© 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. 35776
    Prev
    Next
    INDEX
    Explanations

    utility for Halloween

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

    This neuron detects programming loop constructs (e.g. for, while, do).

    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
    weisung
    -0.81
     topics
    -0.77
    <0xDE>
    -0.76
     zůst
    -0.76
     area
    -0.75
    badlogic
    -0.75
     selections
    -0.74
     řek
    -0.73
    ガジン
    -0.73
    cciones
    -0.71
    POSITIVE LOGITS
     progesterone
    0.90
     wäre
    0.85
     Try
    0.84
    尝试
    0.84
    才是
    0.82
    Try
    0.81
     vuestros
    0.79
    ึ
    0.78
    Attempt
    0.78
    would
    0.78
    Activations Density 0.003%

    No Known Activations