© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 84180
    Prev
    Next
    INDEX
    Explanations

    conditional statements or requirements

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

    The neuron primarily responds to English modal verbs (like “must,” “should,” “can”) that signal obligation or permission.

    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
     preponder
    -1.00
    ₨
    -0.92
     indes
    -0.88
    感受
    -0.87
    INFIN
    -0.86
     preval
    -0.85
     preced
    -0.83
     डि
    -0.80
    MÁS
    -0.80
    ου
    -0.80
    POSITIVE LOGITS
     any
    1.52
     unless
    1.42
     either
    1.40
     if
    1.39
     cannot
    1.39
     otherwise
    1.37
     so
    1.34
     must
    1.29
     other
    1.26
     exception
    1.25
    Activations Density 0.060%

    No Known Activations