© 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. 79999
    Prev
    Next
    INDEX
    Explanations

    rank twists, act as, rest of

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

    This neuron primarily fires on the word “to.”

    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
     enfance
    -1.29
     ſeveral
    -1.17
     riscos
    -1.15
     wither
    -1.14
     varier
    -1.13
     bluz
    -1.13
     нашел
    -1.12
     provocando
    -1.11
     mistura
    -1.11
    也将
    -1.10
    POSITIVE LOGITS
     in
    1.55
     when
    1.49
     from
    1.47
    You
    1.45
     If
    1.38
     When
    1.34
     by
    1.29
    Maybe
    1.26
     के
    1.25
     as
    1.20
    Activations Density 0.013%

    No Known Activations