© 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. 10085
    Prev
    Next
    INDEX
    Explanations

    `"$"` followed by `1` or `original`

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

    This neuron detects hash‐mark tokens (one or more “#” characters) used as Markdown‐style headings.

    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
     %
    0.70
     %\
    0.66
     %(
    0.63
    %
    0.62
     %,
    0.59
     dini
    0.57
    %'
    0.55
    adoff
    0.55
    Ž
    0.55
     %.
    0.54
    POSITIVE LOGITS
    PUBL
    0.66
    rootScope
    0.65
    WIL
    0.63
    FALL
    0.63
    CUSS
    0.62
    KEYWORDS
    0.61
    STATEMENT
    0.61
    HARD
    0.61
    Scenario
    0.59
    康
    0.59
    Activations Density 0.008%

    No Known Activations