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. 6417
    Prev
    Next
    INDEX
    Explanations

    attends to numerical tokens from text tokens that introduce an enumeration

    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
    0:0.19
    1:0.22
    2:0.17
    3:0.06
    4:0.09
    5:0.06
    6:0.05
    7:0.13
    Negative Logits
    SpringBootTest
    -0.29
    шое
    -0.26
    exitRule
    -0.26
    ylabel
    -0.25
     guarn
    -0.24
    alagi
    -0.23
     dzie
    -0.23
     rempliss
    -0.23
    GeneratedValue
    -0.23
    GetType
    -0.22
    POSITIVE LOGITS
    AndEndTag
    0.34
    存于互联网档案馆
    0.34
     rul
    0.33
     Audiodateien
    0.33
     Lordships
    0.32
    GHIJKLM
    0.32
    webElementXpaths
    0.32
    ✨:
    0.31
    CompleteListener
    0.31
    posedge
    0.30
    Activations Density 0.158%

    No Known Activations