© 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. 8826
    Prev
    Next
    INDEX
    Explanations

    optional arguments

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

    The neuron fires on tokens that end a paragraph or line block—that is, it detects end-of-segment (paragraph/sentence-final) tokens.

    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
    หลากหลาย
    0.86
    annya
    0.83
    ро
    0.82
    ه
    0.82
    ฟ
    0.82
     molte
    0.79
    <0xAB>
    0.79
    ถี่
    0.79
     użytk
    0.78
     इत्या
    0.77
    POSITIVE LOGITS
    t
    0.82
     químicos
    0.81
     Watching
    0.80
     Predicting
    0.77
    tar
    0.75
    sters
    0.75
    sos
    0.74
    svm
    0.73
    sab
    0.73
     Flexibility
    0.72
    Activations Density 0.000%

    No Known Activations