© 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. 22-GEMMASCOPE-RES-131K
    7. 35056
    Prev
    Next
    INDEX
    Explanations

    code structure and imports

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

    The neuron activates on numeric literals (integer or floating‐point number tokens).

    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
     administratif
    -1.23
     garan
    -1.20
     patrie
    -1.19
     发表于
    -1.17
     plafon
    -1.11
     compromis
    -1.10
     петербург
    -1.09
     fluide
    -1.07
     revend
    -1.03
     haine
    -1.02
    POSITIVE LOGITS
     often
    1.18
     which
    1.15
     usually
    1.13
     their
    1.08
     project
    1.05
     each
    1.05
     information
    1.03
     whose
    1.02
     this
    1.01
     If
    0.99
    Activations Density 0.014%

    No Known Activations