© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 124143
    Prev
    Next
    INDEX
    Explanations

    punitive and legal context

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

    The neuron activates whenever it sees TeX/LaTeX command tokens (i.e. sequences beginning with a backslash in math mode).

    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
     ktorí
    -2.36
     którzy
    -2.09
    ྜ
    -2.06
    "));
    -2.00
     namun
    -1.97
     kurie
    -1.94
     sondern
    -1.93
    ࿙
    -1.93
     tetapi
    -1.91
     apakah
    -1.91
    POSITIVE LOGITS
    _
    2.30
     .
    1.96
     zusätzliche
    1.95
     said
    1.94
    睺
    1.90
    one
    1.86
     duradero
    1.86
    or
    1.86
    撂
    1.82
     colgante
    1.81
    Activations Density 0.001%

    No Known Activations