© 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
    2. Gemma-2-9B
    3. 20-RES-MATRYOSHKA-DC
    4. 206
    Prev
    Next
    INDEX
    Explanations

    contrastive conjunctions that introduce opposing ideas

    oai_token-act-pair · gpt-4o-mini
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
    transQ
    -0.45
     redor
    -0.44
    eventName
    -0.43
     Richardson
    -0.41
     paire
    -0.41
     entera
    -0.41
     Rueda
    -0.41
     acier
    -0.40
     Acero
    -0.39
    grés
    -0.39
    POSITIVE LOGITS
     but
    1.34
    But
    1.33
     But
    1.24
    but
    1.13
     BUT
    1.05
     لكن
    0.98
     tetapi
    0.97
    但
    0.97
     nhưng
    0.96
    BUT
    0.96
    Activations Density 0.918%

    No Known Activations