© 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. 14763
    Prev
    Next
    INDEX
    Explanations

    a/an followed by a noun

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

    The neuron fires strongly on verbs that are immediately followed by the indefinite article “a” or “an” introducing their object.

    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
     vů
    -1.24
    AGAIN
    -1.21
     esencial
    -1.21
     
    -1.16
     assez
    -1.16
    JHEP
    -1.16
    ,
    -1.13
    laa
    -1.12
    oung
    -1.11
    回到了
    -1.11
    POSITIVE LOGITS
     these
    1.70
     this
    1.66
     it
    1.59
     before
    1.34
     its
    1.32
     in
    1.32
     such
    1.32
     while
    1.32
     any
    1.31
     what
    1.27
    Activations Density 0.060%

    No Known Activations