© 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. 71887
    Prev
    Next
    INDEX
    Explanations

    is a/an/of

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

    The neuron fires on high‐frequency function words (e.g. common tokens like “is,” “to,” “an,” 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
    congrats
    -1.20
    excluding
    -1.16
    遇到了
    -1.14
    hopefully
    -1.09
    honestly
    -1.07
    íně
    -1.06
    btw
    -1.05
    hmmmm
    -1.05
     versions
    -1.05
    hmmm
    -1.05
    POSITIVE LOGITS
     can
    1.54
     its
    1.52
     a
    1.34
    When
    1.30
     būt
    1.29
     could
    1.19
     when
    1.15
     usually
    1.14
     would
    1.12
    lepaskan
    1.08
    Activations Density 0.139%

    No Known Activations