© 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. 1852
    Prev
    Next
    INDEX
    Explanations

    code structure and operations

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

    This neuron activates on multi‐part identifiers or variable/function names in snake_case that contain one or more underscores.

    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
    -
    0.55
    ,
    0.45
    i
    0.43
     the
    0.42
    =
    0.42
     -
    0.41
    j
    0.39
    ُ
    0.38
    .
    0.38
     stre
    0.38
    POSITIVE LOGITS
    এর
    0.55
    鼻子
    0.45
    小孩
    0.44
    এ
    0.44
    风险
    0.43
    在意
    0.43
    <unused336>
    0.43
    <unused411>
    0.42
     बच्ची
    0.42
    ઉન
    0.42
    Activations Density 0.001%

    No Known Activations