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. Attention Out - 16k
    5. 10-GEMMASCOPE-ATT-16K
    6. 6045
    Prev
    Next
    INDEX
    Explanations

    attends to tokens related to diplomatic titles from tokens related to geographic locations

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 10-gemmascope-att-16k
    Configuration
    google/gemma-scope-2b-pt-att/layer_10/width_16k/average_l0_70
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.10.attn.hook_z
    Hook Layer
    10
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Head Attr Weights
    0:0.12
    1:0.19
    2:0.17
    3:0.07
    4:0.10
    5:0.04
    6:0.10
    7:0.18
    Negative Logits
     surla
    -0.40
    Datuak
    -0.40
    ंदीखरीदारी
    -0.36
    tagena
    -0.35
    MMV
    -0.30
    Hauptartikel
    -0.30
     typelib
    -0.30
    värr
    -0.30
    Tikang
    -0.30
    AILABLE
    -0.30
    POSITIVE LOGITS
    RenderAtEndOf
    0.20
     guste
    0.20
     Nark
    0.20
     IEnumerator
    0.20
    var
    0.19
     onOptions
    0.19
    if
    0.19
    izr
    0.19
    let
    0.19
    Dim
    0.19
    Activations Density 0.063%

    No Known Activations