{"args": {"mode": "bp", "dataset": "hierarchical", "depth": 2, "width": 16, "act": "tanh", "residual": 1, "epochs": 25, "batch_size": 128, "train_examples": 0, "val_examples": 5000, "split_seed": 2027, "eval_split": "validation", "task_seed": 0, "task_train_examples": 50000, "task_test_examples": 10000, "task_levels": 6, "task_n_in": 128, "task_classes": 10, "teacher_depth": 8, "teacher_width": 64, "teacher_residual": 1, "eta": 0.03, "eta_A": 0.02, "eta_P": 0.002, "momentum": 0.9, "w_scale": 1.0, "a_scale": 1.0, "feedback_scale": 1.0, "pert_sigma": 0.01, "pert_every": 4, "pert_ndirs": 16, "pert_mode": "simultaneous", "use_residual": 1, "raw_scale_control": "none", "learn_A": 1, "learn_P": 1, "p_neutral": 1, "p_warmup_steps": 200, "p_warmup_eta": 0.05, "nuis_rho": 0.0, "traffic_mode": "soma", "predictor_mode": "diagonal", "normalize_delta": 0, "settle_steps": 0, "kappa": 0.0, "feedback": "error", "seed": 0, "device": "cuda", "log_every": 200, "max_steps": 0, "probe_bs": 512, "outdir": "results", "tag": "depthtask_dev_v1_hierarchical_bp_w16_d2_t0_s0", "n_in": 64, "n_out": 10}, "split": {"dataset": "hierarchical", "task_seed": 0, "train_examples": 45000, "test_examples": 10000, "evaluation_split": "validation", "synthetic_generator": true, "split_seed": 2027, "validation_examples": 5000, "validation_index_sha256": "f5282ed201086357dbe24c23e70cc9eb86deaaf70d99c41553ffc78626ee8d6c", "validation_class_counts": {"0": 500, "1": 500, "2": 500, "3": 500, "4": 500, "5": 500, "6": 500, "7": 500, "8": 500, "9": 500}, "split_from_training_only": true}, "provenance": {"git_commit": "a0696a3d72cc5bf69876e1716b74f7e1ac525dd0", "git_dirty": false}, "steps": [{"step": 0, "epoch": 0, "train_loss": 2.5838396549224854}, {"step": 200, "epoch": 0, "train_loss": 1.780808448791504}, {"epoch_end": 0, "step": 352, "eval_split": "validation", "eval_acc": 0.3292, "eval_loss": 1.9055708740234376, "val_acc": 0.3292, "val_loss": 1.9055708740234376}, {"step": 400, "epoch": 1, "train_loss": 1.7676631212234497}, {"step": 600, "epoch": 1, "train_loss": 1.7367398738861084}, {"epoch_end": 1, "step": 704, "eval_split": "validation", "eval_acc": 0.3436, "eval_loss": 1.8824305419921874, "val_acc": 0.3436, "val_loss": 1.8824305419921874}, {"step": 800, "epoch": 2, "train_loss": 1.6690762042999268}, {"step": 1000, "epoch": 2, "train_loss": 1.7731873989105225}, {"epoch_end": 2, "step": 1056, "eval_split": "validation", "eval_acc": 0.355, "eval_loss": 1.84503623046875, "val_acc": 0.355, "val_loss": 1.84503623046875}, {"step": 1200, "epoch": 3, "train_loss": 1.6389950513839722}, {"step": 1400, "epoch": 3, "train_loss": 1.6875262260437012}, {"epoch_end": 3, "step": 1408, "eval_split": "validation", "eval_acc": 0.3622, "eval_loss": 1.8366450439453126, "val_acc": 0.3622, "val_loss": 1.8366450439453126}, {"step": 1600, "epoch": 4, "train_loss": 1.6472316980361938}, {"epoch_end": 4, "step": 1760, "eval_split": "validation", "eval_acc": 0.371, "eval_loss": 1.81101865234375, "val_acc": 0.371, "val_loss": 1.81101865234375}, {"step": 1800, "epoch": 5, "train_loss": 1.855771541595459}, {"step": 2000, "epoch": 5, "train_loss": 1.6434890031814575}, {"epoch_end": 5, "step": 2112, "eval_split": "validation", "eval_acc": 0.3682, "eval_loss": 1.8016017333984375, "val_acc": 0.3682, "val_loss": 1.8016017333984375}, {"step": 2200, "epoch": 6, "train_loss": 1.614069938659668}, {"step": 2400, "epoch": 6, "train_loss": 1.6868332624435425}, {"epoch_end": 6, "step": 2464, "eval_split": "validation", "eval_acc": 0.373, "eval_loss": 1.7997779541015626, "val_acc": 0.373, "val_loss": 1.7997779541015626}, {"step": 2600, "epoch": 7, "train_loss": 1.6740370988845825}, {"step": 2800, "epoch": 7, "train_loss": 1.7635605335235596}, {"epoch_end": 7, "step": 2816, "eval_split": "validation", "eval_acc": 0.3716, "eval_loss": 1.7907843994140624, "val_acc": 0.3716, "val_loss": 1.7907843994140624}, {"step": 3000, "epoch": 8, "train_loss": 1.6583243608474731}, {"epoch_end": 8, "step": 3168, "eval_split": "validation", "eval_acc": 0.3684, "eval_loss": 1.7946490478515624, "val_acc": 0.3684, "val_loss": 1.7946490478515624}, {"step": 3200, "epoch": 9, "train_loss": 1.6853324174880981}, {"step": 3400, "epoch": 9, "train_loss": 1.5609760284423828}, {"epoch_end": 9, "step": 3520, "eval_split": "validation", "eval_acc": 0.3708, "eval_loss": 1.7879376953125, "val_acc": 0.3708, "val_loss": 1.7879376953125}, {"step": 3600, "epoch": 10, "train_loss": 1.6333789825439453}, {"step": 3800, "epoch": 10, "train_loss": 1.544682264328003}, {"epoch_end": 10, "step": 3872, "eval_split": "validation", "eval_acc": 0.3702, "eval_loss": 1.7904752197265625, "val_acc": 0.3702, "val_loss": 1.7904752197265625}, {"step": 4000, "epoch": 11, "train_loss": 1.6339285373687744}, {"step": 4200, "epoch": 11, "train_loss": 1.6893727779388428}, {"epoch_end": 11, "step": 4224, "eval_split": "validation", "eval_acc": 0.3728, "eval_loss": 1.7810276123046875, "val_acc": 0.3728, "val_loss": 1.7810276123046875}, {"step": 4400, "epoch": 12, "train_loss": 1.833978295326233}, {"epoch_end": 12, "step": 4576, "eval_split": "validation", "eval_acc": 0.3664, "eval_loss": 1.8002841552734374, "val_acc": 0.3664, "val_loss": 1.8002841552734374}, {"step": 4600, "epoch": 13, "train_loss": 1.571160078048706}, {"step": 4800, "epoch": 13, "train_loss": 1.7381309270858765}, {"epoch_end": 13, "step": 4928, "eval_split": "validation", "eval_acc": 0.3666, "eval_loss": 1.792106591796875, "val_acc": 0.3666, "val_loss": 1.792106591796875}, {"step": 5000, "epoch": 14, "train_loss": 1.4158738851547241}, {"step": 5200, "epoch": 14, "train_loss": 1.663922905921936}, {"epoch_end": 14, "step": 5280, "eval_split": "validation", "eval_acc": 0.3628, "eval_loss": 1.8038263427734376, "val_acc": 0.3628, "val_loss": 1.8038263427734376}, {"step": 5400, "epoch": 15, "train_loss": 1.7760043144226074}, {"step": 5600, "epoch": 15, "train_loss": 1.5528504848480225}, {"epoch_end": 15, "step": 5632, "eval_split": "validation", "eval_acc": 0.3656, "eval_loss": 1.792407958984375, "val_acc": 0.3656, "val_loss": 1.792407958984375}, {"step": 5800, "epoch": 16, "train_loss": 1.5818465948104858}, {"epoch_end": 16, "step": 5984, "eval_split": "validation", "eval_acc": 0.371, "eval_loss": 1.7979139404296876, "val_acc": 0.371, "val_loss": 1.7979139404296876}, {"step": 6000, "epoch": 17, "train_loss": 1.5356544256210327}, {"step": 6200, "epoch": 17, "train_loss": 1.4888254404067993}, {"epoch_end": 17, "step": 6336, "eval_split": "validation", "eval_acc": 0.3722, "eval_loss": 1.79617490234375, "val_acc": 0.3722, "val_loss": 1.79617490234375}, {"step": 6400, "epoch": 18, "train_loss": 1.423915982246399}, {"step": 6600, "epoch": 18, "train_loss": 1.594796061515808}, {"epoch_end": 18, "step": 6688, "eval_split": "validation", "eval_acc": 0.373, "eval_loss": 1.79276591796875, "val_acc": 0.373, "val_loss": 1.79276591796875}, {"step": 6800, "epoch": 19, "train_loss": 1.6619716882705688}, {"step": 7000, "epoch": 19, "train_loss": 1.4323354959487915}, {"epoch_end": 19, "step": 7040, "eval_split": "validation", "eval_acc": 0.375, "eval_loss": 1.776577978515625, "val_acc": 0.375, "val_loss": 1.776577978515625}, {"step": 7200, "epoch": 20, "train_loss": 1.6437153816223145}, {"epoch_end": 20, "step": 7392, "eval_split": "validation", "eval_acc": 0.3704, "eval_loss": 1.789465576171875, "val_acc": 0.3704, "val_loss": 1.789465576171875}, {"step": 7400, "epoch": 21, "train_loss": 1.5673000812530518}, {"step": 7600, "epoch": 21, "train_loss": 1.5698800086975098}, {"epoch_end": 21, "step": 7744, "eval_split": "validation", "eval_acc": 0.373, "eval_loss": 1.7942180908203125, "val_acc": 0.373, "val_loss": 1.7942180908203125}, {"step": 7800, "epoch": 22, "train_loss": 1.5791683197021484}, {"step": 8000, "epoch": 22, "train_loss": 1.6318250894546509}, {"epoch_end": 22, "step": 8096, "eval_split": "validation", "eval_acc": 0.3756, "eval_loss": 1.790483203125, "val_acc": 0.3756, "val_loss": 1.790483203125}, {"step": 8200, "epoch": 23, "train_loss": 1.4760571718215942}, {"step": 8400, "epoch": 23, "train_loss": 1.6910078525543213}, {"epoch_end": 23, "step": 8448, "eval_split": "validation", "eval_acc": 0.3762, "eval_loss": 1.7710974609375, "val_acc": 0.3762, "val_loss": 1.7710974609375}, {"step": 8600, "epoch": 24, "train_loss": 1.5565590858459473}, {"epoch_end": 24, "step": 8800, "eval_split": "validation", "eval_acc": 0.379, "eval_loss": 1.7692759521484376, "val_acc": 0.379, "val_loss": 1.7692759521484376}], "final": {"eval_split": "validation", "eval_acc": 0.379, "eval_loss": 1.7692759521484376, "wall_s": 17.724177837371826, "val_acc": 0.379, "val_loss": 1.7692759521484376}}