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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 13791
    Prev
    Next
    INDEX
    Explanations

    phrases indicating sections and organization of a document

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

    Prepositions such as "of", "to", or "on"

    np_acts-logits-general · gemini-2.0-flash

    headings, titles, or structured text labels

    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
    /*
    -0.46
     pouvoit
    -0.45
     nemlig
    -0.44
     enfans
    -0.42
     dessutom
    -0.42
     varandra
    -0.42
     niyang
    -0.41
     parezca
    -0.40
     bahkan
    -0.40
    ówczas
    -0.39
    POSITIVE LOGITS
    StoreMessageInfo
    0.60
    IndentedString
    0.59
    請繼續往下閱讀
    0.59
    XmlAccessType
    0.57
    UrlResolution
    0.52
    ंदीखरीदारी
    0.52
     Italijanski
    0.50
     informée
    0.45
     VIAF
    0.45
     lizenzfreies
    0.45
    Activations Density 0.521%

    No Known Activations