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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 3-GEMMASCOPE-RES-16K
    6. 6123
    Prev
    Next
    INDEX
    Explanations

    nested structures, particularly parentheses and various types of brackets

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 3-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_3/width_16k/average_l0_90
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.3.hook_resid_post
    Hook Layer
    3
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    tencent
    -0.59
     Planes
    -0.58
    出版年
    -0.56
    zumaki
    -0.56
     Plugins
    -0.56
     fileSize
    -0.55
    mpagne
    -0.55
     RelativeLayout
    -0.55
    ToolStripButton
    -0.55
    atchewan
    -0.54
    POSITIVE LOGITS
    AutoScaleMode
    0.43
     nahilalakip
    0.36
    TestingModule
    0.34
    __(/*!
    0.33
    GIVEREF
    0.33
     =
    0.33
    клопе
    0.29
    География
    0.29
     başlad
    0.29
    awtextra
    0.28
    Activations Density 0.235%

    No Known Activations