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. 18-GEMMASCOPE-RES-16K
    6. 14426
    Prev
    Next
    INDEX
    Explanations

    directional movements or transitions

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

    Directional words: up, down

    np_acts-logits-general · gemini-2.0-flash

    up or down

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     MenuView
    -0.62
     OAS
    -0.61
     Olympedia
    -0.54
     azy
    -0.53
     Winfrey
    -0.51
     Cister
    -0.51
     ASF
    -0.51
    Personensuche
    -0.49
    RUnlock
    -0.49
     aught
    -0.49
    POSITIVE LOGITS
     the
    0.75
     toward
    0.74
     into
    0.73
     towards
    0.72
    ViewImports
    0.62
    ContentAsync
    0.62
     onto
    0.60
    towards
    0.55
    
    0.55
    åt
    0.55
    Activations Density 0.166%

    No Known Activations