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. 22-GEMMASCOPE-RES-16K
    6. 13889
    Prev
    Next
    INDEX
    Explanations

    phrases related to university admissions, specifically focusing on statistics and processes

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

    College admissions and scholarships

    np_acts-logits-general · gemini-2.0-flash

    admission and academic scholarships

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Waray
    -0.69
    vacy
    -0.62
    expandindo
    -0.60
    RegistryLite
    -0.60
    quiti
    -0.60
    transQ
    -0.59
     (;;
    -0.57
    GOTREF
    -0.56
    ULUM
    -0.56
     Flick
    -0.54
    POSITIVE LOGITS
     scholarships
    1.33
     scholarship
    1.31
     Scholarships
    1.15
    scholars
    1.11
     Scholarship
    1.09
    Scholars
    1.07
    Scholarship
    1.07
     SCHOLAR
    1.00
     admission
    0.99
     tuition
    0.95
    Activations Density 0.114%

    No Known Activations