© 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. 31-GEMMASCOPE-2-RES-16K
    7. 2605
    Prev
    Next
    INDEX
    Explanations

    1

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

    This neuron fires on blank or whitespace-only tokens marking paragraph or section breaks.

    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
    另一种
    0.67
    Another
    0.64
    另一个
    0.64
     Another
    0.63
     আরেক
    0.63
     மற்றொரு
    0.61
    另一
    0.61
     सार्थक
    0.60
     இன்னொரு
    0.59
    の通販
    0.59
    POSITIVE LOGITS
    Firstly
    1.60
     firstly
    1.59
     ①
    1.58
    一是
    1.56
    ①
    1.55
     Firstly
    1.54
    ⒈
    1.52
     اولا
    1.51
     पहला
    1.41
    ❶
    1.41
    Activations Density 0.709%

    No Known Activations