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 - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 14287
    Prev
    Next
    INDEX
    Explanations

    markup language elements and their attributes in a coding context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     toallas
    -0.54
    RenderAtEndOf
    -0.53
     orejas
    -0.53
     manzanas
    -0.52
    ьажоргаш
    -0.52
    jaqueta
    -0.50
     affari
    -0.50
     manzana
    -0.50
     Trabalho
    -0.49
    chaleco
    -0.49
    POSITIVE LOGITS
     nahilalakip
    0.48
     surla
    0.47
    tagHelperRunner
    0.46
    %")
    0.44
    LikeLiked
    0.43
     kasarigan
    0.39
    mobileqq
    0.39
    !")
    
    0.38
     poffible
    0.38
    localctx
    0.38
    Activations Density 0.033%

    No Known Activations