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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 14738
    Prev
    Next
    INDEX
    Explanations

    references to different groups, cultures, and diversity within contexts

    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
    RegressionTest
    -0.66
    :✨
    -0.66
     surla
    -0.65
     CreateTagHelper
    -0.65
    arangay
    -0.64
    InSection
    -0.62
    Gön
    -0.57
    featureID
    -0.57
     تضيفلها
    -0.56
     OFDb
    -0.56
    POSITIVE LOGITS
     cliente
    0.28
    addSubview
    0.28
     region
    0.28
     orígenes
    0.27
     urgente
    0.27
     yöne
    0.27
    â
    0.27
    cuela
    0.26
     stakeholders
    0.26
    tangentMode
    0.26
    Activations Density 0.603%

    No Known Activations