© 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. 14025
    Prev
    Next
    INDEX
    Explanations

    Politics, country, potential energy

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

    This neuron tends to pick out the first meaningful content word (a non‐stopword noun, verb, or adjective) at the start of each segment or sentence.

    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
    t
    0.76
    s
    0.76
    m
    0.73
    vý
    0.73
    щаем
    0.73
    '.'
    0.71
    vä
    0.71
     precautions
    0.71
    yoruz
    0.71
     ..."
    0.70
    POSITIVE LOGITS
     personalmente
    0.78
    ጶ
    0.78
    esterday
    0.77
     днів
    0.76
     complaining
    0.75
     CPP
    0.75
     justiciable
    0.74
     بھی
    0.74
     قبضہ
    0.74
    LLCATS
    0.73
    Activations Density 0.000%

    No Known Activations