© 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. 4547
    Prev
    Next
    INDEX
    Explanations

    file line

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

    The neuron strongly activates on countable concrete nouns referring to discrete items or events (e.g., matches, leagues, months, maps, cones, groups).

    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
    ဣ
    0.57
     fantastical
    0.53
    idän
    0.52
     heartwarming
    0.49
    ◈
    0.49
     Zayn
    0.48
    𒉰
    0.48
     टाइटल
    0.48
     справед
    0.48
     fairytale
    0.47
    POSITIVE LOGITS
    in
    0.54
     for
    0.50
     I
    0.50
     from
    0.50
    /
    0.48
    ol
    0.47
     with
    0.47
     and
    0.46
    as
    0.46
    o
    0.45
    Activations Density 0.686%

    No Known Activations