© 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-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 3907
    Prev
    Next
    INDEX
    Explanations

    fish with

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

    The neuron is a detector for action verbs—words naming or giving instructions for activities (e.g. install, purchase, list, listen, write, fly, cut, etc.).

    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_24_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
    Nursing
    0.66
    ouvoir
    0.65
    SERVATION
    0.64
    orschungs
    0.63
    ეგისტრ
    0.63
    㫴
    0.62
    Zach
    0.62
     내가
    0.61
    arqu
    0.61
     身
    0.61
    POSITIVE LOGITS
     faster
    0.81
     only
    0.79
     только
    0.75
     тільки
    0.75
     hanya
    0.74
     with
    0.73
     more
    0.72
     lebih
    0.72
     without
    0.70
     via
    0.70
    Activations Density 0.758%

    No Known Activations