© 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
    3. OpenMOSS · Llama Scope: SAEs for Llama-3.1-8B
    5. Llama3.1-8B (Base)
    7. Residual Stream
    9. 30-LLAMASCOPE-RES-32K
    11. 12436
    Prev
    Next
    INDEX
    Explanations

    references to nationality or ethnic identities related to consumers and brands

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.30.hook_resid_post
    Hook Layer
    30
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    bidden
    -0.19
     british
    -0.19
     British
    -0.18
     chinese
    -0.18
     Japanese
    -0.17
     Italian
    -0.17
    ä¸Ńåľĭ
    -0.17
     Danish
    -0.16
     Polish
    -0.16
     French
    -0.16
    POSITIVE LOGITS
    -American
    0.40
    -speaking
    0.38
    -born
    0.37
    -language
    0.34
    -Americans
    0.33
    -made
    0.30
    /E
    0.29
    -Russian
    0.28
    /A
    0.28
    -origin
    0.28
    Activations Density 0.300%

    No Known Activations