© 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. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 131k
    9. 20-GEMMASCOPE-RES-131K
    11. 104818
    Prev
    Next
    INDEX
    Explanations

    instances of the word "but" and variations of contrastive conjunctions

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
    astify
    -0.56
    сылкі
    -0.55
    dyž
    -0.54
     Obrador
    -0.50
    cticut
    -0.50
     still
    -0.48
     useAppContext
    -0.48
     onBackPressed
    -0.48
     Still
    -0.47
     Picchu
    -0.45
    POSITIVE LOGITS
     increasingly
    0.65
     artık
    0.56
     ormai
    0.49
     mittlerweile
    0.49
     progressively
    0.47
     coraz
    0.46
    不再
    0.45
     gradually
    0.45
     zuneh
    0.42
     inzwischen
    0.40
    Activations Density 0.048%

    No Known Activations