© 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. 15953
    Prev
    Next
    INDEX
    Explanations

    AI model documentation

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

    The neuron fires on capitalized tokens—proper nouns, acronyms, names, or sentence-initial uppercase words.

    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
    alloy
    0.80
    GHz
    0.80
    ب
    0.77
    Logging
    0.76
    GH
    0.74
    toggle
    0.72
    mode
    0.71
    locking
    0.70
    په
    0.70
    unsaturated
    0.69
    POSITIVE LOGITS
    ’
    0.76
     Aga
    0.73
     Tava
    0.72
     kiddos
    0.72
     Airbnb
    0.71
    രിച്ചു
    0.71
     couvert
    0.69
     Sheila
    0.69
     Khy
    0.69
     scammers
    0.68
    Activations Density 0.000%

    No Known Activations