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. 6089
    Prev
    Next
    INDEX
    Explanations

    attends to the negation "not" from tokens indicating presence or obligation, including those surrounding the concepts of necessity or inability

    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
    0:0.09
    1:0.10
    2:0.11
    3:0.05
    4:0.03
    5:0.01
    6:0.05
    7:0.52
    Negative Logits
     nor
    -0.53
     anymore
    -0.52
     any
    -0.50
    nor
    -0.46
    任何
    -0.40
    any
    -0.40
     anything
    -0.40
     or
    -0.38
    Any
    -0.37
     jemals
    -0.36
    POSITIVE LOGITS
    aarrggbb
    0.58
    setVerticalGroup
    0.51
    RenderAtEndOf
    0.47
    UrlResolution
    0.47
     unknownFields
    0.47
     autorytatywna
    0.46
    CppMethod
    0.46
    NameInMap
    0.45
     ostavi
    0.45
    IsMutable
    0.44
    Activations Density 0.422%

    No Known Activations