© 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. 12301
    Prev
    Next
    INDEX
    Explanations

    arrang / arrangement / GPT / bathroom

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

    This neuron strongly responds to uppercase alphanumeric codes and identifiers (e.g. product or grade codes like “DQ5”) and similar title‐style 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
    abort
    0.72
    ̿
    0.67
    йно
    0.66
    дый
    0.66
    aguchi
    0.66
    اريات
    0.65
    Artículo
    0.65
    avljen
    0.64
     þing
    0.63
    кую
    0.63
    POSITIVE LOGITS
     arrang
    0.80
    ↵
    0.78
     GL
    0.73
     point
    0.70
     table
    0.70
     toucher
    0.70
     di
    0.68
     compos
    0.67
     TW
    0.67
     by
    0.66
    Activations Density 0.006%

    No Known Activations