© 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. 104559
    Prev
    Next
    INDEX
    Explanations

    to include levels

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

    The neuron fires on prominent content words and key terms in headings, titles, or question‐title phrases.

    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
     when
    -1.09
     for
    -0.96
     Instyt
    -0.92
    濑
    -0.92
     Weiterbildung
    -0.90
     External
    -0.90
     one
    -0.89
    crown
    -0.88
    地域の
    -0.88
    级别的
    -0.88
    POSITIVE LOGITS
     oliveira
    1.11
     Reykjav
    0.99
    也很
    0.97
    也有
    0.95
    チャーシュー
    0.91
     KELLY
    0.91
     zaragoza
    0.91
    「
    0.88
     alemania
    0.88
     ARNOLD
    0.87
    Activations Density 0.279%

    No Known Activations