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-2-2B
    4. Residual Stream - 16k
    5. 18-GEMMASCOPE-RES-16K
    6. 10529
    Prev
    Next
    INDEX
    Explanations

    LaTeX commands or syntax used in mathematical expressions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    исленность
    -0.64
    AccessorTable
    -0.63
     PyTuple
    -0.62
    AndEndTag
    -0.61
    хьтан
    -0.60
    RenderAtEndOf
    -0.60
     fieldNum
    -0.60
    kyou
    -0.58
    Билгалдахарш
    -0.58
    цездатний
    -0.57
    POSITIVE LOGITS
    text
    0.70
    AddHtmlAttribute
    0.63
    MemoryWarning
    0.59
    TEXT
    0.59
    makeText
    0.58
     authService
    0.56
     Newswire
    0.56
     Pingback
    0.55
    DataBind
    0.55
     défin
    0.53
    Activations Density 0.023%

    No Known Activations