© 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. 4451
    Prev
    Next
    INDEX
    Explanations

    explicit content or harmful requests

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

    The neuron chiefly responds to numeric tokens (i.e. numbers or number‐like strings).

    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
     armature
    0.71
     откуда
    0.69
    <unused33>
    0.67
     andRow
    0.65
     причем
    0.65
    आईओएस
    0.65
    વાડી
    0.64
    thebetter
    0.64
     ด่า
    0.64
     fichero
    0.64
    POSITIVE LOGITS
     [
    0.61
    <em>
    0.59
    迷
    0.58
    <strong>
    0.54
    強力
    0.54
    再次
    0.54
    等
    0.54
    自
    0.53
    ↵↵
    0.53
    使用
    0.52
    Activations Density 0.028%

    No Known Activations