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. 6595
    Prev
    Next
    INDEX
    Explanations

    specific chemical or biological terms and their related properties

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

    ending in "y"

    np_acts-logits-general · gemini-2.0-flash

    words ending in y, ry, or ary

    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
    ations
    -0.92
    ם
    -0.77
     ThemeData
    -0.73
    ⇑
    -0.65
    ת
    -0.64
    ATION
    -0.64
    ation
    -0.61
    ally
    -0.61
    ality
    -0.60
    ative
    -0.59
    POSITIVE LOGITS
    yyy
    1.33
    yyyy
    1.25
    yyyyy
    1.03
    yy
    0.98
    ey
    0.86
    outube
    0.79
    eyes
    0.74
     CMY
    0.73
    YYY
    0.72
    esterday
    0.71
    Activations Density 0.766%

    No Known Activations