© 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. 1141
    Prev
    Next
    INDEX
    Explanations

    multi-language non-English words

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

    The neuron activates on tokens in non-Latin scripts (e.g. Japanese katakana or Thai characters).

    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
    к
    0.42
    преде
    0.40
    выя
    0.40
    прос
    0.39
    assembl
    0.39
    ще
    0.39
    ӵ
    0.39
     descriptor
    0.38
     multitude
    0.38
     ذریع
    0.38
    POSITIVE LOGITS
    ชั่น
    0.61
    ियर
    0.58
    煐
    0.58
    해서
    0.56
     Pawar
    0.56
    ซ์
    0.55
    ค์
    0.55
     Kakkar
    0.54
    糖尿
    0.54
     Nava
    0.54
    Activations Density 0.000%

    No Known Activations