© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 12221
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on code-style or technical identifiers—things like file paths, environment variables, build/log commands, and acronyms—rather than ordinary prose.

    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
     roupa
    0.88
     የሚያ
    0.83
     bgColor
    0.80
     decept
    0.77
    𝐃
    0.77
     ripping
    0.76
     препара
    0.73
     दख
    0.73
     magyar
    0.73
     mobilier
    0.73
    POSITIVE LOGITS
    रिडोर
    0.82
    currentIndex
    0.82
    randomIndex
    0.80
    但我
    0.79
    ArrayIndex
    0.77
    ByIndex
    0.76
     каждым
    0.76
    ++]
    0.75
    番号
    0.75
    序号
    0.75
    Activations Density 0.670%

    No Known Activations