© 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. 34-GEMMASCOPE-RES-131K
    7. 108673
    Prev
    Next
    INDEX
    Explanations

    specific named entities following "the"

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

    The neuron strongly activates on numeric tokens (digits and 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_34/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
     one
    -1.53
     certainly
    -1.51
     only
    -1.51
     basically
    -1.45
     probably
    -1.44
     definitely
    -1.42
     mainly
    -1.40
     just
    -1.39
     what
    -1.33
     this
    -1.32
    POSITIVE LOGITS
     Añade
    1.52
    !\!\
    1.35
    Likewise
    1.33
     ซึ่ง
    1.30
     Atención
    1.29
    Which
    1.27
     calificaciones
    1.27
    Similarly
    1.26
    纰
    1.23
     Tienes
    1.21
    Activations Density 0.191%

    No Known Activations