© 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. 84469
    Prev
    Next
    INDEX
    Explanations

    politician, politico, politicize, poliomyelitis, apolipoprotein

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

    the word root "polit" as it appears in words like "political", "politician", and "politics".

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron fires on occurrences of the subword “poli,” marking terms related to politics or politicians.

    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
     
    -3.75
    ’
    -3.48
     You
    -2.97
     To
    -2.73
    除此之外
    -2.70
    と言って
    -2.53
     “
    -2.45
     Now
    -2.42
     Its
    -2.42
     So
    -2.39
    POSITIVE LOGITS
    龱
    3.58
    鋱
    3.50
    }$
    3.17
    畚
    3.13
     pikir
    3.08
    垍
    3.06
    禤
    3.02
    旂
    3.00
     magasiner
    3.00
    ⌵
    2.98
    Activations Density 0.003%

    No Known Activations