© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 128017
    Prev
    Next
    INDEX
    Explanations

    This neuron detects occurrences of the root “teach” (in any form/inflection, e.g. Teach, teach, Teaching, teaching).

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

    teaching

    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
    '
    -3.05
    ’
    -2.50
    ',
    -2.47
     \
    -2.36
     /
    -2.28
     –
    -2.25
    …
    -2.22
      
    -2.16
     '
    -2.08
     действия
    -2.08
    POSITIVE LOGITS
     Saltar
    2.53
     eccellente
    2.45
     affari
    2.45
     strumenti
    2.38
     Mijn
    2.31
     collezione
    2.31
    尴
    2.30
     $_=
    2.28
     Añade
    2.27
     itſelf
    2.22
    Activations Density 0.009%

    No Known Activations