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. 6407
    Prev
    Next
    INDEX
    Explanations

    attends to numeric encoded punctuation tokens from their surrounding textual context

    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
    0:0.33
    1:0.28
    2:0.10
    3:0.07
    4:0.06
    5:0.02
    6:0.04
    7:0.07
    Negative Logits
    });
    
    
    -0.30
    EndContext
    -0.29
    }();
    -0.29
    ")));
    
    -0.28
    CodeAttribute
    -0.27
    "]));
    -0.27
    :+:
    -0.27
    ConstraintMaker
    -0.27
    verwijspagina
    -0.27
    ])));
    -0.26
    POSITIVE LOGITS
    SharedCtor
    0.29
    audiovisuel
    0.28
     مشين
    0.26
     demografica
    0.26
    ้า
    0.25
    amen
    0.25
    CardModule
    0.25
    ClientSize
    0.25
    pezi
    0.25
    holde
    0.25
    Activations Density 1.715%

    No Known Activations