© 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. 75002
    Prev
    Next
    INDEX
    Explanations

    of and the

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

    The neuron fires strongly on terms related to scientific methods and evaluation—words like test, study, evaluate, investigate, model, and related methodological verbs and nouns.

    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
     -¿
    -1.10
     tzw
    -1.08
     некоторы
    -1.05
     Traducción
    -1.05
     alltid
    -1.04
     некоторых
    -1.03
    柔ら
    -1.02
     Estructura
    -1.01
     ſi
    -0.99
     fungerar
    -0.99
    POSITIVE LOGITS
     any
    1.56
     anything
    1.50
     there
    1.27
     things
    1.19
     since
    1.18
    løs
    1.16
     it
    1.09
     here
    1.04
     because
    1.04
     other
    1.03
    Activations Density 0.060%

    No Known Activations