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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Attention Out - 16k
    5. 0-GEMMASCOPE-ATT-16K
    6. 1270
    Prev
    Next
    INDEX
    Explanations

    attends to the first token in each subsequence from corresponding second tokens

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-att/layer_0/width_16k/average_l0_61
    Prompts (Dashboard)
    16,384 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
    0:0.09
    1:0.02
    2:0.01
    3:0.02
    4:0.03
    5:0.03
    6:0.05
    7:0.05
    8:0.07
    9:0.31
    10:0.04
    11:0.05
    12:0.05
    13:0.06
    14:0.03
    15:0.02
    Negative Logits
     informée
    -0.56
     nahilalakip
    -0.56
     allAfrica
    -0.51
     المعيارى
    -0.50
     $_(
    -0.50
    balleur
    -0.49
     getCustomer
    -0.48
    verifyException
    -0.48
     Nigerian
    -0.47
     WWE
    -0.47
    POSITIVE LOGITS
    }"
    0.33
    };
    0.31
     }
    0.31
                            
    0.31
    "}
    0.30
         
    0.30
                        
    0.30
    							
    0.30
    });
    0.30
     },
    0.30
    Activations Density 0.003%

    No Known Activations