© 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. 22-GEMMASCOPE-RES-131K
    7. 15332
    Prev
    Next
    INDEX
    Explanations

    owner/role relation

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

    The neuron fires on parts of programmer-defined identifiers (e.g. “Owner,” “Button,” “input,” “foo”) in source code.

    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
     Pág
    -1.22
     rodas
    -1.15
     Nós
    -1.14
    Wys
    -1.14
     péro
    -1.13
    shrimp
    -1.13
     Nossa
    -1.12
     vulgares
    -1.09
     kalend
    -1.09
     motoru
    -1.08
    POSITIVE LOGITS
    _
    1.34
    r
    1.31
    j
    1.27
    o
    1.24
    f
    1.24
    R
    1.20
    l
    1.20
    p
    1.20
    O
    1.20
    v
    1.19
    Activations Density 0.008%

    No Known Activations