© 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. 24-GEMMASCOPE-2-RES-16K
    4. 518
    Prev
    Next
    INDEX
    Explanations

    what you're referring to

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

    The neuron is triggered by direct second-person addresses—phrases using “you” (and related forms) speaking straight to the reader.

    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
     oraz
    0.70
     மற்றும்
    0.68
     एवं
    0.66
     remarkably
    0.62
     számos
    0.61
    2
    0.60
     ,
    0.58
    6
    0.57
     بسیار
    0.57
     and
    0.56
    POSITIVE LOGITS
     solchen
    0.66
     irgende
    0.65
     그것
    0.64
     alguien
    0.63
     그냥
    0.63
     入っ
    0.62
    そこ
    0.61
    碜
    0.61
     มัน
    0.61
     진짜
    0.61
    Activations Density 12.193%

    No Known Activations