© 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. 67273
    Prev
    Next
    INDEX
    Explanations

    Caffè, Cadix, Causal, Caverta

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

    city names or portions of place names from various languages.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron responds to isolated capitalized letter sequences or proper‐name tokens—e.g. subwords like “Ca,” “Dix,” “Saka,” “GUI,” etc.—i.e. it picks out capitalized names or acronyms.

    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
     parachoque
    -2.75
    菹
    -2.59
    鬭
    -2.55
    蚡
    -2.52
     houſe
    -2.48
    鼷
    -2.47
    赜
    -2.45
    珰
    -2.45
    峤
    -2.44
    諏
    -2.42
    POSITIVE LOGITS
    In
    2.75
    The
    2.72
    )
    2.61
    en
    2.58
    .
    2.56
    To
    2.39
     It
    2.34
    ar
    2.27
     
    2.27
     S
    2.20
    Activations Density 0.001%

    No Known Activations