© 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. 10-GEMMASCOPE-RES-131K
    7. 84360
    Prev
    Next
    INDEX
    Explanations

    lines of asterisks

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

    The neuron activates almost exclusively on long runs of asterisks, i.e. decorative “*****…” separator lines.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    妧
    -2.17
    呿
    -2.16
    dengan
    -2.09
     you
    -2.09
     castigo
    -2.00
    躂
    -1.97
    鼍
    -1.95
    錞
    -1.92
     alivio
    -1.88
    ৄ
    -1.86
    POSITIVE LOGITS
    .
    3.19
    </strong>
    2.70
    7
    2.53
    l
    2.52
    g
    2.42
    4
    2.41
    x
    2.39
    …”
    2.38
    k
    2.38
    us
    2.28
    Activations Density 0.015%

    No Known Activations