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 - 65k
    5. 20-GEMMASCOPE-RES-65K
    6. 1064
    Prev
    Next
    INDEX
    Explanations

    words related to various animals, particularly those associated with clubs or teams

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

    words related to the character "Eagle" in various contexts

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

    words ending in 'le'

    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-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    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
    ing
    -0.95
     ***!
    -0.75
    ی
    -0.68
    으로
    -0.66
    ة
    -0.64
    ים
    -0.62
    ING
    -0.62
    asthan
    -0.61
    ️
    -0.60
    es
    -0.58
    POSITIVE LOGITS
    ization
    1.07
    migrationBuilder
    0.82
    idoscope
    0.79
     ―――――
    0.78
    tonsoft
    0.75
    ized
    0.74
    moveToFirst
    0.70
    tle
    0.69
    таратура
    0.68
     ――――――――
    0.68
    Activations Density 0.052%

    No Known Activations