© 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. 24-GEMMASCOPE-2-RES-16K
    7. 12627
    Prev
    Next
    INDEX
    Explanations

    oldest continuously measured operating data

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

    The neuron fires most strongly on words that immediately follow numeric or measurement expressions (e.g. tokens right after digits, ranges, ratings, or unit counts).

    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
     ved
    0.89
     segundo
    0.85
     inicio
    0.84
    베
    0.84
     primeiro
    0.83
     após
    0.83
     primeira
    0.81
     membro
    0.81
    ن
    0.81
    PR
    0.80
    POSITIVE LOGITS
    用
    0.79
     horribly
    0.72
    руу
    0.72
     Chapters
    0.70
    acquired
    0.70
    ктор
    0.70
     Sites
    0.69
    телни
    0.69
     Efficient
    0.68
     INTRODUCTION
    0.68
    Activations Density 0.000%

    No Known Activations