© 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. 13707
    Prev
    Next
    INDEX
    Explanations

    The neuron fires strongly on the very first content token of each new text segment (i.e. it marks the start of a section).

    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
    ād
    0.83
    ūr
    0.79
    aan
    0.76
    ṇ
    0.76
     Такая
    0.72
     웨
    0.71
    ē
    0.71
     edición
    0.71
    ōu
    0.70
    aterdag
    0.70
    POSITIVE LOGITS
    isRequired
    0.79
    我又
    0.79
    韩
    0.77
    粹
    0.77
    什么是
    0.74
     décerné
    0.72
    ین
    0.70
    价
    0.70
    赵
    0.70
    ান্ট
    0.69
    Activations Density 0.033%

    No Known Activations