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. Residual Stream - 16k
    5. 15-GEMMASCOPE-RES-16K
    6. 1559
    Prev
    Next
    INDEX
    Explanations

    expressions indicating preference and enjoyment

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Hook Layer
    15
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    WebVitals
    -0.52
    )];
    
    -0.47
     CURLOPT
    -0.46
    você
    -0.46
    ródło
    -0.46
    orial
    -0.46
    vog
    -0.45
    NAMESPACE
    -0.44
    aca
    -0.44
     Chwiliwch
    -0.43
    POSITIVE LOGITS
     downside
    1.13
     Allerdings
    1.09
     entanto
    1.09
     However
    1.07
     however
    1.05
     allerdings
    1.03
    However
    1.01
     Unfortunately
    0.96
     Правда
    0.95
     però
    0.92
    Activations Density 0.439%

    No Known Activations