© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 84272
    Prev
    Next
    INDEX
    Explanations

    simple actions like tap or click

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

    This neuron identifies promotional “call‐to‐action” and ease/speed language—words and numbers that emphasize quick, simple actions (e.g. just, easy, single, seconds, tap, click).

    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
     we
    -1.06
     eğer
    -1.02
     version
    -1.01
     for
    -0.98
     but
    -0.96
    čas
    -0.96
     manteiga
    -0.95
     toujours
    -0.88
     zase
    -0.88
     to
    -0.88
    POSITIVE LOGITS
     wenigen
    1.29
     metropolitana
    1.20
    wnątrz
    1.16
    无需
    1.15
     your
    1.12
    ziehungen
    1.09
     kwenye
    1.08
    就能
    1.07
     simply
    1.05
    只需
    1.05
    Activations Density 0.050%

    No Known Activations