© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 9418
    Prev
    Next
    INDEX
    Explanations

    responsive mobile design

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

    The neuron selectively lights up on numeric tokens and punctuation used in technical or code contexts—especially digits, slashes, and version- or fraction-style number sequences (as seen in media-query and “1/0” style syntax).

    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
    frameNStart
    0.56
     அருக
    0.55
    indazole
    0.54
    avirus
    0.54
     terbentuk
    0.54
    葡萄
    0.53
     efecto
    0.52
    wheels
    0.52
    💒
    0.51
     ফ্লাই
    0.51
    POSITIVE LOGITS
    MEDIA
    0.98
     media
    0.98
     Media
    0.97
     MEDIA
    0.97
    Media
    0.96
     responsive
    0.93
    MediaQuery
    0.93
    媒体
    0.89
     médias
    0.89
    Responsive
    0.88
    Activations Density 0.051%

    No Known Activations