© 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. 10882
    Prev
    Next
    INDEX
    Explanations

    its use or value

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

    The neuron fires strongly on the very first substantive token that appears in each new text segment, effectively marking the beginning of each document or paragraph.

    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
    ers
    0.89
    ERS
    0.89
    2
    0.81
    환
    0.80
     Kek
    0.79
    တ်
    0.78
    1
    0.78
    상
    0.77
     Nen
    0.76
    限
    0.75
    POSITIVE LOGITS
     aligns
    0.76
     কখনোই
    0.74
     separa
    0.73
     capitalise
    0.71
     sanitize
    0.70
     dehuman
    0.70
     align
    0.69
     capitalize
    0.68
     gameObject
    0.68
     receivables
    0.68
    Activations Density 0.000%

    No Known Activations