© 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. 22-GEMMASCOPE-RES-131K
    7. 35654
    Prev
    Next
    INDEX
    Explanations

    San followed by place names

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

    The neuron fires on occurrences of the token “San” as part of place names (e.g. San Francisco, San Jose).

    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
    鹮
    -3.36
     If
    -2.94
    厹
    -2.64
     funguje
    -2.58
    に見える
    -2.56
    ؿ
    -2.53
    ctuaries
    -2.42
    al
    -2.41
     nebude
    -2.38
     ถ้า
    -2.38
    POSITIVE LOGITS
    N
    3.72
    D
    3.53
    E
    3.14
    In
    2.91
    L
    2.86
    A
    2.84
    T
    2.78
    F
    2.78
    O
    2.72
    Y
    2.59
    Activations Density 0.016%

    No Known Activations