© 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. 34-GEMMASCOPE-RES-131K
    4. 93866
    Prev
    Next
    INDEX
    Explanations

    court case filings

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

    The neuron strongly activates on sequences of whitespace or indentation (long runs of blank characters).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     Tenggara
    -0.89
     or
    -0.88
     know
    -0.88
     any
    -0.86
     from
    -0.85
    Przeczytaj
    -0.84
     conos
    -0.83
     bring
    -0.82
     dulu
    -0.82
     bringing
    -0.82
    POSITIVE LOGITS
    ^{*}$
    0.96
    はん
    0.94
    ChrTalk
    0.88
    (‘
    0.87
     Lacto
    0.87
     reaffirm
    0.86
    alele
    0.86
    romas
    0.85
    ening
    0.84
    Jährige
    0.83
    Activations Density 0.002%

    No Known Activations