "loss": 0.0, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.49, | |
| "learning_rate": 1.1021804189824713e-05, | |
| "loss": 0.0002, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 4.7, | |
| "learning_rate": 1.0594271056006842e-05, | |
| "loss": 0.0001, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 4.92, | |
| "learning_rate": 1.016673792218897e-05, | |
| "loss": 0.0, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 2.273698555654846e-06, | |
| "eval_runtime": 227.9681, | |
| "eval_samples_per_second": 328.3, | |
| "eval_steps_per_second": 10.26, | |
| "step": 11695 | |
| }, | |
| { | |
| "epoch": 5.13, | |
| "learning_rate": 9.739204788371099e-06, | |
| "loss": 0.0007, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 5.34, | |
| "learning_rate": 9.311671654553228e-06, | |
| "loss": 0.0001, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 5.56, | |
| "learning_rate": 8.884138520735359e-06, | |
| "loss": 0.0, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 5.77, | |
| "learning_rate": 8.456605386917486e-06, | |
| "loss": 0.0001, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 5.99, | |
| "learning_rate": 8.029072253099615e-06, | |
| "loss": 0.0001, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 3.073917014262406e-06, | |
| "eval_runtime": 228.2599, | |
| "eval_samples_per_second": 327.881, | |
| "eval_steps_per_second": 10.247, | |
| "step": 14034 | |
| } | |
| ], | |
| "max_steps": 23390, | |
| "num_train_epochs": 10, | |
| "total_flos": 2.955497070599885e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |