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. 5864
    Prev
    Next
    INDEX
    Explanations

    attends to tokens indicating deferred actions from the subsequent comparison or context tokens

    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
    0:0.11
    1:0.11
    2:0.14
    3:0.14
    4:0.12
    5:0.07
    6:0.14
    7:0.14
    Negative Logits
    LookAnd
    -0.31
    ul
    -0.27
     invokingState
    -0.25
    ism
    -0.25
    im
    -0.25
    obs
    -0.24
    ;
    -0.23
    ,”
    -0.23
    -\
    -0.23
     observed
    -0.23
    POSITIVE LOGITS
    endpush
    0.42
     arşivlendi
    0.41
    Referencer
    0.41
    PointerException
    0.41
    ineno
    0.39
    ########.
    0.38
    clusal
    0.36
     חיצוניים
    0.36
    skosten
    0.34
    riwal
    0.34
    Activations Density 0.002%

    No Known Activations