© 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. 22-GEMMASCOPE-RES-131K
    7. 128343
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on occurrences of the “remove(” function call syntax in code.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    programming functions and variables

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    Not in Any Lists

    No Comments

    Negative Logits
     a
    -1.32
     unmittelbar
    -1.27
     mehrfach
    -1.24
     Everything
    -1.23
     zukünf
    -1.23
     häufig
    -1.20
     grundsätzlich
    -1.20
    的精神
    -1.17
     experiences
    -1.16
     atividades
    -1.16
    POSITIVE LOGITS
    Toeschouwers
    1.46
     Appellants
    1.40
    Даже
    1.38
    够了
    1.34
    Réponse
    1.32
    ranie
    1.30
    薗
    1.29
    ☞
    1.27
    fflush
    1.25
     TAMBIÉN
    1.25
    Activations Density 0.012%

    No Known Activations