© 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. 47911
    Prev
    Next
    INDEX
    Explanations

    street names or avenues

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

    The neuron strongly activates on address-like components—street numbers, street names, and street‐type words (e.g. “Drive,” “Avenue,” “Street”)—i.e. parts of postal addresses.

    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
     päät
    -0.83
    南部
    -0.81
    북
    -0.79
     southward
    -0.79
    }}-\
    -0.79
    venus
    -0.74
     Áng
    -0.73
     mistakes
    -0.73
     unforeseen
    -0.73
     guarantees
    -0.72
    POSITIVE LOGITS
     valid
    0.81
    valid
    0.80
     verbund
    0.75
    ITALY
    0.74
    wester
    0.74
     же
    0.73
    xdb
    0.72
     bewusst
    0.71
    zek
    0.70
     polar
    0.69
    Activations Density 0.024%

    No Known Activations