© 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. 68080
    Prev
    Next
    INDEX
    Explanations

    news articles, code, questions

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

    This neuron fires on news‐style headings and metadata labels (e.g. article titles, section headers like “RELATED ARTICLES,” and image captions such as “Image: Getty”).

    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
    なのですが
    -0.98
    ??)
    -0.93
    ụp
    -0.93
     our
    -0.93
     suéter
    -0.90
     esqueleto
    -0.88
     összefoglal
    -0.87
    着头
    -0.87
     szól
    -0.86
    ấc
    -0.85
    POSITIVE LOGITS
    ，
    1.11
     зру
    0.98
     بخوانید
    0.94
    December
    0.93
     lmao
    0.92
    ást
    0.91
    。
    0.91
     સ્
    0.90
    Black
    0.89
    <strong>
    0.89
    Activations Density 0.000%

    No Known Activations