© 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. 3190
    Prev
    Next
    INDEX
    Explanations

    multi-lingual phrases

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

    This neuron activates on Latin words and phrases, signaling the presence of Latin-language text.

    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
     বেশিরভাগ
    0.86
    ROY
    0.84
    没有任何
    0.82
    DWORD
    0.78
    쐞
    0.78
    splitLength
    0.76
    Polynomial
    0.74
    仯
    0.73
    Ҥ
    0.73
    是一名
    0.73
    POSITIVE LOGITS
     atque
    0.66
    ✨
    0.63
    .
    0.63
     vivere
    0.61
     corporis
    0.60
     moderne
    0.59
     وَ
    0.59
     niemals
    0.59
     καὶ
    0.58
     quae
    0.58
    Activations Density 0.093%

    No Known Activations