© 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. 34-GEMMASCOPE-RES-131K
    7. 90105
    Prev
    Next
    INDEX
    Explanations

    graphics code, data, names

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

    This neuron detects code snippets by firing strongly on programming identifiers, numeric literals, and other syntax tokens typical of source‐code.

    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
    Mucho
    -0.81
     beneficio
    -0.73
    Folk
    -0.72
    缶
    -0.71
     гря
    -0.71
    pegno
    -0.70
    Ւ
    -0.69
    bigsqcup
    -0.68
    yii
    -0.68
     while
    -0.68
    POSITIVE LOGITS
     rotación
    0.99
     clásicos
    0.90
     hessen
    0.84
    bha
    0.83
     صنعت
    0.83
    就是要
    0.79
    mathematical
    0.78
     biến
    0.77
     glut
    0.77
    szer
    0.76
    Activations Density 0.015%

    No Known Activations