© 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. 34024
    Prev
    Next
    INDEX
    Explanations

    provide services and support

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

    The main thing this neuron does is find numeric tokens (numbers and numerical expressions).

    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
     has
    -1.68
     after
    -1.63
     having
    -1.55
     people
    -1.50
     two
    -1.48
     it
    -1.45
     had
    -1.41
     by
    -1.41
     three
    -1.38
     as
    -1.38
    POSITIVE LOGITS
     themselves
    1.80
     arrêter
    1.42
    themselves
    1.41
     assister
    1.40
     nettoy
    1.40
    他们
    1.37
     célè
    1.34
     diriger
    1.30
     mereka
    1.25
     accompag
    1.25
    Activations Density 0.067%

    No Known Activations