© 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. 73241
    Prev
    Next
    INDEX
    Explanations

    code or version separators

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

    The neuron fires on long runs or clusters of underscore characters (e.g. ASCII‐art or decorative underlines).

    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
     it
    -1.82
     from
    -1.59
     in
    -1.54
     by
    -1.54
     an
    -1.52
     both
    -1.50
     so
    -1.50
     with
    -1.48
     like
    -1.44
     one
    -1.41
    POSITIVE LOGITS
     Dinas
    1.46
     strikingly
    1.26
     dilapidated
    1.25
     horrid
    1.24
     deliciously
    1.23
     wondrous
    1.23
     áng
    1.23
    ඍ
    1.23
    蹀
    1.22
    athered
    1.19
    Activations Density 0.004%

    No Known Activations