© 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. 31-GEMMASCOPE-2-RES-16K
    7. 10769
    Prev
    Next
    INDEX
    Explanations

    appearance, appear, or variations

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

    The neuron fires on mentions of “appeared/appearing/appearance” often paired with large numeric values, i.e. tokens that report how many times something “appeared.”

    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_31_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
     Plast
    0.71
     शनि
    0.71
    Testables
    0.65
    ങ്ക
    0.64
     Arnav
    0.64
    សម្រាប់
    0.64
    kali
    0.64
    䊀
    0.63
    dd
    0.62
     Aq
    0.62
    POSITIVE LOGITS
    appear
    1.41
    appearance
    1.33
     appearance
    1.21
     appearances
    1.17
     apareció
    1.17
    appeared
    1.17
     Appear
    1.16
    appearing
    1.15
     appear
    1.14
    Appear
    1.13
    Activations Density 0.154%

    No Known Activations