© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 800
    Prev
    Next
    INDEX
    Explanations

    The neuron detects cardinal‐direction words (like “North,” “South,” “Central,” etc.) when they appear as part of geographic or place names.

    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_24_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
    样
    0.55
     chết
    0.54
    StudentNo
    0.54
    ник
    0.53
    Appuntamento
    0.53
    身边
    0.52
    すべての
    0.51
    Pentru
    0.51
    janje
    0.51
    ಶ
    0.51
    POSITIVE LOGITS
     hemisphere
    0.64
     quadrant
    0.61
     noroeste
    0.60
    most
    0.59
     quadrants
    0.58
     mainland
    0.56
     Hemisphere
    0.55
     highland
    0.55
     extremity
    0.54
    ্ঞ
    0.53
    Activations Density 0.069%

    No Known Activations