{"args": {"mode": "sdil", "dataset": "cifar10", "depth": 10, "width": 64, "act": "tanh", "residual": 1, "epochs": 5, "batch_size": 128, "train_examples": 0, "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": 0.02, "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, "predictor_mode": "diagonal", "normalize_delta": 0, "settle_steps": 0, "kappa": 0.0, "feedback": "error", "seed": 0, "device": "cuda", "log_every": 500, "max_steps": 0, "probe_bs": 256, "outdir": "results", "tag": "scale_v3_cifar10_sdil_w64_d10_s0", "n_in": 3072, "n_out": 10}, "provenance": {"git_commit": "d4aef1a571c21e6e4e4125c90aa58da7f14e62e3", "git_dirty": false}, "steps": [{"step": 0, "epoch": 0, "train_loss": 2.783294200897217, "cos_r_negg": [-0.024411339312791824, -0.07429128140211105, 0.03880922496318817, -0.10908837616443634, -0.029607944190502167, 0.040592096745967865, 0.014943718910217285, -0.03367207199335098, 0.024695973843336105, -0.01185297966003418], "cos_innovation_negg": [-0.024411339312791824, -0.07429128140211105, 0.03880922496318817, -0.10908837616443634, -0.029607944190502167, 0.040592096745967865, 0.014943718910217285, -0.03367207199335098, 0.024695973843336105, -0.01185297966003418], "cos_apical_negg": [-0.024411339312791824, -0.07429128140211105, 0.03880922496318817, -0.10908837616443634, -0.029607944190502167, 0.040592096745967865, 0.014943718910217285, -0.03367207199335098, 0.024695973843336105, -0.01185297966003418], "cos_Ac_negg": [-0.024411339312791824, -0.07429128140211105, 0.03880922496318817, -0.10908837616443634, -0.029607944190502167, 0.040592096745967865, 0.014943718910217285, -0.03367207199335098, 0.024695973843336105, -0.01185297966003418], "r_norm": [2.4589152336120605, 2.4222028255462646, 2.5381217002868652, 2.4834070205688477, 2.627248764038086, 2.4855048656463623, 2.704280376434326, 2.449425220489502, 2.473682403564453, 2.333059787750244], "ldr": {"dL_sdil": -0.04551553726196289, "dL_bp": -0.5264525413513184, "ratio": 0.08645705678451449}}, {"epoch_end": 0, "step": 391, "test_acc": 0.3648, "test_loss": 1.8065431030273438}, {"step": 500, "epoch": 1, "train_loss": 1.6820344924926758, "cos_r_negg": [0.6186816692352295, 0.13329757750034332, 0.05688320845365524, 0.0676628053188324, 0.17456591129302979, 0.1374761164188385, 0.032134220004081726, 0.007959588430821896, 0.08307185769081116, 0.09195937216281891], "cos_innovation_negg": [0.6186816692352295, 0.13329757750034332, 0.05688320845365524, 0.0676628053188324, 0.17456591129302979, 0.1374761164188385, 0.032134220004081726, 0.007959588430821896, 0.08307185769081116, 0.09195937216281891], "cos_apical_negg": [0.6186816692352295, 0.13329757750034332, 0.05688320845365524, 0.0676628053188324, 0.17456591129302979, 0.1374761164188385, 0.032134220004081726, 0.007959588430821896, 0.08307185769081116, 0.09195937216281891], "cos_Ac_negg": [0.6186816692352295, 0.13329757750034332, 0.05688320845365524, 0.0676628053188324, 0.17456591129302979, 0.1374761164188385, 0.032134220004081726, 0.007959588430821896, 0.08307185769081116, 0.09195937216281891], "r_norm": [1.7735178470611572, 1.6732423305511475, 1.7767467498779297, 1.6892765760421753, 1.8081178665161133, 1.7160742282867432, 1.822899341583252, 1.6610676050186157, 1.7158029079437256, 1.5910365581512451]}, {"epoch_end": 1, "step": 782, "test_acc": 0.3956, "test_loss": 1.7484036254882813}, {"step": 1000, "epoch": 2, "train_loss": 1.7308626174926758, "cos_r_negg": [0.7285373210906982, 0.2141266167163849, 0.15224072337150574, 0.11251021921634674, 0.2175149768590927, 0.2475420981645584, 0.09858103096485138, 0.07968410849571228, 0.16850169003009796, 0.15028029680252075], "cos_innovation_negg": [0.7285373210906982, 0.2141266167163849, 0.15224072337150574, 0.11251021921634674, 0.2175149768590927, 0.2475420981645584, 0.09858103096485138, 0.07968410849571228, 0.16850169003009796, 0.15028029680252075], "cos_apical_negg": [0.7285373210906982, 0.2141266167163849, 0.15224072337150574, 0.11251021921634674, 0.2175149768590927, 0.2475420981645584, 0.09858103096485138, 0.07968410849571228, 0.16850169003009796, 0.15028029680252075], "cos_Ac_negg": [0.7285373210906982, 0.2141266167163849, 0.15224072337150574, 0.11251021921634674, 0.2175149768590927, 0.2475420981645584, 0.09858103096485138, 0.07968410849571228, 0.16850169003009796, 0.15028029680252075], "r_norm": [1.496934413909912, 1.346694827079773, 1.4201955795288086, 1.355584740638733, 1.461911678314209, 1.385198950767517, 1.4596062898635864, 1.3296327590942383, 1.3754956722259521, 1.2817555665969849]}, {"epoch_end": 2, "step": 1173, "test_acc": 0.4094, "test_loss": 1.7081601318359374}, {"step": 1500, "epoch": 3, "train_loss": 1.7025033235549927, "cos_r_negg": [0.7352991700172424, 0.31392210721969604, 0.22603045403957367, 0.20909461379051208, 0.31354910135269165, 0.32676202058792114, 0.17813296616077423, 0.2323893904685974, 0.26618701219558716, 0.2948843240737915], "cos_innovation_negg": [0.7352991700172424, 0.31392210721969604, 0.22603045403957367, 0.20909461379051208, 0.31354910135269165, 0.32676202058792114, 0.17813296616077423, 0.2323893904685974, 0.26618701219558716, 0.2948843240737915], "cos_apical_negg": [0.7352991700172424, 0.31392210721969604, 0.22603045403957367, 0.20909461379051208, 0.31354910135269165, 0.32676202058792114, 0.17813296616077423, 0.2323893904685974, 0.26618701219558716, 0.2948843240737915], "cos_Ac_negg": [0.7352991700172424, 0.31392210721969604, 0.22603045403957367, 0.20909461379051208, 0.31354910135269165, 0.32676202058792114, 0.17813296616077423, 0.2323893904685974, 0.26618701219558716, 0.2948843240737915], "r_norm": [1.3232111930847168, 1.1326242685317993, 1.1844547986984253, 1.129760503768921, 1.2296662330627441, 1.1742801666259766, 1.2066210508346558, 1.1059434413909912, 1.15767502784729, 1.07524836063385]}, {"epoch_end": 3, "step": 1564, "test_acc": 0.4103, "test_loss": 1.7133595336914063}, {"epoch_end": 4, "step": 1955, "test_acc": 0.4073, "test_loss": 1.6981572631835937}], "final": {"test_acc": 0.4073, "test_loss": 1.6981572631835937, "wall_s": 24.6966552734375, "cos_r_negg": [0.7890317440032959, 0.415300190448761, 0.32825377583503723, 0.3291832208633423, 0.3790973424911499, 0.4270481467247009, 0.2915228307247162, 0.3314583897590637, 0.3846270442008972, 0.3905963897705078], "cos_innovation_negg": [0.7890317440032959, 0.415300190448761, 0.32825377583503723, 0.3291832208633423, 0.3790973424911499, 0.4270481467247009, 0.2915228307247162, 0.3314583897590637, 0.3846270442008972, 0.3905963897705078], "cos_apical_negg": [0.7890317440032959, 0.415300190448761, 0.32825377583503723, 0.3291832208633423, 0.3790973424911499, 0.4270481467247009, 0.2915228307247162, 0.3314583897590637, 0.3846270442008972, 0.3905963897705078], "cos_Ac_negg": [0.7890317440032959, 0.415300190448761, 0.32825377583503723, 0.3291832208633423, 0.3790973424911499, 0.4270481467247009, 0.2915228307247162, 0.3314583897590637, 0.3846270442008972, 0.3905963897705078]}}