© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 62817
    Prev
    Next
    INDEX
    Explanations

    article A before capitalized words

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

    The neuron responds to the standalone capital letter “A” when it appears as the first word of a title or heading.

    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
    Цент
    -1.03
    颜值
    -0.99
     און
    -0.93
    謁
    -0.92
    struktion
    -0.90
     alfombra
    -0.89
     artesanía
    -0.88
     almohada
    -0.86
    Цели
    -0.85
     dobles
    -0.85
    POSITIVE LOGITS
     Different
    1.21
     Place
    1.10
     Matter
    1.09
     to
    1.06
     History
    1.03
     Study
    1.02
     Very
    1.00
     so
    0.99
     place
    0.98
    Different
    0.98
    Activations Density 0.031%

    No Known Activations