© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 11589
    Prev
    Next
    INDEX
    Explanations

    This neuron selectively activates on occurrences of the substring “kir” (often in names or brand/model tokens).

    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
    ésar
    0.60
     reprehenderit
    0.55
    FER
    0.52
    argu
    0.51
    **)
    0.48
     আশা
    0.48
    Estados
    0.48
     '<
    0.48
    優先
    0.47
    resser
    0.47
    POSITIVE LOGITS
    ўнай
    0.57
     উড়িয়ে
    0.55
     బిల్
    0.55
    maier
    0.55
     offert
    0.53
     करत
    0.53
    oule
    0.52
    কাতা
    0.51
     spacetime
    0.51
     lensing
    0.51
    Activations Density 0.127%

    No Known Activations