© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 15870
    Prev
    Next
    INDEX
    Explanations

    joke

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

    That neuron fires on the “content” word in two‐word or paired constructions—e.g. the main verb in phrasal verbs (“made up,” “stay connected”) and the second term of an “X or Y”/“X and Y” pair (“restricts or inhibits,” “journal.pone”).

    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
    ाई
    0.86
    ,
    0.78
     on
    0.77
    /
    0.75
     
    0.74
     ON
    0.73
    7
    0.73
     z
    0.72
    .
    0.72
     -
    0.71
    POSITIVE LOGITS
     кеңсеси
    1.11
     alken
    1.05
     кеңселер
    1.01
     кеңсе
    1.00
    ʿ
    0.99
    thisTrack
    0.97
    ATEG
    0.93
    အစိတ်အပိုင်း
    0.93
     অর্চন
    0.92
    netz
    0.92
    Activations Density 0.001%

    No Known Activations