© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 3575
    Prev
    Next
    INDEX
    Explanations

    whole thing

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

    The neuron specifically fires on the word “whole,” especially when it’s used emphatically (e.g. “the whole world,” “the whole thing”) to denote entirety.

    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
     γνω
    0.99
     სხვა
    0.96
     amelyek
    0.92
     ০
    0.92
    客様
    0.91
     கிலோ
    0.91
     тысяч
    0.89
     niektor
    0.88
     ktoré
    0.88
    ὦ
    0.88
    POSITIVE LOGITS
     entire
    1.21
     Entire
    0.96
     gamut
    0.93
    hearted
    0.92
    entire
    0.80
    ય
    0.79
    整个
    0.77
    Entire
    0.77
    こと
    0.75
    લ
    0.75
    Activations Density 0.027%

    No Known Activations