© 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. 128191
    Prev
    Next
    INDEX
    Explanations

    The neuron is a strong detector for the word “to” (as in the infinitive marker “to” + verb).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    to + verb

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     unangemessene
    -1.95
    沨
    -1.89
     unangemess
    -1.88
    鳊
    -1.76
    ጧ
    -1.73
    iſter
    -1.71
    èges
    -1.71
    ኗ
    -1.70
    酲
    -1.69
    𓄹
    -1.66
    POSITIVE LOGITS
     be
    2.36
     to
    2.06
     know
    1.59
    t
    1.50
    ra
    1.42
    ↵
    1.38
     '
    1.38
    /*!
    1.36
     ''
    1.35
    n
    1.32
    Activations Density 0.037%

    No Known Activations