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. 2992
    Prev
    Next
    INDEX
    Explanations

    elements related to user interface layout and design in programming.

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

    programming code related to graphical user interfaces and layout management.

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot

    instances of object creation using the "new" keyword.

    oai_token-act-pair · claude-3-5-sonnet-20240620Triggered by @bot

    Follows common words (new, pattern, in, get)

    np_acts-logits-general · gemini-2.0-flash

    new Max, pattern., dioxide one, Get the

    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
     ddelweddau
    -0.94
     Monfieur
    -0.77
     Cæsar
    -0.74
     Shakspeare
    -0.73
     endblock
    -0.72
    енча
    -0.71
    ReusableCell
    -0.71
     Efq
    -0.70
     فريبيس
    -0.69
     itſelf
    -0.66
    POSITIVE LOGITS
     autorytatywna
    0.64
     java
    0.63
    migrationBuilder
    0.58
     point
    0.56
    Autoritní
    0.54
    ">//
    0.53
     Point
    0.50
     J
    0.49
     long
    0.48
    HasAnnotation
    0.48
    Activations Density 0.009%

    No Known Activations