© 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. 13969
    Prev
    Next
    INDEX
    Explanations

    lists and dictionaries

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

    This neuron is tuned to detect square-bracket list or array literals, in particular the closing ‘]’ tokens that end lists.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
     is
    -0.94
     respiración
    -0.91
    ampire
    -0.91
    ímos
    -0.91
     colchón
    -0.90
     αλλά
    -0.90
    IENTS
    -0.89
    elor
    -0.89
    ̣ng
    -0.86
     permitirá
    -0.85
    POSITIVE LOGITS
    Ges
    0.96
    0.95
     campeão
    0.94
     connaissance
    0.93
     különböz
    0.89
     not
    0.88
     diesen
    0.88
    appBar
    0.86
    辈子
    0.86
     diretor
    0.85
    Activations Density 0.006%
    No Known Activations