© 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. 22-GEMMASCOPE-RES-131K
    7. 67814
    Prev
    Next
    INDEX
    Explanations

    comparing different types

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

    The neuron is strongly activated by numeric tokens (integers or decimal measurements) in scientific 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_22/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
     miſ
    -1.26
    eſt
    -1.16
    ︲
    -1.15
     rembourse
    -1.14
     ſi
    -1.13
    UNIDAD
    -1.13
    
    -1.13
     ſon
    -1.13
    ໆ
    -1.13
    ſen
    -1.12
    POSITIVE LOGITS
     more
    1.76
     those
    1.70
     much
    1.44
     less
    1.42
     large
    1.41
     larger
    1.32
     primarily
    1.27
     types
    1.24
     not
    1.22
     most
    1.22
    Activations Density 0.122%

    No Known Activations