© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 40359
    Prev
    Next
    INDEX
    Explanations

    Windy Nicklaus William Winfield

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

    The neuron activates on double‐quote characters, effectively marking the boundaries of string literals in code or text.

    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
     información
    -1.05
     llamado
    -1.02
     lär
    -0.93
    :✨
    -0.91
     Irán
    -0.88
    zeichnet
    -0.88
    𝕩
    -0.85
     “(
    -0.84
     oferece
    -0.83
    !(
    -0.82
    POSITIVE LOGITS
    _
    1.23
    ^{\
    1.09
     each
    1.06
    )^{\
    1.03
    _{\
    1.03
     assolutamente
    0.99
    ната
    0.96
     stwier
    0.95
    Each
    0.95
     estremamente
    0.95
    Activations Density 0.000%

    No Known Activations