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-2-2B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 4926
    Prev
    Next
    INDEX
    Explanations

    information related to astronomical observations and their details

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

    astronomical observations or events

    np_acts-logits-general · gemini-2.0-flash

    celestial events and phenomena

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 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
     مشين
    -0.61
    ItemBackground
    -0.59
     Saxons
    -0.57
     समीक्षक
    -0.52
    Rhestr
    -0.52
     SQLITE
    -0.52
     morphe
    -0.51
     Signalez
    -0.48
     biofilm
    -0.48
     embal
    -0.47
    POSITIVE LOGITS
     astro
    1.05
     astronomers
    1.02
     planets
    1.00
    Astro
    0.95
     planet
    0.94
     Astro
    0.93
     astronomy
    0.93
     astronomer
    0.90
     Astronomy
    0.89
    astro
    0.89
    Activations Density 0.392%

    No Known Activations