© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 4309
    Prev
    Next
    INDEX
    Explanations

    dates and times

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

    The neuron is highly tuned to recognize date‐ and time‐related numeric tokens (e.g. month/day/year timestamps).

    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
    attho
    0.63
    াটো
    0.62
    梩
    0.59
    itth
    0.59
    ్ఞ
    0.59
    代码
    0.59
    ProductName
    0.58
     idempot
    0.57
    ittings
    0.57
    wendungen
    0.56
    POSITIVE LOGITS
     a
    0.59
     Koy
    0.57
     daily
    0.55
     khi
    0.54
     Friday
    0.54
     Paulo
    0.54
     May
    0.52
     Josh
    0.52
     Leslie
    0.51
     Daha
    0.50
    Activations Density 0.022%

    No Known Activations