© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    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. 2721
    Prev
    Next
    INDEX
    Explanations

    references to political repression and movements advocating for democracy in Swaziland

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

    Text after apostrophes or numbers

    np_acts-logits-general · gemini-2.0-flash

    genus Mabuya

    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
    psack
    -0.54
     lenker
    -0.47
    featureID
    -0.46
    ogast
    -0.46
     galo
    -0.45
    Violin
    -0.45
     propeller
    -0.45
     JAXB
    -0.45
     Nutr
    -0.45
    RepeatedField
    -0.44
    POSITIVE LOGITS
    BufferException
    0.95
     StatefulWidget
    0.66
    Skocz
    0.65
    الحياه
    0.61
     gyhoeddwyd
    0.61
     throw
    0.60
    protoimpl
    0.59
    findpost
    0.57
     Snakes
    0.56
    seamnă
    0.56
    Activations Density 0.141%

    No Known Activations