© 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. 130349
    Prev
    Next
    INDEX
    Explanations

    This neuron responds to mid-frequency content words—especially nouns and adjectives carrying the main concepts of a sentence.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    things that aren't true

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     бампер
    -1.19
     tornillos
    -1.19
    ּוֹ
    -1.15
    וּ
    -1.13
     previous
    -1.13
    ֶּ
    -1.13
     weeks
    -1.10
    ライター
    -1.09
    ִּ
    -1.07
    पित
    -1.06
    POSITIVE LOGITS
     suele
    1.45
     cristianismo
    1.28
     seldom
    1.27
     plupart
    1.23
     rarely
    1.21
     revelou
    1.20
     mentales
    1.20
    glück
    1.19
     naturen
    1.15
     medan
    1.14
    Activations Density 0.031%

    No Known Activations