© 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. 109245
    Prev
    Next
    INDEX
    Explanations

    official paper version

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

    the neuron responds to very common, high-frequency words (i.e. the most frequent “stop” words in English).

    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
     Then
    -1.09
     then
    -1.08
     their
    -0.91
    他还
    -0.90
     especially
    -0.88
    niejszych
    -0.86
    Then
    -0.86
     Niem
    -0.85
     Remember
    -0.82
     other
    -0.82
    POSITIVE LOGITS
     associé
    1.03
    ecause
    0.95
     NSCoder
    0.93
    mlink
    0.92
    corrected
    0.90
     byla
    0.89
    carus
    0.89
     ASTM
    0.87
    ournal
    0.86
     Ríos
    0.86
    Activations Density 0.115%

    No Known Activations