© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 36058
    Prev
    Next
    INDEX
    Explanations

    virtual single, chemical character

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

    line breaks or formatting elements in text, particularly when marking paragraph separations.

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

    The neuron is a high‐precision detector for the informal discourse marker “BTW.”

    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
     its
    -1.77
     this
    -1.61
     it
    -1.54
     for
    -1.33
     But
    -1.30
     stolet
    -1.20
    懸命
    -1.20
    ETIES
    -1.20
    綜合報導
    -1.20
     sejarah
    -1.18
    POSITIVE LOGITS
    '
    1.49
    ’
    1.47
     وليس
    1.39
    !
    1.25
     bekom
    1.23
    “
    1.22
    <b>
    1.22
     olup
    1.20
     simplifié
    1.20
    ‘
    1.19
    Activations Density 0.000%

    No Known Activations