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 - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 45840
    Prev
    Next
    INDEX
    Explanations

    references to notable individuals and their actions in the context of entertainment and legal situations

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

    age and taxes

    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-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
     neighbourhood
    -0.37
     Wahrnehmung
    -0.34
     Bergmann
    -0.34
     rapidez
    -0.34
    towym
    -0.34
     ansiedade
    -0.33
     Mumford
    -0.33
     ansiedad
    -0.32
     presentación
    -0.32
     aceptación
    -0.32
    POSITIVE LOGITS
    saraba
    0.71
    ########.
    0.63
    脚注の使い方
    0.62
     المعيارى
    0.58
     retired
    0.57
     Aged
    0.57
     aging
    0.57
     aged
    0.56
    retired
    0.56
     retiring
    0.55
    Activations Density 0.029%

    No Known Activations