© 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-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 9760
    Prev
    Next
    INDEX
    Explanations

    referring to multiple things

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

    The neuron activates on third-person object pronouns (e.g. “it,” “them”).

    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
     unieke
    -2.39
    颡
    -2.23
     kovo
    -2.19
    曏
    -2.19
     verschillen
    -2.19
     gedeelte
    -2.16
     and
    -2.14
    睺
    -2.11
    wachen
    -2.08
     oorspron
    -2.06
    POSITIVE LOGITS
    r
    3.20
    l
    2.88
    se
    2.64
    ter
    2.55
    n
    2.53
     sorts
    2.52
    v
    2.52
    ak
    2.50
    í
    2.41
    ang
    2.39
    Activations Density 0.015%

    No Known Activations