© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-3-27B-IT
    3. 15-GEMMASCOPE-2-TRANSCODER-262K
    4. 198766
    Prev
    Next
    INDEX
    Explanations

    The provided lists are:<MAX_ACTIVATING_TOKENS>:ifALguysprintpleasesamountsIF*,unlock<TOKENS_AFTER_MAX_ACTIVATING_TOKEN>:xPs.text].:forsythe.<TOP_POSITIVE_LOGITS>:tolueneIUnaryახებNxgमाटरHauntedsuddenBLUENRGCHDpopping<TOP_ACTIVATING_TEXTS>: like this: ```latex % Code environment \DefineEnvironment{Code}{ \expandafter\ifx\csname verbatim@noligatures\endcsname\relax \expandafter\let\csname verbatim * **Dark Matter:** WIMPs are losing favor. Current focus: * **Axions/ALPs:** Strongly motivated, many experiments (ADMX, HAYSTAC, etc.). * **Weak to a childhood accident, but it was rarely the focus of jokes – the show generally treated him as just one of the guys. the lines following the first one in the bib environment \defbibenvironment{bibliography} {\list {\printtext[labelnumberwidth]{% \printfield{labelprefix}% \printfield{labelnumber flow with graceful [ease], A quiet joy, a victory hard [won], A digital creation meant to [please]. So let us code, though fraught with toil and [pain], For in this craft, true beauty wesatz über die Rolle von IBM

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/transcoder_all/layer_15_width_262k_l0_small_affine
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     и
    0.63
     आहे
    0.62
     en
    0.61
     і
    0.59
    os
    0.58
     engag
    0.58
     или
    0.57
     connue
    0.57
    ية
    0.56
     ө
    0.55
    POSITIVE LOGITS
    toluene
    0.50
     IUnary
    0.50
    ახებ
    0.49
     Nxg
    0.48
    माटर
    0.47
     Haunted
    0.47
     sudden
    0.46
     BLUENRG
    0.46
    CHD
    0.46
     popping
    0.46
    Activations Density 0.000%

    No Known Activations