© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 65k
    9. 20-GEMMASCOPE-RES-65K
    11. 528
    Prev
    Next
    INDEX
    Explanations

    instances of numerical values and categories

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

    numerical values and data structure related to financial information

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

    parenthetical numerical notation

    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-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    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
     getCount
    -0.52
     roles
    -0.50
    CppMethod
    -0.50
    HEET
    -0.49
     CYTHON
    -0.49
    béco
    -0.48
    QtCore
    -0.48
    PackageManager
    -0.48
     glyphicon
    -0.48
    icka
    -0.47
    POSITIVE LOGITS
     Baillargeon
    0.65
    ंदीखरीदारी
    0.63
     ModelExpression
    0.62
     oublié
    0.59
    PMailer
    0.57
    Източници
    0.57
    StoryboardSegue
    0.55
     htons
    0.54
    RenderAtEndOf
    0.53
    ⓘ
    0.52
    Activations Density 0.026%

    No Known Activations