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. 21-GEMMASCOPE-RES-16K
    6. 7033
    Prev
    Next
    INDEX
    Explanations

    references to community engagement and infrastructure development

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    脚注の使い方
    -0.59
     للاسماء
    -0.58
     &___
    -0.56
    ftagPool
    -0.56
     الحره
    -0.52
    webElementXpaths
    -0.51
     picioare
    -0.49
    Vidite
    -0.49
    رشف
    -0.48
     دیکھیے
    -0.47
    POSITIVE LOGITS
     hospitals
    1.57
     hotels
    1.45
     hospital
    1.36
     museums
    1.34
     schools
    1.31
     restaurants
    1.30
     hotel
    1.26
     Hospitals
    1.26
     airports
    1.24
     factories
    1.22
    Activations Density 0.679%

    No Known Activations