{"args": {"mode": "bp", "dataset": "teacher", "depth": 4, "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_d4_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": 2.4966790676116943}, {"step": 200, "epoch": 0, "train_loss": 0.9647588133811951}, {"epoch_end": 0, "step": 352, "eval_split": "validation", "eval_acc": 0.5676, "eval_loss": 1.14811044921875, "val_acc": 0.5676, "val_loss": 1.14811044921875}, {"step": 400, "epoch": 1, "train_loss": 0.9004788398742676}, {"step": 600, "epoch": 1, "train_loss": 0.858989953994751}, {"epoch_end": 1, "step": 704, "eval_split": "validation", "eval_acc": 0.5852, "eval_loss": 1.1066900390625, "val_acc": 0.5852, "val_loss": 1.1066900390625}, {"step": 800, "epoch": 2, "train_loss": 0.9216505289077759}, {"step": 1000, "epoch": 2, "train_loss": 0.7750069499015808}, {"epoch_end": 2, "step": 1056, "eval_split": "validation", "eval_acc": 0.5792, "eval_loss": 1.1309549072265626, "val_acc": 0.5792, "val_loss": 1.1309549072265626}, {"step": 1200, "epoch": 3, "train_loss": 0.97446608543396}, {"step": 1400, "epoch": 3, "train_loss": 1.0129777193069458}, {"epoch_end": 3, "step": 1408, "eval_split": "validation", "eval_acc": 0.5812, "eval_loss": 1.1271224853515625, "val_acc": 0.5812, "val_loss": 1.1271224853515625}, {"step": 1600, "epoch": 4, "train_loss": 0.8938019871711731}, {"epoch_end": 4, "step": 1760, "eval_split": "validation", "eval_acc": 0.5804, "eval_loss": 1.1166867919921875, "val_acc": 0.5804, "val_loss": 1.1166867919921875}, {"step": 1800, "epoch": 5, "train_loss": 1.0304168462753296}, {"step": 2000, "epoch": 5, "train_loss": 0.884192705154419}, {"epoch_end": 5, "step": 2112, "eval_split": "validation", "eval_acc": 0.5798, "eval_loss": 1.1317115966796876, "val_acc": 0.5798, "val_loss": 1.1317115966796876}, {"step": 2200, "epoch": 6, "train_loss": 1.0397288799285889}, {"step": 2400, "epoch": 6, "train_loss": 0.8639501929283142}, {"epoch_end": 6, "step": 2464, "eval_split": "validation", "eval_acc": 0.593, "eval_loss": 1.085141796875, "val_acc": 0.593, "val_loss": 1.085141796875}, {"step": 2600, "epoch": 7, "train_loss": 0.9758634567260742}, {"step": 2800, "epoch": 7, "train_loss": 1.0234332084655762}, {"epoch_end": 7, "step": 2816, "eval_split": "validation", "eval_acc": 0.5758, "eval_loss": 1.1223146240234374, "val_acc": 0.5758, "val_loss": 1.1223146240234374}, {"step": 3000, "epoch": 8, "train_loss": 0.9471226334571838}, {"epoch_end": 8, "step": 3168, "eval_split": "validation", "eval_acc": 0.5852, "eval_loss": 1.096402392578125, "val_acc": 0.5852, "val_loss": 1.096402392578125}, {"step": 3200, "epoch": 9, "train_loss": 0.858224630355835}, {"step": 3400, "epoch": 9, "train_loss": 0.8620807528495789}, {"epoch_end": 9, "step": 3520, "eval_split": "validation", "eval_acc": 0.58, "eval_loss": 1.116325390625, "val_acc": 0.58, "val_loss": 1.116325390625}, {"step": 3600, "epoch": 10, "train_loss": 0.9411643743515015}, {"step": 3800, "epoch": 10, "train_loss": 1.002820372581482}, {"epoch_end": 10, "step": 3872, "eval_split": "validation", "eval_acc": 0.5904, "eval_loss": 1.0973555908203125, "val_acc": 0.5904, "val_loss": 1.0973555908203125}, {"step": 4000, "epoch": 11, "train_loss": 0.8986897468566895}, {"step": 4200, "epoch": 11, "train_loss": 0.9725231528282166}, {"epoch_end": 11, "step": 4224, "eval_split": "validation", "eval_acc": 0.5848, "eval_loss": 1.1082283935546875, "val_acc": 0.5848, "val_loss": 1.1082283935546875}, {"step": 4400, "epoch": 12, "train_loss": 0.8426617980003357}, {"epoch_end": 12, "step": 4576, "eval_split": "validation", "eval_acc": 0.5868, "eval_loss": 1.1015863525390626, "val_acc": 0.5868, "val_loss": 1.1015863525390626}, {"step": 4600, "epoch": 13, "train_loss": 1.1166877746582031}, {"step": 4800, "epoch": 13, "train_loss": 0.9234418869018555}, {"epoch_end": 13, "step": 4928, "eval_split": "validation", "eval_acc": 0.5838, "eval_loss": 1.1126053955078126, "val_acc": 0.5838, "val_loss": 1.1126053955078126}, {"step": 5000, "epoch": 14, "train_loss": 1.0969547033309937}, {"step": 5200, "epoch": 14, "train_loss": 0.9123508930206299}, {"epoch_end": 14, "step": 5280, "eval_split": "validation", "eval_acc": 0.5834, "eval_loss": 1.0911812133789063, "val_acc": 0.5834, "val_loss": 1.0911812133789063}, {"step": 5400, "epoch": 15, "train_loss": 1.0123803615570068}, {"step": 5600, "epoch": 15, "train_loss": 0.852265477180481}, {"epoch_end": 15, "step": 5632, "eval_split": "validation", "eval_acc": 0.5892, "eval_loss": 1.102031005859375, "val_acc": 0.5892, "val_loss": 1.102031005859375}, {"step": 5800, "epoch": 16, "train_loss": 0.8970330953598022}, {"epoch_end": 16, "step": 5984, "eval_split": "validation", "eval_acc": 0.6008, "eval_loss": 1.0820575073242187, "val_acc": 0.6008, "val_loss": 1.0820575073242187}, {"step": 6000, "epoch": 17, "train_loss": 0.8559246063232422}, {"step": 6200, "epoch": 17, "train_loss": 0.8634406924247742}, {"epoch_end": 17, "step": 6336, "eval_split": "validation", "eval_acc": 0.58, "eval_loss": 1.111931103515625, "val_acc": 0.58, "val_loss": 1.111931103515625}, {"step": 6400, "epoch": 18, "train_loss": 1.0019081830978394}, {"step": 6600, "epoch": 18, "train_loss": 0.9256800413131714}, {"epoch_end": 18, "step": 6688, "eval_split": "validation", "eval_acc": 0.5932, "eval_loss": 1.087797021484375, "val_acc": 0.5932, "val_loss": 1.087797021484375}, {"step": 6800, "epoch": 19, "train_loss": 1.118901252746582}, {"step": 7000, "epoch": 19, "train_loss": 0.7939027547836304}, {"epoch_end": 19, "step": 7040, "eval_split": "validation", "eval_acc": 0.5904, "eval_loss": 1.089816748046875, "val_acc": 0.5904, "val_loss": 1.089816748046875}, {"step": 7200, "epoch": 20, "train_loss": 0.9085933566093445}, {"epoch_end": 20, "step": 7392, "eval_split": "validation", "eval_acc": 0.5858, "eval_loss": 1.090731396484375, "val_acc": 0.5858, "val_loss": 1.090731396484375}, {"step": 7400, "epoch": 21, "train_loss": 1.145005702972412}, {"step": 7600, "epoch": 21, "train_loss": 0.973916232585907}, {"epoch_end": 21, "step": 7744, "eval_split": "validation", "eval_acc": 0.5916, "eval_loss": 1.0808295043945313, "val_acc": 0.5916, "val_loss": 1.0808295043945313}, {"step": 7800, "epoch": 22, "train_loss": 0.9709288477897644}, {"step": 8000, "epoch": 22, "train_loss": 0.8602243661880493}, {"epoch_end": 22, "step": 8096, "eval_split": "validation", "eval_acc": 0.592, "eval_loss": 1.0842331298828125, "val_acc": 0.592, "val_loss": 1.0842331298828125}, {"step": 8200, "epoch": 23, "train_loss": 0.9825401902198792}, {"step": 8400, "epoch": 23, "train_loss": 0.9386957287788391}, {"epoch_end": 23, "step": 8448, "eval_split": "validation", "eval_acc": 0.585, "eval_loss": 1.0979935546875, "val_acc": 0.585, "val_loss": 1.0979935546875}, {"step": 8600, "epoch": 24, "train_loss": 0.8032169342041016}, {"epoch_end": 24, "step": 8800, "eval_split": "validation", "eval_acc": 0.5936, "eval_loss": 1.0907277465820313, "val_acc": 0.5936, "val_loss": 1.0907277465820313}], "final": {"eval_split": "validation", "eval_acc": 0.5936, "eval_loss": 1.0907277465820313, "wall_s": 26.082186222076416, "val_acc": 0.5936, "val_loss": 1.0907277465820313}}