© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 28342
    Prev
    Next
    INDEX
    Explanations

    creating, initializing, or building objects

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

    The neuron flags code tokens that are method or constructor names—especially factory/creation calls like `create`, `rand`, `open`, `buttonWithType`, etc.

    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
    ris
    -1.13
    ane
    -1.12
    ra
    -1.08
    O
    -1.06
    ora
    -1.06
    lin
    -1.04
     idéia
    -1.04
    ath
    -1.04
    pa
    -1.03
    ひた
    -1.03
    POSITIVE LOGITS
     with
    1.42
     from
    1.34
     dekol
    1.26
     kú
    1.20
     kré
    1.20
     koncer
    1.20
     kellett
    1.18
     komfort
    1.17
     kalend
    1.17
     koncep
    1.16
    Activations Density 0.013%

    No Known Activations