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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 12705
    Prev
    Next
    INDEX
    Explanations

    locations and entities related to conflict or military operations in the Democratic Republic of the Congo

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

    Appears before brackets or parentheses

    np_acts-logits-general · gemini-2.0-flash

    Congo, around, trader

    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
     Montevideo
    -0.47
     Puglia
    -0.45
     Maldives
    -0.45
     Amman
    -0.45
    allis
    -0.44
     Maltese
    -0.44
     maroca
    -0.44
     Jacek
    -0.42
     Barbados
    -0.42
    Aller
    -0.42
    POSITIVE LOGITS
    TagMode
    1.01
    ंदीखरीदारी
    0.91
    CppMethod
    0.86
    LookAnd
    0.79
     يتيمه
    0.79
    ValueStyle
    0.78
     surla
    0.74
    +#+#
    0.74
    expandindo
    0.72
     للاسماء
    0.71
    Activations Density 0.104%

    No Known Activations