© 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. 34-GEMMASCOPE-RES-131K
    7. 102957
    Prev
    Next
    INDEX
    Explanations

    Massachusetts towns and geography

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

    The neuron activates strongly on tokens that are part of URLs, file paths, or hyphenated/ dotted identifiers (i.e. tokens containing punctuation like “.”, “/”, or “-”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    材质
    -0.81
    栢
    -0.81
    zoni
    -0.76
     TU
    -0.76
     pel
    -0.75
    TRL
    -0.75
    ¡¡¡
    -0.75
    Swing
    -0.75
    Fork
    -0.73
     LAP
    -0.72
    POSITIVE LOGITS
     Nantucket
    1.09
    Massachusetts
    1.07
     Massachusetts
    0.93
     Martha
    0.92
     MASSACHUSETTS
    0.87
    Mitt
    0.78
     Brahmin
    0.78
    Martha
    0.78
    BOSTON
    0.75
     island
    0.74
    Activations Density 0.009%

    No Known Activations