© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 10045
    Prev
    Next
    INDEX
    Explanations

    punctuation and structural elements in code snippets

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    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
     ModelExpression
    -0.54
     NSCoder
    -0.53
     faſt
    -0.52
     ſche
    -0.50
    存于互联网档案馆
    -0.50
     ſta
    -0.50
     مشين
    -0.48
    IsMutable
    -0.47
     eventType
    -0.47
     ſy
    -0.47
    POSITIVE LOGITS
    xaml
    0.38
    richTextPanel
    0.36
     geweest
    0.36
     seguida
    0.36
    AnchorTagHelper
    0.34
     réservé
    0.34
    accompanied
    0.33
     inspirado
    0.33
    ver
    0.32
     xí
    0.32
    Activations Density 0.150%

    No Known Activations