© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 71720
    Prev
    Next
    INDEX
    Explanations

    programming syntax and keywords

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

    This neuron consistently activates on numeric literals (tokens consisting of numbers), i.e. it finds numbers in the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    }),
    
    -1.38
     ocho
    -1.22
    }));
    -1.19
    ']),
    -1.17
    -};
    -1.16
     khawatir
    -1.13
     трябва
    -1.13
     climático
    -1.13
    šie
    -1.11
     Dě
    -1.08
    POSITIVE LOGITS
    le
    0.97
    qu
    0.97
     peculiarity
    0.96
     compri
    0.93
    8
    0.93
    er
    0.91
     ravis
    0.90
    ковая
    0.90
     lamentable
    0.88
    N
    0.88
    Activations Density 0.015%

    No Known Activations