© 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. 1332
    Prev
    Next
    INDEX
    Explanations

    general attributes and specifications

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

    The neuron fires strongly on the very first real token of each new text chunk, i.e. it detects the start of a new document or paragraph segment.

    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
     freepik
    0.97
     prüfen
    0.88
     überprüft
    0.88
     prepd
    0.86
    𝙶
    0.86
     dargestellt
    0.82
     mencionados
    0.82
     Idha
    0.80
    CodeAttribute
    0.79
     perusahaan
    0.79
    POSITIVE LOGITS
    _
    0.74
    func
    0.71
    edu
    0.64
    weather
    0.64
     AV
    0.64
    black
    0.63
    fang
    0.63
     дистан
    0.63
    wch
    0.62
    Jack
    0.62
    Activations Density 0.000%

    No Known Activations