© 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. 7806
    Prev
    Next
    INDEX
    Explanations

    Enron, None, UI

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

    The neuron activates on numeric tokens, especially numerical measurements and values (including integers and decimals).

    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
     Bhai
    0.79
     nimeni
    0.76
     Dalit
    0.74
     Terbaik
    0.73
     unglaublich
    0.73
     Bollywood
    0.71
     billionaire
    0.71
    𝙨
    0.71
     Millennials
    0.70
     hoodies
    0.70
    POSITIVE LOGITS
    '.
    0.68
    .'
    0.63
    '-
    0.63
    贮
    0.56
     moisture
    0.56
     stairway
    0.55
    <0x04>
    0.54
     জলাবদ্ধ
    0.54
    ,'
    0.54
    .—
    0.54
    Activations Density 0.100%

    No Known Activations