© 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. 12530
    Prev
    Next
    INDEX
    Explanations

    Clar followed by "itin", "ins", or "ifai"

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

    This neuron selectively activates on tokens beginning with the character sequence “Clar,” i.e. words or names starting with “Clar.”

    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
     Classified
    0.62
    MOR
    0.62
     Ecole
    0.57
     LEVEL
    0.57
     Maser
    0.57
     හොඳ
    0.56
     Slavery
    0.56
    왼
    0.56
     OMG
    0.55
    ૫
    0.55
    POSITIVE LOGITS
     പ്രവര്‍ത്തന
    0.59
     covert
    0.54
    entario
    0.52
    enda
    0.50
     Sand
    0.48
    なんて
    0.48
    tram
    0.48
     sand
    0.48
     করুন
    0.47
    시면
    0.47
    Activations Density 0.001%

    No Known Activations