© 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. 43951
    Prev
    Next
    INDEX
    Explanations

    names and titles

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

    The neuron detects news‐story bylines or reporter‐credit headers (e.g. “NPR’s Eleanor Beardsley, byline:”).

    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
     if
    -2.20
     also
    -1.88
     even
    -1.71
     for
    -1.67
     some
    -1.66
     that
    -1.60
    ụp
    -1.59
     there
    -1.57
     because
    -1.57
     although
    -1.55
    POSITIVE LOGITS
     Dinas
    1.90
    撺
    1.84
     vorige
    1.81
    ఽ
    1.73
    
    1.73
    
    1.69
     zul
    1.68
    ⟣
    1.67
     heilig
    1.66
     ENTIRE
    1.66
    Activations Density 0.008%

    No Known Activations