© 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. 31-GEMMASCOPE-2-RES-16K
    4. 10312
    Prev
    Next
    INDEX
    Explanations

    years, numbers, and dates

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

    The neuron fires on numeric tokens that denote years or date‐like four‐digit numbers in the text.

    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
    8
    1.00
    7
    0.86
     ২০২২
    0.84
     ২০২৩
    0.79
    6
    0.77
     ८
    0.75
    𝟖
    0.73
     ఇటీవల
    0.72
     ৮
    0.71
     ২০২১
    0.71
    POSITIVE LOGITS
    絓
    0.61
     contractile
    0.58
     gingerbread
    0.57
    ০১
    0.57
    
    0.57
     خمس
    0.56
    合约
    0.53
    開幕
    0.52
     ምክንያት
    0.52
    нг
    0.52
    Activations Density 0.242%

    No Known Activations