© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 13401
    Prev
    Next
    INDEX
    Explanations

    names like Charles Edward

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

    The neuron primarily fires on numeric tokens (numbers or numeric expressions) 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
    descargar
    -0.98
    깁
    -0.97
    Cześć
    -0.91
    Cevap
    -0.88
    Contr
    -0.85
     ravis
    -0.84
     ravages
    -0.83
    cerrar
    -0.83
    Partager
    -0.83
     poss
    -0.82
    POSITIVE LOGITS
     userRepository
    1.07
     userManager
    0.89
     objectMapper
    0.88
     continue
    0.84
    bsen
    0.81
    ISHMENT
    0.81
     serão
    0.80
    眼下
    0.80
     había
    0.80
    较大
    0.79
    Activations Density 0.019%

    No Known Activations