© 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. 34-GEMMASCOPE-RES-131K
    7. 90531
    Prev
    Next
    INDEX
    Explanations

    exam restrictions and allowed actions

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

    The neuron strongly detects structural or formatting markers in text—especially section‐heading tokens (e.g. “Materials,” “Contact,” “Instructions”), colons, and other layout/markup elements.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     المط
    -0.85
     und
    -0.84
     ancora
    -0.82
     server
    -0.77
    posix
    -0.75
     poquito
    -0.75
     platform
    -0.73
     typ
    -0.73
    ulla
    -0.72
     μετα
    -0.72
    POSITIVE LOGITS
    corporated
    0.94
     calculator
    0.88
    calculator
    0.87
     answers
    0.86
     Calcul
    0.84
     zugelassen
    0.84
     ši
    0.84
    sper
    0.83
    écri
    0.81
     žen
    0.81
    Activations Density 0.008%

    No Known Activations