© 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. 4212
    Prev
    Next
    INDEX
    Explanations

    distance and direction

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

    The neuron detects numeric distance or location measurements (e.g. “six miles north,” “three miles northwest,” “Road,” “miles,” “km”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    phrases describing geographic location using distances, cardinal directions, and proximity to landmarks or roads.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408

    directional and distance information describing geographic locations.

    oai_token-act-pair · claude-4-5-sonnetTriggered 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
    ximation
    -0.94
     onCreateView
    -0.83
     menino
    -0.83
     índice
    -0.82
    ッチン
    -0.82
    izmu
    -0.81
    ()));
    -0.80
    adaptiveStyles
    -0.80
     galinha
    -0.79
    拳
    -0.78
    POSITIVE LOGITS
     south
    2.31
     west
    2.27
     north
    2.23
     east
    2.20
     southeast
    1.88
     northeast
    1.83
     northwest
    1.78
     southwest
    1.59
     between
    1.38
     western
    1.33
    Activations Density 0.134%

    No Known Activations