© 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. 10-GEMMASCOPE-RES-131K
    7. 40881
    Prev
    Next
    INDEX
    Explanations

    numbers and colons

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

    The neuron activates on the character sequence “16,” detecting any token that contains the digits 1 and 6 together.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     entschieden
    -1.57
     about
    -1.51
    tamaris
    -1.40
    身的
    -1.38
     jenem
    -1.36
     aptly
    -1.34
     gefeiert
    -1.34
    ويد
    -1.33
    なんでも
    -1.33
     than
    -1.32
    POSITIVE LOGITS
    鹟
    1.72
     lección
    1.55
     herbe
    1.55
    鹩
    1.52
     flotte
    1.50
    姽
    1.48
    это
    1.45
     andere
    1.44
    鸲
    1.42
    raten
    1.41
    Activations Density 0.012%

    No Known Activations