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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 12039
    Prev
    Next
    INDEX
    Explanations

    references to video and media content

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

    "Video" followed by related media types

    np_acts-logits-general · gemini-2.0-flash

    video games and streaming

    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
    WriteAttribute
    -0.70
     GeoNames
    -0.67
    ритори
    -0.64
    RetentionPolicy
    -0.61
    elemField
    -0.61
    PerformLayout
    -0.61
    })();
    
    -0.60
    tecido
    -0.59
     InputDecoration
    -0.59
     propOrder
    -0.58
    POSITIVE LOGITS
    tape
    0.88
     clip
    0.88
     clips
    0.80
    juego
    0.74
    confer
    0.71
     footage
    0.69
    thora
    0.69
     conferencing
    0.68
     Clip
    0.62
    clip
    0.61
    Activations Density 0.057%

    No Known Activations