© 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. 14384
    Prev
    Next
    INDEX
    Explanations

    legal terms like estoppel and indemnity

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

    The neuron activates on numeric expressions (especially standalone numbers and decimal values).

    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
    (
    -2.45
    i
    -2.30
    and
    -2.30
    y
    -2.09
    al
    -2.09
     akin
    -2.00
    mpagne
    -1.98
    י
    -1.95
     his
    -1.94
     verheir
    -1.90
    POSITIVE LOGITS
    踅
    2.50
     nuest
    2.30
     奶油
    2.19
    ização
    2.06
     油画
    2.03
     perdon
    2.02
    zja
    1.99
     cartier
    1.98
     oca
    1.97
    之为
    1.95
    Activations Density 0.002%

    No Known Activations