© 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. 22-GEMMASCOPE-RES-131K
    4. 84311
    Prev
    Next
    INDEX
    Explanations

    code class definitions

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

    This neuron responds to long runs of whitespace (e.g. deep code indentation).

    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
    Teraz
    -1.77
    Ostat
    -1.74
    corrhi
    -1.61
    АО
    -1.59
     biela
    -1.59
    でございます
    -1.58
     pán
    -1.58
    Przed
    -1.57
    besondere
    -1.56
     filozof
    -1.55
    POSITIVE LOGITS
    e
    1.72
    a
    1.69
    ce
    1.63
     is
    1.63
    ia
    1.59
    ,
    1.57
    th
    1.53
    I
    1.48
    It
    1.47
    $
    1.45
    Activations Density 0.006%

    No Known Activations