Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 24-GEMMASCOPE-RES-16K
    6. 2341
    Prev
    Next
    INDEX
    Explanations

    geographical references to South American flora and fauna, particularly related to crops and their origins

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

    South American locations/culture

    np_acts-logits-general · gemini-2.0-flash

    South America and Peru

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.24.hook_resid_post
    Hook Layer
    24
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Crimea
    -0.52
    helia
    -0.52
    transQ
    -0.52
    Fprintf
    -0.51
     Gees
    -0.50
    клопе
    -0.48
     rubles
    -0.48
     ISRAEL
    -0.48
    ectady
    -0.47
     Irán
    -0.46
    POSITIVE LOGITS
     Peru
    0.97
     Peruvian
    0.93
     Andean
    0.91
     Amazon
    0.86
    Peru
    0.86
     Andes
    0.79
    Amazon
    0.77
     peru
    0.77
     Lima
    0.76
     Cusco
    0.74
    Activations Density 0.215%

    No Known Activations