© 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. 14097
    Prev
    Next
    INDEX
    Explanations

    VRBO, Q:, Eric Arnold, Kanye West, Over the

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

    The neuron activates on the “{” token that opens a new code block (i.e. an opening curly brace).

    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
     desde
    -1.08
    VIAF
    -1.06
    sery
    -1.05
    ktar
    -0.99
     depuis
    -0.98
    nées
    -0.94
     unequiv
    -0.93
    zegovina
    -0.91
    from
    -0.91
    тали
    -0.90
    POSITIVE LOGITS
    ecé
    1.10
     things
    1.10
     and
    1.05
     uñas
    0.99
     gewerb
    0.92
    彼が
    0.92
     wollen
    0.91
    鑾
    0.91
     Erinnerung
    0.91
    limin
    0.90
    Activations Density 0.000%

    No Known Activations