© 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. 95691
    Prev
    Next
    INDEX
    Explanations

    0 or assuming positive

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

    The neuron is triggered by tokens in inline mathematical expressions—especially variable names, fraction slashes “/”, and exponentiation operators.

    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
    IBOutlet
    -0.91
    CLASSES
    -0.90
    šla
    -0.88
    zmir
    -0.86
    ප
    -0.81
    martre
    -0.79
    vability
    -0.79
    了起来
    -0.78
    verna
    -0.78
     möglicherweise
    -0.78
    POSITIVE LOGITS
     redundant
    0.85
    redund
    0.85
    0.82
        
    0.81
      
    0.77
    isObject
    0.77
     teaching
    0.77
     Any
    0.77
     Saratoga
    0.77
     повече
    0.76
    Activations Density 0.100%

    No Known Activations