© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 6105
    Prev
    Next
    INDEX
    Explanations

    race, ethnicity, religion, gender

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

    This neuron isn’t picking out any linguistic feature of the words themselves but rather their position in the input: it reliably spikes on tokens that occur around the same absolute sequence index.

    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
     ഡ
    0.95
     accentuated
    0.94
     tiempo
    0.91
     கிஷ
    0.87
    ՝
    0.84
     conecta
    0.83
    ﺨ
    0.82
     anhydrous
    0.82
    নের
    0.81
    ﮑ
    0.81
    POSITIVE LOGITS
    ES
    0.79
    ούν
    0.72
     Screenshots
    0.72
    ard
    0.70
    禍
    0.69
    ству
    0.68
     Histogram
    0.68
    ある
    0.67
     Travelling
    0.67
    újt
    0.66
    Activations Density 0.000%

    No Known Activations