© 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. 11894
    Prev
    Next
    INDEX
    Explanations

    numeric identifiers or codes related to medical or technical data

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

    Tokens after bracketed citations

    np_acts-logits-general · gemini-2.0-flash

    identifiers with f and digits

    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
     Chwiliwch
    -0.77
    Diweddarwch
    -0.72
     Administrativna
    -0.72
    complexContent
    -0.70
    存于互联网档案馆
    -0.63
    UnusedPrivate
    -0.63
     Geſch
    -0.62
    énario
    -0.61
    DoubleQuotes
    -0.60
    ьаж
    -0.59
    POSITIVE LOGITS
     betweenstory
    0.40
     obsequ
    0.32
    iPhone
    0.31
    GPI
    0.31
    esp
    0.31
    FileList
    0.30
    stateMutability
    0.30
     Contra
    0.30
    woche
    0.30
    Kol
    0.30
    Activations Density 0.256%

    No Known Activations