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. 13828
    Prev
    Next
    INDEX
    Explanations

    concepts related to television news and broadcasting

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

    following "written" or "oral"

    np_acts-logits-general · gemini-2.0-flash

    visual communication

    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
    Not in Any Lists

    No Comments

    Negative Logits
    fromnode
    -0.64
    DockStyle
    -0.57
    InjectAttribute
    -0.53
     karier
    -0.50
    レギュラー
    -0.50
    RectangleBorder
    -0.50
    Rè
    -0.48
     nahilalakip
    -0.48
    hatsu
    -0.47
    Šaltiniai
    -0.46
    POSITIVE LOGITS
     visual
    1.69
     visuals
    1.48
    visual
    1.42
     pictorial
    1.31
     VISUAL
    1.30
     Visual
    1.28
    Visual
    1.28
     verbal
    1.21
     visuel
    1.17
     visually
    1.17
    Activations Density 0.430%

    No Known Activations