© 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-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 130596
    Prev
    Next
    INDEX
    Explanations

    empty or zero values

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

    This neuron activates on identifiers denoting constant or default values (e.g. Empty, None, INT_MIN, INT_MAX, Default, newaxis, etc.).

    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
    秝
    -1.67
     of
    -1.67
    廰
    -1.58
    男主
    -1.47
    bibit
    -1.46
    结语
    -1.42
     paff
    -1.41
     declarat
    -1.38
     from
    -1.38
    我一直
    -1.37
    POSITIVE LOGITS
     estekak
    1.83
    ixante
    1.68
    3
    1.67
     rengi
    1.67
     kaplama
    1.65
     dekoratif
    1.63
    他
    1.59
    6
    1.52
    ぼちゃ
    1.51
     podían
    1.50
    Activations Density 0.005%

    No Known Activations