© 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. 48182
    Prev
    Next
    INDEX
    Explanations

    Class, package, tea, Q:

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

    The neuron specifically activates on numeric tokens (i.e. sequences of digits) 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_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
     PROPOSED
    -0.88
    かし
    -0.86
    mappedBy
    -0.85
    һы
    -0.85
     oblong
    -0.83
     hidratante
    -0.81
     الشيخ
    -0.77
    🇬
    -0.76
     少女
    -0.75
     RECOMMENDED
    -0.73
    POSITIVE LOGITS
    语法
    0.85
     Often
    0.80
     Column
    0.78
    指令
    0.75
     Article
    0.75
     often
    0.74
     lelaki
    0.73
    ет
    0.70
     كافة
    0.69
     Syntax
    0.68
    Activations Density 0.008%

    No Known Activations