© 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. 67215
    Prev
    Next
    INDEX
    Explanations

    knowing you are not alone

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

    The neuron detects personal-pronoun phrases (I, you, they, we) often together with auxiliaries or modals (can, should, don’t, had, etc.) used in self-referential or affirmational statements.

    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
     of
    -3.23
     or
    -2.53
     new
    -1.95
     on
    -1.94
     our
    -1.70
     so
    -1.66
     using
    -1.64
     attuale
    -1.64
     because
    -1.63
     including
    -1.58
    POSITIVE LOGITS
     tremend
    1.73
     anormal
    1.70
    voer
    1.69
     entusi
    1.66
     intrig
    1.63
     ilumin
    1.63
    quilo
    1.62
     alarme
    1.61
     instin
    1.59
     indifer
    1.59
    Activations Density 0.045%

    No Known Activations