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-2-2B
    4. Residual Stream - 65k
    5. 17-GEMMASCOPE-RES-65K
    6. 28438
    Prev
    Next
    INDEX
    Explanations

    conditional statements and loops in programming code

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 17-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_17/width_65k/average_l0_125
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.17.hook_resid_post
    Hook Layer
    17
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    
    -1.34
    )";
    
    -1.22
    )"),
    -1.12
    NUMX
    -1.11
    __':
    
    -1.09
    )");
    
    -1.04
     $_"
    -1.03
     ...]
    -1.02
    ).</
    -1.01
    '},
    
    -1.01
    POSITIVE LOGITS
      
    0.81
       
    0.63
    (
    0.56
    ((
    0.50
    	
    0.49
     my
    0.49
        
    0.46
     kalian
    0.44
     didalam
    0.43
     justiça
    0.42
    Activations Density 0.025%

    No Known Activations