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 - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 73773
    Prev
    Next
    INDEX
    Explanations

    data structures and code syntax in programming or technical documents

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
    
    -0.58
     surla
    -0.57
     StatelessWidget
    -0.51
    webElementXpaths
    -0.49
    himovic
    -0.47
    SequentialGroup
    -0.47
     Италијани
    -0.45
    expandindo
    -0.45
     Мексичка
    -0.44
     Study
    -0.43
    POSITIVE LOGITS
    posedge
    0.47
    AutoField
    0.47
     verb
    0.38
    UseVisualStyle
    0.38
     Chwiliwch
    0.38
    belajaran
    0.36
     pinulongan
    0.36
     vô
    0.36
     paid
    0.36
     label
    0.36
    Activations Density 0.120%

    No Known Activations