© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 8734
    Prev
    Next
    INDEX
    Explanations

    Wal and Wall prefixes

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

    The neuron strongly activates on occurrences of the substring “wall” (e.g. Wallenberg, wall street, wall jump, wall units).

    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
     Gerais
    0.57
     ගන්න
    0.57
    SourceIP
    0.55
    BrowserRouter
    0.52
    مق
    0.52
     맥
    0.52
    ヴァン
    0.52
    кои
    0.50
    subkey
    0.50
    擾
    0.50
    POSITIVE LOGITS
     WALL
    1.02
     Wall
    0.96
    Wall
    0.83
     Wal
    0.82
     Wally
    0.80
     wal
    0.77
     wall
    0.71
    wall
    0.71
     壁
    0.69
     Wale
    0.69
    Activations Density 0.149%

    No Known Activations