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. 667
    Prev
    Next
    INDEX
    Explanations

    specific details regarding the physical attributes and features of images or objects

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

    Image edges, borders, and margins

    np_acts-logits-general · gemini-2.0-flash

    edges and borders

    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
     Diſ
    -0.55
     quæ
    -0.52
     ſche
    -0.52
     ſta
    -0.51
     ſch
    -0.50
     ſte
    -0.49
    dorfer
    -0.48
    iterranée
    -0.48
     juſ
    -0.47
     broche
    -0.47
    POSITIVE LOGITS
     edges
    1.60
     edge
    1.50
     Edges
    1.41
    edges
    1.38
     margins
    1.29
     border
    1.28
     borders
    1.25
     margin
    1.22
    edge
    1.20
     edging
    1.19
    Activations Density 0.302%

    No Known Activations