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. 20-GEMMASCOPE-RES-16K
    6. 7740
    Prev
    Next
    INDEX
    Explanations

    references to various proteins or biological components

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

    beginning with "B" followed by another letter

    np_acts-logits-general · gemini-2.0-flash

    bi ologica, bus

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
    •Marfan Syndrome

    No Comments

    Negative Logits
     Meksiku
    -0.56
    ActionEvent
    -0.50
     chande
    -0.49
     Orto
    -0.49
    yntaxException
    -0.49
     TextFormField
    -0.49
    er
    -0.49
    uple
    -0.48
     Sitten
    -0.48
     tourner
    -0.48
    POSITIVE LOGITS
    bing
    0.76
    bed
    0.73
     Sib
    0.67
     SIB
    0.66
    bbbb
    0.61
    Teb
    0.57
     Ceb
    0.55
    BED
    0.54
    ecause
    0.54
    usiness
    0.54
    Activations Density 0.539%

    No Known Activations