© 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. 4005
    Prev
    Next
    INDEX
    Explanations

    needs access

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

    The neuron fires most strongly on the head noun of formal definition‐style sentences—that is, the key term (e.g. “liability,” “system,” “manner”) being defined or described.

    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
    <unused38>
    0.72
    TextureData
    0.72
    <unused97>
    0.68
    <unused19>
    0.67
    (?)
    0.66
    oretically
    0.64
    ColorBit
    0.63
    <unused13>
    0.63
    <unused91>
    0.62
    TreeNavigator
    0.61
    POSITIVE LOGITS
    ).
    0.88
    Ī
    0.80
    .)
    0.80
    ī
    0.79
     his
    0.77
    ’.
    0.76
    ™.
    0.74
    .’
    0.73
    ?
    0.73
    ?’
    0.72
    Activations Density 4.385%

    No Known Activations