© 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. 71374
    Prev
    Next
    INDEX
    Explanations

    opening parentheses

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

    The neuron strongly activates on standalone three-digit numeric tokens (e.g. “300,” “296,” “292,” “288”).

    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
     '
    -3.27
    </em>
    -3.00
     c
    -2.80
     is
    -2.75
     of
    -2.52
     _
    -2.50
     et
    -2.50
    </h4>
    -2.39
     $
    -2.38
     n
    -2.31
    POSITIVE LOGITS
    銲
    2.59
    there
    2.47
    鶺
    2.45
    when
    2.44
    Dengan
    2.42
     插画
    2.39
    錼
    2.36
    
    2.34
    踔
    2.31
     maría
    2.30
    Activations Density 0.006%

    No Known Activations