© 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. 34-GEMMASCOPE-RES-131K
    7. 111048
    Prev
    Next
    INDEX
    Explanations

    .NET code

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

    The neuron responds to numeric tokens (standalone digits and numbers, such as version numbers).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     its
    -0.87
     several
    -0.78
    ことができる
    -0.78
    ところに
    -0.77
     Learn
    -0.76
    しております
    -0.73
    うまい
    -0.71
    learn
    -0.70
     Normally
    -0.70
    图片
    -0.69
    POSITIVE LOGITS
     TED
    0.97
    š
    0.94
    Forbes
    0.92
     McQu
    0.92
     près
    0.90
    ırken
    0.89
     '':
    0.85
     desir
    0.85
     långt
    0.85
    USC
    0.84
    Activations Density 0.001%

    No Known Activations