© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 27986
    Prev
    Next
    INDEX
    Explanations

    presence or falsehood

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

    The neuron fires on first-person self-references and introspective language (e.g. “I,” “myself,” “am,” “convincing myself,” expressions of personal feeling).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
     Coutinho
    -0.77
    ďaka
    -0.73
    им
    -0.72
     "'");
    -0.72
    letting
    -0.70
    lp
    -0.70
     Compact
    -0.69
    omas
    -0.69
    Detached
    -0.69
     Eddie
    -0.68
    POSITIVE LOGITS
     when
    1.05
     something
    0.96
     nonexistent
    0.95
    YOND
    0.90
     inexist
    0.86
     which
    0.84
    false
    0.84
    يسة
    0.84
     false
    0.82
     jornadas
    0.82
    Activations Density 0.048%

    No Known Activations