© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 107220
    Prev
    Next
    INDEX
    Explanations

    self-reflection and writing

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

    This neuron responds to first-person self-references, especially the pronoun “I” and associated personal statements.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     flesta
    -0.98
    indeki
    -0.89
     kérd
    -0.88
    "}";
    -0.85
    utilisons
    -0.83
    zehnten
    -0.82
    IPPI
    -0.81
     notícias
    -0.80
     Definition
    -0.80
     samochod
    -0.77
    POSITIVE LOGITS
     writing
    4.97
     write
    4.06
    Writing
    3.45
    writing
    3.38
     wrote
    3.28
     Writing
    3.27
     typing
    3.20
    write
    3.11
     writes
    3.06
     escribir
    2.83
    Activations Density 0.091%

    No Known Activations