© 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. 14385
    Prev
    Next
    INDEX
    Explanations

    numbers and citations

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

    This neuron strongly responds to very short, high‐frequency tokens—common function words like “the,” “to,” the possessive “’s,” and standalone numeric values.

    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
     already
    -1.66
    ner
    -1.44
    ure
    -1.42
    菒
    -1.36
    ara
    -1.36
    
    -1.34
     lading
    -1.30
     effectuer
    -1.30
     diverses
    -1.27
     平台
    -1.27
    POSITIVE LOGITS
     people
    1.63
     new
    1.43
     dozens
    1.38
     two
    1.34
     nine
    1.33
     news
    1.27
     feststellen
    1.26
     việc
    1.25
     some
    1.24
     mascotas
    1.23
    Activations Density 0.085%

    No Known Activations