© 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. 6508
    Prev
    Next
    INDEX
    Explanations

    Answer options or definitions

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

    The neuron strongly activates on long spans of whitespace or blank‐line tokens, i.e. large empty gaps in the text.

    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
     (
    0.71
     aforesaid
    0.70
     preprint
    0.69
     glassware
    0.68
     schist
    0.66
     ($\
    0.66
     Yukawa
    0.66
     assorted
    0.65
     müm
    0.65
     wherever
    0.65
    POSITIVE LOGITS
    髗
    0.79
    rivez
    0.78
    definition
    0.75
    descripcion
    0.75
    Descripcion
    0.75
    От
    0.73
    definitions
    0.73
    Ін
    0.73
    categories
    0.72
    projeto
    0.72
    Activations Density 0.000%

    No Known Activations