{"args": {"mode": "bp", "dataset": "teacher", "depth": 8, "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": 8, "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_teacher_bp_w16_d8_t0_s0", "n_in": 128, "n_out": 10}, "split": {"dataset": "teacher", "task_seed": 0, "train_examples": 45000, "test_examples": 10000, "evaluation_split": "validation", "synthetic_generator": true, "split_seed": 2027, "validation_examples": 5000, "validation_index_sha256": "0ec30707a69784626a2d17128e5408bb8a8a8964e2ea753ef8e54b35ffb36355", "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": 2.4545867443084717}, {"step": 200, "epoch": 0, "train_loss": 0.9655951261520386}, {"epoch_end": 0, "step": 352, "eval_split": "validation", "eval_acc": 0.5674, "eval_loss": 1.1627013671875, "val_acc": 0.5674, "val_loss": 1.1627013671875}, {"step": 400, "epoch": 1, "train_loss": 0.921790599822998}, {"step": 600, "epoch": 1, "train_loss": 0.81261146068573}, {"epoch_end": 1, "step": 704, "eval_split": "validation", "eval_acc": 0.5892, "eval_loss": 1.12190517578125, "val_acc": 0.5892, "val_loss": 1.12190517578125}, {"step": 800, "epoch": 2, "train_loss": 0.8968669176101685}, {"step": 1000, "epoch": 2, "train_loss": 0.8287943601608276}, {"epoch_end": 2, "step": 1056, "eval_split": "validation", "eval_acc": 0.5836, "eval_loss": 1.1352845703125, "val_acc": 0.5836, "val_loss": 1.1352845703125}, {"step": 1200, "epoch": 3, "train_loss": 1.045316457748413}, {"step": 1400, "epoch": 3, "train_loss": 0.9924063682556152}, {"epoch_end": 3, "step": 1408, "eval_split": "validation", "eval_acc": 0.5778, "eval_loss": 1.1459072265625, "val_acc": 0.5778, "val_loss": 1.1459072265625}, {"step": 1600, "epoch": 4, "train_loss": 0.87221759557724}, {"epoch_end": 4, "step": 1760, "eval_split": "validation", "eval_acc": 0.5846, "eval_loss": 1.11298935546875, "val_acc": 0.5846, "val_loss": 1.11298935546875}, {"step": 1800, "epoch": 5, "train_loss": 0.983815610408783}, {"step": 2000, "epoch": 5, "train_loss": 0.8368768692016602}, {"epoch_end": 5, "step": 2112, "eval_split": "validation", "eval_acc": 0.5886, "eval_loss": 1.1216568359375, "val_acc": 0.5886, "val_loss": 1.1216568359375}, {"step": 2200, "epoch": 6, "train_loss": 1.0421881675720215}, {"step": 2400, "epoch": 6, "train_loss": 0.9238089323043823}, {"epoch_end": 6, "step": 2464, "eval_split": "validation", "eval_acc": 0.5902, "eval_loss": 1.093966552734375, "val_acc": 0.5902, "val_loss": 1.093966552734375}, {"step": 2600, "epoch": 7, "train_loss": 0.9274234175682068}, {"step": 2800, "epoch": 7, "train_loss": 0.9723529815673828}, {"epoch_end": 7, "step": 2816, "eval_split": "validation", "eval_acc": 0.59, "eval_loss": 1.1151231201171874, "val_acc": 0.59, "val_loss": 1.1151231201171874}, {"step": 3000, "epoch": 8, "train_loss": 0.9794726371765137}, {"epoch_end": 8, "step": 3168, "eval_split": "validation", "eval_acc": 0.5872, "eval_loss": 1.1117497314453124, "val_acc": 0.5872, "val_loss": 1.1117497314453124}, {"step": 3200, "epoch": 9, "train_loss": 0.7986926436424255}, {"step": 3400, "epoch": 9, "train_loss": 0.7764381766319275}, {"epoch_end": 9, "step": 3520, "eval_split": "validation", "eval_acc": 0.5902, "eval_loss": 1.1044780029296875, "val_acc": 0.5902, "val_loss": 1.1044780029296875}, {"step": 3600, "epoch": 10, "train_loss": 0.9259852766990662}, {"step": 3800, "epoch": 10, "train_loss": 0.939167320728302}, {"epoch_end": 10, "step": 3872, "eval_split": "validation", "eval_acc": 0.586, "eval_loss": 1.1029083984375, "val_acc": 0.586, "val_loss": 1.1029083984375}, {"step": 4000, "epoch": 11, "train_loss": 0.9145900011062622}, {"step": 4200, "epoch": 11, "train_loss": 1.0672698020935059}, {"epoch_end": 11, "step": 4224, "eval_split": "validation", "eval_acc": 0.581, "eval_loss": 1.123863525390625, "val_acc": 0.581, "val_loss": 1.123863525390625}, {"step": 4400, "epoch": 12, "train_loss": 0.8724120259284973}, {"epoch_end": 12, "step": 4576, "eval_split": "validation", "eval_acc": 0.5936, "eval_loss": 1.0961650390625, "val_acc": 0.5936, "val_loss": 1.0961650390625}, {"step": 4600, "epoch": 13, "train_loss": 0.9762986898422241}, {"step": 4800, "epoch": 13, "train_loss": 0.9933286905288696}, {"epoch_end": 13, "step": 4928, "eval_split": "validation", "eval_acc": 0.5746, "eval_loss": 1.128486279296875, "val_acc": 0.5746, "val_loss": 1.128486279296875}, {"step": 5000, "epoch": 14, "train_loss": 0.9770033955574036}, {"step": 5200, "epoch": 14, "train_loss": 0.7995492815971375}, {"epoch_end": 14, "step": 5280, "eval_split": "validation", "eval_acc": 0.588, "eval_loss": 1.1073376220703126, "val_acc": 0.588, "val_loss": 1.1073376220703126}, {"step": 5400, "epoch": 15, "train_loss": 0.9432471394538879}, {"step": 5600, "epoch": 15, "train_loss": 0.9147461652755737}, {"epoch_end": 15, "step": 5632, "eval_split": "validation", "eval_acc": 0.5764, "eval_loss": 1.13014189453125, "val_acc": 0.5764, "val_loss": 1.13014189453125}, {"step": 5800, "epoch": 16, "train_loss": 0.9600958228111267}, {"epoch_end": 16, "step": 5984, "eval_split": "validation", "eval_acc": 0.5874, "eval_loss": 1.09278125, "val_acc": 0.5874, "val_loss": 1.09278125}, {"step": 6000, "epoch": 17, "train_loss": 0.9386426210403442}, {"step": 6200, "epoch": 17, "train_loss": 0.8959528803825378}, {"epoch_end": 17, "step": 6336, "eval_split": "validation", "eval_acc": 0.5846, "eval_loss": 1.1262297607421874, "val_acc": 0.5846, "val_loss": 1.1262297607421874}, {"step": 6400, "epoch": 18, "train_loss": 1.0170623064041138}, {"step": 6600, "epoch": 18, "train_loss": 0.9944022297859192}, {"epoch_end": 18, "step": 6688, "eval_split": "validation", "eval_acc": 0.578, "eval_loss": 1.1140205810546875, "val_acc": 0.578, "val_loss": 1.1140205810546875}, {"step": 6800, "epoch": 19, "train_loss": 1.1006981134414673}, {"step": 7000, "epoch": 19, "train_loss": 0.8122886419296265}, {"epoch_end": 19, "step": 7040, "eval_split": "validation", "eval_acc": 0.5838, "eval_loss": 1.1186203857421875, "val_acc": 0.5838, "val_loss": 1.1186203857421875}, {"step": 7200, "epoch": 20, "train_loss": 0.7818880081176758}, {"epoch_end": 20, "step": 7392, "eval_split": "validation", "eval_acc": 0.5844, "eval_loss": 1.1130032958984375, "val_acc": 0.5844, "val_loss": 1.1130032958984375}, {"step": 7400, "epoch": 21, "train_loss": 1.0730682611465454}, {"step": 7600, "epoch": 21, "train_loss": 0.9104151725769043}, {"epoch_end": 21, "step": 7744, "eval_split": "validation", "eval_acc": 0.5846, "eval_loss": 1.11394580078125, "val_acc": 0.5846, "val_loss": 1.11394580078125}, {"step": 7800, "epoch": 22, "train_loss": 0.9913809299468994}, {"step": 8000, "epoch": 22, "train_loss": 0.7919666171073914}, {"epoch_end": 22, "step": 8096, "eval_split": "validation", "eval_acc": 0.5766, "eval_loss": 1.12707822265625, "val_acc": 0.5766, "val_loss": 1.12707822265625}, {"step": 8200, "epoch": 23, "train_loss": 0.9503212571144104}, {"step": 8400, "epoch": 23, "train_loss": 0.8721488118171692}, {"epoch_end": 23, "step": 8448, "eval_split": "validation", "eval_acc": 0.5784, "eval_loss": 1.129735498046875, "val_acc": 0.5784, "val_loss": 1.129735498046875}, {"step": 8600, "epoch": 24, "train_loss": 0.7976495027542114}, {"epoch_end": 24, "step": 8800, "eval_split": "validation", "eval_acc": 0.5818, "eval_loss": 1.1297189208984375, "val_acc": 0.5818, "val_loss": 1.1297189208984375}], "final": {"eval_split": "validation", "eval_acc": 0.5818, "eval_loss": 1.1297189208984375, "wall_s": 41.0548882484436, "val_acc": 0.5818, "val_loss": 1.1297189208984375}}