© 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. 34-GEMMASCOPE-RES-131K
    4. 90574
    Prev
    Next
    INDEX
    Explanations

    Kevin Schneider

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

    This neuron specifically responds to mentions of the Benghazi CIA story, peaking strongest on “Benghazi.”

    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
    ={
    
    -0.78
     pelas
    -0.77
    argli
    -0.74
     junction
    -0.69
    valry
    -0.69
     груп
    -0.69
     csv
    -0.68
    ategorias
    -0.68
    ɦ
    -0.66
     setelah
    -0.65
    POSITIVE LOGITS
     Beng
    1.90
    Beng
    1.62
     beng
    1.13
     consulate
    1.07
     embassy
    1.04
     tháng
    0.88
     Bengali
    0.88
    Ambassador
    0.88
     Comer
    0.87
     Accountability
    0.87
    Activations Density 0.026%

    No Known Activations