© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 14304
    Prev
    Next
    INDEX
    Explanations

    GUI elements and code structures

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

    The neuron fires on identifiers for library or API classes/objects (e.g. Stage, Parent, scene, renderer) in 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
     wiem
    -1.17
     our
    -1.12
     all
    -1.11
     each
    -1.06
    Referenties
    -1.04
     mów
    -1.02
     treino
    -1.02
    viale
    -1.00
    auré
    -1.00
    biography
    -0.98
    POSITIVE LOGITS
    <bos>
    1.16
    ianum
    1.11
    jando
    1.10
     uno
    1.05
     proble
    1.05
    ェ
    1.05
    了一個
    1.02
     DU
    1.00
    这部
    0.98
     fó
    0.97
    Activations Density 0.001%

    No Known Activations