© 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. 22-GEMMASCOPE-RES-131K
    7. 64308
    Prev
    Next
    INDEX
    Explanations

    Beckenheim, Ayenda, adhaý

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

    The neuron responds to longer, multi‐letter tokens—especially capitalized words that are parts of proper names or titles.

    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_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
    '
    -1.45
     only
    -1.40
    "
    -1.38
    C
    -1.24
    ,
    -1.23
     nahezu
    -1.21
    以前的
    -1.20
     "
    -1.17
    P
    -1.12
    昔の
    -1.11
    POSITIVE LOGITS
    ValueStyle
    1.70
    ReusableCell
    1.48
     และ
    1.47
    InputBorder
    1.47
    WidgetItem
    1.46
    ſſen
    1.41
    และ
    1.40
    adhyay
    1.38
    TargetException
    1.38
    titial
    1.36
    Activations Density 0.059%

    No Known Activations