© 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. 28547
    Prev
    Next
    INDEX
    Explanations

    qualities of service and products

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

    The neuron fires on positive evaluative words—mainly service‐ or product‐quality adjectives and related descriptors conveying benefits (e.g. realistic, affordable, compassionate, individualized).

    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
    そうですが
    -1.13
    zahlt
    -1.13
     sema
    -1.05
    がありますが
    -1.04
     nível
    -1.02
    amblea
    -1.02
     という
    -1.02
    いは
    -1.01
     utilizados
    -1.00
     associé
    -1.00
    POSITIVE LOGITS
     and
    1.91
     when
    1.23
     our
    1.14
    ‼
    1.13
     What
    1.10
     what
    1.09
     This
    1.09
     When
    1.08
    anthes
    1.07
     There
    1.04
    Activations Density 0.039%

    No Known Activations