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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 0-GEMMASCOPE-ATT-16K
    6. 6099
    Prev
    Next
    INDEX
    Explanations

    attends to tokens indicating a state, condition, or place from tokens referencing a related time

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 0-gemmascope-att-16k
    Configuration
    google/gemma-scope-2b-pt-att/layer_0/width_16k/average_l0_104
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.0.attn.hook_z
    Hook Layer
    0
    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
    0:0.06
    1:0.02
    2:0.02
    3:0.02
    4:0.18
    5:0.61
    6:0.02
    7:0.04
    Negative Logits
    ,
    -0.35
    回
    -0.29
    www
    -0.29
    <eos>
    -0.28
    ary
    -0.28
    求
    -0.28
    、
    -0.27
    ://
    -0.27
     
    -0.27
    ine
    -0.26
    POSITIVE LOGITS
    ̈́
    0.46
    ^(@)
    0.44
    PYX
    0.43
     Majefty
    0.42
    CloseOperation
    0.42
    UseVisualStyle
    0.42
     незавершена
    0.40
     auffi
    0.40
    ſelf
    0.40
    HtmlAttribute
    0.39
    Activations Density 0.936%

    No Known Activations