© 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. 54201
    Prev
    Next
    INDEX
    Explanations

    commitment to quality and service

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

    The neuron is highly sensitive to marketing superlatives and guarantee‐style benefit words—strong positive adjectives and verbs that promise quality, superiority, or assured performance.

    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
     but
    -1.29
     two
    -1.26
     estampado
    -1.14
     or
    -1.11
     tearDown
    -1.05
    𝘩
    -1.05
     zanah
    -1.05
    rientes
    -1.04
     enticing
    -1.04
     three
    -1.02
    POSITIVE LOGITS
     our
    1.46
     kesi
    1.38
     sako
    1.32
     each
    1.30
     miele
    1.27
     filosof
    1.27
     saan
    1.27
     mahd
    1.27
     kela
    1.26
     daga
    1.26
    Activations Density 0.032%

    No Known Activations