{"args": {"mode": "bp", "dataset": "teacher", "depth": 12, "width": 64, "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_w64_d12_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": 2.4617087841033936}, {"step": 200, "epoch": 0, "train_loss": 0.902694821357727}, {"epoch_end": 0, "step": 352, "eval_split": "validation", "eval_acc": 0.6118, "eval_loss": 1.03099111328125, "val_acc": 0.6118, "val_loss": 1.03099111328125}, {"step": 400, "epoch": 1, "train_loss": 0.7640933990478516}, {"step": 600, "epoch": 1, "train_loss": 0.7631095051765442}, {"epoch_end": 1, "step": 704, "eval_split": "validation", "eval_acc": 0.6158, "eval_loss": 1.0279851928710937, "val_acc": 0.6158, "val_loss": 1.0279851928710937}, {"step": 800, "epoch": 2, "train_loss": 0.6875019073486328}, {"step": 1000, "epoch": 2, "train_loss": 0.7486509680747986}, {"epoch_end": 2, "step": 1056, "eval_split": "validation", "eval_acc": 0.6082, "eval_loss": 1.0466720336914062, "val_acc": 0.6082, "val_loss": 1.0466720336914062}, {"step": 1200, "epoch": 3, "train_loss": 0.8495298624038696}, {"step": 1400, "epoch": 3, "train_loss": 0.9343891143798828}, {"epoch_end": 3, "step": 1408, "eval_split": "validation", "eval_acc": 0.6126, "eval_loss": 1.0547455078125, "val_acc": 0.6126, "val_loss": 1.0547455078125}, {"step": 1600, "epoch": 4, "train_loss": 0.7209277153015137}, {"epoch_end": 4, "step": 1760, "eval_split": "validation", "eval_acc": 0.6168, "eval_loss": 1.0449613891601564, "val_acc": 0.6168, "val_loss": 1.0449613891601564}, {"step": 1800, "epoch": 5, "train_loss": 0.7454065084457397}, {"step": 2000, "epoch": 5, "train_loss": 0.6858469843864441}, {"epoch_end": 5, "step": 2112, "eval_split": "validation", "eval_acc": 0.608, "eval_loss": 1.076257861328125, "val_acc": 0.608, "val_loss": 1.076257861328125}, {"step": 2200, "epoch": 6, "train_loss": 0.8187124729156494}, {"step": 2400, "epoch": 6, "train_loss": 0.6559977531433105}, {"epoch_end": 6, "step": 2464, "eval_split": "validation", "eval_acc": 0.624, "eval_loss": 1.0495591796875, "val_acc": 0.624, "val_loss": 1.0495591796875}, {"step": 2600, "epoch": 7, "train_loss": 0.7761538624763489}, {"step": 2800, "epoch": 7, "train_loss": 0.7890281081199646}, {"epoch_end": 7, "step": 2816, "eval_split": "validation", "eval_acc": 0.5998, "eval_loss": 1.1207243408203125, "val_acc": 0.5998, "val_loss": 1.1207243408203125}, {"step": 3000, "epoch": 8, "train_loss": 0.6752810478210449}, {"epoch_end": 8, "step": 3168, "eval_split": "validation", "eval_acc": 0.6138, "eval_loss": 1.09807646484375, "val_acc": 0.6138, "val_loss": 1.09807646484375}, {"step": 3200, "epoch": 9, "train_loss": 0.6439822316169739}, {"step": 3400, "epoch": 9, "train_loss": 0.6429063081741333}, {"epoch_end": 9, "step": 3520, "eval_split": "validation", "eval_acc": 0.6086, "eval_loss": 1.1253881591796875, "val_acc": 0.6086, "val_loss": 1.1253881591796875}, {"step": 3600, "epoch": 10, "train_loss": 0.5866493582725525}, {"step": 3800, "epoch": 10, "train_loss": 0.6085690259933472}, {"epoch_end": 10, "step": 3872, "eval_split": "validation", "eval_acc": 0.6132, "eval_loss": 1.1205505859375, "val_acc": 0.6132, "val_loss": 1.1205505859375}, {"step": 4000, "epoch": 11, "train_loss": 0.6126564145088196}, {"step": 4200, "epoch": 11, "train_loss": 0.8024030923843384}, {"epoch_end": 11, "step": 4224, "eval_split": "validation", "eval_acc": 0.6018, "eval_loss": 1.17735126953125, "val_acc": 0.6018, "val_loss": 1.17735126953125}, {"step": 4400, "epoch": 12, "train_loss": 0.633552074432373}, {"epoch_end": 12, "step": 4576, "eval_split": "validation", "eval_acc": 0.6066, "eval_loss": 1.1778637451171876, "val_acc": 0.6066, "val_loss": 1.1778637451171876}, {"step": 4600, "epoch": 13, "train_loss": 0.6782782077789307}, {"step": 4800, "epoch": 13, "train_loss": 0.61345374584198}, {"epoch_end": 13, "step": 4928, "eval_split": "validation", "eval_acc": 0.6028, "eval_loss": 1.2229530029296876, "val_acc": 0.6028, "val_loss": 1.2229530029296876}, {"step": 5000, "epoch": 14, "train_loss": 0.5067461133003235}, {"step": 5200, "epoch": 14, "train_loss": 0.5135768055915833}, {"epoch_end": 14, "step": 5280, "eval_split": "validation", "eval_acc": 0.5984, "eval_loss": 1.246344970703125, "val_acc": 0.5984, "val_loss": 1.246344970703125}, {"step": 5400, "epoch": 15, "train_loss": 0.5423197150230408}, {"step": 5600, "epoch": 15, "train_loss": 0.5771681666374207}, {"epoch_end": 15, "step": 5632, "eval_split": "validation", "eval_acc": 0.59, "eval_loss": 1.292396142578125, "val_acc": 0.59, "val_loss": 1.292396142578125}, {"step": 5800, "epoch": 16, "train_loss": 0.5234283804893494}, {"epoch_end": 16, "step": 5984, "eval_split": "validation", "eval_acc": 0.5934, "eval_loss": 1.297881494140625, "val_acc": 0.5934, "val_loss": 1.297881494140625}, {"step": 6000, "epoch": 17, "train_loss": 0.6049938797950745}, {"step": 6200, "epoch": 17, "train_loss": 0.5433491468429565}, {"epoch_end": 17, "step": 6336, "eval_split": "validation", "eval_acc": 0.582, "eval_loss": 1.3660036865234375, "val_acc": 0.582, "val_loss": 1.3660036865234375}, {"step": 6400, "epoch": 18, "train_loss": 0.38602903485298157}, {"step": 6600, "epoch": 18, "train_loss": 0.4970954656600952}, {"epoch_end": 18, "step": 6688, "eval_split": "validation", "eval_acc": 0.5854, "eval_loss": 1.3779427978515626, "val_acc": 0.5854, "val_loss": 1.3779427978515626}, {"step": 6800, "epoch": 19, "train_loss": 0.5171346068382263}, {"step": 7000, "epoch": 19, "train_loss": 0.30657118558883667}, {"epoch_end": 19, "step": 7040, "eval_split": "validation", "eval_acc": 0.5748, "eval_loss": 1.439207568359375, "val_acc": 0.5748, "val_loss": 1.439207568359375}, {"step": 7200, "epoch": 20, "train_loss": 0.4765183925628662}, {"epoch_end": 20, "step": 7392, "eval_split": "validation", "eval_acc": 0.583, "eval_loss": 1.4810153076171875, "val_acc": 0.583, "val_loss": 1.4810153076171875}, {"step": 7400, "epoch": 21, "train_loss": 0.5005311965942383}, {"step": 7600, "epoch": 21, "train_loss": 0.3740580976009369}, {"epoch_end": 21, "step": 7744, "eval_split": "validation", "eval_acc": 0.5828, "eval_loss": 1.5324156494140626, "val_acc": 0.5828, "val_loss": 1.5324156494140626}, {"step": 7800, "epoch": 22, "train_loss": 0.38728761672973633}, {"step": 8000, "epoch": 22, "train_loss": 0.3829094469547272}, {"epoch_end": 22, "step": 8096, "eval_split": "validation", "eval_acc": 0.5746, "eval_loss": 1.6113326904296874, "val_acc": 0.5746, "val_loss": 1.6113326904296874}, {"step": 8200, "epoch": 23, "train_loss": 0.2806970477104187}, {"step": 8400, "epoch": 23, "train_loss": 0.4131513237953186}, {"epoch_end": 23, "step": 8448, "eval_split": "validation", "eval_acc": 0.5686, "eval_loss": 1.6380853759765626, "val_acc": 0.5686, "val_loss": 1.6380853759765626}, {"step": 8600, "epoch": 24, "train_loss": 0.2939942479133606}, {"epoch_end": 24, "step": 8800, "eval_split": "validation", "eval_acc": 0.5702, "eval_loss": 1.6876431396484375, "val_acc": 0.5702, "val_loss": 1.6876431396484375}], "final": {"eval_split": "validation", "eval_acc": 0.5702, "eval_loss": 1.6876431396484375, "wall_s": 57.20587205886841, "val_acc": 0.5702, "val_loss": 1.6876431396484375}}