© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 111635
    Prev
    Next
    INDEX
    Explanations

    native inhabitants

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

    This neuron fires on the key “head” word at the end of multiword descriptive or idiomatic expressions (e.g. granted in “take for granted,” better/of/you in “get the better of you,” life in “larger-than-life,” native in “Raleigh native”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     persino
    -2.05
    ⢎
    -1.96
     studen
    -1.87
     Проци
    -1.80
     jednej
    -1.80
     certamente
    -1.80
    焢
    -1.74
     addirittura
    -1.73
    вгений
    -1.72
     всѣ
    -1.70
    POSITIVE LOGITS
    ’
    2.09
     {
    1.79
    —
    1.77
     —
    1.76
     for
    1.66
    /
    1.63
    ?
    1.63
    又是一
    1.62
    the
    1.60
     apprehensive
    1.59
    Activations Density 0.015%

    No Known Activations