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 - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 12170
    Prev
    Next
    INDEX
    Explanations

    references to health-related challenges and the involvement of adolescents in health service utilization

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Whilst
    -1.32
     Whilst
    -1.27
     Amongst
    -1.02
     apologise
    -0.99
    Firstly
    -0.99
     whilst
    -0.98
     localisation
    -0.94
     Firstly
    -0.91
    の為
    -0.89
     realises
    -0.89
    POSITIVE LOGITS
    ​—
    0.94
     eſſ
    0.84
     vérit
    0.81
     ―――――
    0.80
     uſ
    0.79
     quæ
    0.79
     imidlertid
    0.78
     guère
    0.77
     auffi
    0.77
    
    0.77
    Activations Density 2.869%

    No Known Activations