1
|
{"args": {"mode": "dfa", "dataset": "mnist", "depth": 1, "width": 500, "act": "tanh", "residual": 0, "epochs": 25, "batch_size": 128, "train_examples": 50000, "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": 0.05, "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": 3, "device": "cuda", "log_every": 50, "max_steps": 0, "probe_bs": 512, "outdir": "results", "tag": "ep_archmatch_v1_mnist_dfa_w500_d1_s3", "n_in": 784, "n_out": 10}, "provenance": {"git_commit": "d4aef1a571c21e6e4e4125c90aa58da7f14e62e3", "git_dirty": false}, "steps": [{"step": 0, "epoch": 0, "train_loss": 2.442439317703247, "cos_r_negg": [0.008548112586140633], "cos_innovation_negg": [0.008548112586140633], "cos_apical_negg": [0.008548112586140633], "cos_Ac_negg": [0.008548112586140633], "r_norm": [6.77165412902832]}, {"step": 50, "epoch": 0, "train_loss": 0.5363419055938721, "cos_r_negg": [0.4376137852668762], "cos_innovation_negg": [0.4376137852668762], "cos_apical_negg": [0.4376137852668762], "cos_Ac_negg": [0.4376137852668762], "r_norm": [1.3919880390167236]}, {"step": 100, "epoch": 0, "train_loss": 0.1216859221458435, "cos_r_negg": [0.4438024163246155], "cos_innovation_negg": [0.4438024163246155], "cos_apical_negg": [0.4438024163246155], "cos_Ac_negg": [0.4438024163246155], "r_norm": [1.1661627292633057]}, {"step": 150, "epoch": 0, "train_loss": 0.1311318725347519, "cos_r_negg": [0.4501838684082031], "cos_innovation_negg": [0.4501838684082031], "cos_apical_negg": [0.4501838684082031], "cos_Ac_negg": [0.4501838684082031], "r_norm": [0.9205038547515869]}, {"step": 200, "epoch": 0, "train_loss": 0.23071861267089844, "cos_r_negg": [0.4427683651447296], "cos_innovation_negg": [0.4427683651447296], "cos_apical_negg": [0.4427683651447296], "cos_Ac_negg": [0.4427683651447296], "r_norm": [0.8930307626724243]}, {"step": 250, "epoch": 0, "train_loss": 0.3257734477519989, "cos_r_negg": [0.46152177453041077], "cos_innovation_negg": [0.46152177453041077], "cos_apical_negg": [0.46152177453041077], "cos_Ac_negg": [0.46152177453041077], "r_norm": [0.869844377040863]}, {"step": 300, "epoch": 0, "train_loss": 0.156560480594635, "cos_r_negg": [0.4561105966567993], "cos_innovation_negg": [0.4561105966567993], "cos_apical_negg": [0.4561105966567993], "cos_Ac_negg": [0.4561105966567993], "r_norm": [0.7620288133621216]}, {"step": 350, "epoch": 0, "train_loss": 0.23274347186088562, "cos_r_negg": [0.4531751871109009], "cos_innovation_negg": [0.4531751871109009], "cos_apical_negg": [0.4531751871109009], "cos_Ac_negg": [0.4531751871109009], "r_norm": [0.6907886266708374]}, {"epoch_end": 0, "step": 391, "test_acc": 0.9503, "test_loss": 0.16692773971557617}, {"step": 400, "epoch": 1, "train_loss": 0.13416726887226105, "cos_r_negg": [0.49407586455345154], "cos_innovation_negg": [0.49407586455345154], "cos_apical_negg": [0.49407586455345154], "cos_Ac_negg": [0.49407586455345154], "r_norm": [0.6247698664665222]}, {"step": 450, "epoch": 1, "train_loss": 0.13773851096630096, "cos_r_negg": [0.4886684715747833], "cos_innovation_negg": [0.4886684715747833], "cos_apical_negg": [0.4886684715747833], "cos_Ac_negg": [0.4886684715747833], "r_norm": [0.5645762085914612]}, {"step": 500, "epoch": 1, "train_loss": 0.20568495988845825, "cos_r_negg": [0.4984704554080963], "cos_innovation_negg": [0.4984704554080963], "cos_apical_negg": [0.4984704554080963], "cos_Ac_negg": [0.4984704554080963], "r_norm": [0.4723920524120331]}, {"step": 550, "epoch": 1, "train_loss": 0.08550918102264404, "cos_r_negg": [0.5000975728034973], "cos_innovation_negg": [0.5000975728034973], "cos_apical_negg": [0.5000975728034973], "cos_Ac_negg": [0.5000975728034973], "r_norm": [0.5834252834320068]}, {"step": 600, "epoch": 1, "train_loss": 0.07054577767848969, "cos_r_negg": [0.5248688459396362], "cos_innovation_negg": [0.5248688459396362], "cos_apical_negg": [0.5248688459396362], "cos_Ac_negg": [0.5248688459396362], "r_norm": [0.5585659146308899]}, {"step": 650, "epoch": 1, "train_loss": 0.10482070595026016, "cos_r_negg": [0.5035796761512756], "cos_innovation_negg": [0.5035796761512756], "cos_apical_negg": [0.5035796761512756], "cos_Ac_negg": [0.5035796761512756], "r_norm": [0.5744994878768921]}, {"step": 700, "epoch": 1, "train_loss": 0.15410007536411285, "cos_r_negg": [0.500011682510376], "cos_innovation_negg": [0.500011682510376], "cos_apical_negg": [0.500011682510376], "cos_Ac_negg": [0.500011682510376], "r_norm": [0.5410553216934204]}, {"step": 750, "epoch": 1, "train_loss": 0.15145228803157806, "cos_r_negg": [0.51210618019104], "cos_innovation_negg": [0.51210618019104], "cos_apical_negg": [0.51210618019104], "cos_Ac_negg": [0.51210618019104], "r_norm": [0.6504545211791992]}, {"epoch_end": 1, "step": 782, "test_acc": 0.9559, "test_loss": 0.14620965728759766}, {"step": 800, "epoch": 2, "train_loss": 0.10612640529870987, "cos_r_negg": [0.5219205021858215], "cos_innovation_negg": [0.5219205021858215], "cos_apical_negg": [0.5219205021858215], "cos_Ac_negg": [0.5219205021858215], "r_norm": [0.4923829734325409]}, {"step": 850, "epoch": 2, "train_loss": 0.06960354000329971, "cos_r_negg": [0.5385115146636963], "cos_innovation_negg": [0.5385115146636963], "cos_apical_negg": [0.5385115146636963], "cos_Ac_negg": [0.5385115146636963], "r_norm": [0.46809083223342896]}, {"step": 900, "epoch": 2, "train_loss": 0.07304694503545761, "cos_r_negg": [0.541561484336853], "cos_innovation_negg": [0.541561484336853], "cos_apical_negg": [0.541561484336853], "cos_Ac_negg": [0.541561484336853], "r_norm": [0.47019800543785095]}, {"step": 950, "epoch": 2, "train_loss": 0.08335687965154648, "cos_r_negg": [0.5382529497146606], "cos_innovation_negg": [0.5382529497146606], "cos_apical_negg": [0.5382529497146606], "cos_Ac_negg": [0.5382529497146606], "r_norm": [0.3802074193954468]}, {"step": 1000, "epoch": 2, "train_loss": 0.07608257234096527, "cos_r_negg": [0.5502356290817261], "cos_innovation_negg": [0.5502356290817261], "cos_apical_negg": [0.5502356290817261], "cos_Ac_negg": [0.5502356290817261], "r_norm": [0.44931650161743164]}, {"step": 1050, "epoch": 2, "train_loss": 0.06301358342170715, "cos_r_negg": [0.5453894138336182], "cos_innovation_negg": [0.5453894138336182], "cos_apical_negg": [0.5453894138336182], "cos_Ac_negg": [0.5453894138336182], "r_norm": [0.3921957314014435]}, {"step": 1100, "epoch": 2, "train_loss": 0.14986667037010193, "cos_r_negg": [0.5441368222236633], "cos_innovation_negg": [0.5441368222236633], "cos_apical_negg": [0.5441368222236633], "cos_Ac_negg": [0.5441368222236633], "r_norm": [0.35281893610954285]}, {"step": 1150, "epoch": 2, "train_loss": 0.0765186995267868, "cos_r_negg": [0.556448757648468], "cos_innovation_negg": [0.556448757648468], "cos_apical_negg": [0.556448757648468], "cos_Ac_negg": [0.556448757648468], "r_norm": [0.3307812213897705]}, {"epoch_end": 2, "step": 1173, "test_acc": 0.9688, "test_loss": 0.10577053813934326}, {"step": 1200, "epoch": 3, "train_loss": 0.06769527494907379, "cos_r_negg": [0.5597448348999023], "cos_innovation_negg": [0.5597448348999023], "cos_apical_negg": [0.5597448348999023], "cos_Ac_negg": [0.5597448348999023], "r_norm": [0.4475802481174469]}, {"step": 1250, "epoch": 3, "train_loss": 0.060645073652267456, "cos_r_negg": [0.5574368238449097], "cos_innovation_negg": [0.5574368238449097], "cos_apical_negg": [0.5574368238449097], "cos_Ac_negg": [0.5574368238449097], "r_norm": [0.36504653096199036]}, {"step": 1300, "epoch": 3, "train_loss": 0.010290195234119892, "cos_r_negg": [0.5699120759963989], "cos_innovation_negg": [0.5699120759963989], "cos_apical_negg": [0.5699120759963989], "cos_Ac_negg": [0.5699120759963989], "r_norm": [0.33620917797088623]}, {"step": 1350, "epoch": 3, "train_loss": 0.028610646724700928, "cos_r_negg": [0.5648489594459534], "cos_innovation_negg": [0.5648489594459534], "cos_apical_negg": [0.5648489594459534], "cos_Ac_negg": [0.5648489594459534], "r_norm": [0.32549506425857544]}, {"step": 1400, "epoch": 3, "train_loss": 0.03743978589773178, "cos_r_negg": [0.5771533846855164], "cos_innovation_negg": [0.5771533846855164], "cos_apical_negg": [0.5771533846855164], "cos_Ac_negg": [0.5771533846855164], "r_norm": [0.41055676341056824]}, {"step": 1450, "epoch": 3, "train_loss": 0.03383064270019531, "cos_r_negg": [0.5811811685562134], "cos_innovation_negg": [0.5811811685562134], "cos_apical_negg": [0.5811811685562134], "cos_Ac_negg": [0.5811811685562134], "r_norm": [0.4641897678375244]}, {"step": 1500, "epoch": 3, "train_loss": 0.09725669026374817, "cos_r_negg": [0.5892876982688904], "cos_innovation_negg": [0.5892876982688904], "cos_apical_negg": [0.5892876982688904], "cos_Ac_negg": [0.5892876982688904], "r_norm": [0.30729448795318604]}, {"step": 1550, "epoch": 3, "train_loss": 0.0516163595020771, "cos_r_negg": [0.5763197541236877], "cos_innovation_negg": [0.5763197541236877], "cos_apical_negg": [0.5763197541236877], "cos_Ac_negg": [0.5763197541236877], "r_norm": [0.3487250506877899]}, {"epoch_end": 3, "step": 1564, "test_acc": 0.9681, "test_loss": 0.1060694658279419}, {"step": 1600, "epoch": 4, "train_loss": 0.06377555429935455, "cos_r_negg": [0.5885501503944397], "cos_innovation_negg": [0.5885501503944397], "cos_apical_negg": [0.5885501503944397], "cos_Ac_negg": [0.5885501503944397], "r_norm": [0.2843015789985657]}, {"step": 1650, "epoch": 4, "train_loss": 0.07234866172075272, "cos_r_negg": [0.5844722986221313], "cos_innovation_negg": [0.5844722986221313], "cos_apical_negg": [0.5844722986221313], "cos_Ac_negg": [0.5844722986221313], "r_norm": [0.29238879680633545]}, {"step": 1700, "epoch": 4, "train_loss": 0.04979585111141205, "cos_r_negg": [0.5935534238815308], "cos_innovation_negg": [0.5935534238815308], "cos_apical_negg": [0.5935534238815308], "cos_Ac_negg": [0.5935534238815308], "r_norm": [0.3028966188430786]}, {"step": 1750, "epoch": 4, "train_loss": 0.029761599376797676, "cos_r_negg": [0.600528359413147], "cos_innovation_negg": [0.600528359413147], "cos_apical_negg": [0.600528359413147], "cos_Ac_negg": [0.600528359413147], "r_norm": [0.2917250990867615]}, {"step": 1800, "epoch": 4, "train_loss": 0.02162511833012104, "cos_r_negg": [0.5946288704872131], "cos_innovation_negg": [0.5946288704872131], "cos_apical_negg": [0.5946288704872131], "cos_Ac_negg": [0.5946288704872131], "r_norm": [0.3227265179157257]}, {"step": 1850, "epoch": 4, "train_loss": 0.08971917629241943, "cos_r_negg": [0.5956673622131348], "cos_innovation_negg": [0.5956673622131348], "cos_apical_negg": [0.5956673622131348], "cos_Ac_negg": [0.5956673622131348], "r_norm": [0.28730106353759766]}, {"step": 1900, "epoch": 4, "train_loss": 0.0074749416671693325, "cos_r_negg": [0.6076840162277222], "cos_innovation_negg": [0.6076840162277222], "cos_apical_negg": [0.6076840162277222], "cos_Ac_negg": [0.6076840162277222], "r_norm": [0.27146175503730774]}, {"step": 1950, "epoch": 4, "train_loss": 0.06698575615882874, "cos_r_negg": [0.6047966480255127], "cos_innovation_negg": [0.6047966480255127], "cos_apical_negg": [0.6047966480255127], "cos_Ac_negg": [0.6047966480255127], "r_norm": [0.2758903503417969]}, {"epoch_end": 4, "step": 1955, "test_acc": 0.9737, "test_loss": 0.0915428316116333}, {"step": 2000, "epoch": 5, "train_loss": 0.03715651109814644, "cos_r_negg": [0.6262306571006775], "cos_innovation_negg": [0.6262306571006775], "cos_apical_negg": [0.6262306571006775], "cos_Ac_negg": [0.6262306571006775], "r_norm": [0.3062130808830261]}, {"step": 2050, "epoch": 5, "train_loss": 0.01559838093817234, "cos_r_negg": [0.6118366122245789], "cos_innovation_negg": [0.6118366122245789], "cos_apical_negg": [0.6118366122245789], "cos_Ac_negg": [0.6118366122245789], "r_norm": [0.28920838236808777]}, {"step": 2100, "epoch": 5, "train_loss": 0.020354852080345154, "cos_r_negg": [0.6169546842575073], "cos_innovation_negg": [0.6169546842575073], "cos_apical_negg": [0.6169546842575073], "cos_Ac_negg": [0.6169546842575073], "r_norm": [0.2608243227005005]}, {"step": 2150, "epoch": 5, "train_loss": 0.005694162100553513, "cos_r_negg": [0.6161947250366211], "cos_innovation_negg": [0.6161947250366211], "cos_apical_negg": [0.6161947250366211], "cos_Ac_negg": [0.6161947250366211], "r_norm": [0.2769538164138794]}, {"step": 2200, "epoch": 5, "train_loss": 0.031961146742105484, "cos_r_negg": [0.6140713691711426], "cos_innovation_negg": [0.6140713691711426], "cos_apical_negg": [0.6140713691711426], "cos_Ac_negg": [0.6140713691711426], "r_norm": [0.24277204275131226]}, {"step": 2250, "epoch": 5, "train_loss": 0.01551815401762724, "cos_r_negg": [0.6204153299331665], "cos_innovation_negg": [0.6204153299331665], "cos_apical_negg": [0.6204153299331665], "cos_Ac_negg": [0.6204153299331665], "r_norm": [0.25323212146759033]}, {"step": 2300, "epoch": 5, "train_loss": 0.017334941774606705, "cos_r_negg": [0.618844211101532], "cos_innovation_negg": [0.618844211101532], "cos_apical_negg": [0.618844211101532], "cos_Ac_negg": [0.618844211101532], "r_norm": [0.24020928144454956]}, {"epoch_end": 5, "step": 2346, "test_acc": 0.9719, "test_loss": 0.0941448320388794}, {"step": 2350, "epoch": 6, "train_loss": 0.01014021784067154, "cos_r_negg": [0.6194796562194824], "cos_innovation_negg": [0.6194796562194824], "cos_apical_negg": [0.6194796562194824], "cos_Ac_negg": [0.6194796562194824], "r_norm": [0.23130422830581665]}, {"step": 2400, "epoch": 6, "train_loss": 0.008165987208485603, "cos_r_negg": [0.6260712146759033], "cos_innovation_negg": [0.6260712146759033], "cos_apical_negg": [0.6260712146759033], "cos_Ac_negg": [0.6260712146759033], "r_norm": [0.23445481061935425]}, {"step": 2450, "epoch": 6, "train_loss": 0.024763470515608788, "cos_r_negg": [0.6184272766113281], "cos_innovation_negg": [0.6184272766113281], "cos_apical_negg": [0.6184272766113281], "cos_Ac_negg": [0.6184272766113281], "r_norm": [0.23025111854076385]}, {"step": 2500, "epoch": 6, "train_loss": 0.009604115039110184, "cos_r_negg": [0.6274337768554688], "cos_innovation_negg": [0.6274337768554688], "cos_apical_negg": [0.6274337768554688], "cos_Ac_negg": [0.6274337768554688], "r_norm": [0.21369673311710358]}, {"step": 2550, "epoch": 6, "train_loss": 0.02136237546801567, "cos_r_negg": [0.6231203079223633], "cos_innovation_negg": [0.6231203079223633], "cos_apical_negg": [0.6231203079223633], "cos_Ac_negg": [0.6231203079223633], "r_norm": [0.22462376952171326]}, {"step": 2600, "epoch": 6, "train_loss": 0.016291538253426552, "cos_r_negg": [0.6236748695373535], "cos_innovation_negg": [0.6236748695373535], "cos_apical_negg": [0.6236748695373535], "cos_Ac_negg": [0.6236748695373535], "r_norm": [0.24315427243709564]}, {"step": 2650, "epoch": 6, "train_loss": 0.02006765455007553, "cos_r_negg": [0.6304926872253418], "cos_innovation_negg": [0.6304926872253418], "cos_apical_negg": [0.6304926872253418], "cos_Ac_negg": [0.6304926872253418], "r_norm": [0.21973562240600586]}, {"step": 2700, "epoch": 6, "train_loss": 0.007136902771890163, "cos_r_negg": [0.6280251741409302], "cos_innovation_negg": [0.6280251741409302], "cos_apical_negg": [0.6280251741409302], "cos_Ac_negg": [0.6280251741409302], "r_norm": [0.25129562616348267]}, {"epoch_end": 6, "step": 2737, "test_acc": 0.9751, "test_loss": 0.08822272167205811}, {"step": 2750, "epoch": 7, "train_loss": 0.0030623339116573334, "cos_r_negg": [0.6366771459579468], "cos_innovation_negg": [0.6366771459579468], "cos_apical_negg": [0.6366771459579468], "cos_Ac_negg": [0.6366771459579468], "r_norm": [0.22319546341896057]}, {"step": 2800, "epoch": 7, "train_loss": 0.010890500620007515, "cos_r_negg": [0.6421500444412231], "cos_innovation_negg": [0.6421500444412231], "cos_apical_negg": [0.6421500444412231], "cos_Ac_negg": [0.6421500444412231], "r_norm": [0.20071640610694885]}, {"step": 2850, "epoch": 7, "train_loss": 0.007570589892566204, "cos_r_negg": [0.6394366025924683], "cos_innovation_negg": [0.6394366025924683], "cos_apical_negg": [0.6394366025924683], "cos_Ac_negg": [0.6394366025924683], "r_norm": [0.2162337303161621]}, {"step": 2900, "epoch": 7, "train_loss": 0.003252351190894842, "cos_r_negg": [0.6404790282249451], "cos_innovation_negg": [0.6404790282249451], "cos_apical_negg": [0.6404790282249451], "cos_Ac_negg": [0.6404790282249451], "r_norm": [0.20290347933769226]}, {"step": 2950, "epoch": 7, "train_loss": 0.004013213329017162, "cos_r_negg": [0.6417880058288574], "cos_innovation_negg": [0.6417880058288574], "cos_apical_negg": [0.6417880058288574], "cos_Ac_negg": [0.6417880058288574], "r_norm": [0.20410795509815216]}, {"step": 3000, "epoch": 7, "train_loss": 0.00501081719994545, "cos_r_negg": [0.6377766728401184], "cos_innovation_negg": [0.6377766728401184], "cos_apical_negg": [0.6377766728401184], "cos_Ac_negg": [0.6377766728401184], "r_norm": [0.1983664631843567]}, {"step": 3050, "epoch": 7, "train_loss": 0.004919731058180332, "cos_r_negg": [0.6433807611465454], "cos_innovation_negg": [0.6433807611465454], "cos_apical_negg": [0.6433807611465454], "cos_Ac_negg": [0.6433807611465454], "r_norm": [0.1903722882270813]}, {"step": 3100, "epoch": 7, "train_loss": 0.0022773887030780315, "cos_r_negg": [0.6466061472892761], "cos_innovation_negg": [0.6466061472892761], "cos_apical_negg": [0.6466061472892761], "cos_Ac_negg": [0.6466061472892761], "r_norm": [0.19517894089221954]}, {"epoch_end": 7, "step": 3128, "test_acc": 0.9745, "test_loss": 0.08891878757476807}, {"step": 3150, "epoch": 8, "train_loss": 0.004336378071457148, "cos_r_negg": [0.6444277763366699], "cos_innovation_negg": [0.6444277763366699], "cos_apical_negg": [0.6444277763366699], "cos_Ac_negg": [0.6444277763366699], "r_norm": [0.20238903164863586]}, {"step": 3200, "epoch": 8, "train_loss": 0.010351760312914848, "cos_r_negg": [0.6447922587394714], "cos_innovation_negg": [0.6447922587394714], "cos_apical_negg": [0.6447922587394714], "cos_Ac_negg": [0.6447922587394714], "r_norm": [0.20848368108272552]}, {"step": 3250, "epoch": 8, "train_loss": 0.0028710688930004835, "cos_r_negg": [0.6454180479049683], "cos_innovation_negg": [0.6454180479049683], "cos_apical_negg": [0.6454180479049683], "cos_Ac_negg": [0.6454180479049683], "r_norm": [0.21021220088005066]}, {"step": 3300, "epoch": 8, "train_loss": 0.004988059401512146, "cos_r_negg": [0.6489178538322449], "cos_innovation_negg": [0.6489178538322449], "cos_apical_negg": [0.6489178538322449], "cos_Ac_negg": [0.6489178538322449], "r_norm": [0.19697944819927216]}, {"step": 3350, "epoch": 8, "train_loss": 0.012388825416564941, "cos_r_negg": [0.6492448449134827], "cos_innovation_negg": [0.6492448449134827], "cos_apical_negg": [0.6492448449134827], "cos_Ac_negg": [0.6492448449134827], "r_norm": [0.19435536861419678]}, {"step": 3400, "epoch": 8, "train_loss": 0.003946593962609768, "cos_r_negg": [0.6433913707733154], "cos_innovation_negg": [0.6433913707733154], "cos_apical_negg": [0.6433913707733154], "cos_Ac_negg": [0.6433913707733154], "r_norm": [0.21129077672958374]}, {"step": 3450, "epoch": 8, "train_loss": 0.0032899288926273584, "cos_r_negg": [0.647679328918457], "cos_innovation_negg": [0.647679328918457], "cos_apical_negg": [0.647679328918457], "cos_Ac_negg": [0.647679328918457], "r_norm": [0.19783155620098114]}, {"step": 3500, "epoch": 8, "train_loss": 0.010790111497044563, "cos_r_negg": [0.650292158126831], "cos_innovation_negg": [0.650292158126831], "cos_apical_negg": [0.650292158126831], "cos_Ac_negg": [0.650292158126831], "r_norm": [0.19501425325870514]}, {"epoch_end": 8, "step": 3519, "test_acc": 0.9764, "test_loss": 0.08624833240509033}, {"step": 3550, "epoch": 9, "train_loss": 0.00415037153288722, "cos_r_negg": [0.6443274021148682], "cos_innovation_negg": [0.6443274021148682], "cos_apical_negg": [0.6443274021148682], "cos_Ac_negg": [0.6443274021148682], "r_norm": [0.19695767760276794]}, {"step": 3600, "epoch": 9, "train_loss": 0.002783270087093115, "cos_r_negg": [0.6474475860595703], "cos_innovation_negg": [0.6474475860595703], "cos_apical_negg": [0.6474475860595703], "cos_Ac_negg": [0.6474475860595703], "r_norm": [0.18732833862304688]}, {"step": 3650, "epoch": 9, "train_loss": 0.004318370949476957, "cos_r_negg": [0.6503322124481201], "cos_innovation_negg": [0.6503322124481201], "cos_apical_negg": [0.6503322124481201], "cos_Ac_negg": [0.6503322124481201], "r_norm": [0.18886592984199524]}, {"step": 3700, "epoch": 9, "train_loss": 0.007296190597116947, "cos_r_negg": [0.6492711901664734], "cos_innovation_negg": [0.6492711901664734], "cos_apical_negg": [0.6492711901664734], "cos_Ac_negg": [0.6492711901664734], "r_norm": [0.1957516372203827]}, {"step": 3750, "epoch": 9, "train_loss": 0.005999542772769928, "cos_r_negg": [0.6541937589645386], "cos_innovation_negg": [0.6541937589645386], "cos_apical_negg": [0.6541937589645386], "cos_Ac_negg": [0.6541937589645386], "r_norm": [0.19850140810012817]}, {"step": 3800, "epoch": 9, "train_loss": 0.0019062224309891462, "cos_r_negg": [0.6500388383865356], "cos_innovation_negg": [0.6500388383865356], "cos_apical_negg": [0.6500388383865356], "cos_Ac_negg": [0.6500388383865356], "r_norm": [0.18814346194267273]}, {"step": 3850, "epoch": 9, "train_loss": 0.006979479920119047, "cos_r_negg": [0.65274977684021], "cos_innovation_negg": [0.65274977684021], "cos_apical_negg": [0.65274977684021], "cos_Ac_negg": [0.65274977684021], "r_norm": [0.20159852504730225]}, {"step": 3900, "epoch": 9, "train_loss": 0.017795085906982422, "cos_r_negg": [0.65329509973526], "cos_innovation_negg": [0.65329509973526], "cos_apical_negg": [0.65329509973526], "cos_Ac_negg": [0.65329509973526], "r_norm": [0.18815459311008453]}, {"epoch_end": 9, "step": 3910, "test_acc": 0.9764, "test_loss": 0.08635039100646973}, {"step": 3950, "epoch": 10, "train_loss": 0.004156068898737431, "cos_r_negg": [0.6538388133049011], "cos_innovation_negg": [0.6538388133049011], "cos_apical_negg": [0.6538388133049011], "cos_Ac_negg": [0.6538388133049011], "r_norm": [0.21263061463832855]}, {"step": 4000, "epoch": 10, "train_loss": 0.0023026023991405964, "cos_r_negg": [0.6548007130622864], "cos_innovation_negg": [0.6548007130622864], "cos_apical_negg": [0.6548007130622864], "cos_Ac_negg": [0.6548007130622864], "r_norm": [0.1913701593875885]}, {"step": 4050, "epoch": 10, "train_loss": 0.0015156329609453678, "cos_r_negg": [0.6540789604187012], "cos_innovation_negg": [0.6540789604187012], "cos_apical_negg": [0.6540789604187012], "cos_Ac_negg": [0.6540789604187012], "r_norm": [0.19870451092720032]}, {"step": 4100, "epoch": 10, "train_loss": 0.003098537679761648, "cos_r_negg": [0.6531257629394531], "cos_innovation_negg": [0.6531257629394531], "cos_apical_negg": [0.6531257629394531], "cos_Ac_negg": [0.6531257629394531], "r_norm": [0.2000044584274292]}, {"step": 4150, "epoch": 10, "train_loss": 0.0019474627915769815, "cos_r_negg": [0.6530778408050537], "cos_innovation_negg": [0.6530778408050537], "cos_apical_negg": [0.6530778408050537], "cos_Ac_negg": [0.6530778408050537], "r_norm": [0.18807224929332733]}, {"step": 4200, "epoch": 10, "train_loss": 0.004173523746430874, "cos_r_negg": [0.6563830375671387], "cos_innovation_negg": [0.6563830375671387], "cos_apical_negg": [0.6563830375671387], "cos_Ac_negg": [0.6563830375671387], "r_norm": [0.19230778515338898]}, {"step": 4250, "epoch": 10, "train_loss": 0.005850169807672501, "cos_r_negg": [0.6605209708213806], "cos_innovation_negg": [0.6605209708213806], "cos_apical_negg": [0.6605209708213806], "cos_Ac_negg": [0.6605209708213806], "r_norm": [0.1888607293367386]}, {"step": 4300, "epoch": 10, "train_loss": 0.0018562866607680917, "cos_r_negg": [0.6599082350730896], "cos_innovation_negg": [0.6599082350730896], "cos_apical_negg": [0.6599082350730896], "cos_Ac_negg": [0.6599082350730896], "r_norm": [0.19036468863487244]}, {"epoch_end": 10, "step": 4301, "test_acc": 0.9755, "test_loss": 0.0883064962387085}, {"step": 4350, "epoch": 11, "train_loss": 0.003945862874388695, "cos_r_negg": [0.6616725921630859], "cos_innovation_negg": [0.6616725921630859], "cos_apical_negg": [0.6616725921630859], "cos_Ac_negg": [0.6616725921630859], "r_norm": [0.19148670136928558]}, {"step": 4400, "epoch": 11, "train_loss": 0.00252861762419343, "cos_r_negg": [0.6590904593467712], "cos_innovation_negg": [0.6590904593467712], "cos_apical_negg": [0.6590904593467712], "cos_Ac_negg": [0.6590904593467712], "r_norm": [0.18406416475772858]}, {"step": 4450, "epoch": 11, "train_loss": 0.008483943529427052, "cos_r_negg": [0.662162721157074], "cos_innovation_negg": [0.662162721157074], "cos_apical_negg": [0.662162721157074], "cos_Ac_negg": [0.662162721157074], "r_norm": [0.1862124353647232]}, {"step": 4500, "epoch": 11, "train_loss": 0.0004408641834743321, "cos_r_negg": [0.6604998111724854], "cos_innovation_negg": [0.6604998111724854], "cos_apical_negg": [0.6604998111724854], "cos_Ac_negg": [0.6604998111724854], "r_norm": [0.19248756766319275]}, {"step": 4550, "epoch": 11, "train_loss": 0.005936149507761002, "cos_r_negg": [0.6583750247955322], "cos_innovation_negg": [0.6583750247955322], "cos_apical_negg": [0.6583750247955322], "cos_Ac_negg": [0.6583750247955322], "r_norm": [0.18951238691806793]}, {"step": 4600, "epoch": 11, "train_loss": 0.0011925961589440703, "cos_r_negg": [0.6599512100219727], "cos_innovation_negg": [0.6599512100219727], "cos_apical_negg": [0.6599512100219727], "cos_Ac_negg": [0.6599512100219727], "r_norm": [0.1931493729352951]}, {"step": 4650, "epoch": 11, "train_loss": 0.0035241572186350822, "cos_r_negg": [0.6626706123352051], "cos_innovation_negg": [0.6626706123352051], "cos_apical_negg": [0.6626706123352051], "cos_Ac_negg": [0.6626706123352051], "r_norm": [0.19371065497398376]}, {"epoch_end": 11, "step": 4692, "test_acc": 0.9767, "test_loss": 0.08754708909988404}, {"step": 4700, "epoch": 12, "train_loss": 0.0032634595409035683, "cos_r_negg": [0.6621760725975037], "cos_innovation_negg": [0.6621760725975037], "cos_apical_negg": [0.6621760725975037], "cos_Ac_negg": [0.6621760725975037], "r_norm": [0.19149097800254822]}, {"step": 4750, "epoch": 12, "train_loss": 0.0025233339983969927, "cos_r_negg": [0.6616387963294983], "cos_innovation_negg": [0.6616387963294983], "cos_apical_negg": [0.6616387963294983], "cos_Ac_negg": [0.6616387963294983], "r_norm": [0.1847679167985916]}, {"step": 4800, "epoch": 12, "train_loss": 0.0017325032968074083, "cos_r_negg": [0.6621156930923462], "cos_innovation_negg": [0.6621156930923462], "cos_apical_negg": [0.6621156930923462], "cos_Ac_negg": [0.6621156930923462], "r_norm": [0.19214680790901184]}, {"step": 4850, "epoch": 12, "train_loss": 0.0015416943933814764, "cos_r_negg": [0.6622645258903503], "cos_innovation_negg": [0.6622645258903503], "cos_apical_negg": [0.6622645258903503], "cos_Ac_negg": [0.6622645258903503], "r_norm": [0.1910938024520874]}, {"step": 4900, "epoch": 12, "train_loss": 0.002132368739694357, "cos_r_negg": [0.6633182764053345], "cos_innovation_negg": [0.6633182764053345], "cos_apical_negg": [0.6633182764053345], "cos_Ac_negg": [0.6633182764053345], "r_norm": [0.188226580619812]}, {"step": 4950, "epoch": 12, "train_loss": 0.0016303692245855927, "cos_r_negg": [0.6629526615142822], "cos_innovation_negg": [0.6629526615142822], "cos_apical_negg": [0.6629526615142822], "cos_Ac_negg": [0.6629526615142822], "r_norm": [0.1891656219959259]}, {"step": 5000, "epoch": 12, "train_loss": 0.001205473905429244, "cos_r_negg": [0.6645966172218323], "cos_innovation_negg": [0.6645966172218323], "cos_apical_negg": [0.6645966172218323], "cos_Ac_negg": [0.6645966172218323], "r_norm": [0.19087006151676178]}, {"step": 5050, "epoch": 12, "train_loss": 0.0020330441184341908, "cos_r_negg": [0.6632612943649292], "cos_innovation_negg": [0.6632612943649292], "cos_apical_negg": [0.6632612943649292], "cos_Ac_negg": [0.6632612943649292], "r_norm": [0.19048205018043518]}, {"epoch_end": 12, "step": 5083, "test_acc": 0.9765, "test_loss": 0.0897450662612915}, {"step": 5100, "epoch": 13, "train_loss": 0.0015689498977735639, "cos_r_negg": [0.6632356643676758], "cos_innovation_negg": [0.6632356643676758], "cos_apical_negg": [0.6632356643676758], "cos_Ac_negg": [0.6632356643676758], "r_norm": [0.19515526294708252]}, {"step": 5150, "epoch": 13, "train_loss": 0.0032305759377777576, "cos_r_negg": [0.6647845506668091], "cos_innovation_negg": [0.6647845506668091], "cos_apical_negg": [0.6647845506668091], "cos_Ac_negg": [0.6647845506668091], "r_norm": [0.19155773520469666]}, {"step": 5200, "epoch": 13, "train_loss": 0.0021498212590813637, "cos_r_negg": [0.6653106808662415], "cos_innovation_negg": [0.6653106808662415], "cos_apical_negg": [0.6653106808662415], "cos_Ac_negg": [0.6653106808662415], "r_norm": [0.1925402730703354]}, {"step": 5250, "epoch": 13, "train_loss": 0.0013341844314709306, "cos_r_negg": [0.6644632816314697], "cos_innovation_negg": [0.6644632816314697], "cos_apical_negg": [0.6644632816314697], "cos_Ac_negg": [0.6644632816314697], "r_norm": [0.1913897544145584]}, {"step": 5300, "epoch": 13, "train_loss": 0.00297095300629735, "cos_r_negg": [0.6649411916732788], "cos_innovation_negg": [0.6649411916732788], "cos_apical_negg": [0.6649411916732788], "cos_Ac_negg": [0.6649411916732788], "r_norm": [0.18792349100112915]}, {"step": 5350, "epoch": 13, "train_loss": 0.0038473668973892927, "cos_r_negg": [0.664700984954834], "cos_innovation_negg": [0.664700984954834], "cos_apical_negg": [0.664700984954834], "cos_Ac_negg": [0.664700984954834], "r_norm": [0.19175377488136292]}, {"step": 5400, "epoch": 13, "train_loss": 0.0031616524793207645, "cos_r_negg": [0.6671086549758911], "cos_innovation_negg": [0.6671086549758911], "cos_apical_negg": [0.6671086549758911], "cos_Ac_negg": [0.6671086549758911], "r_norm": [0.1898212432861328]}, {"step": 5450, "epoch": 13, "train_loss": 0.0024428064934909344, "cos_r_negg": [0.6663897037506104], "cos_innovation_negg": [0.6663897037506104], "cos_apical_negg": [0.6663897037506104], "cos_Ac_negg": [0.6663897037506104], "r_norm": [0.18423236906528473]}, {"epoch_end": 13, "step": 5474, "test_acc": 0.9761, "test_loss": 0.09059316177368164}, {"step": 5500, "epoch": 14, "train_loss": 0.0011226912029087543, "cos_r_negg": [0.6657710075378418], "cos_innovation_negg": [0.6657710075378418], "cos_apical_negg": [0.6657710075378418], "cos_Ac_negg": [0.6657710075378418], "r_norm": [0.19469454884529114]}, {"step": 5550, "epoch": 14, "train_loss": 0.0013064880622550845, "cos_r_negg": [0.6653034687042236], "cos_innovation_negg": [0.6653034687042236], "cos_apical_negg": [0.6653034687042236], "cos_Ac_negg": [0.6653034687042236], "r_norm": [0.1871453821659088]}, {"step": 5600, "epoch": 14, "train_loss": 0.002314056269824505, "cos_r_negg": [0.6679279804229736], "cos_innovation_negg": [0.6679279804229736], "cos_apical_negg": [0.6679279804229736], "cos_Ac_negg": [0.6679279804229736], "r_norm": [0.1868903934955597]}, {"step": 5650, "epoch": 14, "train_loss": 0.0010740846628323197, "cos_r_negg": [0.6664503812789917], "cos_innovation_negg": [0.6664503812789917], "cos_apical_negg": [0.6664503812789917], "cos_Ac_negg": [0.6664503812789917], "r_norm": [0.18747372925281525]}, {"step": 5700, "epoch": 14, "train_loss": 0.0006839900161139667, "cos_r_negg": [0.6664535999298096], "cos_innovation_negg": [0.6664535999298096], "cos_apical_negg": [0.6664535999298096], "cos_Ac_negg": [0.6664535999298096], "r_norm": [0.19066667556762695]}, {"step": 5750, "epoch": 14, "train_loss": 0.0017767411191016436, "cos_r_negg": [0.6681963205337524], "cos_innovation_negg": [0.6681963205337524], "cos_apical_negg": [0.6681963205337524], "cos_Ac_negg": [0.6681963205337524], "r_norm": [0.18777838349342346]}, {"step": 5800, "epoch": 14, "train_loss": 0.004349137656390667, "cos_r_negg": [0.66842120885849], "cos_innovation_negg": [0.66842120885849], "cos_apical_negg": [0.66842120885849], "cos_Ac_negg": [0.66842120885849], "r_norm": [0.18292754888534546]}, {"step": 5850, "epoch": 14, "train_loss": 0.0020366176031529903, "cos_r_negg": [0.667387843132019], "cos_innovation_negg": [0.667387843132019], "cos_apical_negg": [0.667387843132019], "cos_Ac_negg": [0.667387843132019], "r_norm": [0.18801191449165344]}, {"epoch_end": 14, "step": 5865, "test_acc": 0.9767, "test_loss": 0.08959818105697631}, {"step": 5900, "epoch": 15, "train_loss": 0.0015641085337847471, "cos_r_negg": [0.6683471202850342], "cos_innovation_negg": [0.6683471202850342], "cos_apical_negg": [0.6683471202850342], "cos_Ac_negg": [0.6683471202850342], "r_norm": [0.18830448389053345]}, {"step": 5950, "epoch": 15, "train_loss": 0.0017050523310899734, "cos_r_negg": [0.6678061485290527], "cos_innovation_negg": [0.6678061485290527], "cos_apical_negg": [0.6678061485290527], "cos_Ac_negg": [0.6678061485290527], "r_norm": [0.18552029132843018]}, {"step": 6000, "epoch": 15, "train_loss": 0.0030467016622424126, "cos_r_negg": [0.6681082844734192], "cos_innovation_negg": [0.6681082844734192], "cos_apical_negg": [0.6681082844734192], "cos_Ac_negg": [0.6681082844734192], "r_norm": [0.18875452876091003]}, {"step": 6050, "epoch": 15, "train_loss": 0.0028290008194744587, "cos_r_negg": [0.6685616970062256], "cos_innovation_negg": [0.6685616970062256], "cos_apical_negg": [0.6685616970062256], "cos_Ac_negg": [0.6685616970062256], "r_norm": [0.18562611937522888]}, {"step": 6100, "epoch": 15, "train_loss": 0.0019021242624148726, "cos_r_negg": [0.6682847738265991], "cos_innovation_negg": [0.6682847738265991], "cos_apical_negg": [0.6682847738265991], "cos_Ac_negg": [0.6682847738265991], "r_norm": [0.18805864453315735]}, {"step": 6150, "epoch": 15, "train_loss": 0.0016579189104959369, "cos_r_negg": [0.6697932481765747], "cos_innovation_negg": [0.6697932481765747], "cos_apical_negg": [0.6697932481765747], "cos_Ac_negg": [0.6697932481765747], "r_norm": [0.1929086148738861]}, {"step": 6200, "epoch": 15, "train_loss": 0.0026102494448423386, "cos_r_negg": [0.6687422394752502], "cos_innovation_negg": [0.6687422394752502], "cos_apical_negg": [0.6687422394752502], "cos_Ac_negg": [0.6687422394752502], "r_norm": [0.1908755600452423]}, {"step": 6250, "epoch": 15, "train_loss": 0.0015464629977941513, "cos_r_negg": [0.670972466468811], "cos_innovation_negg": [0.670972466468811], "cos_apical_negg": [0.670972466468811], "cos_Ac_negg": [0.670972466468811], "r_norm": [0.18564239144325256]}, {"epoch_end": 15, "step": 6256, "test_acc": 0.9765, "test_loss": 0.08959217720031738}, {"step": 6300, "epoch": 16, "train_loss": 0.003177294274792075, "cos_r_negg": [0.6713321208953857], "cos_innovation_negg": [0.6713321208953857], "cos_apical_negg": [0.6713321208953857], "cos_Ac_negg": [0.6713321208953857], "r_norm": [0.1880650520324707]}, {"step": 6350, "epoch": 16, "train_loss": 0.0009205355308949947, "cos_r_negg": [0.6703627109527588], "cos_innovation_negg": [0.6703627109527588], "cos_apical_negg": [0.6703627109527588], "cos_Ac_negg": [0.6703627109527588], "r_norm": [0.18405726552009583]}, {"step": 6400, "epoch": 16, "train_loss": 0.0018161828629672527, "cos_r_negg": [0.6709251403808594], "cos_innovation_negg": [0.6709251403808594], "cos_apical_negg": [0.6709251403808594], "cos_Ac_negg": [0.6709251403808594], "r_norm": [0.18353164196014404]}, {"step": 6450, "epoch": 16, "train_loss": 0.0012991197872906923, "cos_r_negg": [0.6692332625389099], "cos_innovation_negg": [0.6692332625389099], "cos_apical_negg": [0.6692332625389099], "cos_Ac_negg": [0.6692332625389099], "r_norm": [0.1804175078868866]}, {"step": 6500, "epoch": 16, "train_loss": 0.001308799022808671, "cos_r_negg": [0.6709599494934082], "cos_innovation_negg": [0.6709599494934082], "cos_apical_negg": [0.6709599494934082], "cos_Ac_negg": [0.6709599494934082], "r_norm": [0.18312883377075195]}, {"step": 6550, "epoch": 16, "train_loss": 0.0017033701296895742, "cos_r_negg": [0.6704203486442566], "cos_innovation_negg": [0.6704203486442566], "cos_apical_negg": [0.6704203486442566], "cos_Ac_negg": [0.6704203486442566], "r_norm": [0.1911018043756485]}, {"step": 6600, "epoch": 16, "train_loss": 0.0012333118356764317, "cos_r_negg": [0.6712208986282349], "cos_innovation_negg": [0.6712208986282349], "cos_apical_negg": [0.6712208986282349], "cos_Ac_negg": [0.6712208986282349], "r_norm": [0.1841079592704773]}, {"epoch_end": 16, "step": 6647, "test_acc": 0.9762, "test_loss": 0.09051880359649658}, {"step": 6650, "epoch": 17, "train_loss": 0.0011065974831581116, "cos_r_negg": [0.6707524061203003], "cos_innovation_negg": [0.6707524061203003], "cos_apical_negg": [0.6707524061203003], "cos_Ac_negg": [0.6707524061203003], "r_norm": [0.190361887216568]}, {"step": 6700, "epoch": 17, "train_loss": 0.0010079825296998024, "cos_r_negg": [0.6701521873474121], "cos_innovation_negg": [0.6701521873474121], "cos_apical_negg": [0.6701521873474121], "cos_Ac_negg": [0.6701521873474121], "r_norm": [0.183818057179451]}, {"step": 6750, "epoch": 17, "train_loss": 0.0021076167467981577, "cos_r_negg": [0.6722524166107178], "cos_innovation_negg": [0.6722524166107178], "cos_apical_negg": [0.6722524166107178], "cos_Ac_negg": [0.6722524166107178], "r_norm": [0.18282531201839447]}, {"step": 6800, "epoch": 17, "train_loss": 0.0020436591003090143, "cos_r_negg": [0.6717022657394409], "cos_innovation_negg": [0.6717022657394409], "cos_apical_negg": [0.6717022657394409], "cos_Ac_negg": [0.6717022657394409], "r_norm": [0.18591180443763733]}, {"step": 6850, "epoch": 17, "train_loss": 0.0009509872179478407, "cos_r_negg": [0.6710649132728577], "cos_innovation_negg": [0.6710649132728577], "cos_apical_negg": [0.6710649132728577], "cos_Ac_negg": [0.6710649132728577], "r_norm": [0.18384113907814026]}, {"step": 6900, "epoch": 17, "train_loss": 0.0016628298908472061, "cos_r_negg": [0.6716196537017822], "cos_innovation_negg": [0.6716196537017822], "cos_apical_negg": [0.6716196537017822], "cos_Ac_negg": [0.6716196537017822], "r_norm": [0.18868836760520935]}, {"step": 6950, "epoch": 17, "train_loss": 0.0024520400911569595, "cos_r_negg": [0.6711359024047852], "cos_innovation_negg": [0.6711359024047852], "cos_apical_negg": [0.6711359024047852], "cos_Ac_negg": [0.6711359024047852], "r_norm": [0.19121608138084412]}, {"step": 7000, "epoch": 17, "train_loss": 0.001846964005380869, "cos_r_negg": [0.6719766855239868], "cos_innovation_negg": [0.6719766855239868], "cos_apical_negg": [0.6719766855239868], "cos_Ac_negg": [0.6719766855239868], "r_norm": [0.18608233332633972]}, {"epoch_end": 17, "step": 7038, "test_acc": 0.9765, "test_loss": 0.09158344593048096}, {"step": 7050, "epoch": 18, "train_loss": 0.0006504197372123599, "cos_r_negg": [0.6722675561904907], "cos_innovation_negg": [0.6722675561904907], "cos_apical_negg": [0.6722675561904907], "cos_Ac_negg": [0.6722675561904907], "r_norm": [0.18477511405944824]}, {"step": 7100, "epoch": 18, "train_loss": 0.001657924149185419, "cos_r_negg": [0.6710620522499084], "cos_innovation_negg": [0.6710620522499084], "cos_apical_negg": [0.6710620522499084], "cos_Ac_negg": [0.6710620522499084], "r_norm": [0.18965701758861542]}, {"step": 7150, "epoch": 18, "train_loss": 0.0005512891802936792, "cos_r_negg": [0.6717802286148071], "cos_innovation_negg": [0.6717802286148071], "cos_apical_negg": [0.6717802286148071], "cos_Ac_negg": [0.6717802286148071], "r_norm": [0.1871088296175003]}, {"step": 7200, "epoch": 18, "train_loss": 0.0014886315912008286, "cos_r_negg": [0.6731422543525696], "cos_innovation_negg": [0.6731422543525696], "cos_apical_negg": [0.6731422543525696], "cos_Ac_negg": [0.6731422543525696], "r_norm": [0.18091577291488647]}, {"step": 7250, "epoch": 18, "train_loss": 0.001611289451830089, "cos_r_negg": [0.6723629236221313], "cos_innovation_negg": [0.6723629236221313], "cos_apical_negg": [0.6723629236221313], "cos_Ac_negg": [0.6723629236221313], "r_norm": [0.1808919608592987]}, {"step": 7300, "epoch": 18, "train_loss": 0.0005207016947679222, "cos_r_negg": [0.6733918786048889], "cos_innovation_negg": [0.6733918786048889], "cos_apical_negg": [0.6733918786048889], "cos_Ac_negg": [0.6733918786048889], "r_norm": [0.1826336681842804]}, {"step": 7350, "epoch": 18, "train_loss": 0.001265919883735478, "cos_r_negg": [0.6740373373031616], "cos_innovation_negg": [0.6740373373031616], "cos_apical_negg": [0.6740373373031616], "cos_Ac_negg": [0.6740373373031616], "r_norm": [0.1859230101108551]}, {"step": 7400, "epoch": 18, "train_loss": 0.0005655413842760026, "cos_r_negg": [0.6732701063156128], "cos_innovation_negg": [0.6732701063156128], "cos_apical_negg": [0.6732701063156128], "cos_Ac_negg": [0.6732701063156128], "r_norm": [0.18533606827259064]}, {"epoch_end": 18, "step": 7429, "test_acc": 0.9761, "test_loss": 0.09127200088500977}, {"step": 7450, "epoch": 19, "train_loss": 0.001061121467500925, "cos_r_negg": [0.67383873462677], "cos_innovation_negg": [0.67383873462677], "cos_apical_negg": [0.67383873462677], "cos_Ac_negg": [0.67383873462677], "r_norm": [0.1806277185678482]}, {"step": 7500, "epoch": 19, "train_loss": 0.0009110849350690842, "cos_r_negg": [0.6734578609466553], "cos_innovation_negg": [0.6734578609466553], "cos_apical_negg": [0.6734578609466553], "cos_Ac_negg": [0.6734578609466553], "r_norm": [0.18546360731124878]}, {"step": 7550, "epoch": 19, "train_loss": 0.000802901922725141, "cos_r_negg": [0.6739841103553772], "cos_innovation_negg": [0.6739841103553772], "cos_apical_negg": [0.6739841103553772], "cos_Ac_negg": [0.6739841103553772], "r_norm": [0.1846744269132614]}, {"step": 7600, "epoch": 19, "train_loss": 0.0012915183324366808, "cos_r_negg": [0.6741548776626587], "cos_innovation_negg": [0.6741548776626587], "cos_apical_negg": [0.6741548776626587], "cos_Ac_negg": [0.6741548776626587], "r_norm": [0.18603235483169556]}, {"step": 7650, "epoch": 19, "train_loss": 0.001798536628484726, "cos_r_negg": [0.6733959317207336], "cos_innovation_negg": [0.6733959317207336], "cos_apical_negg": [0.6733959317207336], "cos_Ac_negg": [0.6733959317207336], "r_norm": [0.18958942592144012]}, {"step": 7700, "epoch": 19, "train_loss": 0.002252830658107996, "cos_r_negg": [0.672349214553833], "cos_innovation_negg": [0.672349214553833], "cos_apical_negg": [0.672349214553833], "cos_Ac_negg": [0.672349214553833], "r_norm": [0.178838849067688]}, {"step": 7750, "epoch": 19, "train_loss": 0.0019049308029934764, "cos_r_negg": [0.6738132238388062], "cos_innovation_negg": [0.6738132238388062], "cos_apical_negg": [0.6738132238388062], "cos_Ac_negg": [0.6738132238388062], "r_norm": [0.18273049592971802]}, {"step": 7800, "epoch": 19, "train_loss": 0.0024760360829532146, "cos_r_negg": [0.6734486818313599], "cos_innovation_negg": [0.6734486818313599], "cos_apical_negg": [0.6734486818313599], "cos_Ac_negg": [0.6734486818313599], "r_norm": [0.18798071146011353]}, {"epoch_end": 19, "step": 7820, "test_acc": 0.9763, "test_loss": 0.09233485288619996}, {"step": 7850, "epoch": 20, "train_loss": 0.0017160282004624605, "cos_r_negg": [0.6750516891479492], "cos_innovation_negg": [0.6750516891479492], "cos_apical_negg": [0.6750516891479492], "cos_Ac_negg": [0.6750516891479492], "r_norm": [0.1821345090866089]}, {"step": 7900, "epoch": 20, "train_loss": 0.0012108697555959225, "cos_r_negg": [0.6732663512229919], "cos_innovation_negg": [0.6732663512229919], "cos_apical_negg": [0.6732663512229919], "cos_Ac_negg": [0.6732663512229919], "r_norm": [0.18891817331314087]}, {"step": 7950, "epoch": 20, "train_loss": 0.0010040879715234041, "cos_r_negg": [0.6739406585693359], "cos_innovation_negg": [0.6739406585693359], "cos_apical_negg": [0.6739406585693359], "cos_Ac_negg": [0.6739406585693359], "r_norm": [0.18470434844493866]}, {"step": 8000, "epoch": 20, "train_loss": 0.002193755004554987, "cos_r_negg": [0.673803985118866], "cos_innovation_negg": [0.673803985118866], "cos_apical_negg": [0.673803985118866], "cos_Ac_negg": [0.673803985118866], "r_norm": [0.18491744995117188]}, {"step": 8050, "epoch": 20, "train_loss": 0.002146694343537092, "cos_r_negg": [0.6729025840759277], "cos_innovation_negg": [0.6729025840759277], "cos_apical_negg": [0.6729025840759277], "cos_Ac_negg": [0.6729025840759277], "r_norm": [0.18356776237487793]}, {"step": 8100, "epoch": 20, "train_loss": 0.0015185822267085314, "cos_r_negg": [0.6743298172950745], "cos_innovation_negg": [0.6743298172950745], "cos_apical_negg": [0.6743298172950745], "cos_Ac_negg": [0.6743298172950745], "r_norm": [0.1841581165790558]}, {"step": 8150, "epoch": 20, "train_loss": 0.0009880242869257927, "cos_r_negg": [0.674461841583252], "cos_innovation_negg": [0.674461841583252], "cos_apical_negg": [0.674461841583252], "cos_Ac_negg": [0.674461841583252], "r_norm": [0.18146878480911255]}, {"step": 8200, "epoch": 20, "train_loss": 0.0006425578612834215, "cos_r_negg": [0.6744028329849243], "cos_innovation_negg": [0.6744028329849243], "cos_apical_negg": [0.6744028329849243], "cos_Ac_negg": [0.6744028329849243], "r_norm": [0.1845739632844925]}, {"epoch_end": 20, "step": 8211, "test_acc": 0.9765, "test_loss": 0.0917292818069458}, {"step": 8250, "epoch": 21, "train_loss": 0.0016986222472041845, "cos_r_negg": [0.6758262515068054], "cos_innovation_negg": [0.6758262515068054], "cos_apical_negg": [0.6758262515068054], "cos_Ac_negg": [0.6758262515068054], "r_norm": [0.18208380043506622]}, {"step": 8300, "epoch": 21, "train_loss": 0.0013625193387269974, "cos_r_negg": [0.6755895614624023], "cos_innovation_negg": [0.6755895614624023], "cos_apical_negg": [0.6755895614624023], "cos_Ac_negg": [0.6755895614624023], "r_norm": [0.17714960873126984]}, {"step": 8350, "epoch": 21, "train_loss": 0.001160505460575223, "cos_r_negg": [0.6755126714706421], "cos_innovation_negg": [0.6755126714706421], "cos_apical_negg": [0.6755126714706421], "cos_Ac_negg": [0.6755126714706421], "r_norm": [0.17934376001358032]}, {"step": 8400, "epoch": 21, "train_loss": 0.0012507388601079583, "cos_r_negg": [0.6752291321754456], "cos_innovation_negg": [0.6752291321754456], "cos_apical_negg": [0.6752291321754456], "cos_Ac_negg": [0.6752291321754456], "r_norm": [0.17855310440063477]}, {"step": 8450, "epoch": 21, "train_loss": 0.0014778455952182412, "cos_r_negg": [0.674967885017395], "cos_innovation_negg": [0.674967885017395], "cos_apical_negg": [0.674967885017395], "cos_Ac_negg": [0.674967885017395], "r_norm": [0.17693959176540375]}, {"step": 8500, "epoch": 21, "train_loss": 0.0010109890718013048, "cos_r_negg": [0.6755441427230835], "cos_innovation_negg": [0.6755441427230835], "cos_apical_negg": [0.6755441427230835], "cos_Ac_negg": [0.6755441427230835], "r_norm": [0.18128079175949097]}, {"step": 8550, "epoch": 21, "train_loss": 0.0006054211989976466, "cos_r_negg": [0.6765226721763611], "cos_innovation_negg": [0.6765226721763611], "cos_apical_negg": [0.6765226721763611], "cos_Ac_negg": [0.6765226721763611], "r_norm": [0.18221010267734528]}, {"step": 8600, "epoch": 21, "train_loss": 0.0033144697081297636, "cos_r_negg": [0.6759686470031738], "cos_innovation_negg": [0.6759686470031738], "cos_apical_negg": [0.6759686470031738], "cos_Ac_negg": [0.6759686470031738], "r_norm": [0.1840457171201706]}, {"epoch_end": 21, "step": 8602, "test_acc": 0.9768, "test_loss": 0.09318850193023681}, {"step": 8650, "epoch": 22, "train_loss": 0.0007331828819587827, "cos_r_negg": [0.675412654876709], "cos_innovation_negg": [0.675412654876709], "cos_apical_negg": [0.675412654876709], "cos_Ac_negg": [0.675412654876709], "r_norm": [0.1873997300863266]}, {"step": 8700, "epoch": 22, "train_loss": 0.0011605904437601566, "cos_r_negg": [0.6757599711418152], "cos_innovation_negg": [0.6757599711418152], "cos_apical_negg": [0.6757599711418152], "cos_Ac_negg": [0.6757599711418152], "r_norm": [0.18548086285591125]}, {"step": 8750, "epoch": 22, "train_loss": 0.0007738434360362589, "cos_r_negg": [0.67568439245224], "cos_innovation_negg": [0.67568439245224], "cos_apical_negg": [0.67568439245224], "cos_Ac_negg": [0.67568439245224], "r_norm": [0.1830318123102188]}, {"step": 8800, "epoch": 22, "train_loss": 0.0008718061726540327, "cos_r_negg": [0.6765251755714417], "cos_innovation_negg": [0.6765251755714417], "cos_apical_negg": [0.6765251755714417], "cos_Ac_negg": [0.6765251755714417], "r_norm": [0.18194422125816345]}, {"step": 8850, "epoch": 22, "train_loss": 0.0010430978145450354, "cos_r_negg": [0.6770457029342651], "cos_innovation_negg": [0.6770457029342651], "cos_apical_negg": [0.6770457029342651], "cos_Ac_negg": [0.6770457029342651], "r_norm": [0.18583367764949799]}, {"step": 8900, "epoch": 22, "train_loss": 0.002213382860645652, "cos_r_negg": [0.676719069480896], "cos_innovation_negg": [0.676719069480896], "cos_apical_negg": [0.676719069480896], "cos_Ac_negg": [0.676719069480896], "r_norm": [0.18495997786521912]}, {"step": 8950, "epoch": 22, "train_loss": 0.0010738542769104242, "cos_r_negg": [0.6768510341644287], "cos_innovation_negg": [0.6768510341644287], "cos_apical_negg": [0.6768510341644287], "cos_Ac_negg": [0.6768510341644287], "r_norm": [0.18300531804561615]}, {"epoch_end": 22, "step": 8993, "test_acc": 0.9769, "test_loss": 0.09264851064682007}, {"step": 9000, "epoch": 23, "train_loss": 0.000845407834276557, "cos_r_negg": [0.6777102947235107], "cos_innovation_negg": [0.6777102947235107], "cos_apical_negg": [0.6777102947235107], "cos_Ac_negg": [0.6777102947235107], "r_norm": [0.18136712908744812]}, {"step": 9050, "epoch": 23, "train_loss": 0.001263210317119956, "cos_r_negg": [0.6765074133872986], "cos_innovation_negg": [0.6765074133872986], "cos_apical_negg": [0.6765074133872986], "cos_Ac_negg": [0.6765074133872986], "r_norm": [0.18011383712291718]}, {"step": 9100, "epoch": 23, "train_loss": 0.0006898348219692707, "cos_r_negg": [0.6774665713310242], "cos_innovation_negg": [0.6774665713310242], "cos_apical_negg": [0.6774665713310242], "cos_Ac_negg": [0.6774665713310242], "r_norm": [0.1836041510105133]}, {"step": 9150, "epoch": 23, "train_loss": 0.0016093901358544827, "cos_r_negg": [0.6768931746482849], "cos_innovation_negg": [0.6768931746482849], "cos_apical_negg": [0.6768931746482849], "cos_Ac_negg": [0.6768931746482849], "r_norm": [0.18405351042747498]}, {"step": 9200, "epoch": 23, "train_loss": 0.0021341536194086075, "cos_r_negg": [0.6770271062850952], "cos_innovation_negg": [0.6770271062850952], "cos_apical_negg": [0.6770271062850952], "cos_Ac_negg": [0.6770271062850952], "r_norm": [0.17932364344596863]}, {"step": 9250, "epoch": 23, "train_loss": 0.0009358518291264772, "cos_r_negg": [0.6771856546401978], "cos_innovation_negg": [0.6771856546401978], "cos_apical_negg": [0.6771856546401978], "cos_Ac_negg": [0.6771856546401978], "r_norm": [0.18240931630134583]}, {"step": 9300, "epoch": 23, "train_loss": 0.0009725701529532671, "cos_r_negg": [0.6766215562820435], "cos_innovation_negg": [0.6766215562820435], "cos_apical_negg": [0.6766215562820435], "cos_Ac_negg": [0.6766215562820435], "r_norm": [0.17748074233531952]}, {"step": 9350, "epoch": 23, "train_loss": 0.0006379790720529854, "cos_r_negg": [0.6773027181625366], "cos_innovation_negg": [0.6773027181625366], "cos_apical_negg": [0.6773027181625366], "cos_Ac_negg": [0.6773027181625366], "r_norm": [0.18186412751674652]}, {"epoch_end": 23, "step": 9384, "test_acc": 0.9767, "test_loss": 0.09364033155441284}, {"step": 9400, "epoch": 24, "train_loss": 0.0009541834006085992, "cos_r_negg": [0.6765420436859131], "cos_innovation_negg": [0.6765420436859131], "cos_apical_negg": [0.6765420436859131], "cos_Ac_negg": [0.6765420436859131], "r_norm": [0.1849023997783661]}, {"step": 9450, "epoch": 24, "train_loss": 0.0004028431139886379, "cos_r_negg": [0.677538275718689], "cos_innovation_negg": [0.677538275718689], "cos_apical_negg": [0.677538275718689], "cos_Ac_negg": [0.677538275718689], "r_norm": [0.18501731753349304]}, {"step": 9500, "epoch": 24, "train_loss": 0.0011584380408748984, "cos_r_negg": [0.6779544353485107], "cos_innovation_negg": [0.6779544353485107], "cos_apical_negg": [0.6779544353485107], "cos_Ac_negg": [0.6779544353485107], "r_norm": [0.18292036652565002]}, {"step": 9550, "epoch": 24, "train_loss": 0.0008456854848191142, "cos_r_negg": [0.6771849393844604], "cos_innovation_negg": [0.6771849393844604], "cos_apical_negg": [0.6771849393844604], "cos_Ac_negg": [0.6771849393844604], "r_norm": [0.1810213178396225]}, {"step": 9600, "epoch": 24, "train_loss": 0.000599990482442081, "cos_r_negg": [0.6773977279663086], "cos_innovation_negg": [0.6773977279663086], "cos_apical_negg": [0.6773977279663086], "cos_Ac_negg": [0.6773977279663086], "r_norm": [0.17862775921821594]}, {"step": 9650, "epoch": 24, "train_loss": 0.0014905506977811456, "cos_r_negg": [0.678149402141571], "cos_innovation_negg": [0.678149402141571], "cos_apical_negg": [0.678149402141571], "cos_Ac_negg": [0.678149402141571], "r_norm": [0.17745369672775269]}, {"step": 9700, "epoch": 24, "train_loss": 0.0007170581957325339, "cos_r_negg": [0.6775041818618774], "cos_innovation_negg": [0.6775041818618774], "cos_apical_negg": [0.6775041818618774], "cos_Ac_negg": [0.6775041818618774], "r_norm": [0.18083257973194122]}, {"step": 9750, "epoch": 24, "train_loss": 0.0020477729849517345, "cos_r_negg": [0.6772384643554688], "cos_innovation_negg": [0.6772384643554688], "cos_apical_negg": [0.6772384643554688], "cos_Ac_negg": [0.6772384643554688], "r_norm": [0.18001912534236908]}, {"epoch_end": 24, "step": 9775, "test_acc": 0.9767, "test_loss": 0.09284619626998901}], "final": {"test_acc": 0.9767, "test_loss": 0.09284619626998901, "wall_s": 13.517838716506958, "cos_r_negg": [0.6772804260253906], "cos_innovation_negg": [0.6772804260253906], "cos_apical_negg": [0.6772804260253906], "cos_Ac_negg": [0.6772804260253906]}}
|