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. 20-GEMMASCOPE-RES-16K
    6. 15638
    Prev
    Next
    INDEX
    Explanations

    culinary and restaurant-related terms

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

    Following types or versions of something

    np_acts-logits-general · gemini-2.0-flash

    types of things

    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
    RegistryLite
    -0.58
    はじめに
    -0.56
     betweenstory
    -0.54
     judiciaire
    -0.53
     ویکی‌پدیا
    -0.53
     للمعارف
    -0.52
     سكانية
    -0.51
     SqlParameter
    -0.51
    gnore
    -0.51
    iosamente
    -0.50
    POSITIVE LOGITS
     ones
    1.95
     version
    1.50
    のもの
    1.30
     versions
    1.24
    なもの
    1.24
     вариант
    1.13
     counterpart
    1.10
     theirs
    1.08
     counterparts
    1.05
     option
    1.04
    Activations Density 0.770%

    No Known Activations