© 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. 10-GEMMASCOPE-RES-131K
    7. 85443
    Prev
    Next
    INDEX
    Explanations

    strength, technology, administration

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

    The neuron activates on negative numeric tokens (numbers with a leading minus sign, including decimals).

    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
     in
    -1.85
     altering
    -1.70
     bolster
    -1.69
     If
    -1.63
     Unlike
    -1.63
     While
    -1.62
     garner
    -1.56
     Despite
    -1.55
     nestled
    -1.53
     To
    -1.52
    POSITIVE LOGITS
    that
    1.84
     magnificent
    1.81
     masterful
    1.67
     of
    1.64
    鲩
    1.63
     quintessential
    1.61
     wonderfully
    1.59
    another
    1.59
     praticamente
    1.59
     immense
    1.58
    Activations Density 0.019%

    No Known Activations