© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 93373
    Prev
    Next
    INDEX
    Explanations

    biology, clinical, trademarks

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

    The neuron primarily activates on periods or dots in structured text, especially in technical documents and programming code.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron specifically fires on the standalone numeral “6.”

    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
    歺
    -1.75
    ܇
    -1.64
     gewel
    -1.61
     superbes
    -1.60
     Hinblick
    -1.57
    
    -1.57
     emblematic
    -1.54
    這些
    -1.53
     reagiert
    -1.52
     mawar
    -1.45
    POSITIVE LOGITS
     is
    2.11
     (
    1.77
    m
    1.70
    a
    1.67
    t
    1.67
     as
    1.57
    n
    1.56
     there
    1.53
    u
    1.51
    N
    1.47
    Activations Density 0.000%

    No Known Activations