© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 71238
    Prev
    Next
    INDEX
    Explanations

    code assignment and variables

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

    The neuron predominantly activates on numeric literals (digits and numbers) in the text.

    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
     especially
    -1.49
     particularly
    -1.30
     besonders
    -1.30
     whereas
    -1.17
     different
    -1.17
     although
    -1.13
     Besonders
    -1.11
    ၉
    -1.10
     особенно
    -1.04
     dankbar
    -1.04
    POSITIVE LOGITS
     demselben
    1.12
    gehenden
    1.05
    やってる
    0.98
    ódó
    0.98
     =
    0.94
     zei
    0.94
    تمبر
    0.93
     занимается
    0.93
    合わせる
    0.93
     приводит
    0.93
    Activations Density 0.005%

    No Known Activations