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. 6040
    Prev
    Next
    INDEX
    Explanations

    attends to tokens related to "from" from tokens related to "on," emphasizing the relationship between efforts and considerations in varied contexts

    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
    0:0.10
    1:0.12
    2:0.11
    3:0.06
    4:0.05
    5:0.02
    6:0.06
    7:0.43
    Negative Logits
    XmlAccessType
    -0.32
     IActionResult
    -0.32
    ReusableCell
    -0.30
    modb
    -0.29
    enteuer
    -0.29
    jednoc
    -0.28
    rzez
    -0.28
    ItemIndex
    -0.27
    наче
    -0.27
     IconData
    -0.27
    POSITIVE LOGITS
    VIAF
    0.43
     VIAF
    0.32
    UserScript
    0.29
    NameInMap
    0.28
    Extension
    0.28
    reciate
    0.28
    HasAnnotation
    0.28
     Extension
    0.28
    فرم
    0.26
     EXTENSION
    0.26
    Activations Density 0.357%

    No Known Activations