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-2-2B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 15608
    Prev
    Next
    INDEX
    Explanations

    details about administration and procedural elements in legal or bureaucratic contexts

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

    "during" (or similar temporal conjunctions)

    np_acts-logits-general · gemini-2.0-flash

    while or during events

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
     const
    -0.54
    =?";
    -0.50
    IVAS
    -0.49
    biography
    -0.48
    czej
    -0.47
     EconPapers
    -0.47
    unie
    -0.47
    emiah
    -0.46
    uanya
    -0.46
    ijo
    -0.46
    POSITIVE LOGITS
     while
    2.97
    while
    2.71
     WHILE
    2.44
     enquanto
    2.41
     mientras
    2.39
     whilst
    2.33
    While
    2.29
     While
    2.22
     mentre
    2.22
    WHILE
    2.21
    Activations Density 0.635%

    No Known Activations