© 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. 1536
    Prev
    Next
    INDEX
    Explanations

    data structure or state

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

    The neuron strongly activates on multiword proper‐name sequences—especially street addresses or place names composed of consecutive capitalized words.

    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
    其他
    0.79
     условии
    0.77
    و
    0.76
    Options
    0.75
    dB
    0.75
    EB
    0.74
    𝒟
    0.74
    AF
    0.73
    EndPoint
    0.73
    CallOptions
    0.73
    POSITIVE LOGITS
     cujo
    0.70
     toaster
    0.66
    ↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵
    0.65
     vragen
    0.61
     victoria
    0.61
     domande
    0.61
     seeding
    0.60
    ">-->
    0.60
     packer
    0.60
    larınız
    0.59
    Activations Density 0.032%

    No Known Activations