© 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. 92565
    Prev
    Next
    INDEX
    Explanations

    comparing quantities

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

    The neuron strongly activates on standalone numeric literals (i.e. tokens that are 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
    锚
    -0.91
     became
    -0.90
    ността
    -0.88
     exquisitely
    -0.86
     censiti
    -0.81
    ˢ
    -0.79
     lange
    -0.79
     disemb
    -0.78
     revital
    -0.77
    hbs
    -0.77
    POSITIVE LOGITS
    ᅧ
    1.03
     vivace
    1.00
    Demografía
    0.97
     doctrina
    0.97
    憷
    0.96
     certaines
    0.94
     poë
    0.94
    typeof
    0.94
     económicos
    0.94
     typeof
    0.93
    Activations Density 0.003%

    No Known Activations