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. 3242
    Prev
    Next
    INDEX
    Explanations

    information related to corruption and bribery involving public officials

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

    China or Chinese topics

    np_acts-logits-general · gemini-2.0-flash

    China and Chinese

    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
    లాలు
    -0.66
    Kaieteur
    -0.60
    toBeDefined
    -0.56
    
    -0.54
     EClass
    -0.52
     **)
    -0.52
     الحره
    -0.49
    НИК
    -0.47
    MLLoader
    -0.47
    CAPTURE
    -0.46
    POSITIVE LOGITS
     China
    1.46
     Chinese
    1.36
    China
    1.32
     Beijing
    1.28
    Chinese
    1.20
     Shanghai
    1.14
     CHINA
    1.12
    Beijing
    1.08
     china
    1.08
     CHINESE
    1.07
    Activations Density 0.487%

    No Known Activations