© 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. 36-GEMMASCOPE-RES-16K
    11. 4953
    Prev
    Next
    INDEX
    Explanations

    specific names and terms related to various subjects and categories

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 36-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_36/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.36.hook_resid_post
    Hook Layer
    36
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
     myſelf
    -0.69
    iſter
    -0.69
     juſ
    -0.67
     ſou
    -0.67
     Monfieur
    -0.64
     ſever
    -0.64
     snippetHide
    -0.63
     ſol
    -0.62
     ſtate
    -0.62
     viſ
    -0.61
    POSITIVE LOGITS
    es
    1.11
    ses
    1.05
    ES
    0.82
    ess
    0.78
    ness
    0.65
    sed
    0.62
    hes
    0.61
    esa
    0.56
    sing
    0.55
    ès
    0.54
    Activations Density 0.953%

    No Known Activations