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. 987
    Prev
    Next
    INDEX
    Explanations

    mentions of geographical locations and nationality

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

    countries

    np_acts-logits-general · gemini-2.0-flash

    countries and locations

    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
    genstein
    -0.32
     ocasião
    -0.31
     béisbol
    -0.31
     occasione
    -0.31
     Infórmanos
    -0.29
     pronom
    -0.29
     INTERESAR
    -0.28
    AddAttribute
    -0.28
     História
    -0.27
    Toponymie
    -0.27
    POSITIVE LOGITS
    principalColumn
    0.54
    fillType
    0.53
    舺
    0.53
    UrlResolution
    0.52
    CloseOperation
    0.52
    contentLoaded
    0.52
    SharedCtor
    0.52
    yntaxException
    0.52
     betweenstory
    0.52
    elemField
    0.50
    Activations Density 0.639%

    No Known Activations