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. 1181
    Prev
    Next
    INDEX
    Explanations

    specific references to societal structures and historical contexts involving race and community

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

    before names/titles

    np_acts-logits-general · gemini-2.0-flash

    name, proper noun, or acronym

    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-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
    httphttps
    -0.70
    RenderAtEndOf
    -0.70
    CppMethod
    -0.68
    MockMvc
    -0.65
    fromnode
    -0.65
    rrggbb
    -0.63
     Wikimedijinoj
    -0.59
     resourceCulture
    -0.59
    Alike
    -0.59
    ViewFeatures
    -0.59
    POSITIVE LOGITS
     Stö
    0.30
     الذات
    0.29
    ATEGORIES
    0.29
    nes
    0.28
     itself
    0.27
     selber
    0.27
     Ward
    0.27
     câ
    0.26
     Clor
    0.26
     participaron
    0.25
    Activations Density 1.166%

    No Known Activations