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. 14527
    Prev
    Next
    INDEX
    Explanations

    terms related to software implementation and programming

    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
     snippetHide
    -0.57
    rungsseite
    -0.56
    layoutControl
    -0.50
     myſelf
    -0.50
     himſelf
    -0.49
    webElementXpaths
    -0.49
    setVerticalGroup
    -0.48
     <=",
    -0.47
     kasarigan
    -0.47
     reaſon
    -0.46
    POSITIVE LOGITS
     lâmpada
    0.40
    AntiForgeryToken
    0.39
     croissance
    0.38
     growth
    0.38
    råd
    0.36
     procédures
    0.36
     cebola
    0.36
     Growth
    0.35
     vidro
    0.35
     storage
    0.34
    Activations Density 0.235%

    No Known Activations