© 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. 67346
    Prev
    Next
    INDEX
    Explanations

    descriptive phrases or actions

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

    The neuron is strongly sensitive to common function words—especially determiners (the), subordinators/relativizers (which), infinitival to, auxiliary be, pronouns (you), 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_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
     esimerkiksi
    -1.61
    seven
    -1.60
    ités
    -1.59
    gheny
    -1.58
     empêcher
    -1.57
     você
    -1.54
    five
    -1.53
     améli
    -1.51
     AMAZING
    -1.50
     noved
    -1.49
    POSITIVE LOGITS
     this
    3.02
     to
    2.69
     and
    2.69
     that
    2.64
     which
    2.56
     with
    2.36
     where
    2.25
     it
    2.16
     these
    2.02
     what
    1.79
    Activations Density 0.455%

    No Known Activations