© 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. 34-GEMMASCOPE-RES-131K
    7. 106166
    Prev
    Next
    INDEX
    Explanations

    URL fragments and alphanumeric sequences

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

    The neuron flags long, dense alphanumeric or symbol‐heavy runs that look like encoded data blobs (e.g. Base64 segments, URL percent‐escapes, or “\u” Unicode escapes).

    oai_token-act-pair · o4-mini
    Triggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
    Rouge
    -1.09
    ÉD
    -0.95
     japones
    -0.95
     guitarra
    -0.93
     andra
    -0.91
     hukuman
    -0.90
    дется
    -0.90
     comandante
    -0.90
     kabar
    -0.90
     vokal
    -0.89
    POSITIVE LOGITS
     inmueble
    0.97
     подоб
    0.94
     legte
    0.93
    is
    0.91
     complejos
    0.90
    }^{*
    0.90
    you
    0.89
     establecimiento
    0.88
    んですよね
    0.88
     注
    0.88
    Activations Density 0.003%

    No Known Activations