© 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-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 19300
    Prev
    Next
    INDEX
    Explanations

    things that existed

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

    The neuron detects words and phrases expressing the realization or discovery that something exists (e.g. “existed,” “exists,” “I didn’t even know,” “I had no idea”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     first
    -1.20
     while
    -1.19
     that
    -1.16
     sappi
    -1.09
     laurel
    -1.04
     lupo
    -1.03
    ما
    -1.02
     ahu
    -1.00
     号
    -0.99
    vação
    -0.98
    POSITIVE LOGITS
    こんなに
    1.45
    still
    1.34
    匂
    1.28
     wonderful
    1.27
    そも
    1.25
    eight
    1.24
     coloridas
    1.24
    躾
    1.23
    おかしい
    1.23
     rojos
    1.22
    Activations Density 0.070%

    No Known Activations