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
    4. Residual Stream - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 11504
    Prev
    Next
    INDEX
    Explanations

    code snippets related to API definitions and requests

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     işe
    -0.32
    super
    -0.31
    Original
    -0.29
     altında
    -0.29
    Super
    -0.29
    sendRedirect
    -0.29
    kadang
    -0.28
    aleta
    -0.28
    prit
    -0.28
    link
    -0.27
    POSITIVE LOGITS
    AndEndTag
    0.78
    RenderAtEndOf
    0.72
     المعيارى
    0.68
     препратки
    0.65
    msgTypes
    0.65
     Infórmanos
    0.64
    EndContext
    0.64
     '\\;'
    0.63
     GenerationType
    0.63
     שוליים
    0.63
    Activations Density 0.146%

    No Known Activations