© 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. 3072
    Prev
    Next
    INDEX
    Explanations

    programming language

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

    The neuron fires on tokens that are items in enumerated option lists—e.g. numeric sizes, deadlines, colors, languages, or other product/configuration choices.

    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
    在我们
    0.70
     sonra
    0.69
    斯
    0.66
    予想
    0.63
    从
    0.63
    不久
    0.62
     언급
    0.62
    在我
    0.62
     знаю
    0.61
     sebab
    0.61
    POSITIVE LOGITS
     יותר
    0.72
     മറ്റ്
    0.68
    גים
    0.68
     කරයි
    0.67
    DICTION
    0.65
    Poppins
    0.65
    applique
    0.64
    ไม่มี
    0.64
     सामान्य
    0.62
     മാറ
    0.61
    Activations Density 0.002%

    No Known Activations