© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 16015
    Prev
    Next
    INDEX
    Explanations

    URLs with slashes

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

    This neuron detects subword pieces that are part of hyphen-joined compounds.

    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
     socalled
    0.73
    <unused60>
    0.65
     ——
    0.61
    Mult
    0.57
    ;
    0.56
    されない
    0.55
    <unused61>
    0.54
    Array
    0.54
    ᱨ
    0.52
    ныгы
    0.51
    POSITIVE LOGITS
    -
    3.22
    -...
    2.27
    -‘
    2.09
    -'
    1.97
    -*
    1.95
    -[
    1.95
    -{
    1.91
    -(
    1.90
    -/
    1.84
    -]
    1.84
    Activations Density 0.330%

    No Known Activations