© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 14030
    Prev
    Next
    INDEX
    Explanations

    questions about code

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

    The neuron is detecting numeric literals (especially numbers with decimals) 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
     in
    -1.65
     or
    -1.48
     eleições
    -1.12
     what
    -1.11
     should
    -1.09
     during
    -1.06
     tỏi
    -1.04
     упаковка
    -1.01
     our
    -0.99
     for
    -0.97
    POSITIVE LOGITS
    𝒙
    1.05
     choses
    1.04
     montrer
    1.03
     شوید
    1.02
     '').
    1.02
    を見ながら
    0.98
    Ikr
    0.98
     evtl
    0.97
     commandés
    0.96
    ままで
    0.96
    Activations Density 0.000%

    No Known Activations