© 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. 34-GEMMASCOPE-RES-131K
    7. 57887
    Prev
    Next
    INDEX
    Explanations

    information in tables

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

    Explanation of neuron 4 behavior: the main thing this neuron does is find references to tables (e.g. “shown in the table below,” “depicted in the following table”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    figsize
    -0.83
    FileVersion
    -0.78
    ILabel
    -0.78
    UserManager
    -0.77
     visualize
    -0.76
    Illustrated
    -0.75
     üzer
    -0.73
     featured
    -0.73
    ぐらい
    -0.72
     jugend
    -0.72
    POSITIVE LOGITS
     table
    5.22
     Table
    5.19
    Table
    4.97
     tables
    3.69
    table
    3.67
     Tables
    3.59
    TABLE
    3.39
    Tables
    3.19
     TABLE
    3.19
     табли
    2.86
    Activations Density 0.010%

    No Known Activations