© 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. 10-GEMMASCOPE-RES-131K
    7. 130216
    Prev
    Next
    INDEX
    Explanations

    numbers followed by operators or punctuation

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

    The neuron is primarily detecting numeric tokens (standalone numbers and digits).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    -
    -1.53
    !
    -1.44
    ︸
    -1.36
    United
    -1.34
     United
    -1.33
    sometimes
    -1.30
    those
    -1.29
    eight
    -1.29
     laajenta
    -1.27
    although
    -1.27
    POSITIVE LOGITS
     autorización
    1.85
     both
    1.74
     semejante
    1.61
     them
    1.59
     cárcel
    1.59
     apaixon
    1.57
     města
    1.55
     fundación
    1.54
     sowohl
    1.54
     also
    1.54
    Activations Density 0.094%

    No Known Activations