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. 10291
    Prev
    Next
    INDEX
    Explanations

    programming constructs related to object instantiation and method calling

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

    Following "=" or "Lett" followed by non-English words

    np_acts-logits-general · gemini-2.0-flash

    new in multiple languages

    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
    AccessException
    -0.56
    phazard
    -0.56
    anmoins
    -0.56
    polation
    -0.54
    bibfnamefont
    -0.52
    ValueStyle
    -0.51
    argli
    -0.51
    revet
    -0.49
    urit
    -0.48
    DeleteMapping
    -0.48
    POSITIVE LOGITS
     new
    2.42
    new
    2.20
     nuevos
    1.54
     neuen
    1.52
     nuevas
    1.52
     nieuwe
    1.50
     nuovi
    1.47
    新的
    1.47
     neue
    1.45
     nueva
    1.42
    Activations Density 0.129%

    No Known Activations