© 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. 73620
    Prev
    Next
    INDEX
    Explanations

    switch case or control flow

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

    The neuron is detecting occurrences of the word “switch.”

    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
     
    -3.66
    0
    -2.83
    en
    -2.64
    ’
    -2.59
    1
    -2.59
    {
    -2.55
    9
    -2.47
     \
    -2.39
    為に
    -2.38
    8
    -2.36
    POSITIVE LOGITS
    豨
    2.77
    LLocation
    2.67
    ፝
    2.48
    驺
    2.47
    putExtra
    2.47
    鋱
    2.47
    镚
    2.45
    玧
    2.45
     ſou
    2.44
    傩
    2.42
    Activations Density 0.014%

    No Known Activations