© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 85389
    Prev
    Next
    INDEX
    Explanations

    downtown city names

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

    The neuron fires on occurrences of the word “downtown.”

    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
     Kiedy
    -2.52
     pierde
    -2.47
     Może
    -2.44
     Wszyst
    -2.39
     Ostat
    -2.39
    躄
    -2.38
     две
    -2.34
     ก็
    -2.34
     Jeśli
    -2.33
    多么
    -2.28
    POSITIVE LOGITS
    鷓
    2.69
    s
    2.56
    唯美
    2.52
     vertes
    2.38
     sowohl
    2.36
    ."
    2.34
     strategis
    2.34
    鬈
    2.31
     sehari
    2.30
    ”、“
    2.30
    Activations Density 0.004%

    No Known Activations