© 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. 79709
    Prev
    Next
    INDEX
    Explanations

    confirmations and questions

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

    The neuron detects discourse markers and interjections (e.g. “Yes,” “Well,” “Hmm,” “Hell,” etc.) that open or emphasize a comment.

    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
     advogado
    -1.34
     również
    -1.25
     juiz
    -1.24
     my
    -1.18
    もとも
    -1.17
     I
    -1.13
    𝑈
    -1.13
    其他的
    -1.09
     reported
    -1.08
     Ursachen
    -1.07
    POSITIVE LOGITS
    淆
    1.21
     计算
    1.04
    黃
    1.02
     mathvariant
    1.01
    struzioni
    1.00
     عليهم
    1.00
    吧
    1.00
    чков
    0.99
    它
    0.99
     idő
    0.98
    Activations Density 0.061%

    No Known Activations