© 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. 102396
    Prev
    Next
    INDEX
    Explanations

    help your business

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

    The neuron responds to personal or relational words—especially first- and second-person pronouns (e.g. “you,” “my”) and kinship terms like “family”—marking when the text speaks from or to a personal perspective.

    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
    '
    -2.03
    ſelves
    -1.66
    了你
    -1.56
     for
    -1.48
    ём
    -1.42
    中医
    -1.39
     permitió
    -1.37
    丨
    -1.31
    чня
    -1.30
     maduro
    -1.30
    POSITIVE LOGITS
    ↵↵
    2.28
    5
    2.16
    7
    2.02
    4
    1.79
    他の
    1.77
    9
    1.71
     другим
    1.67
     =
    1.66
    8
    1.63
    no
    1.62
    Activations Density 0.036%

    No Known Activations