© 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. 71593
    Prev
    Next
    INDEX
    Explanations

    the following

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

    The neuron responds to list‐introduction phrases that announce or label a following set of items (e.g. “the following … are,” “documents must be submitted,” “systems are affected,” etc.).

    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
    スポンサーリンク
    -1.00
     abund
    -0.98
     otra
    -0.98
    Here
    -0.97
     muchas
    -0.96
     banyak
    -0.94
    сяг
    -0.93
     memang
    -0.89
     Haft
    -0.89
     doble
    -0.88
    POSITIVE LOGITS
    :
    1.48
     respectively
    1.32
     which
    1.23
     are
    1.18
     onsdag
    1.10
     säkert
    1.07
     informó
    1.05
     for
    1.05
     respective
    1.02
    vestidos
    1.02
    Activations Density 0.064%

    No Known Activations