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-2-2B
    4. Residual Stream - 65k
    5. 16-GEMMASCOPE-RES-65K
    6. 53372
    Prev
    Next
    INDEX
    Explanations

    segment markers or tokens indicating the structure of documents

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

    Ho, Go, App, with

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Theſe
    -0.85
     '\\;'
    -0.85
     Monfieur
    -0.81
    MessageOf
    -0.80
    ValueStyle
    -0.79
     فريبيس
    -0.78
    parsedMessage
    -0.77
     المعيارى
    -0.77
    PerformLayout
    -0.75
     betweenstory
    -0.75
    POSITIVE LOGITS
    
    0.54
     $^{\
    0.53
     tra
    0.49
     Vertra
    0.49
    semantics
    0.48
     $\
    0.48
     Sa
    0.47
     Tra
    0.46
    insky
    0.46
     $(\
    0.45
    Activations Density 0.190%

    No Known Activations