© 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. 15763
    Prev
    Next
    INDEX
    Explanations

    stop or go away

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

    The neuron lights up on short, direct imperative commands (e.g. “stop it,” “go away,” “grow up,” “back off,” etc.).

    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
     to
    -1.16
     through
    -1.09
     but
    -1.02
     decided
    -1.02
     experience
    -0.98
     efektif
    -0.94
    kable
    -0.91
    lossene
    -0.90
     possible
    -0.90
     vetorial
    -0.90
    POSITIVE LOGITS
    !
    1.49
     сейчас
    1.23
    แล้ว
    1.21
     artık
    1.05
     simplement
    1.05
    ！
    1.01
     proprietary
    1.00
     bund
    0.99
     rồi
    0.98
    lrrr
    0.98
    Activations Density 0.014%

    No Known Activations