© 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. 127809
    Prev
    Next
    INDEX
    Explanations

    unofficial, Afro, FTSE, non

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

    This neuron fires on uncommon, low-frequency tokens—especially proper names, foreign or specialized terms, and other rare words.

    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
     they
    -2.80
     addition
    -2.59
      
    -2.52
    嘥
    -2.45
    諼
    -2.38
    alties
    -2.34
    䜣
    -2.28
    ℟
    -2.23
     agreeable
    -2.14
    le
    -2.13
    POSITIVE LOGITS
    After
    2.77
    with
    2.73
    According
    2.67
    ，
    2.66
    Despite
    2.61
    }$
    2.52
    的小
    2.50
    The
    2.48
    When
    2.47
    While
    2.45
    Activations Density 0.003%

    No Known Activations