© 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. 22-GEMMASCOPE-RES-131K
    4. 80018
    Prev
    Next
    INDEX
    Explanations

    publisher names, company suffixes, or place names

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

    The neuron activates on publisher names (i.e. the proper‐noun strings naming publishing houses in bibliographic entries).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    :
    -1.16
    ciąż
    -1.10
     vollen
    -1.03
     Antrieb
    -1.01
     extremely
    -1.00
     exceptionally
    -1.00
    のだろう
    -0.99
    んでしょう
    -0.99
     remarkably
    -0.98
     significantly
    -0.97
    POSITIVE LOGITS
    دانلود
    1.16
    توضیحات
    1.13
     NACIONAL
    1.11
    معرفی
    1.09
    верь
    1.08
    there
    1.04
    Suara
    1.02
     MUNDO
    0.99
    쏟
    0.97
     while
    0.96
    Activations Density 0.004%

    No Known Activations