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-2-2B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 2121
    Prev
    Next
    INDEX
    Explanations

    references to significant events and personnel related to World Wars

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

    wars and conflicts

    np_acts-logits-general · gemini-2.0-flash

    war or wars

    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
     CURIAM
    -0.43
     AttributeSet
    -0.42
    にあ
    -0.41
    styrelsen
    -0.41
    colum
    -0.41
    andır
    -0.41
    NECT
    -0.40
    LogFactory
    -0.40
    cader
    -0.40
    hach
    -0.39
    POSITIVE LOGITS
     war
    1.86
     wars
    1.70
     guerra
    1.51
     War
    1.50
     guerre
    1.39
    War
    1.37
    war
    1.35
     WAR
    1.33
     Wars
    1.29
     guerras
    1.23
    Activations Density 0.285%

    No Known Activations