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. 23-GEMMASCOPE-RES-16K
    6. 7605
    Prev
    Next
    INDEX
    Explanations

    references to food items and their descriptions

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

    food-related terms, particularly focusing on types of bread and sandwiches.

    oai_token-act-pair · gpt-4oTriggered by @noracai0625
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 23-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_23/width_16k/average_l0_75
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.23.hook_resid_post
    Hook Layer
    23
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     ocu
    -0.53
     estekak
    -0.50
    werking
    -0.50
     Toxicol
    -0.49
    edding
    -0.48
     Durga
    -0.47
    RenderTarget
    -0.46
     glaucoma
    -0.45
     scher
    -0.45
    urare
    -0.45
    POSITIVE LOGITS
     bread
    1.20
     Bread
    1.01
     breads
    0.99
     loaf
    0.98
     loaves
    0.96
    Bread
    0.94
     BREAD
    0.93
     baguette
    0.93
    bread
    0.92
     sourdough
    0.84
    Activations Density 0.146%

    No Known Activations