© 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. 40220
    Prev
    Next
    INDEX
    Explanations

    praise for individuals

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

    The neuron detects direct second-person address in testimonial or praising statements—specifically sentences that speak to “you.”

    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
    міні
    -0.98
     then
    -0.93
     berkomunikasi
    -0.90
     their
    -0.90
     kesulitan
    -0.89
     when
    -0.89
    setFirstName
    -0.88
    ßig
    -0.87
    ✤
    -0.87
    tamanya
    -0.86
    POSITIVE LOGITS
     should
    1.48
     deserve
    1.34
     deserves
    1.31
     seem
    1.25
     sounds
    1.23
    Should
    1.16
     certamente
    1.13
     powinien
    1.09
     seems
    1.06
     powinna
    1.05
    Activations Density 0.016%

    No Known Activations