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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 1137
    Prev
    Next
    INDEX
    Explanations

    file names, code snippets

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

    The neuron is strongly activated by tokens containing digit sequences (i.e. numbers or identifiers with numerals).

    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
    estries
    -0.82
    💝
    -0.81
    =*
    -0.81
    🍻
    -0.80
     minister
    -0.80
    ($)
    -0.80
     〜
    -0.79
    áž
    -0.78
     and
    -0.78
     lubang
    -0.77
    POSITIVE LOGITS
    经过
    0.96
    このような
    0.94
    sorry
    0.90
    ここ
    0.86
    doesn
    0.86
     héroe
    0.81
    vinner
    0.81
    жено
    0.80
     Winding
    0.80
    おしゃれ
    0.79
    Activations Density 0.285%

    No Known Activations