© 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. 31-GEMMASCOPE-2-RES-16K
    7. 2806
    Prev
    Next
    INDEX
    Explanations

    terr, terra, ter, terror

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

    This neuron fires on the place name “Terengganu” (often split into token parts Ter-eng-gan-u).

    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
     результатов
    0.55
    yczny
    0.55
     conform
    0.54
     xer
    0.54
    teil
    0.53
    启用
    0.53
    ಮ್
    0.52
    注意的是
    0.51
    的区别
    0.51
     augment
    0.50
    POSITIVE LOGITS
     тер
    0.71
    pyridine
    0.69
    peuta
    0.67
     Тер
    0.65
    acotta
    0.65
     Ter
    0.64
    Ter
    0.64
    Terr
    0.63
    тера
    0.63
    itory
    0.60
    Activations Density 0.182%

    No Known Activations