© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 26770
    Prev
    Next
    INDEX
    Explanations

    lists of things and concepts

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

    The neuron activates on the coordinating conjunction “and” (including ampersands) when it links items or phrases, especially in lists.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    を
    -1.05
    は
    -1.02
    だが
    -0.99
    が
    -0.95
    ซึ่ง
    -0.95
    での
    -0.92
    による
    -0.90
     $-
    -0.89
    mó
    -0.89
    بحث
    -0.89
    POSITIVE LOGITS
     Marinette
    0.88
    Sulfur
    0.82
    Almond
    0.82
    に入れる
    0.81
    Aquatic
    0.81
    Caffeine
    0.81
    símbolo
    0.81
     künf
    0.80
     Decisions
    0.80
     cliquer
    0.79
    Activations Density 0.073%

    No Known Activations