© 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 - 16k
    9. 35-GEMMASCOPE-RES-16K
    11. 10448
    Prev
    Next
    INDEX
    Explanations

    variations of the prefix "wh" in different word contexts

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 35-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_35/width_16k/average_l0_120
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.35.hook_resid_post
    Hook Layer
    35
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
    absent
    -0.40
     StatefulWidget
    -0.39
    scryfall
    -0.37
     Trot
    -0.37
     Cuth
    -0.36
     Gug
    -0.36
     Hartmann
    -0.36
    RTGC
    -0.35
     mesh
    -0.35
    ophen
    -0.34
    POSITIVE LOGITS
     wh
    0.91
    Wh
    0.86
     WH
    0.85
     Wh
    0.84
    wh
    0.80
     Whi
    0.70
    Whi
    0.68
    WH
    0.66
    whi
    0.61
     whi
    0.60
    Activations Density 0.139%

    No Known Activations