Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 5732
    Prev
    Next
    INDEX
    Explanations

    attends to numeric values preceding the corresponding sequence number from designated markers appearing later in the sequence

    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
    0:0.13
    1:0.14
    2:0.16
    3:0.16
    4:0.13
    5:0.06
    6:0.07
    7:0.11
    Negative Logits
    expandindo
    -0.38
    @[+][
    -0.35
    */].
    -0.35
     mxArray
    -0.35
    ])):
    -0.32
    ]));
    
    -0.32
     VizieR
    -0.31
     دیکھیے
    -0.30
    IsMutable
    -0.30
     CanadaChoose
    -0.30
    POSITIVE LOGITS
     Advent
    0.29
    wein
    0.29
    onOptions
    0.28
     faits
    0.26
    Advent
    0.26
    stderr
    0.26
    ​​
    0.26
    findpost
    0.26
    ofold
    0.26
    using
    0.25
    Activations Density 0.290%

    No Known Activations