© 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. 65570
    Prev
    Next
    INDEX
    Explanations

    numbers and code keywords

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

    The neuron is primarily detecting numeric tokens (i.e. sequences of digits representing numbers).

    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
     on
    -1.11
     giving
    -0.95
     three
    -0.93
     gave
    -0.91
     makeshift
    -0.90
    職人
    -0.89
     people
    -0.89
     it
    -0.85
     two
    -0.84
     S
    -0.83
    POSITIVE LOGITS
     kangaroo
    1.02
     ZEN
    1.01
     exprimer
    0.98
    ecie
    0.98
    LOGICAL
    0.96
    nachweise
    0.95
    Calla
    0.95
     curent
    0.94
     AUS
    0.94
     Badan
    0.94
    Activations Density 0.049%

    No Known Activations