© 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. 8312
    Prev
    Next
    INDEX
    Explanations

    crime or victim of

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

    This neuron fires on occurrences of the preposition “of” introducing noun‐of‐noun phrases (e.g. “robberies of motorists,” “murder of members”).

    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
    没有了
    -1.61
     tilbud
    -1.56
    獁
    -1.50
     offers
    -1.49
     what
    -1.48
     スタッフ
    -1.46
     aktivitet
    -1.39
    ramenta
    -1.39
     bidra
    -1.35
     available
    -1.34
    POSITIVE LOGITS
    那一
    1.64
     by
    1.64
    you
    1.63
    those
    1.62
    Those
    1.55
    :
    1.54
    YOU
    1.52
    }
    1.49
    ]$.
    1.47
     us
    1.45
    Activations Density 0.119%

    No Known Activations