© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 130199
    Prev
    Next
    INDEX
    Explanations

    This neuron detects the opening square‐bracket tokens used in code to unpack or assign multiple outputs (e.g. “[” in a multi‐output assignment).

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

    tuple assignment `[`

    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
     to
    -1.79
     aprobación
    -1.34
     continúa
    -1.29
     legislación
    -1.24
     conocido
    -1.22
     específicas
    -1.22
     esempi
    -1.21
     mempunyai
    -1.18
     gleichen
    -1.18
    
    -1.16
    POSITIVE LOGITS
    ,
    1.21
    🥉
    1.11
     one
    1.10
    Dimensioni
    1.10
     Weltkrieg
    1.08
    歲
    1.07
     lesquelles
    1.04
     börjar
    1.02
     _,
    1.02
     overall
    1.01
    Activations Density 0.003%

    No Known Activations