© 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. 34-GEMMASCOPE-RES-131K
    7. 93072
    Prev
    Next
    INDEX
    Explanations

    Echocardiographic

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

    version numbers in software dependencies and package specifications.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron primarily activates on numeric version tokens (e.g. digits and dot‐delimited version numbers).

    oai_token-act-pair · o4-miniTriggered 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
    Not in Any Lists

    No Comments

    Negative Logits
    ならば
    -0.93
    兲
    -0.89
     ALLEN
    -0.88
    respectively
    -0.88
    之际
    -0.87
     vort
    -0.84
    bringer
    -0.83
    🚮
    -0.82
     pierres
    -0.82
    のでしょうか
    -0.81
    POSITIVE LOGITS
     but
    0.95
     Always
    0.93
     aunque
    0.91
     debido
    0.90
    Mã
    0.88
    ||
    0.87
     kuitenkin
    0.86
    啻
    0.86
    とにかく
    0.86
     siempre
    0.86
    Activations Density 0.021%

    No Known Activations