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. 23412
    Prev
    Next
    INDEX
    Explanations

    negative or unsupported statistical results in medical studies

    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
    untamente
    -0.38
     zudem
    -0.36
    ówczas
    -0.36
    서는
    -0.36
    насе
    -0.35
     accomplish
    -0.34
     contribute
    -0.34
     then
    -0.34
     ingresar
    -0.34
     ferner
    -0.33
    POSITIVE LOGITS
     autorytatywna
    0.78
     Signalez
    0.74
     فريبيس
    0.71
     المعيارى
    0.71
     архивлан
    0.62
    лтамалар
    0.59
     препратки
    0.58
     Савезне
    0.58
     ویکی‌پدی
    0.56
     AppCompatTheme
    0.55
    Activations Density 0.028%

    No Known Activations