© 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. 22-GEMMASCOPE-RES-131K
    7. 72575
    Prev
    Next
    INDEX
    Explanations

    code delimiters and non-English"}," or "code delimiters and multilingual"

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

    This neuron detects instances of “in … order” phrases (e.g. “in descending order,” “in increasing order,” etc.) that specify how to sort or arrange items.

    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
    گون
    -0.81
     chạm
    -0.77
     рассматри
    -0.77
     asistir
    -0.74
    mittances
    -0.74
     harshly
    -0.73
     settore
    -0.73
     Planung
    -0.72
     работу
    -0.72
     oxidase
    -0.72
    POSITIVE LOGITS
     magát
    0.79
    ⇗
    0.78
    cce
    0.78
     sonrası
    0.76
     gutted
    0.75
    olunt
    0.75
     Venez
    0.74
     nogle
    0.74
     akik
    0.74
     zuerst
    0.73
    Activations Density 0.021%

    No Known Activations