© 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. 10-GEMMASCOPE-RES-131K
    7. 40944
    Prev
    Next
    INDEX
    Explanations

    titles and officials

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

    This neuron strongly activates on the definite article “the” when it begins a speech‐attribution phrase naming a speaker or entity (e.g. “the CEO said,” “the governor said,” “the group’s leader wrote”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     gainera
    -1.80
     sahiptir
    -1.63
     potreb
    -1.59
    ܊
    -1.59
    色々
    -1.56
     idén
    -1.53
    Gön
    -1.51
     články
    -1.51
    巍
    -1.50
     cuadrada
    -1.48
    POSITIVE LOGITS
     of
    2.09
    5
    2.02
    of
    1.95
    6
    1.87
    7
    1.84
     \
    1.83
    },
    1.76
    9
    1.64
     or
    1.59
     from
    1.59
    Activations Density 0.006%

    No Known Activations