© 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. 113844
    Prev
    Next
    INDEX
    Explanations

    in behalf or by favor

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

    This neuron fires on formal “of”-phrases—especially idiomatic prepositional constructs like “on behalf of,” “in the wake of,” or “in favour of.”

    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
     you
    -1.80
    \}.
    -1.74
     siku
    -1.64
     wilayah
    -1.56
    łaś
    -1.55
     texinfo
    -1.55
     COURSES
    -1.52
    >`;
    -1.52
     americana
    -1.50
    '';
    -1.50
    POSITIVE LOGITS
    4
    2.30
    5
    2.16
    8
    2.09
    </h2>
    1.99
    7
    1.91
    9
    1.84
    this
    1.84
    6
    1.76
    3
    1.73
    很久
    1.70
    Activations Density 0.034%

    No Known Activations