© 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. 10-GEMMASCOPE-RES-131K
    7. 59134
    Prev
    Next
    INDEX
    Explanations

    describing methods, results, or interpretations

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

    The neuron fires on those high‐level exposition verbs (e.g. “describe,” “present,” “show”) that introduce or summarize a paper’s contributions.

    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
    :
    -1.58
     a
    -1.53
    there
    -1.38
    it
    -1.37
    !
    -1.33
    i
    -1.32
     from
    -1.32
    a
    -1.30
     you
    -1.28
     D
    -1.27
    POSITIVE LOGITS
    ’”
    1.52
     Glaube
    1.50
    Leírás
    1.49
    sont
    1.48
    ’;
    1.48
     délais
    1.47
     高等
    1.45
    
    1.45
     Tetapi
    1.43
     NTIS
    1.42
    Activations Density 0.083%

    No Known Activations