© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 90060
    Prev
    Next
    INDEX
    Explanations

    German place names and suffixes

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

    The neuron specializes in detecting substrings common to European place names (especially Dutch and German toponym affixes like “burger,” “se,” “dwars,” etc.).

    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
     freue
    -0.88
     Regionen
    -0.83
    恬
    -0.80
     že
    -0.78
     administrativos
    -0.77
     anderen
    -0.75
     químicos
    -0.74
    smi
    -0.73
    jacent
    -0.73
     Wettbewerb
    -0.73
    POSITIVE LOGITS
     Idiot
    0.83
    Italy
    0.82
    zij
    0.82
    IALES
    0.82
    Cou
    0.79
    Switzerland
    0.78
     Antara
    0.78
    Cola
    0.78
    seits
    0.77
     Horiz
    0.77
    Activations Density 0.027%

    No Known Activations