© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 127736
    Prev
    Next
    INDEX
    Explanations

    The neuron fires strongly on terms naming or referring to data‐container sequences (like “array,” “list,” “sequence,” etc.) that are being manipulated or sorted.

    oai_token-act-pair · o4-mini
    Triggered by @jyhe0408

    sorting and ordering

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    Negative Logits
     mög
    -1.07
     kön
    -0.94
    köp
    -0.94
    här
    -0.94
     during
    -0.92
    väl
    -0.91
    новременно
    -0.90
    garanti
    -0.88
     necessary
    -0.88
     permitido
    -0.87
    POSITIVE LOGITS
    rentina
    0.91
    cited
    0.90
    J
    0.87
     according
    0.85
    taminophen
    0.85
    ән
    0.84
     különböző
    0.84
    D
    0.82
    因为它
    0.82
    according
    0.81
    Activations Density 0.019%

    No Known Activations