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. 645
    Prev
    Next
    INDEX
    Explanations

    references to the Virgin Islands or related terms

    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
    einander
    -0.52
    kään
    -0.47
    Show
    -0.46
     Maladies
    -0.45
    asnya
    -0.45
    ctp
    -0.44
     thù
    -0.44
     Bender
    -0.44
    ecki
    -0.43
    moving
    -0.43
    POSITIVE LOGITS
     Virgin
    1.22
    Virgin
    1.16
     virgin
    1.02
    virgin
    1.01
     Virginie
    0.74
     Virginia
    0.74
     virginia
    0.71
    Virginia
    0.67
     virgins
    0.66
     virginity
    0.64
    Activations Density 0.006%

    No Known Activations