© 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. 84507
    Prev
    Next
    INDEX
    Explanations

    modern decor, promotion, accuracy

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

    The neuron responds to runs of the same short phrase repeated over and over (e.g. “Modern Living Room Decor” appearing in repeated succession).

    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
    Vaata
    -0.98
     détru
    -0.97
     több
    -0.96
     armado
    -0.96
     swojego
    -0.91
    菩萨
    -0.91
    beyaz
    -0.89
     Responses
    -0.89
    фильм
    -0.89
    supra
    -0.87
    POSITIVE LOGITS
     vanta
    0.88
     Kontakte
    0.76
    bergs
    0.73
     Interesse
    0.72
     belangrijk
    0.71
    معرفی
    0.70
    MST
    0.70
     penggemar
    0.70
     prend
    0.69
     balan
    0.69
    Activations Density 0.135%

    No Known Activations