© 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
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 8880
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly responds to obscure or unusual tokens (i.e. rare words or rare character sequences).

    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
    στά
    0.87
    놨
    0.86
    s
    0.82
     gatos
    0.82
    ς
    0.80
    स्ताना
    0.79
    anggap
    0.77
     muertos
    0.75
     gripe
    0.73
     Dollar
    0.72
    POSITIVE LOGITS
    Το
    0.84
    เป็น
    0.77
    で
    0.75
     کے
    0.74
    辂
    0.73
     또한
    0.72
    近日
    0.72
    قية
    0.71
    また
    0.70
    看来
    0.70
    Activations Density 0.000%

    No Known Activations