© 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. 10473
    Prev
    Next
    INDEX
    Explanations

    *

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

    The neuron fires strongly on tokens that contain numbers or numeric‐like strings (e.g. phone numbers, “234.3928”, “1”), as well as on URL prefixes (“http://”).

    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
    iu
    0.99
    ar
    0.98
    <em>
    0.96
    ir
    0.91
    +:
    0.91
    вото
    0.89
    رو
    0.85
    ಉ
    0.84
    erve
    0.83
    iyat
    0.82
    POSITIVE LOGITS
     لكن
    0.91
    теру
    0.87
     LMB
    0.87
    請問
    0.85
    рка
    0.84
    方式
    0.84
    startswith
    0.82
     وهذه
    0.82
    其
    0.82
     OMB
    0.80
    Activations Density 0.029%

    No Known Activations