© 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. 84869
    Prev
    Next
    INDEX
    Explanations

    curl and curly suffixes

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

    The neuron fires on any token containing the “cur…/curl…/curly…” sequence (e.g. “curly,” “curl,” “krøll,” etc.).

    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
     But
    -2.56
    袮
    -2.13
     nettsted
    -2.00
     medier
    -1.99
    CURLOPT
    -1.98
     erforderlichen
    -1.98
    鈈
    -1.95
     Ukraina
    -1.94
    蹓
    -1.93
     Presiden
    -1.91
    POSITIVE LOGITS
    ll
    2.63
    of
    2.34
    我先
    2.30
    艫
    2.25
    }$
    2.23
    I
    2.16
      
    2.09
     tempt
    2.08
    Y
    2.06
     ‘
    2.05
    Activations Density 0.007%

    No Known Activations