© 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. 97
    Prev
    Next
    INDEX
    Explanations

    instructions and warnings

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

    The neuron fires on advisory or directive language—phrases urging or instructing people to do (or not do) something.

    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
    ဋ
    -1.23
    /}.
    -1.12
    vieve
    -1.07
     prêtres
    -1.05
     for
    -1.02
     ibland
    -1.02
    ཙ
    -1.02
     aunque
    -1.01
    }],
    -1.00
    дні
    -1.00
    POSITIVE LOGITS
    ﻻ
    1.38
     bemer
    1.34
    ⢕
    1.32
     verhind
    1.31
    Familien
    1.28
    anstal
    1.27
    Glück
    1.25
     déposer
    1.23
    他们
    1.21
     Wednesday
    1.21
    Activations Density 0.015%

    No Known Activations