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. 11074
    Prev
    Next
    INDEX
    Explanations

    references to choices or selections regarding various options

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

    preceding "choose" or its variations

    np_acts-logits-general · gemini-2.0-flash

    choose and select

    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
    ViewFeatures
    -0.73
    MLLoader
    -0.70
    ExtendWith
    -0.66
    tagHelperRunner
    -0.61
    ⟬
    -0.60
    ViewImports
    -0.58
    RunWith
    -0.57
    )++;
    -0.56
     faune
    -0.55
    titleMargin
    -0.54
    POSITIVE LOGITS
     choice
    3.18
     choose
    3.03
     selection
    3.01
     choosing
    2.94
     chosen
    2.93
     selected
    2.84
     selecting
    2.81
     select
    2.74
     chooses
    2.73
     chose
    2.71
    Activations Density 0.887%

    No Known Activations