© 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. 15655
    Prev
    Next
    INDEX
    Explanations

    numbers and lists

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

    The neuron is triggered by large multi-digit numbers (e.g. four-digit year or year-like values) 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
    vislosti
    0.84
    ыр
    0.82
    льзу
    0.79
    を中心に
    0.79
    ឯ
    0.79
    勹
    0.78
     detd
    0.77
     determinados
    0.77
     Producing
    0.76
     Encoding
    0.76
    POSITIVE LOGITS
    but
    0.87
    ball
    0.76
    ingly
    0.75
    Okay
    0.75
    Awesome
    0.74
    joe
    0.69
    Cloud
    0.67
    Regardless
    0.67
    †
    0.67
    Tu
    0.66
    Activations Density 0.000%

    No Known Activations