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. 6605
    Prev
    Next
    INDEX
    Explanations

    attends to process-related tokens from comparative 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
    0:0.08
    1:0.09
    2:0.09
    3:0.12
    4:0.12
    5:0.08
    6:0.22
    7:0.16
    Negative Logits
     still
    -0.28
     stands
    -0.27
     very
    -0.27
     said
    -0.25
     abs
    -0.25
     examples
    -0.25
    メ
    -0.24
     example
    -0.24
    ssa
    -0.24
     was
    -0.23
    POSITIVE LOGITS
    AddTagHelper
    0.51
    findpost
    0.43
    StoryboardSegue
    0.42
     препратки
    0.42
     незавершена
    0.42
     "..\..\
    0.41
     AssemblyCulture
    0.41
    }],
    
    0.40
     actionMode
    0.40
    MigrationBuilder
    0.40
    Activations Density 0.875%

    No Known Activations