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. 3-GEMMASCOPE-RES-16K
    6. 6023
    Prev
    Next
    INDEX
    Explanations

    references to specific countries and political contexts in Central and East Africa

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.3.hook_resid_post
    Hook Layer
    3
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ínsula
    -0.55
    Aix
    -0.54
     PyTuple
    -0.53
    Espan
    -0.52
     QName
    -0.51
    clientY
    -0.51
     deriva
    -0.50
    DNEY
    -0.50
     hObject
    -0.49
    Népesség
    -0.48
    POSITIVE LOGITS
     Haiti
    1.44
     Afghanistan
    1.40
     Liberia
    1.39
     Rwanda
    1.37
    Haiti
    1.36
     Bhutan
    1.36
     Burundi
    1.32
     Kosovo
    1.32
     Afghan
    1.29
     Haitian
    1.27
    Activations Density 0.422%

    No Known Activations