© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 67927
    Prev
    Next
    INDEX
    Explanations

    Q: commands and confessions

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

    The neuron detects liturgical or devotional prayer text—especially lines from traditional Catholic hymns and prayers.

    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
     that
    -1.56
     this
    -1.48
    Interestingly
    -1.35
     Interestingly
    -1.30
    !!!!
    -1.27
     ((
    -1.16
    !!
    -1.15
     very
    -1.13
     going
    -1.13
     now
    -1.13
    POSITIVE LOGITS
     ロゴ
    1.33
     tyrant
    1.29
    arm
    1.27
    ir
    1.27
    ể
    1.25
    ara
    1.20
     joyful
    1.20
    culable
    1.19
     ストレッチ
    1.18
    cken
    1.17
    Activations Density 0.015%

    No Known Activations