© 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. 67615
    Prev
    Next
    INDEX
    Explanations

    activities and actions

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

    The neuron fires on and highlights main action verbs—words naming the activities people do (e.g. relax, unwind, have fun, buy, rest, watch).

    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
     instead
    -1.21
     每
    -1.07
    瞅
    -1.06
     the
    -1.05
    下意识
    -1.05
    /.../
    -1.03
    勉强
    -1.00
    Instead
    -0.99
    可能会
    -0.99
     書
    -0.98
    POSITIVE LOGITS
     azonban
    1.48
     yourself
    1.48
     your
    1.41
     entweder
    1.39
     akár
    1.34
     összefoglaló
    1.30
     either
    1.30
     both
    1.28
     zarówno
    1.24
     összefoglal
    1.22
    Activations Density 0.021%

    No Known Activations