© 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. 40974
    Prev
    Next
    INDEX
    Explanations

    www. followed by dot

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

    The neuron is chiefly firing on lines or tokens with large amounts of leading whitespace (“indentation”) in code.

    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
    ₁)
    -1.42
     llavero
    -1.39
     cordón
    -1.38
    bracelet
    -1.37
    cribes
    -1.33
     casada
    -1.33
     cukru
    -1.33
    camiseta
    -1.32
     béisbol
    -1.31
     cojín
    -1.31
    POSITIVE LOGITS
     any
    2.16
     is
    1.70
    \
    1.69
     or
    1.59
    任何
    1.59
     qualquer
    1.49
     of
    1.48
     you
    1.45
     qualsiasi
    1.45
    ;
    1.44
    Activations Density 0.256%

    No Known Activations