© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26
    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 58109
    Prev
    Next
    INDEX
    Explanations

    of [place names]

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

    The neuron activates on numeric tokens, effectively detecting and outputting the value of numbers appearing in the text.

    oai_token-act-pair · o4-mini
    Triggered by @jyhe0408

    the preposition “of” linking nouns, especially within geographic or locational descriptions.

    oai_token-act-pair · gpt-5
    Triggered by @jyhe0408

    The neuron activates strongly for the word "of" when it appears in geographical or locational phrases, particularly those describing coastal locations or places (e.g., "coast of Honduras," "coast of Korea," "suburbs of Chicago").

    oai_token-act-pair · claude-4-5-sonnet
    Triggered 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
    Negative Logits
     your
    -1.52
    ⬮
    -1.51
     a
    -1.45
     nového
    -1.43
    specified
    -1.37
    宬
    -1.33
     similar
    -1.32
     on
    -1.32
    salomon
    -1.30
     this
    -1.29
    POSITIVE LOGITS
     of
    2.23
    「
    1.77
     作り方
    1.72
     этих
    1.62
     здесь
    1.61
     даже
    1.61
     как
    1.56
     очень
    1.56
     их
    1.52
     bernama
    1.52
    Activations Density 0.062%

    No Known Activations