© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 130473
    Prev
    Next
    INDEX
    Explanations

    the neuron fires on words and phrases that signal quoted speech or news-attribution (e.g. said, stated, mentioned, reporters, addressing).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    conference reporters speaking

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    を読んで
    -1.07
    捅
    -0.97
    ecause
    -0.95
    arture
    -0.92
    starting
    -0.91
    looks
    -0.90
    asily
    -0.90
     ayudas
    -0.89
    Opening
    -0.88
    oment
    -0.88
    POSITIVE LOGITS
     on
    1.76
     at
    1.41
     in
    1.23
    וּ
    1.13
    ☆☆☆
    1.08
     exquisitely
    1.05
     before
    1.03
     recently
    1.02
     чле
    1.01
     during
    0.98
    Activations Density 0.014%

    No Known Activations