© 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. 6439
    Prev
    Next
    INDEX
    Explanations

    method signatures with throws

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

    The neuron detects blocks of pseudo-Latin “Lorem ipsum”–style placeholder 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_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
    ){
    0.63
    ){}
    0.61
    UInt
    0.59
     =>
    0.56
    fromCharCode
    0.55
     ())
    0.54
    бою
    0.52
    ){//
    0.52
    વન
    0.52
    =>$
    0.51
    POSITIVE LOGITS
     throws
    0.83
    throws
    0.83
     IOException
    0.62
     develops
    0.61
     burns
    0.59
    IOException
    0.59
     revital
    0.59
     qualifies
    0.58
     猫
    0.55
    elius
    0.54
    Activations Density 0.015%

    No Known Activations