summaryrefslogtreecommitdiff
path: root/results/ep_archmatch_v1_mnist_dfa_w500_d1_s0.json
blob: dc324181fe477776b171a5c35f36cc2cf4bf8baa (plain)
1
{"args": {"mode": "dfa", "dataset": "mnist", "depth": 1, "width": 500, "act": "tanh", "residual": 0, "epochs": 25, "batch_size": 128, "train_examples": 50000, "eta": 0.05, "eta_A": 0.02, "eta_P": 0.002, "momentum": 0.9, "w_scale": 1.0, "a_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": 50, "max_steps": 0, "probe_bs": 512, "outdir": "results", "tag": "ep_archmatch_v1_mnist_dfa_w500_d1_s0", "n_in": 784}, "provenance": {"git_commit": "262c6d9ef3f0461218c1a5e632fc7dd6357e0a20", "git_dirty": false}, "steps": [{"step": 0, "epoch": 0, "train_loss": 2.529653549194336, "cos_r_negg": [0.007927153259515762], "cos_innovation_negg": [0.007927153259515762], "cos_apical_negg": [0.007927153259515762], "cos_Ac_negg": [0.007927153259515762], "r_norm": [6.919894218444824]}, {"step": 50, "epoch": 0, "train_loss": 0.5378985404968262, "cos_r_negg": [0.446338027715683], "cos_innovation_negg": [0.446338027715683], "cos_apical_negg": [0.446338027715683], "cos_Ac_negg": [0.446338027715683], "r_norm": [1.4515876770019531]}, {"step": 100, "epoch": 0, "train_loss": 0.15611110627651215, "cos_r_negg": [0.4595390260219574], "cos_innovation_negg": [0.4595390260219574], "cos_apical_negg": [0.4595390260219574], "cos_Ac_negg": [0.4595390260219574], "r_norm": [1.1902644634246826]}, {"step": 150, "epoch": 0, "train_loss": 0.17079968750476837, "cos_r_negg": [0.46849602460861206], "cos_innovation_negg": [0.46849602460861206], "cos_apical_negg": [0.46849602460861206], "cos_Ac_negg": [0.46849602460861206], "r_norm": [1.005897045135498]}, {"step": 200, "epoch": 0, "train_loss": 0.3053852319717407, "cos_r_negg": [0.47044897079467773], "cos_innovation_negg": [0.47044897079467773], "cos_apical_negg": [0.47044897079467773], "cos_Ac_negg": [0.47044897079467773], "r_norm": [1.0187816619873047]}, {"step": 250, "epoch": 0, "train_loss": 0.2931833863258362, "cos_r_negg": [0.4832371175289154], "cos_innovation_negg": [0.4832371175289154], "cos_apical_negg": [0.4832371175289154], "cos_Ac_negg": [0.4832371175289154], "r_norm": [0.8806146383285522]}, {"step": 300, "epoch": 0, "train_loss": 0.1320013403892517, "cos_r_negg": [0.4842876195907593], "cos_innovation_negg": [0.4842876195907593], "cos_apical_negg": [0.4842876195907593], "cos_Ac_negg": [0.4842876195907593], "r_norm": [0.788491427898407]}, {"step": 350, "epoch": 0, "train_loss": 0.23352736234664917, "cos_r_negg": [0.5069469809532166], "cos_innovation_negg": [0.5069469809532166], "cos_apical_negg": [0.5069469809532166], "cos_Ac_negg": [0.5069469809532166], "r_norm": [0.6924450397491455]}, {"epoch_end": 0, "step": 391, "test_acc": 0.9526, "test_loss": 0.16421128311157226}, {"step": 400, "epoch": 1, "train_loss": 0.12783637642860413, "cos_r_negg": [0.5114192962646484], "cos_innovation_negg": [0.5114192962646484], "cos_apical_negg": [0.5114192962646484], "cos_Ac_negg": [0.5114192962646484], "r_norm": [0.6613460779190063]}, {"step": 450, "epoch": 1, "train_loss": 0.13752709329128265, "cos_r_negg": [0.5126492381095886], "cos_innovation_negg": [0.5126492381095886], "cos_apical_negg": [0.5126492381095886], "cos_Ac_negg": [0.5126492381095886], "r_norm": [0.5940003395080566]}, {"step": 500, "epoch": 1, "train_loss": 0.207707479596138, "cos_r_negg": [0.5260812640190125], "cos_innovation_negg": [0.5260812640190125], "cos_apical_negg": [0.5260812640190125], "cos_Ac_negg": [0.5260812640190125], "r_norm": [0.41802847385406494]}, {"step": 550, "epoch": 1, "train_loss": 0.09028483927249908, "cos_r_negg": [0.5275427103042603], "cos_innovation_negg": [0.5275427103042603], "cos_apical_negg": [0.5275427103042603], "cos_Ac_negg": [0.5275427103042603], "r_norm": [0.5970697999000549]}, {"step": 600, "epoch": 1, "train_loss": 0.056863557547330856, "cos_r_negg": [0.5313643217086792], "cos_innovation_negg": [0.5313643217086792], "cos_apical_negg": [0.5313643217086792], "cos_Ac_negg": [0.5313643217086792], "r_norm": [0.4569489061832428]}, {"step": 650, "epoch": 1, "train_loss": 0.05803891643881798, "cos_r_negg": [0.540709376335144], "cos_innovation_negg": [0.540709376335144], "cos_apical_negg": [0.540709376335144], "cos_Ac_negg": [0.540709376335144], "r_norm": [0.4741382896900177]}, {"step": 700, "epoch": 1, "train_loss": 0.08346912264823914, "cos_r_negg": [0.5343747138977051], "cos_innovation_negg": [0.5343747138977051], "cos_apical_negg": [0.5343747138977051], "cos_Ac_negg": [0.5343747138977051], "r_norm": [0.5042672157287598]}, {"step": 750, "epoch": 1, "train_loss": 0.15645092725753784, "cos_r_negg": [0.5503103733062744], "cos_innovation_negg": [0.5503103733062744], "cos_apical_negg": [0.5503103733062744], "cos_Ac_negg": [0.5503103733062744], "r_norm": [0.6418039798736572]}, {"epoch_end": 1, "step": 782, "test_acc": 0.9588, "test_loss": 0.13556932945251465}, {"step": 800, "epoch": 2, "train_loss": 0.09077601879835129, "cos_r_negg": [0.5468267798423767], "cos_innovation_negg": [0.5468267798423767], "cos_apical_negg": [0.5468267798423767], "cos_Ac_negg": [0.5468267798423767], "r_norm": [0.45994025468826294]}, {"step": 850, "epoch": 2, "train_loss": 0.04424097761511803, "cos_r_negg": [0.5566318035125732], "cos_innovation_negg": [0.5566318035125732], "cos_apical_negg": [0.5566318035125732], "cos_Ac_negg": [0.5566318035125732], "r_norm": [0.42781031131744385]}, {"step": 900, "epoch": 2, "train_loss": 0.07376974076032639, "cos_r_negg": [0.5656780004501343], "cos_innovation_negg": [0.5656780004501343], "cos_apical_negg": [0.5656780004501343], "cos_Ac_negg": [0.5656780004501343], "r_norm": [0.4573560655117035]}, {"step": 950, "epoch": 2, "train_loss": 0.08151404559612274, "cos_r_negg": [0.5647845268249512], "cos_innovation_negg": [0.5647845268249512], "cos_apical_negg": [0.5647845268249512], "cos_Ac_negg": [0.5647845268249512], "r_norm": [0.36703014373779297]}, {"step": 1000, "epoch": 2, "train_loss": 0.06810769438743591, "cos_r_negg": [0.5567589998245239], "cos_innovation_negg": [0.5567589998245239], "cos_apical_negg": [0.5567589998245239], "cos_Ac_negg": [0.5567589998245239], "r_norm": [0.38407331705093384]}, {"step": 1050, "epoch": 2, "train_loss": 0.03570346534252167, "cos_r_negg": [0.5609099864959717], "cos_innovation_negg": [0.5609099864959717], "cos_apical_negg": [0.5609099864959717], "cos_Ac_negg": [0.5609099864959717], "r_norm": [0.4029432535171509]}, {"step": 1100, "epoch": 2, "train_loss": 0.13104398548603058, "cos_r_negg": [0.5644813179969788], "cos_innovation_negg": [0.5644813179969788], "cos_apical_negg": [0.5644813179969788], "cos_Ac_negg": [0.5644813179969788], "r_norm": [0.3724943697452545]}, {"step": 1150, "epoch": 2, "train_loss": 0.0787217915058136, "cos_r_negg": [0.5776868462562561], "cos_innovation_negg": [0.5776868462562561], "cos_apical_negg": [0.5776868462562561], "cos_Ac_negg": [0.5776868462562561], "r_norm": [0.3190084993839264]}, {"epoch_end": 2, "step": 1173, "test_acc": 0.9687, "test_loss": 0.10461063003540039}, {"step": 1200, "epoch": 3, "train_loss": 0.03989540785551071, "cos_r_negg": [0.5671805739402771], "cos_innovation_negg": [0.5671805739402771], "cos_apical_negg": [0.5671805739402771], "cos_Ac_negg": [0.5671805739402771], "r_norm": [0.3714447617530823]}, {"step": 1250, "epoch": 3, "train_loss": 0.07401026785373688, "cos_r_negg": [0.5758885145187378], "cos_innovation_negg": [0.5758885145187378], "cos_apical_negg": [0.5758885145187378], "cos_Ac_negg": [0.5758885145187378], "r_norm": [0.34470659494400024]}, {"step": 1300, "epoch": 3, "train_loss": 0.0295491311699152, "cos_r_negg": [0.5792849063873291], "cos_innovation_negg": [0.5792849063873291], "cos_apical_negg": [0.5792849063873291], "cos_Ac_negg": [0.5792849063873291], "r_norm": [0.33078885078430176]}, {"step": 1350, "epoch": 3, "train_loss": 0.013916345313191414, "cos_r_negg": [0.5819554328918457], "cos_innovation_negg": [0.5819554328918457], "cos_apical_negg": [0.5819554328918457], "cos_Ac_negg": [0.5819554328918457], "r_norm": [0.30006662011146545]}, {"step": 1400, "epoch": 3, "train_loss": 0.037473201751708984, "cos_r_negg": [0.5953552722930908], "cos_innovation_negg": [0.5953552722930908], "cos_apical_negg": [0.5953552722930908], "cos_Ac_negg": [0.5953552722930908], "r_norm": [0.34922927618026733]}, {"step": 1450, "epoch": 3, "train_loss": 0.018525030463933945, "cos_r_negg": [0.5993934273719788], "cos_innovation_negg": [0.5993934273719788], "cos_apical_negg": [0.5993934273719788], "cos_Ac_negg": [0.5993934273719788], "r_norm": [0.344095379114151]}, {"step": 1500, "epoch": 3, "train_loss": 0.07173670828342438, "cos_r_negg": [0.5991964340209961], "cos_innovation_negg": [0.5991964340209961], "cos_apical_negg": [0.5991964340209961], "cos_Ac_negg": [0.5991964340209961], "r_norm": [0.24706979095935822]}, {"step": 1550, "epoch": 3, "train_loss": 0.03505013510584831, "cos_r_negg": [0.5985463261604309], "cos_innovation_negg": [0.5985463261604309], "cos_apical_negg": [0.5985463261604309], "cos_Ac_negg": [0.5985463261604309], "r_norm": [0.28853869438171387]}, {"epoch_end": 3, "step": 1564, "test_acc": 0.9668, "test_loss": 0.11504416694641113}, {"step": 1600, "epoch": 4, "train_loss": 0.06070321053266525, "cos_r_negg": [0.6009326577186584], "cos_innovation_negg": [0.6009326577186584], "cos_apical_negg": [0.6009326577186584], "cos_Ac_negg": [0.6009326577186584], "r_norm": [0.3139006793498993]}, {"step": 1650, "epoch": 4, "train_loss": 0.0654536783695221, "cos_r_negg": [0.6011091470718384], "cos_innovation_negg": [0.6011091470718384], "cos_apical_negg": [0.6011091470718384], "cos_Ac_negg": [0.6011091470718384], "r_norm": [0.2972111999988556]}, {"step": 1700, "epoch": 4, "train_loss": 0.058433350175619125, "cos_r_negg": [0.6047418117523193], "cos_innovation_negg": [0.6047418117523193], "cos_apical_negg": [0.6047418117523193], "cos_Ac_negg": [0.6047418117523193], "r_norm": [0.2860991656780243]}, {"step": 1750, "epoch": 4, "train_loss": 0.04591287672519684, "cos_r_negg": [0.6114698648452759], "cos_innovation_negg": [0.6114698648452759], "cos_apical_negg": [0.6114698648452759], "cos_Ac_negg": [0.6114698648452759], "r_norm": [0.2644854485988617]}, {"step": 1800, "epoch": 4, "train_loss": 0.019722694531083107, "cos_r_negg": [0.6019526720046997], "cos_innovation_negg": [0.6019526720046997], "cos_apical_negg": [0.6019526720046997], "cos_Ac_negg": [0.6019526720046997], "r_norm": [0.24991106986999512]}, {"step": 1850, "epoch": 4, "train_loss": 0.07621756196022034, "cos_r_negg": [0.6084581017494202], "cos_innovation_negg": [0.6084581017494202], "cos_apical_negg": [0.6084581017494202], "cos_Ac_negg": [0.6084581017494202], "r_norm": [0.2757892906665802]}, {"step": 1900, "epoch": 4, "train_loss": 0.00765196792781353, "cos_r_negg": [0.6242289543151855], "cos_innovation_negg": [0.6242289543151855], "cos_apical_negg": [0.6242289543151855], "cos_Ac_negg": [0.6242289543151855], "r_norm": [0.22341588139533997]}, {"step": 1950, "epoch": 4, "train_loss": 0.07084986567497253, "cos_r_negg": [0.6056423187255859], "cos_innovation_negg": [0.6056423187255859], "cos_apical_negg": [0.6056423187255859], "cos_Ac_negg": [0.6056423187255859], "r_norm": [0.2361248880624771]}, {"epoch_end": 4, "step": 1955, "test_acc": 0.9723, "test_loss": 0.09193439445495606}, {"step": 2000, "epoch": 5, "train_loss": 0.022296397015452385, "cos_r_negg": [0.6212233304977417], "cos_innovation_negg": [0.6212233304977417], "cos_apical_negg": [0.6212233304977417], "cos_Ac_negg": [0.6212233304977417], "r_norm": [0.23196306824684143]}, {"step": 2050, "epoch": 5, "train_loss": 0.013410639949142933, "cos_r_negg": [0.6127545237541199], "cos_innovation_negg": [0.6127545237541199], "cos_apical_negg": [0.6127545237541199], "cos_Ac_negg": [0.6127545237541199], "r_norm": [0.25504547357559204]}, {"step": 2100, "epoch": 5, "train_loss": 0.010225128382444382, "cos_r_negg": [0.6228626370429993], "cos_innovation_negg": [0.6228626370429993], "cos_apical_negg": [0.6228626370429993], "cos_Ac_negg": [0.6228626370429993], "r_norm": [0.23373067378997803]}, {"step": 2150, "epoch": 5, "train_loss": 0.0036907277535647154, "cos_r_negg": [0.6145119667053223], "cos_innovation_negg": [0.6145119667053223], "cos_apical_negg": [0.6145119667053223], "cos_Ac_negg": [0.6145119667053223], "r_norm": [0.2328505963087082]}, {"step": 2200, "epoch": 5, "train_loss": 0.01594649627804756, "cos_r_negg": [0.6221743822097778], "cos_innovation_negg": [0.6221743822097778], "cos_apical_negg": [0.6221743822097778], "cos_Ac_negg": [0.6221743822097778], "r_norm": [0.24157926440238953]}, {"step": 2250, "epoch": 5, "train_loss": 0.01261073350906372, "cos_r_negg": [0.6334763169288635], "cos_innovation_negg": [0.6334763169288635], "cos_apical_negg": [0.6334763169288635], "cos_Ac_negg": [0.6334763169288635], "r_norm": [0.2142297923564911]}, {"step": 2300, "epoch": 5, "train_loss": 0.011496011167764664, "cos_r_negg": [0.6314460039138794], "cos_innovation_negg": [0.6314460039138794], "cos_apical_negg": [0.6314460039138794], "cos_Ac_negg": [0.6314460039138794], "r_norm": [0.2083812803030014]}, {"epoch_end": 5, "step": 2346, "test_acc": 0.9736, "test_loss": 0.09129757633209229}, {"step": 2350, "epoch": 6, "train_loss": 0.016033321619033813, "cos_r_negg": [0.6286594867706299], "cos_innovation_negg": [0.6286594867706299], "cos_apical_negg": [0.6286594867706299], "cos_Ac_negg": [0.6286594867706299], "r_norm": [0.21533192694187164]}, {"step": 2400, "epoch": 6, "train_loss": 0.0075647542253136635, "cos_r_negg": [0.6392087340354919], "cos_innovation_negg": [0.6392087340354919], "cos_apical_negg": [0.6392087340354919], "cos_Ac_negg": [0.6392087340354919], "r_norm": [0.23080016672611237]}, {"step": 2450, "epoch": 6, "train_loss": 0.01589258573949337, "cos_r_negg": [0.6323435306549072], "cos_innovation_negg": [0.6323435306549072], "cos_apical_negg": [0.6323435306549072], "cos_Ac_negg": [0.6323435306549072], "r_norm": [0.23569393157958984]}, {"step": 2500, "epoch": 6, "train_loss": 0.013036523014307022, "cos_r_negg": [0.6303362250328064], "cos_innovation_negg": [0.6303362250328064], "cos_apical_negg": [0.6303362250328064], "cos_Ac_negg": [0.6303362250328064], "r_norm": [0.2163347452878952]}, {"step": 2550, "epoch": 6, "train_loss": 0.01871556043624878, "cos_r_negg": [0.6301670074462891], "cos_innovation_negg": [0.6301670074462891], "cos_apical_negg": [0.6301670074462891], "cos_Ac_negg": [0.6301670074462891], "r_norm": [0.18959662318229675]}, {"step": 2600, "epoch": 6, "train_loss": 0.007934422232210636, "cos_r_negg": [0.6363259553909302], "cos_innovation_negg": [0.6363259553909302], "cos_apical_negg": [0.6363259553909302], "cos_Ac_negg": [0.6363259553909302], "r_norm": [0.2006957232952118]}, {"step": 2650, "epoch": 6, "train_loss": 0.02632422000169754, "cos_r_negg": [0.6353095769882202], "cos_innovation_negg": [0.6353095769882202], "cos_apical_negg": [0.6353095769882202], "cos_Ac_negg": [0.6353095769882202], "r_norm": [0.21200628578662872]}, {"step": 2700, "epoch": 6, "train_loss": 0.0036291200667619705, "cos_r_negg": [0.6354855298995972], "cos_innovation_negg": [0.6354855298995972], "cos_apical_negg": [0.6354855298995972], "cos_Ac_negg": [0.6354855298995972], "r_norm": [0.24552160501480103]}, {"epoch_end": 6, "step": 2737, "test_acc": 0.9764, "test_loss": 0.08520364227294921}, {"step": 2750, "epoch": 7, "train_loss": 0.0025059834588319063, "cos_r_negg": [0.6370502710342407], "cos_innovation_negg": [0.6370502710342407], "cos_apical_negg": [0.6370502710342407], "cos_Ac_negg": [0.6370502710342407], "r_norm": [0.21230348944664001]}, {"step": 2800, "epoch": 7, "train_loss": 0.0032542741391807795, "cos_r_negg": [0.6403782367706299], "cos_innovation_negg": [0.6403782367706299], "cos_apical_negg": [0.6403782367706299], "cos_Ac_negg": [0.6403782367706299], "r_norm": [0.21699409186840057]}, {"step": 2850, "epoch": 7, "train_loss": 0.003031367203220725, "cos_r_negg": [0.6405731439590454], "cos_innovation_negg": [0.6405731439590454], "cos_apical_negg": [0.6405731439590454], "cos_Ac_negg": [0.6405731439590454], "r_norm": [0.19857990741729736]}, {"step": 2900, "epoch": 7, "train_loss": 0.004416514188051224, "cos_r_negg": [0.6357875466346741], "cos_innovation_negg": [0.6357875466346741], "cos_apical_negg": [0.6357875466346741], "cos_Ac_negg": [0.6357875466346741], "r_norm": [0.1913098394870758]}, {"step": 2950, "epoch": 7, "train_loss": 0.0033339718356728554, "cos_r_negg": [0.6328017115592957], "cos_innovation_negg": [0.6328017115592957], "cos_apical_negg": [0.6328017115592957], "cos_Ac_negg": [0.6328017115592957], "r_norm": [0.19515562057495117]}, {"step": 3000, "epoch": 7, "train_loss": 0.008366161957383156, "cos_r_negg": [0.6361695528030396], "cos_innovation_negg": [0.6361695528030396], "cos_apical_negg": [0.6361695528030396], "cos_Ac_negg": [0.6361695528030396], "r_norm": [0.21515579521656036]}, {"step": 3050, "epoch": 7, "train_loss": 0.004386312328279018, "cos_r_negg": [0.6459784507751465], "cos_innovation_negg": [0.6459784507751465], "cos_apical_negg": [0.6459784507751465], "cos_Ac_negg": [0.6459784507751465], "r_norm": [0.21083509922027588]}, {"step": 3100, "epoch": 7, "train_loss": 0.0035264105536043644, "cos_r_negg": [0.6453342437744141], "cos_innovation_negg": [0.6453342437744141], "cos_apical_negg": [0.6453342437744141], "cos_Ac_negg": [0.6453342437744141], "r_norm": [0.1780584156513214]}, {"epoch_end": 7, "step": 3128, "test_acc": 0.9753, "test_loss": 0.08862885894775391}, {"step": 3150, "epoch": 8, "train_loss": 0.003042601514607668, "cos_r_negg": [0.6433063745498657], "cos_innovation_negg": [0.6433063745498657], "cos_apical_negg": [0.6433063745498657], "cos_Ac_negg": [0.6433063745498657], "r_norm": [0.19454246759414673]}, {"step": 3200, "epoch": 8, "train_loss": 0.011032410897314548, "cos_r_negg": [0.644295871257782], "cos_innovation_negg": [0.644295871257782], "cos_apical_negg": [0.644295871257782], "cos_Ac_negg": [0.644295871257782], "r_norm": [0.20710989832878113]}, {"step": 3250, "epoch": 8, "train_loss": 0.0029919464141130447, "cos_r_negg": [0.6445097923278809], "cos_innovation_negg": [0.6445097923278809], "cos_apical_negg": [0.6445097923278809], "cos_Ac_negg": [0.6445097923278809], "r_norm": [0.20775106549263]}, {"step": 3300, "epoch": 8, "train_loss": 0.0023679169826209545, "cos_r_negg": [0.644466757774353], "cos_innovation_negg": [0.644466757774353], "cos_apical_negg": [0.644466757774353], "cos_Ac_negg": [0.644466757774353], "r_norm": [0.191869854927063]}, {"step": 3350, "epoch": 8, "train_loss": 0.006431836634874344, "cos_r_negg": [0.648067057132721], "cos_innovation_negg": [0.648067057132721], "cos_apical_negg": [0.648067057132721], "cos_Ac_negg": [0.648067057132721], "r_norm": [0.19135840237140656]}, {"step": 3400, "epoch": 8, "train_loss": 0.00398094579577446, "cos_r_negg": [0.6451742649078369], "cos_innovation_negg": [0.6451742649078369], "cos_apical_negg": [0.6451742649078369], "cos_Ac_negg": [0.6451742649078369], "r_norm": [0.20446622371673584]}, {"step": 3450, "epoch": 8, "train_loss": 0.014922883361577988, "cos_r_negg": [0.6469951868057251], "cos_innovation_negg": [0.6469951868057251], "cos_apical_negg": [0.6469951868057251], "cos_Ac_negg": [0.6469951868057251], "r_norm": [0.2097967118024826]}, {"step": 3500, "epoch": 8, "train_loss": 0.0067122867330908775, "cos_r_negg": [0.6454347372055054], "cos_innovation_negg": [0.6454347372055054], "cos_apical_negg": [0.6454347372055054], "cos_Ac_negg": [0.6454347372055054], "r_norm": [0.19009284675121307]}, {"epoch_end": 8, "step": 3519, "test_acc": 0.9755, "test_loss": 0.08785459270477294}, {"step": 3550, "epoch": 9, "train_loss": 0.003050497267395258, "cos_r_negg": [0.6495440602302551], "cos_innovation_negg": [0.6495440602302551], "cos_apical_negg": [0.6495440602302551], "cos_Ac_negg": [0.6495440602302551], "r_norm": [0.20186209678649902]}, {"step": 3600, "epoch": 9, "train_loss": 0.003470212686806917, "cos_r_negg": [0.650248646736145], "cos_innovation_negg": [0.650248646736145], "cos_apical_negg": [0.650248646736145], "cos_Ac_negg": [0.650248646736145], "r_norm": [0.18642909824848175]}, {"step": 3650, "epoch": 9, "train_loss": 0.005127946846187115, "cos_r_negg": [0.6513925194740295], "cos_innovation_negg": [0.6513925194740295], "cos_apical_negg": [0.6513925194740295], "cos_Ac_negg": [0.6513925194740295], "r_norm": [0.17918813228607178]}, {"step": 3700, "epoch": 9, "train_loss": 0.0054059759713709354, "cos_r_negg": [0.6479052305221558], "cos_innovation_negg": [0.6479052305221558], "cos_apical_negg": [0.6479052305221558], "cos_Ac_negg": [0.6479052305221558], "r_norm": [0.1881476640701294]}, {"step": 3750, "epoch": 9, "train_loss": 0.005067600402981043, "cos_r_negg": [0.6526027321815491], "cos_innovation_negg": [0.6526027321815491], "cos_apical_negg": [0.6526027321815491], "cos_Ac_negg": [0.6526027321815491], "r_norm": [0.1800059676170349]}, {"step": 3800, "epoch": 9, "train_loss": 0.002939029596745968, "cos_r_negg": [0.6529317498207092], "cos_innovation_negg": [0.6529317498207092], "cos_apical_negg": [0.6529317498207092], "cos_Ac_negg": [0.6529317498207092], "r_norm": [0.18275636434555054]}, {"step": 3850, "epoch": 9, "train_loss": 0.00386594096198678, "cos_r_negg": [0.6545953750610352], "cos_innovation_negg": [0.6545953750610352], "cos_apical_negg": [0.6545953750610352], "cos_Ac_negg": [0.6545953750610352], "r_norm": [0.19248118996620178]}, {"step": 3900, "epoch": 9, "train_loss": 0.004301756154745817, "cos_r_negg": [0.649564266204834], "cos_innovation_negg": [0.649564266204834], "cos_apical_negg": [0.649564266204834], "cos_Ac_negg": [0.649564266204834], "r_norm": [0.17353592813014984]}, {"epoch_end": 9, "step": 3910, "test_acc": 0.9755, "test_loss": 0.08616431102752685}, {"step": 3950, "epoch": 10, "train_loss": 0.002714196452870965, "cos_r_negg": [0.6493533849716187], "cos_innovation_negg": [0.6493533849716187], "cos_apical_negg": [0.6493533849716187], "cos_Ac_negg": [0.6493533849716187], "r_norm": [0.19229425489902496]}, {"step": 4000, "epoch": 10, "train_loss": 0.003034464782103896, "cos_r_negg": [0.6500505208969116], "cos_innovation_negg": [0.6500505208969116], "cos_apical_negg": [0.6500505208969116], "cos_Ac_negg": [0.6500505208969116], "r_norm": [0.19248169660568237]}, {"step": 4050, "epoch": 10, "train_loss": 0.0018771615577861667, "cos_r_negg": [0.6483029127120972], "cos_innovation_negg": [0.6483029127120972], "cos_apical_negg": [0.6483029127120972], "cos_Ac_negg": [0.6483029127120972], "r_norm": [0.18626996874809265]}, {"step": 4100, "epoch": 10, "train_loss": 0.0024207872338593006, "cos_r_negg": [0.6539337038993835], "cos_innovation_negg": [0.6539337038993835], "cos_apical_negg": [0.6539337038993835], "cos_Ac_negg": [0.6539337038993835], "r_norm": [0.18321529030799866]}, {"step": 4150, "epoch": 10, "train_loss": 0.001863198122009635, "cos_r_negg": [0.6543848514556885], "cos_innovation_negg": [0.6543848514556885], "cos_apical_negg": [0.6543848514556885], "cos_Ac_negg": [0.6543848514556885], "r_norm": [0.16975359618663788]}, {"step": 4200, "epoch": 10, "train_loss": 0.004063147585839033, "cos_r_negg": [0.655256450176239], "cos_innovation_negg": [0.655256450176239], "cos_apical_negg": [0.655256450176239], "cos_Ac_negg": [0.655256450176239], "r_norm": [0.17035937309265137]}, {"step": 4250, "epoch": 10, "train_loss": 0.003506137989461422, "cos_r_negg": [0.6565347909927368], "cos_innovation_negg": [0.6565347909927368], "cos_apical_negg": [0.6565347909927368], "cos_Ac_negg": [0.6565347909927368], "r_norm": [0.18507790565490723]}, {"step": 4300, "epoch": 10, "train_loss": 0.0017705544596537948, "cos_r_negg": [0.6559028625488281], "cos_innovation_negg": [0.6559028625488281], "cos_apical_negg": [0.6559028625488281], "cos_Ac_negg": [0.6559028625488281], "r_norm": [0.17871400713920593]}, {"epoch_end": 10, "step": 4301, "test_acc": 0.9765, "test_loss": 0.08781707572937011}, {"step": 4350, "epoch": 11, "train_loss": 0.004004799760878086, "cos_r_negg": [0.6556627750396729], "cos_innovation_negg": [0.6556627750396729], "cos_apical_negg": [0.6556627750396729], "cos_Ac_negg": [0.6556627750396729], "r_norm": [0.17836390435695648]}, {"step": 4400, "epoch": 11, "train_loss": 0.001953459344804287, "cos_r_negg": [0.6563931703567505], "cos_innovation_negg": [0.6563931703567505], "cos_apical_negg": [0.6563931703567505], "cos_Ac_negg": [0.6563931703567505], "r_norm": [0.1793738603591919]}, {"step": 4450, "epoch": 11, "train_loss": 0.0035812892019748688, "cos_r_negg": [0.656152069568634], "cos_innovation_negg": [0.656152069568634], "cos_apical_negg": [0.656152069568634], "cos_Ac_negg": [0.656152069568634], "r_norm": [0.17429597675800323]}, {"step": 4500, "epoch": 11, "train_loss": 0.001329883118160069, "cos_r_negg": [0.658604621887207], "cos_innovation_negg": [0.658604621887207], "cos_apical_negg": [0.658604621887207], "cos_Ac_negg": [0.658604621887207], "r_norm": [0.18984994292259216]}, {"step": 4550, "epoch": 11, "train_loss": 0.004240573383867741, "cos_r_negg": [0.656351625919342], "cos_innovation_negg": [0.656351625919342], "cos_apical_negg": [0.656351625919342], "cos_Ac_negg": [0.656351625919342], "r_norm": [0.19123215973377228]}, {"step": 4600, "epoch": 11, "train_loss": 0.0010742396116256714, "cos_r_negg": [0.656906247138977], "cos_innovation_negg": [0.656906247138977], "cos_apical_negg": [0.656906247138977], "cos_Ac_negg": [0.656906247138977], "r_norm": [0.17487865686416626]}, {"step": 4650, "epoch": 11, "train_loss": 0.002154822926968336, "cos_r_negg": [0.6589900255203247], "cos_innovation_negg": [0.6589900255203247], "cos_apical_negg": [0.6589900255203247], "cos_Ac_negg": [0.6589900255203247], "r_norm": [0.18067440390586853]}, {"epoch_end": 11, "step": 4692, "test_acc": 0.9758, "test_loss": 0.0886783088684082}, {"step": 4700, "epoch": 12, "train_loss": 0.0032127376180142164, "cos_r_negg": [0.6574692726135254], "cos_innovation_negg": [0.6574692726135254], "cos_apical_negg": [0.6574692726135254], "cos_Ac_negg": [0.6574692726135254], "r_norm": [0.17950871586799622]}, {"step": 4750, "epoch": 12, "train_loss": 0.002027628244832158, "cos_r_negg": [0.659750759601593], "cos_innovation_negg": [0.659750759601593], "cos_apical_negg": [0.659750759601593], "cos_Ac_negg": [0.659750759601593], "r_norm": [0.18223021924495697]}, {"step": 4800, "epoch": 12, "train_loss": 0.002229973440989852, "cos_r_negg": [0.6575031876564026], "cos_innovation_negg": [0.6575031876564026], "cos_apical_negg": [0.6575031876564026], "cos_Ac_negg": [0.6575031876564026], "r_norm": [0.18359892070293427]}, {"step": 4850, "epoch": 12, "train_loss": 0.001674095168709755, "cos_r_negg": [0.6567223072052002], "cos_innovation_negg": [0.6567223072052002], "cos_apical_negg": [0.6567223072052002], "cos_Ac_negg": [0.6567223072052002], "r_norm": [0.18486355245113373]}, {"step": 4900, "epoch": 12, "train_loss": 0.001528049586340785, "cos_r_negg": [0.6567079424858093], "cos_innovation_negg": [0.6567079424858093], "cos_apical_negg": [0.6567079424858093], "cos_Ac_negg": [0.6567079424858093], "r_norm": [0.18663716316223145]}, {"step": 4950, "epoch": 12, "train_loss": 0.00165956886485219, "cos_r_negg": [0.6557912826538086], "cos_innovation_negg": [0.6557912826538086], "cos_apical_negg": [0.6557912826538086], "cos_Ac_negg": [0.6557912826538086], "r_norm": [0.17783990502357483]}, {"step": 5000, "epoch": 12, "train_loss": 0.0017242147587239742, "cos_r_negg": [0.6599199771881104], "cos_innovation_negg": [0.6599199771881104], "cos_apical_negg": [0.6599199771881104], "cos_Ac_negg": [0.6599199771881104], "r_norm": [0.17558735609054565]}, {"step": 5050, "epoch": 12, "train_loss": 0.0022029599640518427, "cos_r_negg": [0.659331202507019], "cos_innovation_negg": [0.659331202507019], "cos_apical_negg": [0.659331202507019], "cos_Ac_negg": [0.659331202507019], "r_norm": [0.17106519639492035]}, {"epoch_end": 12, "step": 5083, "test_acc": 0.9761, "test_loss": 0.08887008666992187}, {"step": 5100, "epoch": 13, "train_loss": 0.001803007093258202, "cos_r_negg": [0.6594165563583374], "cos_innovation_negg": [0.6594165563583374], "cos_apical_negg": [0.6594165563583374], "cos_Ac_negg": [0.6594165563583374], "r_norm": [0.17785564064979553]}, {"step": 5150, "epoch": 13, "train_loss": 0.004073751624673605, "cos_r_negg": [0.661300539970398], "cos_innovation_negg": [0.661300539970398], "cos_apical_negg": [0.661300539970398], "cos_Ac_negg": [0.661300539970398], "r_norm": [0.1819925457239151]}, {"step": 5200, "epoch": 13, "train_loss": 0.0028512233402580023, "cos_r_negg": [0.6621579527854919], "cos_innovation_negg": [0.6621579527854919], "cos_apical_negg": [0.6621579527854919], "cos_Ac_negg": [0.6621579527854919], "r_norm": [0.1874617636203766]}, {"step": 5250, "epoch": 13, "train_loss": 0.0018842148128896952, "cos_r_negg": [0.6602969169616699], "cos_innovation_negg": [0.6602969169616699], "cos_apical_negg": [0.6602969169616699], "cos_Ac_negg": [0.6602969169616699], "r_norm": [0.18164794147014618]}, {"step": 5300, "epoch": 13, "train_loss": 0.001935745240189135, "cos_r_negg": [0.6624356508255005], "cos_innovation_negg": [0.6624356508255005], "cos_apical_negg": [0.6624356508255005], "cos_Ac_negg": [0.6624356508255005], "r_norm": [0.17963793873786926]}, {"step": 5350, "epoch": 13, "train_loss": 0.002686596941202879, "cos_r_negg": [0.6628423929214478], "cos_innovation_negg": [0.6628423929214478], "cos_apical_negg": [0.6628423929214478], "cos_Ac_negg": [0.6628423929214478], "r_norm": [0.1857079118490219]}, {"step": 5400, "epoch": 13, "train_loss": 0.0015342553379014134, "cos_r_negg": [0.6614811420440674], "cos_innovation_negg": [0.6614811420440674], "cos_apical_negg": [0.6614811420440674], "cos_Ac_negg": [0.6614811420440674], "r_norm": [0.1764272153377533]}, {"step": 5450, "epoch": 13, "train_loss": 0.001627167803235352, "cos_r_negg": [0.6598830819129944], "cos_innovation_negg": [0.6598830819129944], "cos_apical_negg": [0.6598830819129944], "cos_Ac_negg": [0.6598830819129944], "r_norm": [0.17324164509773254]}, {"epoch_end": 13, "step": 5474, "test_acc": 0.9767, "test_loss": 0.08944097690582276}, {"step": 5500, "epoch": 14, "train_loss": 0.0014794734306633472, "cos_r_negg": [0.6616568565368652], "cos_innovation_negg": [0.6616568565368652], "cos_apical_negg": [0.6616568565368652], "cos_Ac_negg": [0.6616568565368652], "r_norm": [0.17759844660758972]}, {"step": 5550, "epoch": 14, "train_loss": 0.0009711239254102111, "cos_r_negg": [0.6609523892402649], "cos_innovation_negg": [0.6609523892402649], "cos_apical_negg": [0.6609523892402649], "cos_Ac_negg": [0.6609523892402649], "r_norm": [0.17489919066429138]}, {"step": 5600, "epoch": 14, "train_loss": 0.002977547235786915, "cos_r_negg": [0.6612458229064941], "cos_innovation_negg": [0.6612458229064941], "cos_apical_negg": [0.6612458229064941], "cos_Ac_negg": [0.6612458229064941], "r_norm": [0.17257159948349]}, {"step": 5650, "epoch": 14, "train_loss": 0.0009674192988313735, "cos_r_negg": [0.6620712280273438], "cos_innovation_negg": [0.6620712280273438], "cos_apical_negg": [0.6620712280273438], "cos_Ac_negg": [0.6620712280273438], "r_norm": [0.1730516105890274]}, {"step": 5700, "epoch": 14, "train_loss": 0.0004970199661329389, "cos_r_negg": [0.6602256298065186], "cos_innovation_negg": [0.6602256298065186], "cos_apical_negg": [0.6602256298065186], "cos_Ac_negg": [0.6602256298065186], "r_norm": [0.16998662054538727]}, {"step": 5750, "epoch": 14, "train_loss": 0.002579448278993368, "cos_r_negg": [0.6612381935119629], "cos_innovation_negg": [0.6612381935119629], "cos_apical_negg": [0.6612381935119629], "cos_Ac_negg": [0.6612381935119629], "r_norm": [0.17411211133003235]}, {"step": 5800, "epoch": 14, "train_loss": 0.004208316095173359, "cos_r_negg": [0.6595847606658936], "cos_innovation_negg": [0.6595847606658936], "cos_apical_negg": [0.6595847606658936], "cos_Ac_negg": [0.6595847606658936], "r_norm": [0.18091073632240295]}, {"step": 5850, "epoch": 14, "train_loss": 0.001818808726966381, "cos_r_negg": [0.6611254215240479], "cos_innovation_negg": [0.6611254215240479], "cos_apical_negg": [0.6611254215240479], "cos_Ac_negg": [0.6611254215240479], "r_norm": [0.17458906769752502]}, {"epoch_end": 14, "step": 5865, "test_acc": 0.9763, "test_loss": 0.08962259559631347}, {"step": 5900, "epoch": 15, "train_loss": 0.001702244975604117, "cos_r_negg": [0.6622937321662903], "cos_innovation_negg": [0.6622937321662903], "cos_apical_negg": [0.6622937321662903], "cos_Ac_negg": [0.6622937321662903], "r_norm": [0.17746537923812866]}, {"step": 5950, "epoch": 15, "train_loss": 0.0025866953656077385, "cos_r_negg": [0.6620649099349976], "cos_innovation_negg": [0.6620649099349976], "cos_apical_negg": [0.6620649099349976], "cos_Ac_negg": [0.6620649099349976], "r_norm": [0.17555402219295502]}, {"step": 6000, "epoch": 15, "train_loss": 0.0031983363442122936, "cos_r_negg": [0.6604093909263611], "cos_innovation_negg": [0.6604093909263611], "cos_apical_negg": [0.6604093909263611], "cos_Ac_negg": [0.6604093909263611], "r_norm": [0.17582987248897552]}, {"step": 6050, "epoch": 15, "train_loss": 0.004987574648112059, "cos_r_negg": [0.6616648435592651], "cos_innovation_negg": [0.6616648435592651], "cos_apical_negg": [0.6616648435592651], "cos_Ac_negg": [0.6616648435592651], "r_norm": [0.17288324236869812]}, {"step": 6100, "epoch": 15, "train_loss": 0.002388033317402005, "cos_r_negg": [0.6601161956787109], "cos_innovation_negg": [0.6601161956787109], "cos_apical_negg": [0.6601161956787109], "cos_Ac_negg": [0.6601161956787109], "r_norm": [0.17045314610004425]}, {"step": 6150, "epoch": 15, "train_loss": 0.001701981294900179, "cos_r_negg": [0.6634101867675781], "cos_innovation_negg": [0.6634101867675781], "cos_apical_negg": [0.6634101867675781], "cos_Ac_negg": [0.6634101867675781], "r_norm": [0.16769972443580627]}, {"step": 6200, "epoch": 15, "train_loss": 0.0015846432652324438, "cos_r_negg": [0.6644010543823242], "cos_innovation_negg": [0.6644010543823242], "cos_apical_negg": [0.6644010543823242], "cos_Ac_negg": [0.6644010543823242], "r_norm": [0.17465442419052124]}, {"step": 6250, "epoch": 15, "train_loss": 0.0010502494405955076, "cos_r_negg": [0.6645565032958984], "cos_innovation_negg": [0.6645565032958984], "cos_apical_negg": [0.6645565032958984], "cos_Ac_negg": [0.6645565032958984], "r_norm": [0.16563507914543152]}, {"epoch_end": 15, "step": 6256, "test_acc": 0.9774, "test_loss": 0.08891616344451904}, {"step": 6300, "epoch": 16, "train_loss": 0.002964187879115343, "cos_r_negg": [0.6626226902008057], "cos_innovation_negg": [0.6626226902008057], "cos_apical_negg": [0.6626226902008057], "cos_Ac_negg": [0.6626226902008057], "r_norm": [0.1781245470046997]}, {"step": 6350, "epoch": 16, "train_loss": 0.0007718633860349655, "cos_r_negg": [0.6630537509918213], "cos_innovation_negg": [0.6630537509918213], "cos_apical_negg": [0.6630537509918213], "cos_Ac_negg": [0.6630537509918213], "r_norm": [0.17614293098449707]}, {"step": 6400, "epoch": 16, "train_loss": 0.0027924366295337677, "cos_r_negg": [0.6617437601089478], "cos_innovation_negg": [0.6617437601089478], "cos_apical_negg": [0.6617437601089478], "cos_Ac_negg": [0.6617437601089478], "r_norm": [0.1721075177192688]}, {"step": 6450, "epoch": 16, "train_loss": 0.000959826516918838, "cos_r_negg": [0.6613703370094299], "cos_innovation_negg": [0.6613703370094299], "cos_apical_negg": [0.6613703370094299], "cos_Ac_negg": [0.6613703370094299], "r_norm": [0.176346093416214]}, {"step": 6500, "epoch": 16, "train_loss": 0.0013776915147900581, "cos_r_negg": [0.662568986415863], "cos_innovation_negg": [0.662568986415863], "cos_apical_negg": [0.662568986415863], "cos_Ac_negg": [0.662568986415863], "r_norm": [0.16943183541297913]}, {"step": 6550, "epoch": 16, "train_loss": 0.0019063070649281144, "cos_r_negg": [0.6650741100311279], "cos_innovation_negg": [0.6650741100311279], "cos_apical_negg": [0.6650741100311279], "cos_Ac_negg": [0.6650741100311279], "r_norm": [0.1697850227355957]}, {"step": 6600, "epoch": 16, "train_loss": 0.0013843283522874117, "cos_r_negg": [0.6621103286743164], "cos_innovation_negg": [0.6621103286743164], "cos_apical_negg": [0.6621103286743164], "cos_Ac_negg": [0.6621103286743164], "r_norm": [0.1635584980249405]}, {"epoch_end": 16, "step": 6647, "test_acc": 0.977, "test_loss": 0.08964291744232178}, {"step": 6650, "epoch": 17, "train_loss": 0.0011154257226735353, "cos_r_negg": [0.6619985103607178], "cos_innovation_negg": [0.6619985103607178], "cos_apical_negg": [0.6619985103607178], "cos_Ac_negg": [0.6619985103607178], "r_norm": [0.16848492622375488]}, {"step": 6700, "epoch": 17, "train_loss": 0.0007620551041327417, "cos_r_negg": [0.6607497334480286], "cos_innovation_negg": [0.6607497334480286], "cos_apical_negg": [0.6607497334480286], "cos_Ac_negg": [0.6607497334480286], "r_norm": [0.17597457766532898]}, {"step": 6750, "epoch": 17, "train_loss": 0.0018879538401961327, "cos_r_negg": [0.6646029949188232], "cos_innovation_negg": [0.6646029949188232], "cos_apical_negg": [0.6646029949188232], "cos_Ac_negg": [0.6646029949188232], "r_norm": [0.176978200674057]}, {"step": 6800, "epoch": 17, "train_loss": 0.0020793303847312927, "cos_r_negg": [0.6638383269309998], "cos_innovation_negg": [0.6638383269309998], "cos_apical_negg": [0.6638383269309998], "cos_Ac_negg": [0.6638383269309998], "r_norm": [0.17357929050922394]}, {"step": 6850, "epoch": 17, "train_loss": 0.0012102670734748244, "cos_r_negg": [0.6635543704032898], "cos_innovation_negg": [0.6635543704032898], "cos_apical_negg": [0.6635543704032898], "cos_Ac_negg": [0.6635543704032898], "r_norm": [0.1714283972978592]}, {"step": 6900, "epoch": 17, "train_loss": 0.0018079549772664905, "cos_r_negg": [0.6638709306716919], "cos_innovation_negg": [0.6638709306716919], "cos_apical_negg": [0.6638709306716919], "cos_Ac_negg": [0.6638709306716919], "r_norm": [0.16890369355678558]}, {"step": 6950, "epoch": 17, "train_loss": 0.001843736506998539, "cos_r_negg": [0.6620208621025085], "cos_innovation_negg": [0.6620208621025085], "cos_apical_negg": [0.6620208621025085], "cos_Ac_negg": [0.6620208621025085], "r_norm": [0.17306850850582123]}, {"step": 7000, "epoch": 17, "train_loss": 0.0020277579315006733, "cos_r_negg": [0.6619917750358582], "cos_innovation_negg": [0.6619917750358582], "cos_apical_negg": [0.6619917750358582], "cos_Ac_negg": [0.6619917750358582], "r_norm": [0.16570419073104858]}, {"epoch_end": 17, "step": 7038, "test_acc": 0.9756, "test_loss": 0.09115127239227296}, {"step": 7050, "epoch": 18, "train_loss": 0.0007009297842159867, "cos_r_negg": [0.6638060808181763], "cos_innovation_negg": [0.6638060808181763], "cos_apical_negg": [0.6638060808181763], "cos_Ac_negg": [0.6638060808181763], "r_norm": [0.1716570407152176]}, {"step": 7100, "epoch": 18, "train_loss": 0.0012758707161992788, "cos_r_negg": [0.6635274887084961], "cos_innovation_negg": [0.6635274887084961], "cos_apical_negg": [0.6635274887084961], "cos_Ac_negg": [0.6635274887084961], "r_norm": [0.17431879043579102]}, {"step": 7150, "epoch": 18, "train_loss": 0.0007745464681647718, "cos_r_negg": [0.6642131209373474], "cos_innovation_negg": [0.6642131209373474], "cos_apical_negg": [0.6642131209373474], "cos_Ac_negg": [0.6642131209373474], "r_norm": [0.17175468802452087]}, {"step": 7200, "epoch": 18, "train_loss": 0.001303626224398613, "cos_r_negg": [0.6641005277633667], "cos_innovation_negg": [0.6641005277633667], "cos_apical_negg": [0.6641005277633667], "cos_Ac_negg": [0.6641005277633667], "r_norm": [0.17121271789073944]}, {"step": 7250, "epoch": 18, "train_loss": 0.0012902661692351103, "cos_r_negg": [0.6636227369308472], "cos_innovation_negg": [0.6636227369308472], "cos_apical_negg": [0.6636227369308472], "cos_Ac_negg": [0.6636227369308472], "r_norm": [0.16889938712120056]}, {"step": 7300, "epoch": 18, "train_loss": 0.0007635349757038057, "cos_r_negg": [0.6639586687088013], "cos_innovation_negg": [0.6639586687088013], "cos_apical_negg": [0.6639586687088013], "cos_Ac_negg": [0.6639586687088013], "r_norm": [0.16924510896205902]}, {"step": 7350, "epoch": 18, "train_loss": 0.003067204263061285, "cos_r_negg": [0.6645622253417969], "cos_innovation_negg": [0.6645622253417969], "cos_apical_negg": [0.6645622253417969], "cos_Ac_negg": [0.6645622253417969], "r_norm": [0.17142267525196075]}, {"step": 7400, "epoch": 18, "train_loss": 0.0007239767583087087, "cos_r_negg": [0.6640191078186035], "cos_innovation_negg": [0.6640191078186035], "cos_apical_negg": [0.6640191078186035], "cos_Ac_negg": [0.6640191078186035], "r_norm": [0.1669405996799469]}, {"epoch_end": 18, "step": 7429, "test_acc": 0.9776, "test_loss": 0.09094000663757325}, {"step": 7450, "epoch": 19, "train_loss": 0.0006831792998127639, "cos_r_negg": [0.6619687080383301], "cos_innovation_negg": [0.6619687080383301], "cos_apical_negg": [0.6619687080383301], "cos_Ac_negg": [0.6619687080383301], "r_norm": [0.16538608074188232]}, {"step": 7500, "epoch": 19, "train_loss": 0.0010655021760612726, "cos_r_negg": [0.6635773777961731], "cos_innovation_negg": [0.6635773777961731], "cos_apical_negg": [0.6635773777961731], "cos_Ac_negg": [0.6635773777961731], "r_norm": [0.16893739998340607]}, {"step": 7550, "epoch": 19, "train_loss": 0.0006806849269196391, "cos_r_negg": [0.6641970872879028], "cos_innovation_negg": [0.6641970872879028], "cos_apical_negg": [0.6641970872879028], "cos_Ac_negg": [0.6641970872879028], "r_norm": [0.16580957174301147]}, {"step": 7600, "epoch": 19, "train_loss": 0.000440245377831161, "cos_r_negg": [0.6643680334091187], "cos_innovation_negg": [0.6643680334091187], "cos_apical_negg": [0.6643680334091187], "cos_Ac_negg": [0.6643680334091187], "r_norm": [0.17028790712356567]}, {"step": 7650, "epoch": 19, "train_loss": 0.001604728982783854, "cos_r_negg": [0.6654733419418335], "cos_innovation_negg": [0.6654733419418335], "cos_apical_negg": [0.6654733419418335], "cos_Ac_negg": [0.6654733419418335], "r_norm": [0.17399537563323975]}, {"step": 7700, "epoch": 19, "train_loss": 0.002592981094494462, "cos_r_negg": [0.6648432016372681], "cos_innovation_negg": [0.6648432016372681], "cos_apical_negg": [0.6648432016372681], "cos_Ac_negg": [0.6648432016372681], "r_norm": [0.16562968492507935]}, {"step": 7750, "epoch": 19, "train_loss": 0.0011266012443229556, "cos_r_negg": [0.6646178960800171], "cos_innovation_negg": [0.6646178960800171], "cos_apical_negg": [0.6646178960800171], "cos_Ac_negg": [0.6646178960800171], "r_norm": [0.1692628711462021]}, {"step": 7800, "epoch": 19, "train_loss": 0.0018553495174273849, "cos_r_negg": [0.6663547158241272], "cos_innovation_negg": [0.6663547158241272], "cos_apical_negg": [0.6663547158241272], "cos_Ac_negg": [0.6663547158241272], "r_norm": [0.17162495851516724]}, {"epoch_end": 19, "step": 7820, "test_acc": 0.9769, "test_loss": 0.0922179651260376}, {"step": 7850, "epoch": 20, "train_loss": 0.0014949343167245388, "cos_r_negg": [0.664222002029419], "cos_innovation_negg": [0.664222002029419], "cos_apical_negg": [0.664222002029419], "cos_Ac_negg": [0.664222002029419], "r_norm": [0.1665872037410736]}, {"step": 7900, "epoch": 20, "train_loss": 0.001578298513777554, "cos_r_negg": [0.6649748086929321], "cos_innovation_negg": [0.6649748086929321], "cos_apical_negg": [0.6649748086929321], "cos_Ac_negg": [0.6649748086929321], "r_norm": [0.1707608699798584]}, {"step": 7950, "epoch": 20, "train_loss": 0.0008258823654614389, "cos_r_negg": [0.6636658906936646], "cos_innovation_negg": [0.6636658906936646], "cos_apical_negg": [0.6636658906936646], "cos_Ac_negg": [0.6636658906936646], "r_norm": [0.17021676898002625]}, {"step": 8000, "epoch": 20, "train_loss": 0.0012938324362039566, "cos_r_negg": [0.6655979156494141], "cos_innovation_negg": [0.6655979156494141], "cos_apical_negg": [0.6655979156494141], "cos_Ac_negg": [0.6655979156494141], "r_norm": [0.16982626914978027]}, {"step": 8050, "epoch": 20, "train_loss": 0.0021154372952878475, "cos_r_negg": [0.6659412980079651], "cos_innovation_negg": [0.6659412980079651], "cos_apical_negg": [0.6659412980079651], "cos_Ac_negg": [0.6659412980079651], "r_norm": [0.17043495178222656]}, {"step": 8100, "epoch": 20, "train_loss": 0.0015141437761485577, "cos_r_negg": [0.6652356386184692], "cos_innovation_negg": [0.6652356386184692], "cos_apical_negg": [0.6652356386184692], "cos_Ac_negg": [0.6652356386184692], "r_norm": [0.16815754771232605]}, {"step": 8150, "epoch": 20, "train_loss": 0.001185504486784339, "cos_r_negg": [0.6645083427429199], "cos_innovation_negg": [0.6645083427429199], "cos_apical_negg": [0.6645083427429199], "cos_Ac_negg": [0.6645083427429199], "r_norm": [0.16986188292503357]}, {"step": 8200, "epoch": 20, "train_loss": 0.0007164583075791597, "cos_r_negg": [0.6649080514907837], "cos_innovation_negg": [0.6649080514907837], "cos_apical_negg": [0.6649080514907837], "cos_Ac_negg": [0.6649080514907837], "r_norm": [0.17069581151008606]}, {"epoch_end": 20, "step": 8211, "test_acc": 0.9764, "test_loss": 0.09235961103439332}, {"step": 8250, "epoch": 21, "train_loss": 0.0012270014267414808, "cos_r_negg": [0.6644504070281982], "cos_innovation_negg": [0.6644504070281982], "cos_apical_negg": [0.6644504070281982], "cos_Ac_negg": [0.6644504070281982], "r_norm": [0.17279984056949615]}, {"step": 8300, "epoch": 21, "train_loss": 0.0013807707000523806, "cos_r_negg": [0.6651556491851807], "cos_innovation_negg": [0.6651556491851807], "cos_apical_negg": [0.6651556491851807], "cos_Ac_negg": [0.6651556491851807], "r_norm": [0.16744640469551086]}, {"step": 8350, "epoch": 21, "train_loss": 0.0013558367500081658, "cos_r_negg": [0.6657724976539612], "cos_innovation_negg": [0.6657724976539612], "cos_apical_negg": [0.6657724976539612], "cos_Ac_negg": [0.6657724976539612], "r_norm": [0.16786617040634155]}, {"step": 8400, "epoch": 21, "train_loss": 0.0017479443922638893, "cos_r_negg": [0.6657877564430237], "cos_innovation_negg": [0.6657877564430237], "cos_apical_negg": [0.6657877564430237], "cos_Ac_negg": [0.6657877564430237], "r_norm": [0.16394180059432983]}, {"step": 8450, "epoch": 21, "train_loss": 0.001413336955010891, "cos_r_negg": [0.6661390662193298], "cos_innovation_negg": [0.6661390662193298], "cos_apical_negg": [0.6661390662193298], "cos_Ac_negg": [0.6661390662193298], "r_norm": [0.16530540585517883]}, {"step": 8500, "epoch": 21, "train_loss": 0.0015957646537572145, "cos_r_negg": [0.6660447716712952], "cos_innovation_negg": [0.6660447716712952], "cos_apical_negg": [0.6660447716712952], "cos_Ac_negg": [0.6660447716712952], "r_norm": [0.16783112287521362]}, {"step": 8550, "epoch": 21, "train_loss": 0.0010656488593667746, "cos_r_negg": [0.6666637063026428], "cos_innovation_negg": [0.6666637063026428], "cos_apical_negg": [0.6666637063026428], "cos_Ac_negg": [0.6666637063026428], "r_norm": [0.1656591147184372]}, {"step": 8600, "epoch": 21, "train_loss": 0.002240352798253298, "cos_r_negg": [0.6665427684783936], "cos_innovation_negg": [0.6665427684783936], "cos_apical_negg": [0.6665427684783936], "cos_Ac_negg": [0.6665427684783936], "r_norm": [0.1695120632648468]}, {"epoch_end": 21, "step": 8602, "test_acc": 0.976, "test_loss": 0.09303749122619628}, {"step": 8650, "epoch": 22, "train_loss": 0.0006289861630648375, "cos_r_negg": [0.666278600692749], "cos_innovation_negg": [0.666278600692749], "cos_apical_negg": [0.666278600692749], "cos_Ac_negg": [0.666278600692749], "r_norm": [0.17096863687038422]}, {"step": 8700, "epoch": 22, "train_loss": 0.0007678767433390021, "cos_r_negg": [0.6659321188926697], "cos_innovation_negg": [0.6659321188926697], "cos_apical_negg": [0.6659321188926697], "cos_Ac_negg": [0.6659321188926697], "r_norm": [0.17015340924263]}, {"step": 8750, "epoch": 22, "train_loss": 0.0008804360404610634, "cos_r_negg": [0.6653702259063721], "cos_innovation_negg": [0.6653702259063721], "cos_apical_negg": [0.6653702259063721], "cos_Ac_negg": [0.6653702259063721], "r_norm": [0.1693313866853714]}, {"step": 8800, "epoch": 22, "train_loss": 0.0016957317711785436, "cos_r_negg": [0.6660256385803223], "cos_innovation_negg": [0.6660256385803223], "cos_apical_negg": [0.6660256385803223], "cos_Ac_negg": [0.6660256385803223], "r_norm": [0.16689372062683105]}, {"step": 8850, "epoch": 22, "train_loss": 0.0007832589326426387, "cos_r_negg": [0.6687551736831665], "cos_innovation_negg": [0.6687551736831665], "cos_apical_negg": [0.6687551736831665], "cos_Ac_negg": [0.6687551736831665], "r_norm": [0.16506756842136383]}, {"step": 8900, "epoch": 22, "train_loss": 0.002062832470983267, "cos_r_negg": [0.6666590571403503], "cos_innovation_negg": [0.6666590571403503], "cos_apical_negg": [0.6666590571403503], "cos_Ac_negg": [0.6666590571403503], "r_norm": [0.16975219547748566]}, {"step": 8950, "epoch": 22, "train_loss": 0.0009238877100870013, "cos_r_negg": [0.6662837266921997], "cos_innovation_negg": [0.6662837266921997], "cos_apical_negg": [0.6662837266921997], "cos_Ac_negg": [0.6662837266921997], "r_norm": [0.17363396286964417]}, {"epoch_end": 22, "step": 8993, "test_acc": 0.9769, "test_loss": 0.09275986576080322}, {"step": 9000, "epoch": 23, "train_loss": 0.0005426039570011199, "cos_r_negg": [0.6663780212402344], "cos_innovation_negg": [0.6663780212402344], "cos_apical_negg": [0.6663780212402344], "cos_Ac_negg": [0.6663780212402344], "r_norm": [0.16741646826267242]}, {"step": 9050, "epoch": 23, "train_loss": 0.001080094138160348, "cos_r_negg": [0.6670243740081787], "cos_innovation_negg": [0.6670243740081787], "cos_apical_negg": [0.6670243740081787], "cos_Ac_negg": [0.6670243740081787], "r_norm": [0.1692594587802887]}, {"step": 9100, "epoch": 23, "train_loss": 0.0008646490750834346, "cos_r_negg": [0.6668412685394287], "cos_innovation_negg": [0.6668412685394287], "cos_apical_negg": [0.6668412685394287], "cos_Ac_negg": [0.6668412685394287], "r_norm": [0.16468298435211182]}, {"step": 9150, "epoch": 23, "train_loss": 0.0016141673550009727, "cos_r_negg": [0.6663103103637695], "cos_innovation_negg": [0.6663103103637695], "cos_apical_negg": [0.6663103103637695], "cos_Ac_negg": [0.6663103103637695], "r_norm": [0.16689661145210266]}, {"step": 9200, "epoch": 23, "train_loss": 0.0017156414687633514, "cos_r_negg": [0.6665346622467041], "cos_innovation_negg": [0.6665346622467041], "cos_apical_negg": [0.6665346622467041], "cos_Ac_negg": [0.6665346622467041], "r_norm": [0.1663321852684021]}, {"step": 9250, "epoch": 23, "train_loss": 0.0010491367429494858, "cos_r_negg": [0.6658576726913452], "cos_innovation_negg": [0.6658576726913452], "cos_apical_negg": [0.6658576726913452], "cos_Ac_negg": [0.6658576726913452], "r_norm": [0.16685517132282257]}, {"step": 9300, "epoch": 23, "train_loss": 0.0009037284180521965, "cos_r_negg": [0.6654970049858093], "cos_innovation_negg": [0.6654970049858093], "cos_apical_negg": [0.6654970049858093], "cos_Ac_negg": [0.6654970049858093], "r_norm": [0.16875602304935455]}, {"step": 9350, "epoch": 23, "train_loss": 0.0011300718178972602, "cos_r_negg": [0.6664285063743591], "cos_innovation_negg": [0.6664285063743591], "cos_apical_negg": [0.6664285063743591], "cos_Ac_negg": [0.6664285063743591], "r_norm": [0.16760197281837463]}, {"epoch_end": 23, "step": 9384, "test_acc": 0.9763, "test_loss": 0.09372999019622803}, {"step": 9400, "epoch": 24, "train_loss": 0.0007356866262853146, "cos_r_negg": [0.6667789220809937], "cos_innovation_negg": [0.6667789220809937], "cos_apical_negg": [0.6667789220809937], "cos_Ac_negg": [0.6667789220809937], "r_norm": [0.16868163645267487]}, {"step": 9450, "epoch": 24, "train_loss": 0.0007303430465981364, "cos_r_negg": [0.6672382950782776], "cos_innovation_negg": [0.6672382950782776], "cos_apical_negg": [0.6672382950782776], "cos_Ac_negg": [0.6672382950782776], "r_norm": [0.16808977723121643]}, {"step": 9500, "epoch": 24, "train_loss": 0.0012442534789443016, "cos_r_negg": [0.6675103902816772], "cos_innovation_negg": [0.6675103902816772], "cos_apical_negg": [0.6675103902816772], "cos_Ac_negg": [0.6675103902816772], "r_norm": [0.16589809954166412]}, {"step": 9550, "epoch": 24, "train_loss": 0.0010358005529269576, "cos_r_negg": [0.667627215385437], "cos_innovation_negg": [0.667627215385437], "cos_apical_negg": [0.667627215385437], "cos_Ac_negg": [0.667627215385437], "r_norm": [0.16577830910682678]}, {"step": 9600, "epoch": 24, "train_loss": 0.0007997080683708191, "cos_r_negg": [0.6659653186798096], "cos_innovation_negg": [0.6659653186798096], "cos_apical_negg": [0.6659653186798096], "cos_Ac_negg": [0.6659653186798096], "r_norm": [0.16646955907344818]}, {"step": 9650, "epoch": 24, "train_loss": 0.0015334270428866148, "cos_r_negg": [0.6684226989746094], "cos_innovation_negg": [0.6684226989746094], "cos_apical_negg": [0.6684226989746094], "cos_Ac_negg": [0.6684226989746094], "r_norm": [0.1656925082206726]}, {"step": 9700, "epoch": 24, "train_loss": 0.0007564777042716742, "cos_r_negg": [0.6681210994720459], "cos_innovation_negg": [0.6681210994720459], "cos_apical_negg": [0.6681210994720459], "cos_Ac_negg": [0.6681210994720459], "r_norm": [0.16559502482414246]}, {"step": 9750, "epoch": 24, "train_loss": 0.0022185430862009525, "cos_r_negg": [0.6660330891609192], "cos_innovation_negg": [0.6660330891609192], "cos_apical_negg": [0.6660330891609192], "cos_Ac_negg": [0.6660330891609192], "r_norm": [0.16597822308540344]}, {"epoch_end": 24, "step": 9775, "test_acc": 0.9767, "test_loss": 0.09350334692001343}], "final": {"test_acc": 0.9767, "test_loss": 0.09350334692001343, "wall_s": 9.961609363555908, "cos_r_negg": [0.6665401458740234], "cos_innovation_negg": [0.6665401458740234], "cos_apical_negg": [0.6665401458740234], "cos_Ac_negg": [0.6665401458740234]}}