© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 16207
    Prev
    Next
    INDEX
    Explanations

    interruption of expectation

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

    The neuron fires on social‐sharing or user‐action verbs and related “call‐to‐action” terms (e.g. share, posted, wear, gift, cycle begins, etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    statements indicating content being shared or posted online, typical of download or blog page boilerplate.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408

    generic web content and boilerplate text, particularly from app download pages, product descriptions, and blog posts.

    oai_token-act-pair · claude-4-5-sonnetTriggered 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
     troubleshoot
    0.89
    ান
    0.78
     chromatography
    0.74
    בודה
    0.74
     counterparts
    0.72
    ektion
    0.71
     tidal
    0.70
     procurement
    0.70
     sectarian
    0.68
     fields
    0.67
    POSITIVE LOGITS
    <0x90>
    0.77
     toată
    0.76
     ос
    0.70
     appré
    0.70
     చేశ
    0.70
    дие
    0.70
     эту
    0.70
     ঘোষিত
    0.70
     după
    0.69
     કરે
    0.68
    Activations Density 0.000%

    No Known Activations