© 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. 22-GEMMASCOPE-RES-131K
    7. 26614
    Prev
    Next
    INDEX
    Explanations

    signed original copy

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

    The neuron responds strongly to numeric tokens (digits and numbers, such as years, quantities, or standalone numeral characters).

    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
     new
    -1.23
    -1.13
          
    -1.11
    begin
    -1.09
    $

    -1.08
            
    -1.05
     create
    -1.04
         
    -1.04
    Test
    -1.03
     into
    -1.02
    POSITIVE LOGITS
     fechar
    1.29
     rarer
    1.23
     rare
    1.22
     оригі
    1.15
     съдържа
    1.13
     réus
    1.12
     estuche
    1.11
     ricevere
    1.10
     organizaciones
    1.09
     sellado
    1.09
    Activations Density 0.017%
    No Known Activations