© 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. 12640
    Prev
    Next
    INDEX
    Explanations

    go imports

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

    The neuron appears completely inactive and does not reliably detect any particular pattern or token.

    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
     Leisure
    0.66
     Enlightenment
    0.63
    穿
    0.63
     Great
    0.61
    WILLIAM
    0.61
     Object
    0.61
     ‘‘
    0.59
     Elizabethan
    0.59
    你会
    0.58
    Great
    0.57
    POSITIVE LOGITS
    。(
    0.69
    presentableId
    0.67
     🙏
    0.66
    loride
    0.65
     যেহেতু
    0.64
     первого
    0.64
     (−
    0.64
    éro
    0.63
     گردد
    0.63
     första
    0.63
    Activations Density 0.000%

    No Known Activations