© 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. 22-LLAMASCOPE-RES-32K
    11. 3603
    Prev
    Next
    INDEX
    Explanations

    references to social norms and traditional practices

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    éĽĦ
    -0.14
    DOI
    -0.14
    даÑħ
    -0.13
    OrCreate
    -0.13
     recurring
    -0.13
    evil
    -0.13
     feed
    -0.12
    ãģ¤ãģ¶
    -0.12
    ahas
    -0.12
    anki
    -0.12
    POSITIVE LOGITS
     etiquette
    0.36
     etiqu
    0.31
     protocol
    0.30
     manners
    0.28
     courtesy
    0.26
     Courtesy
    0.24
     respect
    0.24
    protocol
    0.24
     polite
    0.24
    Courtesy
    0.24
    Activations Density 0.259%

    No Known Activations