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. 18-GEMMASCOPE-RES-16K
    6. 3185
    Prev
    Next
    INDEX
    Explanations

    symbols and punctuation related to racial and ethnic minority topics

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

    Characters surrounding forward slashes

    np_acts-logits-general · gemini-2.0-flash

    racial/ethnic minority groups

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Билгалдахарш
    -0.96
     Италијани
    -0.93
     kasarigan
    -0.92
    Hauptartikel
    -0.89
    rungsseite
    -0.89
    InjectAttribute
    -0.89
     الحره
    -0.87
     незавершена
    -0.86
    NameInMap
    -0.84
     kaarangay
    -0.81
    POSITIVE LOGITS
    <eos>
    0.67
    ↵↵
    0.54
    ‘
    0.50
    ...
    0.48
    ​
    0.47
      
    0.46
    </strong>
    0.45
    <em>
    0.45
    !
    0.44
    ↵↵↵
    0.42
    Activations Density 0.367%

    No Known Activations