© 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. 22-GEMMASCOPE-RES-131K
    7. 73243
    Prev
    Next
    INDEX
    Explanations

    reporting speech

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

    The neuron primarily responds to instances of the reporting verb “said,” marking speaker-attribution in news text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
     and
    -1.48
     estadounidenses
    -1.27
     sebagainya
    -1.20
    のもの
    -1.19
    把它
    -1.17
     пользователем
    -1.15
    ...»
    -1.15
    DEL
    -1.14
     тщательно
    -1.13
    न
    -1.13
    POSITIVE LOGITS
     solicitó
    1.26
    8
    1.22
     further
    1.17
     explicó
    1.15
    CIUDAD
    1.09
     explicou
    1.09
     unlike
    1.07
    MADRID
    1.05
     dará
    1.05
     passaram
    1.05
    Activations Density 0.009%

    No Known Activations