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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 56758
    Prev
    Next
    INDEX
    Explanations

    mathematical symbols and parameters related to functions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    mathematical expressions and variables

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    GEBURTSDATUM
    -0.73
    ArgsConstructor
    -0.68
    MigrationBuilder
    -0.66
    richTextPanel
    -0.66
     Theſe
    -0.64
    Autoritní
    -0.64
     незавершена
    -0.63
    Gambas
    -0.62
     beſt
    -0.62
    enderror
    -0.62
    POSITIVE LOGITS
    heder
    0.28
    なって
    0.27
     hinh
    0.27
    RIEL
    0.26
    ค์
    0.26
    aires
    0.26
     vore
    0.25
    ư
    0.25
    Dependency
    0.25
     Trost
    0.24
    Activations Density 0.421%

    No Known Activations