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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 105270
    Prev
    Next
    INDEX
    Explanations

    statements of personal action

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

    The neuron fires on coordinated first‐person subject phrases of the form “X and I,” e.g. “my wife and I,” “my husband and I,” “my bf and I.”

    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
     or
    -1.66
     for
    -1.63
     to
    -1.58
     and
    -1.52
     what
    -1.45
    怜
    -1.41
    了不少
    -1.40
    ;
    -1.39
     penting
    -1.38
    executeUpdate
    -1.37
    POSITIVE LOGITS
     purview
    1.78
     飾り
    1.46
    ケモン
    1.46
     sentiers
    1.45
    Hôtel
    1.43
    IZZA
    1.41
     poignée
    1.39
    escalier
    1.38
    possibly
    1.37
     campagnes
    1.36
    Activations Density 0.005%

    No Known Activations