© 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. 98274
    Prev
    Next
    INDEX
    Explanations

    other items or names

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

    The neuron fires on enumerations or item‐listing contexts—for example, when a phrase is listing brands, sites, dates, or other items in sequence.

    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
    cheduling
    -0.88
     złot
    -0.82
     Dinas
    -0.80
    IFICATION
    -0.79
     abenço
    -0.77
     déco
    -0.77
     lodging
    -0.77
     éché
    -0.76
    owali
    -0.76
    obacz
    -0.75
    POSITIVE LOGITS
     another
    1.05
     other
    1.05
    Another
    1.02
    虜
    0.97
    bbean
    0.89
     Another
    0.88
    同樣
    0.84
    另一个
    0.83
     Yorker
    0.83
     ..
    0.82
    Activations Density 0.109%

    No Known Activations