© 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. 60096
    Prev
    Next
    INDEX
    Explanations

    gench.edu.cn

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

    The neuron fires on numeric tokens (numbers, years, or other all‐digit strings).

    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
     Eduard
    -0.95
    Med
    -0.85
     Eduardo
    -0.84
     Elé
    -0.84
     electrochemical
    -0.83
     EDF
    -0.81
     med
    -0.80
    Becker
    -0.80
     Becker
    -0.80
     Eddy
    -0.78
    POSITIVE LOGITS
     Yer
    0.82
    Mesa
    0.81
    dert
    0.81
     desorption
    0.79
     sermon
    0.78
    lerp
    0.77
     Mersey
    0.77
    resnet
    0.75
     Sermon
    0.75
     less
    0.75
    Activations Density 0.015%

    No Known Activations