Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 62852
    Prev
    Next
    INDEX
    Explanations

    references to geometrical shapes and measurements related to circular and cylindrical structures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    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
    Not in Any Lists

    No Comments

    Negative Logits
     koltuk
    -0.34
     cuarzo
    -0.32
     boyunca
    -0.32
     puentes
    -0.32
     zuges
    -0.30
     bankası
    -0.30
    ümüş
    -0.30
     endblock
    -0.29
    klärt
    -0.28
     cartón
    -0.28
    POSITIVE LOGITS
     radius
    0.87
     spherical
    0.84
     circle
    0.83
     circles
    0.82
     concentric
    0.81
     Spherical
    0.81
     Circle
    0.79
    circles
    0.79
    radius
    0.77
    Circle
    0.77
    Activations Density 0.366%

    No Known Activations