© 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. 22-GEMMASCOPE-RES-131K
    7. 73837
    Prev
    Next
    INDEX
    Explanations

    personal pronouns

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

    The neuron fires on direct-address pronouns—especially "you" (and sometimes inclusive "us") used to speak to the reader or audience.

    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
     /
    -1.55
     it
    -1.45
     If
    -1.38
    をします
    -1.35
     jotka
    -1.34
    為に
    -1.34
    After
    -1.32
    ができます
    -1.30
    無い
    -1.29
    Other
    -1.28
    POSITIVE LOGITS
     to
    2.23
     into
    1.84
    seemed
    1.61
     排骨
    1.59
    
    1.53
     foton
    1.52
     菇
    1.52
     frança
    1.49
    などを
    1.45
     素描
    1.45
    Activations Density 0.037%

    No Known Activations