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 - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 8553
    Prev
    Next
    INDEX
    Explanations

    elements of structured communication, such as email headers and document metadata

    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-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 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
     يتيمه
    -0.59
    RepeatedField
    -0.55
     ویکی‌پدی
    -0.54
     Italijanski
    -0.54
    LabelTagHelper
    -0.54
    charCodeAt
    -0.52
     postId
    -0.52
    تقاوى
    -0.52
     &___
    -0.51
     виправивши
    -0.51
    POSITIVE LOGITS
     AssemblyCompany
    0.49
     initComponents
    0.49
    endphp
    0.36
    GOTREF
    0.35
    gridx
    0.34
    adaptiveStyles
    0.33
     preghiera
    0.33
     pluie
    0.32
     persil
    0.31
    exión
    0.31
    Activations Density 0.304%

    No Known Activations