© 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. 24-GEMMASCOPE-2-RES-16K
    7. 1504
    Prev
    Next
    INDEX
    Explanations

    explanation: `*` followed by text

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

    The neuron fires on isolated, content‐bearing words—especially thematic nouns or adjectives like “fun,” “vintage,” or “book.”

    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_24_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
    ል
    0.74
     разрабаты
    0.73
     Franck
    0.72
     fortunate
    0.71
     Каждый
    0.70
     Cálculo
    0.70
     разработан
    0.69
     организова
    0.68
     باسک
    0.68
     departmental
    0.66
    POSITIVE LOGITS
    CAR
    0.87
    enthic
    0.82
    akang
    0.79
    conten
    0.79
    ގ
    0.79
    DEN
    0.77
    ethylsulfanyl
    0.75
    intens
    0.74
    req
    0.73
     відчу
    0.73
    Activations Density 0.000%

    No Known Activations