© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 69955
    Prev
    Next
    INDEX
    Explanations

    guests

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

    The neuron consistently activates on the token “guests,” i.e. references to people being hosted.

    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
    ”
    -2.70
    }$
    -2.64
    </strong>
    -2.61
    “
    -2.61
    </h4>
    -2.36
    It
    -2.28
    <em>
    -2.25
    」
    -2.25
    ”？
    -2.20
    𓐍
    -2.19
    POSITIVE LOGITS
    饪
    2.89
    gucig
    2.64
    理由は
    2.63
    ࿙
    2.58
    
    2.55
    簠
    2.55
     guests
    2.41
    铼
    2.39
    guests
    2.38
     lidí
    2.36
    Activations Density 0.004%

    No Known Activations