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. 10-GEMMASCOPE-ATT-16K
    6. 5557
    Prev
    Next
    INDEX
    Explanations

    attends to closing function call tokens from preceding opening function call tokens

    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
    0:0.17
    1:0.14
    2:0.41
    3:0.05
    4:0.05
    5:0.03
    6:0.03
    7:0.07
    Negative Logits
     purpoſe
    -0.43
     pleaſure
    -0.40
    ArgsConstructor
    -0.40
     }}"></
    -0.38
     doubtnut
    -0.38
    }],
    
    -0.38
     greateſt
    -0.38
     doInBackground
    -0.37
     FetchType
    -0.37
    dehyde
    -0.36
    POSITIVE LOGITS
    сшта
    0.24
     enfant
    0.23
    sizePolicy
    0.22
     &&
    0.22
     comme
    0.22
    TabStop
    0.21
    kembang
    0.21
     étn
    0.21
     +
    0.21
     &
    0.20
    Activations Density 0.182%

    No Known Activations