© 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. 10-GEMMASCOPE-RES-131K
    7. 39723
    Prev
    Next
    INDEX
    Explanations

    code structure and commands

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

    The neuron flags programming-language and markup command tokens (e.g. function names, library or macro identifiers).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    0
    -2.03
     of
    -1.91
     /
    -1.71
    4
    -1.52
    2
    -1.45
    8
    -1.45
    3
    -1.41
     or
    -1.38
    //
    -1.34
     =
    -1.32
    POSITIVE LOGITS
     notori
    1.68
     literalmente
    1.55
     ところ
    1.52
     cozin
    1.48
     Vgl
    1.38
    lentes
    1.34
     imprimi
    1.29
    詎
    1.27
    -
    1.26
     plát
    1.24
    Activations Density 0.018%

    No Known Activations