© 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. 876
    Prev
    Next
    INDEX
    Explanations

    for example

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

    This neuron activates on words at the very start of sentences (i.e. the first token after a sentence boundary).

    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
    톄
    0.62
    氧化
    0.60
    눙
    0.59
    DEFGHIJKLMNOP
    0.57
     চিকিৎ
    0.57
    的研究
    0.57
    裁剪
    0.56
    旪
    0.56
    聝
    0.55
    慓
    0.55
    POSITIVE LOGITS
     However
    0.77
    However
    0.73
     tetapi
    0.73
     however
    0.70
     I
    0.69
    ?
    0.69
     But
    0.69
     They
    0.68
    It
    0.67
     While
    0.66
    Activations Density 0.013%

    No Known Activations