© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 7116
    Prev
    Next
    INDEX
    Explanations

    paved the way

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

    This neuron activates on formal/legal preamble phrases that reserve rights or establish conditions, such as “reserve(s) the right to …” and “pave(s) the way for ….”

    oai_token-act-pair · o4-miniTriggered 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
    Not in Any Lists

    No Comments

    Negative Logits
     a
    -1.73
    哢
    -1.59
    gyz
    -1.58
     жовтня
    -1.58
    tvguide
    -1.55
     修理
    -1.55
    知道了
    -1.54
     our
    -1.54
     Constitucional
    -1.54
     cápsulas
    -1.53
    POSITIVE LOGITS
    4
    2.30
    8
    2.25
    6
    2.22
    1
    2.19
    9
    2.08
    5
    2.06
    2
    1.99
    3
    1.99
    c
    1.80
     to
    1.67
    Activations Density 0.007%

    No Known Activations