© 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. 98192
    Prev
    Next
    INDEX
    Explanations

    special characters and operators

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

    The neuron consistently flags isolated symbol-only tokens—especially mathematical or programming operators (e.g. >>, =, x, +, −)—in the text.

    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
     peut
    -0.87
     krä
    -0.82
    endimiento
    -0.81
     afirma
    -0.78
     autor
    -0.77
     exig
    -0.75
     mą
    -0.75
     друго
    -0.74
     grip
    -0.73
    Paola
    -0.72
    POSITIVE LOGITS
     между
    1.05
     between
    0.99
    ooth
    0.91
     nimic
    0.91
    between
    0.89
     }
    
    0.86
     pomiędzy
    0.86
     ملع
    0.85
    thodox
    0.85
     tabulated
    0.85
    Activations Density 0.028%

    No Known Activations