© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 69674
    Prev
    Next
    INDEX
    Explanations

    `\` escapes and string manipulation

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

    This neuron activates on backslash characters (i.e. string/regex escape sequences).

    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
    _
    -2.52
    缡
    -2.52
     рынке
    -2.28
    珹
    -2.27
    みたいです
    -2.27
     telles
    -2.27
    
    -2.27
    ⟣
    -2.23
     ſou
    -2.22
    泚
    -2.20
    POSITIVE LOGITS
    to
    3.33
    ↵
    3.23
    and
    2.56
     we
    2.31
    To
    2.19
     To
    2.16
    b
    2.14
    ?"
    2.11
    2.09
     nudge
    2.08
    Activations Density 0.002%

    No Known Activations