© 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. 8506
    Prev
    Next
    INDEX
    Explanations

    main function entry point

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

    The neuron strongly activates on numeric data tokens—especially floating‐point numbers or similar numeric references (e.g. measurement values, page/volume numbers, DOIs).

    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
    includes
    0.64
    Laravel
    0.63
    TikTok
    0.61
    AnyCPU
    0.59
    Bit
    0.55
    ChatGPT
    0.55
    LinkedIn
    0.55
    ​​​​
    0.54
    Adobe
    0.54
    Bitmap
    0.53
    POSITIVE LOGITS
     treatise
    0.57
     पंतप्रधान
    0.57
     começo
    0.57
     inizio
    0.55
     prologue
    0.55
     primeras
    0.54
     начали
    0.54
    开始了
    0.54
     pre
    0.54
     beginnen
    0.54
    Activations Density 0.009%

    No Known Activations