© 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. 7755
    Prev
    Next
    INDEX
    Explanations

    Estrogen

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

    The neuron strongly activates on metadata‐style tokens and numeric identifiers—things like “Tags,” “References,” model years (e.g. “2018,” “2020”), and similar label or number markers.

    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
     λοι
    0.83
    ja
    0.77
     kere
    0.77
    nej
    0.76
    ahiye
    0.73
    ikia
    0.72
     തമ്മ
    0.72
    ε
    0.71
     estavam
    0.70
     arka
    0.69
    POSITIVE LOGITS
    ת
    0.87
     rbuf
    0.86
    +'
    0.86
    ל
    0.86
    ',\
    0.85
    FindAsync
    0.78
    ValueOf
    0.76
    짊
    0.75
     filesystem
    0.75
    рный
    0.75
    Activations Density 0.002%

    No Known Activations