| { | |
| "best_metric": 1.5115073919296265, | |
| "best_model_checkpoint": "tam_test_out_drug_data_large_test/checkpoint-40325", | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 40325, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 606205.1875, | |
| "learning_rate": 5.925604463732176e-05, | |
| "loss": 2.1094, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 535000.3125, | |
| "learning_rate": 5.851208927464352e-05, | |
| "loss": 2.0692, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 756629.3125, | |
| "learning_rate": 5.7768133911965284e-05, | |
| "loss": 2.0081, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 652686.0, | |
| "learning_rate": 5.7024178549287045e-05, | |
| "loss": 1.9094, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 713175.3125, | |
| "learning_rate": 5.6280223186608805e-05, | |
| "loss": 1.8251, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 514924.375, | |
| "learning_rate": 5.5536267823930566e-05, | |
| "loss": 1.8088, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 598626.25, | |
| "learning_rate": 5.479231246125233e-05, | |
| "loss": 1.7972, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 455352.53125, | |
| "learning_rate": 5.404835709857409e-05, | |
| "loss": 1.7607, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 498720.8125, | |
| "learning_rate": 5.330440173589585e-05, | |
| "loss": 1.7936, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 416274.09375, | |
| "learning_rate": 5.256044637321761e-05, | |
| "loss": 1.7587, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 426318.875, | |
| "learning_rate": 5.181649101053937e-05, | |
| "loss": 1.778, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 245576.5, | |
| "learning_rate": 5.107253564786113e-05, | |
| "loss": 1.7174, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 343734.96875, | |
| "learning_rate": 5.032858028518289e-05, | |
| "loss": 1.7171, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 543570.625, | |
| "learning_rate": 4.958462492250465e-05, | |
| "loss": 1.7212, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 364161.46875, | |
| "learning_rate": 4.884066955982641e-05, | |
| "loss": 1.6634, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 363151.9375, | |
| "learning_rate": 4.8096714197148174e-05, | |
| "loss": 1.6824, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 584323.75, | |
| "learning_rate": 4.7352758834469934e-05, | |
| "loss": 1.683, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 517269.78125, | |
| "learning_rate": 4.6608803471791695e-05, | |
| "loss": 1.6421, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 589548.0, | |
| "learning_rate": 4.5864848109113456e-05, | |
| "loss": 1.629, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 516179.65625, | |
| "learning_rate": 4.5120892746435217e-05, | |
| "loss": 1.6111, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 719886.4375, | |
| "learning_rate": 4.437693738375697e-05, | |
| "loss": 1.6725, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 374763.96875, | |
| "learning_rate": 4.363298202107873e-05, | |
| "loss": 1.6587, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 539950.75, | |
| "learning_rate": 4.288902665840049e-05, | |
| "loss": 1.5891, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 841417.125, | |
| "learning_rate": 4.214507129572225e-05, | |
| "loss": 1.6487, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 413180.6875, | |
| "learning_rate": 4.1401115933044014e-05, | |
| "loss": 1.6599, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 724763.75, | |
| "learning_rate": 4.0657160570365774e-05, | |
| "loss": 1.6468, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 630071.0, | |
| "learning_rate": 3.9913205207687535e-05, | |
| "loss": 1.6445, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 413867.3125, | |
| "learning_rate": 3.9169249845009296e-05, | |
| "loss": 1.6255, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 431392.0625, | |
| "learning_rate": 3.8425294482331057e-05, | |
| "loss": 1.607, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 465831.90625, | |
| "learning_rate": 3.768133911965282e-05, | |
| "loss": 1.6187, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 578294.4375, | |
| "learning_rate": 3.693738375697458e-05, | |
| "loss": 1.6044, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 463151.03125, | |
| "learning_rate": 3.619342839429634e-05, | |
| "loss": 1.6294, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 394089.59375, | |
| "learning_rate": 3.54494730316181e-05, | |
| "loss": 1.6219, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 808582.75, | |
| "learning_rate": 3.470551766893986e-05, | |
| "loss": 1.5825, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 166829.9375, | |
| "learning_rate": 3.396156230626162e-05, | |
| "loss": 1.5696, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 755760.125, | |
| "learning_rate": 3.321760694358338e-05, | |
| "loss": 1.6148, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 894633.5625, | |
| "learning_rate": 3.247365158090515e-05, | |
| "loss": 1.6118, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 445554.59375, | |
| "learning_rate": 3.172969621822691e-05, | |
| "loss": 1.6123, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 458144.21875, | |
| "learning_rate": 3.098574085554867e-05, | |
| "loss": 1.628, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 382592.625, | |
| "learning_rate": 3.0241785492870428e-05, | |
| "loss": 1.6075, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 343228.125, | |
| "learning_rate": 2.949783013019219e-05, | |
| "loss": 1.6086, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 432318.65625, | |
| "learning_rate": 2.875387476751395e-05, | |
| "loss": 1.5723, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 264779.78125, | |
| "learning_rate": 2.800991940483571e-05, | |
| "loss": 1.6276, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 583042.0625, | |
| "learning_rate": 2.726596404215747e-05, | |
| "loss": 1.583, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 402780.90625, | |
| "learning_rate": 2.6522008679479232e-05, | |
| "loss": 1.5676, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 528016.6875, | |
| "learning_rate": 2.5778053316800993e-05, | |
| "loss": 1.5991, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 442681.25, | |
| "learning_rate": 2.5034097954122753e-05, | |
| "loss": 1.5809, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 820616.5625, | |
| "learning_rate": 2.4290142591444514e-05, | |
| "loss": 1.5583, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 527036.1875, | |
| "learning_rate": 2.3546187228766275e-05, | |
| "loss": 1.5536, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 425855.125, | |
| "learning_rate": 2.2802231866088036e-05, | |
| "loss": 1.604, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 477485.53125, | |
| "learning_rate": 2.2058276503409793e-05, | |
| "loss": 1.5564, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 684294.125, | |
| "learning_rate": 2.1314321140731554e-05, | |
| "loss": 1.5505, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 687388.875, | |
| "learning_rate": 2.0570365778053315e-05, | |
| "loss": 1.5636, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 827780.125, | |
| "learning_rate": 1.9826410415375075e-05, | |
| "loss": 1.5526, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 488844.8125, | |
| "learning_rate": 1.908245505269684e-05, | |
| "loss": 1.5192, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "grad_norm": 463870.6875, | |
| "learning_rate": 1.83384996900186e-05, | |
| "loss": 1.5441, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 906024.625, | |
| "learning_rate": 1.759454432734036e-05, | |
| "loss": 1.5328, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 539641.375, | |
| "learning_rate": 1.6850588964662122e-05, | |
| "loss": 1.5892, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 773464.4375, | |
| "learning_rate": 1.6106633601983882e-05, | |
| "loss": 1.5384, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 939030.25, | |
| "learning_rate": 1.5362678239305643e-05, | |
| "loss": 1.588, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 715276.5, | |
| "learning_rate": 1.4618722876627402e-05, | |
| "loss": 1.5575, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 478972.75, | |
| "learning_rate": 1.3874767513949163e-05, | |
| "loss": 1.5229, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 666075.0, | |
| "learning_rate": 1.3130812151270924e-05, | |
| "loss": 1.5491, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 706126.125, | |
| "learning_rate": 1.2386856788592685e-05, | |
| "loss": 1.5522, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 380624.65625, | |
| "learning_rate": 1.1642901425914447e-05, | |
| "loss": 1.5408, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 436867.90625, | |
| "learning_rate": 1.0898946063236206e-05, | |
| "loss": 1.5076, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "grad_norm": 495405.875, | |
| "learning_rate": 1.0154990700557967e-05, | |
| "loss": 1.5355, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 298892.375, | |
| "learning_rate": 9.411035337879728e-06, | |
| "loss": 1.4743, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 455935.5, | |
| "learning_rate": 8.667079975201488e-06, | |
| "loss": 1.5539, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 513792.59375, | |
| "learning_rate": 7.923124612523249e-06, | |
| "loss": 1.5538, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 156134.75, | |
| "learning_rate": 7.17916924984501e-06, | |
| "loss": 1.5211, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 522846.1875, | |
| "learning_rate": 6.4352138871667705e-06, | |
| "loss": 1.5259, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 373932.15625, | |
| "learning_rate": 5.691258524488531e-06, | |
| "loss": 1.4907, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 747137.0625, | |
| "learning_rate": 4.947303161810291e-06, | |
| "loss": 1.5323, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 354971.9375, | |
| "learning_rate": 4.203347799132052e-06, | |
| "loss": 1.5479, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 623187.875, | |
| "learning_rate": 3.459392436453813e-06, | |
| "loss": 1.5263, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 348824.09375, | |
| "learning_rate": 2.7154370737755734e-06, | |
| "loss": 1.4994, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 538969.25, | |
| "learning_rate": 1.971481711097334e-06, | |
| "loss": 1.5384, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "grad_norm": 965487.0625, | |
| "learning_rate": 1.2275263484190947e-06, | |
| "loss": 1.5025, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 504319.8125, | |
| "learning_rate": 4.835709857408556e-07, | |
| "loss": 1.5344, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.4533618525062773, | |
| "eval_loss": 1.5115073919296265, | |
| "eval_runtime": 520.995, | |
| "eval_samples_per_second": 20.639, | |
| "eval_steps_per_second": 5.161, | |
| "step": 40325 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 40325, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "total_flos": 2.532444326342784e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |