© 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. 13275
    Prev
    Next
    INDEX
    Explanations

    according to, Tony explained, slam dunk, Victoria's Secret

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

    The neuron responds to standalone numbers or numeric strings (e.g. episode numbers, years, and other digit sequences).

    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
     with
    -1.45
     will
    -1.32
     until
    -1.19
     while
    -1.16
    𝒙
    -1.14
    AfterViewInit
    -1.11
    colours
    -1.10
    انظر
    -1.10
     this
    -1.09
     nine
    -1.08
    POSITIVE LOGITS
    According
    1.46
    according
    1.38
    ​​​​
    1.27
    なんだか
    1.24
     according
    1.23
     According
    1.23
     didn
    1.17
    ​​​​​​​
    1.17
     pesso
    1.16
     PEOPLE
    1.16
    Activations Density 0.002%

    No Known Activations