Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 8240
    Prev
    Next
    INDEX
    Explanations

    references to individuals and their familial or significant relationships

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

    Names with titles/initials

    np_acts-logits-general · gemini-2.0-flash

    names and job 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
    INCREF
    -0.32
     potem
    -0.32
     aspect
    -0.30
     dabei
    -0.30
     PLATES
    -0.30
     pu
    -0.30
     twist
    -0.29
     plate
    -0.29
    SaveChangesAsync
    -0.29
     factor
    -0.29
    POSITIVE LOGITS
    localctx
    0.60
    NewGuid
    0.59
    :✨
    0.54
    bootstrapcdn
    0.52
    PMailer
    0.52
    EDEFAULT
    0.50
    onis
    0.48
     nahilalakip
    0.48
    terscotch
    0.48
    ferencia
    0.47
    Activations Density 0.019%

    No Known Activations