© 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. 24-GEMMASCOPE-2-RES-16K
    7. 9687
    Prev
    Next
    INDEX
    Explanations

    tortoise, ultimate, straight,urches, team

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

    The neuron fires most strongly on words immediately surrounding explicit numeric data (decimal numbers), i.e. it’s detecting tokens in contexts where precise numerical values appear.

    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
     %\
    0.80
     ede
    0.73
     *(
    0.73
     '*
    0.72
     ছো
    0.70
     >>=
    0.68
     \%,
    0.68
     *((
    0.68
     UInt
    0.67
     WORDS
    0.67
    POSITIVE LOGITS
    钼
    0.86
     পড়েছে
    0.86
    纶
    0.82
    ều
    0.79
    罗马
    0.79
    ंति
    0.77
     politely
    0.76
    岚
    0.76
    घव
    0.76
    维
    0.76
    Activations Density 0.000%

    No Known Activations