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-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 61933
    Prev
    Next
    INDEX
    Explanations

    programming keywords and class definitions specific to a certain framework

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    `user` input in code/technical contexts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    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
     autorytatywna
    -0.98
    DockStyle
    -0.77
     GenerationType
    -0.72
    səhifə
    -0.69
    sizeCache
    -0.69
    ChildScrollView
    -0.68
    存于互联网档案馆
    -0.66
     queſta
    -0.66
     AssemblyCulture
    -0.65
     препратки
    -0.65
    POSITIVE LOGITS
     Ridley
    0.33
     Clean
    0.31
     Jan
    0.30
     Sau
    0.29
    virtual
    0.29
     Mar
    0.28
    Deploy
    0.28
    WORK
    0.28
     motion
    0.28
     <<
    0.28
    Activations Density 0.007%

    No Known Activations