© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 22573
    Prev
    Next
    INDEX
    Explanations

    string conversion or string manipulation

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

    The neuron fires whenever the text is talking about “string” data (string, strings, str, ToString, text, etc.), i.e. references to string‐type content or operations.

    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
     какую
    -1.86
    三个
    -1.84
     другую
    -1.75
    ኧ
    -1.73
    也不
    -1.70
     ira
    -1.68
    或者
    -1.65
    Warna
    -1.62
     какой
    -1.61
    也都
    -1.61
    POSITIVE LOGITS
    </h3>
    1.93
    :
    1.77
     continuará
    1.76
    的我
    1.65
     budú
    1.63
     deberán
    1.61
     všetky
    1.59
    ～～～
    1.59
     is
    1.57
    activité
    1.55
    Activations Density 0.052%

    No Known Activations