© 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. 2655
    Prev
    Next
    INDEX
    Explanations

    cl / cl / kl variations

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

    The neuron strongly fires on sequences of uppercase letters—that is, on tokens that form proper-name or acronym fragments.

    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
    س
    0.67
     நாள்
    0.61
    ='\
    0.61
    ="\
    0.59
    நாள்
    0.58
    ння
    0.57
    ਸ
    0.57
     المره
    0.57
     nominally
    0.57
     জিৎ
    0.57
    POSITIVE LOGITS
    felter
    0.86
    rimazole
    0.72
     클
    0.71
     cl
    0.69
    udius
    0.68
     Cl
    0.64
     CL
    0.64
    DLLEXPORT
    0.64
     Cli
    0.63
    苎
    0.62
    Activations Density 0.328%

    No Known Activations