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. 2493
    Prev
    Next
    INDEX
    Explanations

    the beginning of a document or section

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

    Preceding numbers or special characters

    np_acts-logits-general · gemini-2.0-flash

    mathematical notation

    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
    <eos>
    -0.60
    ↵
    -0.59
    by
    -0.57
    ab
    -0.56
    ask
    -0.53
     kasarigan
    -0.51
    SuppressMessage
    -0.51
    ho
    -0.51
     [
    -0.49
    ↵↵
    -0.49
    POSITIVE LOGITS
     itſelf
    0.94
     Мексичка
    0.94
     myſelf
    0.91
     Efq
    0.91
     himſelf
    0.86
    RectangleBorder
    0.84
     ſeveral
    0.83
     themſelves
    0.83
    ^(@)
    0.80
     Jefus
    0.78
    Activations Density 0.065%

    No Known Activations