© 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. 75656
    Prev
    Next
    INDEX
    Explanations

    directional US regions and states

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

    The neuron detects geographic region names, especially those combining a directional or positional qualifier (e.g. “Eastern,” “Mid,” “Southeastern,” “West”) with a place name.

    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
     salad
    -0.80
     Intervention
    -0.79
    Significado
    -0.76
    ತಿ
    -0.76
    味が
    -0.75
    誘惑
    -0.75
    らかに
    -0.75
    salvar
    -0.75
    odyne
    -0.75
    whit
    -0.73
    POSITIVE LOGITS
     Ohio
    1.61
     Florida
    1.54
     Texas
    1.48
     Kentucky
    1.43
     Michigan
    1.41
     Arkansas
    1.41
     Louisiana
    1.38
     Tennessee
    1.32
     Alabama
    1.32
     Illinois
    1.31
    Activations Density 0.055%

    No Known Activations