© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 64915
    Prev
    Next
    INDEX
    Explanations

    variable assignment or member access

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron is keyed to capitalized identifiers or class‐style names (tokens beginning with an uppercase letter).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    事です
    -1.14
     Fußballspieler
    -1.14
    luğu
    -1.10
    tymologie
    -1.09
    色々な
    -1.08
    #%%
    -1.05
     desn
    -1.05
    楽しめます
    -1.04
    Weakness
    -1.04
    又は
    -1.04
    POSITIVE LOGITS
     will
    1.82
     =
    1.54
     cómics
    1.40
     must
    1.31
    _
    1.27
     should
    1.17
     still
    1.15
    providedIn
    1.14
     feitas
    1.14
     had
    1.09
    Activations Density 0.010%

    No Known Activations