© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 20794
    Prev
    Next
    INDEX
    Explanations

    defining role or purpose

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

    This neuron detects high-salience framing words (e.g. primary, main, chief, sole, true, first, greatest) used in declarative statements of purpose, priority, or passion.

    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
     of
    -1.72
     they
    -1.50
     always
    -1.28
     your
    -1.27
     usually
    -1.26
     there
    -1.22
    aClass
    -1.21
     we
    -1.20
    罴
    -1.16
    PHeader
    -1.15
    POSITIVE LOGITS
    to
    1.46
    他の
    1.41
     beiden
    1.41
    ]
    1.40
    許多
    1.30
    }
    1.30
    as
    1.29
    ()
    1.27
     being
    1.27
     letzten
    1.26
    Activations Density 0.062%

    No Known Activations