© 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. 7114
    Prev
    Next
    INDEX
    Explanations

    most common

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

    The neuron fires on degree-modifying words—common intensifiers (like “very,” “most,” “ever”) and the negative prefix “non-.”

    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
    e
    0.79
    d
    0.77
    a
    0.75
    x
    0.72
    For
    0.71
    &
    0.70
    o
    0.69
    y
    0.67
    Enter
    0.67
    Cl
    0.65
    POSITIVE LOGITS
     reinvigor
    0.67
     idealistic
    0.64
    τροπ
    0.61
     undeniable
    0.60
     irrevocable
    0.59
     heartfelt
    0.59
    ियारों
    0.58
     adherents
    0.57
     utopian
    0.57
    <unused535>
    0.57
    Activations Density 0.075%

    No Known Activations