© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 10019
    Prev
    Next
    INDEX
    Explanations

    action or actions

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

    This neuron responds strongly to imperative “call‐to‐action” phrases—especially instances of “take action,” “take X,” or similar direct commands urging the reader to act.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    pyridin
    0.55
    imarães
    0.54
     वर्कशीट
    0.53
    თვის
    0.52
    bete
    0.52
    👩
    0.52
     verir
    0.50
    즘
    0.50
    مپ
    0.50
    Ữ
    0.50
    POSITIVE LOGITS
     action
    2.69
     actions
    2.39
    action
    2.33
     Action
    2.31
     acción
    2.28
     Actions
    2.27
    アクション
    2.25
     ACTION
    2.23
    行动
    2.22
    Action
    2.19
    Activations Density 0.382%

    No Known Activations