© 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. 10857
    Prev
    Next
    INDEX
    Explanations

    response or output related to function calls

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

    This neuron fires strongly on tokens that start a new paragraph or line (i.e. tokens immediately following a blank‐line or double-space break).

    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
     flavoring
    0.80
    outube
    0.79
    Ꮘ
    0.79
    ação
    0.76
     disastrous
    0.76
     như
    0.75
    iptic
    0.75
    ться
    0.74
     antagonist
    0.74
     राशिफल
    0.74
    POSITIVE LOGITS
    slick
    0.80
    ع
    0.73
    stub
    0.72
    speakers
    0.70
    spacer
    0.69
    writelines
    0.69
     Think
    0.68
    b
    0.68
     শতাধিক
    0.67
    Think
    0.66
    Activations Density 0.000%

    No Known Activations