© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 87410
    Prev
    Next
    INDEX
    Explanations

    word-of-word or word_of_word structures

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

    The neuron activates specifically on the preposition “of.”

    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
     and
    -2.88
    );
    -2.48
     gets
    -2.41
     considers
    -2.33
     regarded
    -2.23
    )
    -2.22
     takes
    -2.20
     on
    -2.16
    ):
    -2.13
     pensado
    -2.11
    POSITIVE LOGITS
    ’
    3.42
    /
    2.88
    鶿
    2.77
    𖣔
    2.72
    '
    2.66
    “
    2.59
    5
    2.52
    9
    2.50
    *
    2.50
    of
    2.50
    Activations Density 0.010%

    No Known Activations