© 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. 28498
    Prev
    Next
    INDEX
    Explanations

    being yourself

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

    The neuron is looking for first‐person, self‐referential statements (e.g. “I…,” “I am…”) indicating a confessional or personal‐voice context.

    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
     bedeuten
    -1.00
    openssl
    -1.00
     Địa
    -1.00
    ּוֹ
    -1.00
    יסטוריה
    -0.96
    aderie
    -0.95
    TouchListener
    -0.94
     oog
    -0.91
     المسؤول
    -0.91
     함께
    -0.90
    POSITIVE LOGITS
     conformity
    1.39
    Attempts
    1.15
    conform
    1.15
     identitas
    1.07
     diverse
    1.06
    pearance
    1.04
     what
    1.04
     unique
    1.02
     conforme
    1.02
     individuality
    0.99
    Activations Density 0.044%

    No Known Activations