© 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. 39102
    Prev
    Next
    INDEX
    Explanations

    code constructs and libraries

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

    The neuron fires on programming‐style identifiers (e.g. CamelCase names, snake_case names, namespaced tokens with underscores or separators).

    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
    ！！！！
    -1.43
     only
    -1.35
     in
    -1.27
     first
    -1.27
     on
    -1.23
     spurred
    -1.21
    ？？
    -1.20
    ？！
    -1.20
     undeniably
    -1.19
     from
    -1.18
    POSITIVE LOGITS
    久し
    1.35
     ferocious
    1.30
     adecuado
    1.30
     contenedores
    1.28
     adhesivo
    1.25
     adequado
    1.24
     adecuados
    1.23
     anunció
    1.20
     electricidad
    1.20
     licencias
    1.19
    Activations Density 0.016%

    No Known Activations