© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 58524
    Prev
    Next
    INDEX
    Explanations

    Boost.Asio C++ code

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

    The neuron is detecting programming‐style code syntax—especially C/C++ constructs such as preprocessor directives, file paths/includes, punctuation (slashes, angle brackets), and version numbers.

    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
     trustworthiness
    -0.84
    CHREIBUNG
    -0.78
    Horses
    -0.77
    Aldo
    -0.75
    Substrate
    -0.74
    Bett
    -0.72
    𝒯
    -0.72
    Buildable
    -0.71
    まる
    -0.71
    inam
    -0.71
    POSITIVE LOGITS
     asio
    0.85
    strand
    0.82
     subnet
    0.81
    court
    0.76
     Petersburg
    0.75
     vähän
    0.74
    relser
    0.73
    ônimo
    0.73
    write
    0.73
    ious
    0.73
    Activations Density 0.075%

    No Known Activations