© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 78658
    Prev
    Next
    INDEX
    Explanations

    image generation and text processing

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

    The neuron is primarily detecting numeric tokens—instances of digits (e.g. numbers or numeric constants) 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
    遷
    -0.76
     devrez
    -0.75
     Naw
    -0.75
     suelo
    -0.72
    inin
    -0.71
     beverage
    -0.71
     geschehen
    -0.71
     Algebraic
    -0.70
    terase
    -0.70
    ຳ
    -0.70
    POSITIVE LOGITS
     output
    0.81
     chimique
    0.78
     dischar
    0.73
    いが
    0.72
     CIV
    0.72
     Donner
    0.72
     Clarity
    0.71
    輸出
    0.71
    OST
    0.70
    置いて
    0.70
    Activations Density 0.028%

    No Known Activations