© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 8321
    Prev
    Next
    INDEX
    Explanations

    distance units

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

    The neuron strongly activates on numeric distance or measurement tokens (e.g. mileages, kilometers, numbered street abbreviations) in route‐style instructions.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     sacs
    0.55
     lengan
    0.55
     invasions
    0.54
     alcanz
    0.53
     unterstützt
    0.52
     ходить
    0.52
     pens
    0.52
    步
    0.52
     ноги
    0.51
     torso
    0.49
    POSITIVE LOGITS
     mile
    1.79
     miles
    1.70
    mile
    1.64
    Miles
    1.52
     kilometre
    1.52
     Miles
    1.52
     kilometres
    1.52
    Mile
    1.50
    miles
    1.50
     kilometers
    1.46
    Activations Density 0.178%

    No Known Activations