© 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-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 102513
    Prev
    Next
    INDEX
    Explanations

    properties, type, scale, idiom

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

    The neuron strongly responds to tokens that are part of quoted string literals (e.g. JSON keys or code include filenames).

    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
    providedIn
    -3.58
    To
    -2.98
    With
    -2.88
    We
    -2.73
    齮
    -2.69
    That
    -2.67
    After
    -2.63
    While
    -2.63
    Before
    -2.61
    Of
    -2.59
    POSITIVE LOGITS
    ↵
    2.95
     zejména
    2.58
     แต่
    2.47
     for
    2.45
    ৄ
    2.41
    ৢ
    2.39
     of
    2.36
     protože
    2.36
     $
    2.31
    ziu
    2.31
    Activations Density 0.085%

    No Known Activations