© 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-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 11879
    Prev
    Next
    INDEX
    Explanations

    executing functions or operations

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

    The neuron strongly activates on contentful action verbs (especially longer ones) throughout the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    verbs describing actions of providing, giving, or supplying something to someone.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    common action verbs indicating someone performing or receiving an action, often in second-person constructions (e.g., “you” doing something).

    oai_token-act-pair · gpt-5Triggered 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
    ！)
    0.85
    Enough
    0.78
     somew
    0.77
    หลาย
    0.76
    Glue
    0.75
    कैसे
    0.73
     ጉ
    0.73
    कई
    0.73
    EXPECT
    0.73
    ANTE
    0.72
    POSITIVE LOGITS
     encompasses
    0.77
     supremac
    0.75
     represents
    0.75
     had
    0.74
     are
    0.74
     will
    0.73
    പ്പെടുന്ന
    0.73
     contentType
    0.72
     Disabilities
    0.71
     morphogenesis
    0.70
    Activations Density 0.738%

    No Known Activations