© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 13633
    Prev
    Next
    INDEX
    Explanations

    empty

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

    The neuron fires on tokens that represent paragraph or line breaks (including indent/whitespace markers) delimiting sections of the text.

    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
     调用
    0.79
    调用
    0.76
    ):["
    0.70
     нажа
    0.68
     进行
    0.68
    初始化
    0.66
    wParam
    0.66
    localObject
    0.64
    تی
    0.63
    WIDTH
    0.63
    POSITIVE LOGITS
     decarbon
    0.74
     NPCs
    0.68
     reimag
    0.67
     🥰
    0.67
     लष्
    0.66
     sociales
    0.66
     souvenirs
    0.64
     societal
    0.64
     नव्या
    0.63
     사회
    0.63
    Activations Density 0.141%

    No Known Activations