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. 80774
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    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
     viſ
    -0.50
    buttonBar
    -0.50
    AsUp
    -0.47
    ulemon
    -0.47
    }{$\
    -0.46
    textView
    -0.45
     Diſ
    -0.45
    queryInterface
    -0.44
    ventud
    -0.44
    }{#
    -0.44
    POSITIVE LOGITS
     مؤرشف
    0.88
     محفوظة
    0.82
     الأصل
    0.52
     Италијани
    0.48
    原始内容存档于
    0.42
    ValueStyle
    0.42
    Spoljašnje
    0.41
     atualmente
    0.40
     InputDecoration
    0.40
    معلومات
    0.40
    Activations Density 0.000%

    No Known Activations

    This feature has no known activations.