© 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. 34475
    Prev
    Next
    INDEX
    Explanations

    pronouns followed by articles or quantifiers

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

    The neuron fires on objective‐case pronouns (e.g. “you,” “him,” “me”).

    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
     you
    -2.00
    }$.
    -1.79
     it
    -1.61
    lında
    -1.58
     there
    -1.48
     οποία
    -1.48
     intentar
    -1.43
    unción
    -1.34
     tevreden
    -1.32
    潜力
    -1.30
    POSITIVE LOGITS
     a
    3.52
     something
    1.90
     the
    1.86
     an
    1.80
     some
    1.77
     another
    1.56
    注意事
    1.55
     óculos
    1.53
     pensamento
    1.47
     plenty
    1.46
    Activations Density 0.024%

    No Known Activations