© 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. 34-GEMMASCOPE-RES-131K
    7. 47509
    Prev
    Next
    INDEX
    Explanations

    you + modal verbs

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

    The neuron fires on second-person address terms—especially “you” (and related pronouns) directing speech at the reader.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     is
    -1.19
    odine
    -1.08
    cky
    -1.05
     frattempo
    -0.96
     WANTED
    -0.94
     usta
    -0.94
    şk
    -0.93
    beca
    -0.90
    ӧ
    -0.89
     lyk
    -0.88
    POSITIVE LOGITS
     should
    1.89
     will
    1.78
     can
    1.76
     need
    1.73
     may
    1.59
     would
    1.58
     must
    1.58
     might
    1.44
     shall
    1.37
     gotta
    1.29
    Activations Density 0.041%

    No Known Activations