© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 25903
    Prev
    Next
    INDEX
    Explanations

    links to causation or state changes

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

    The neuron detects the central action verbs or predicates that express the main relationships in a sentence (e.g. “leads,” “is,” “caused,” “combined,” “override,” 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
     when
    -1.88
     in
    -1.62
     then
    -1.52
     therefore
    -1.45
    それから
    -1.43
     this
    -1.39
     should
    -1.38
     will
    -1.35
     here
    -1.33
     these
    -1.31
    POSITIVE LOGITS
     ohne
    1.15
     appunto
    1.14
     zamiast
    1.12
     gdzie
    1.12
    ###########
    1.09
    ϴ
    1.09
     anstatt
    1.08
    digde
    1.08
    TBD
    1.07
     zonder
    1.05
    Activations Density 0.114%

    No Known Activations