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. 62236
    Prev
    Next
    INDEX
    Explanations

    email headers and message formatting elements

    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
    RenderAtEndOf
    -0.70
    ValueStyle
    -0.63
     Infórmanos
    -0.56
    IntoConstraints
    -0.55
     فريبيس
    -0.53
    帖最后由
    -0.53
    InjectAttribute
    -0.51
    GTCX
    -0.50
    AndEndTag
    -0.50
    ContentAsync
    -0.49
    POSITIVE LOGITS
     disambiguazione
    0.38
    xmlhttp
    0.35
    baomidou
    0.34
     Noto
    0.34
    Pron
    0.33
    
    0.33
    干
    0.32
    fahrt
    0.32
     gamba
    0.32
     enquire
    0.32
    Activations Density 0.352%

    No Known Activations