© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 9360
    Prev
    Next
    INDEX
    Explanations

    this point

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

    The neuron activates on tokens that form date- and time-stamp elements (hours, minutes, “am”/“pm,” and similar timestamp fragments).

    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
     at
    0.79
    于
    0.77
     于
    0.75
    於
    0.75
     atthe
    0.69
     ở
    0.67
    อยู่ที่
    0.63
     Ở
    0.61
     tại
    0.59
    AtIndex
    0.58
    POSITIVE LOGITS
     glance
    0.98
     level
    0.88
     juncture
    0.81
     expense
    0.76
     crossroads
    0.75
     behest
    0.75
    level
    0.74
     स्तर
    0.72
     discretion
    0.71
    expense
    0.65
    Activations Density 0.314%

    No Known Activations