© 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. 26244
    Prev
    Next
    INDEX
    Explanations

    questions about correctness

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

    This neuron detects confirmation-seeking question phrasing, e.g. “Is that correct?”, “Did I understand this correctly?”

    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
     Gerät
    -0.96
    salty
    -0.91
    Открыть
    -0.91
    Teb
    -0.88
    ısında
    -0.87
    inspiration
    -0.87
     Anpass
    -0.85
     slant
    -0.84
     apareció
    -0.84
    相关文章
    -0.84
    POSITIVE LOGITS
     correct
    1.54
    correct
    1.33
    Correct
    1.03
    ?
    1.03
     correcto
    0.99
    ใช่
    0.96
     correcta
    0.96
     then
    0.94
     Correct
    0.93
    それで
    0.92
    Activations Density 0.122%

    No Known Activations