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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    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. 747
    Prev
    Next
    INDEX
    Explanations

    references to statistics and projections related to health or demographic changes

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

    Mentions a projected increase/doubling

    np_acts-logits-general · gemini-2.0-flash

    doubling pace or convincing evidence

    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
     Meksiku
    -0.53
     ویکی‌پدی
    -0.45
    nonatomic
    -0.41
     Walkover
    -0.38
     nonatomic
    -0.37
    __);
    -0.36
     iprot
    -0.36
     snapped
    -0.35
     Italijanski
    -0.35
     ESS
    -0.35
    POSITIVE LOGITS
    WaitGroup
    0.66
     logros
    0.56
    的目标
    0.54
     éché
    0.53
     laikā
    0.52
     goal
    0.52
     attainment
    0.51
     Warszawie
    0.51
     goals
    0.51
     TRABAJO
    0.50
    Activations Density 0.140%

    No Known Activations