© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 10164
    Prev
    Next
    INDEX
    Explanations

    phrases or elements related to news reporting and media sources, particularly focusing on exclusive or breaking news.

    oai_token-act-pair · claude-3-5-sonnet-20240620Triggered by @myinzi123

    references to well-known personalities and media reports about them

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

    followed by punctuation or capitalization

    np_acts-logits-general · gemini-2.0-flash

    news reports and exclusive content

    np_acts-logits-general · gemini-2.5-flash-lite

    mentions typical of news articles—capitalized proper names and acronyms, references to media outlets or reporting, and associated dates/punctuation.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    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
    occup
    -0.40
     تضيفلها
    -0.36
     Sterne
    -0.34
     Fische
    -0.32
     herbs
    -0.32
     Reisen
    -0.32
    好多
    -0.31
    蔦
    -0.31
     Grä
    -0.31
     colonialism
    -0.31
    POSITIVE LOGITS
     Италијани
    0.88
    хьтан
    0.73
    adaptiveStyles
    0.73
    NameInMap
    0.73
    hyrchwyd
    0.71
    parsedMessage
    0.70
    verifyException
    0.68
    msgTypes
    0.68
    windowFixed
    0.66
     Italijani
    0.66
    Activations Density 0.388%

    No Known Activations