© 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. 64780
    Prev
    Next
    INDEX
    Explanations

    list delimiters and separators

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

    Explanation of neuron 4 behavior: the main thing this neuron does is find numeric tokens (digits and numbers) in code or data tables.

    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
    Tél
    -1.39
    也将
    -1.29
    Parfait
    -1.27
    Extrait
    -1.26
     OPPORT
    -1.25
    Traduction
    -1.25
    Transk
    -1.24
    Détails
    -1.23
    Publié
    -1.20
     luxem
    -1.20
    POSITIVE LOGITS
    قال
    0.96
    You
    0.96
     möchtest
    0.95
     }^{
    0.94
    For
    0.93
                 
    0.91
    getReference
    0.90
    sendLogPayload
    0.89
     Gerechtigkeit
    0.89
     منذ
    0.89
    Activations Density 0.081%

    No Known Activations