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. 9438
    Prev
    Next
    INDEX
    Explanations

    terms related to locations or places

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

    a location

    np_acts-logits-general · gemini-2.0-flash

    somewhere, anywhere, nowhere

    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
    日閲覧
    -0.56
    dotenv
    -0.52
    ung
    -0.51
    s
    -0.49
    //
    -0.49
    峙
    -0.48
     many
    -0.48
    en
    -0.48
     GenerationType
    -0.48
    a
    -0.48
    POSITIVE LOGITS
     else
    1.03
    下载附件
    0.69
     Else
    0.66
     Wikispecies
    0.66
    󠁢
    0.64
     Nowhere
    0.64
    Else
    0.63
    Jîn
    0.60
     afo
    0.59
     anytime
    0.59
    Activations Density 0.063%

    No Known Activations