© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 72171
    Prev
    Next
    INDEX
    Explanations

    the following

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

    The neuron fires on the common function‐word pattern of question stems—especially words like “Which,” “of,” and “the” in phrases such as “Which of the following.”

    oai_token-act-pair · o4-mini
    Triggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
     状态
    -1.54
    zember
    -1.52
     it
    -1.51
     who
    -1.47
    ''''
    -1.42
     was
    -1.41
     Jovi
    -1.40
    comings
    -1.39
     配件
    -1.38
    ſſen
    -1.36
    POSITIVE LOGITS
    </em>
    2.02
     -
    1.80
     چیست
    1.57
    CHREIB
    1.56
    álise
    1.52
    Leírás
    1.52
    geral
    1.45
    人たち
    1.44
    .
    1.41
     całej
    1.40
    Activations Density 0.048%

    No Known Activations