© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 9759
    Prev
    Next
    INDEX
    Explanations

    listing numbers, punctuation, code

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

    The neuron strongly lights up on technical fire‐investigation terminology—references to the origin area of a fire, ignition sources, and titles like “Investigators” or “Association of Fires Investigators.”

    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_31_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
    —
    0.96
    सुद्धा
    0.90
    id
    0.87
    en
    0.85
    Type
    0.84
    /
    0.83
    Page
    0.82
    e
    0.82
    ஃப்
    0.81
    no
    0.80
    POSITIVE LOGITS
     ,
    2.83
     ，
    2.20
     ,[
    2.20
     부터
    2.19
     ،
    2.13
     ,"
    2.11
    ॔
    2.11
     :"
    2.00
    ॕ
    2.00
     ۔
    1.98
    Activations Density 1.694%

    No Known Activations