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. MLP Out - 65k
    5. 0-GEMMASCOPE-MLP-65K
    6. 60998
    Prev
    Next
    INDEX
    Explanations

    instances of the word "listened"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 0-gemmascope-mlp-65k
    Configuration
    google/gemma-scope-2b-pt-mlp/layer_0/width_65k/average_l0_72
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.0.hook_mlp_out
    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

    Negative Logits
     listened
    -1.27
    '));
    
    -1.19
    ']))
    
    -1.18
    rungsseite
    -1.13
    '))
    
    -1.09
    ']);
    
    -1.09
     estekak
    -1.09
    '],
    
    -1.08
    '])){
    
    -1.05
    NameInMap
    -1.03
    POSITIVE LOGITS
     Fällen
    0.56
    ات
    0.53
    hire
    0.52
     Mandela
    0.51
    hankelijk
    0.51
     marca
    0.50
     Anspruch
    0.50
    zitter
    0.50
     latihan
    0.50
     Jaeger
    0.49
    Activations Density 0.017%

    No Known Activations