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-IT
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 1025
    Prev
    Next
    INDEX
    Explanations

    references to royalty and kingship

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    associated with royalty

    np_acts-logits-general · gemini-2.0-flash

    royal family and titles

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 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
    fromLTRB
    -0.47
    Nebraska
    -0.41
    noscript
    -0.40
     Grünen
    -0.39
     RTE
    -0.39
    testdata
    -0.37
     Leverkusen
    -0.37
    
    -0.37
    rapnel
    -0.37
     labourer
    -0.36
    POSITIVE LOGITS
     King
    0.80
     Queen
    0.77
    👑
    0.77
     Kings
    0.75
     Kral
    0.74
    King
    0.73
     KINGS
    0.73
     royal
    0.72
     Royal
    0.72
     king
    0.71
    Activations Density 0.280%

    No Known Activations