© 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. 16-LLAMASCOPE-RES-32K
    11. 8123
    Prev
    Next
    INDEX
    Explanations

    references to personal experiences and professional journeys

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    elm
    -0.18
    ег
    -0.15
     Mec
    -0.15
     us
    -0.14
    odo
    -0.14
    ilia
    -0.14
    plx
    -0.14
    ien
    -0.14
    egas
    -0.14
    pedia
    -0.14
    POSITIVE LOGITS
     my
    0.30
     myself
    0.29
    æĪijçļĦ
    0.28
     saya
    0.25
    æĪij
    0.23
     minha
    0.23
     mijn
    0.23
    ï¼ĮæĪij
    0.22
    my
    0.21
     моÑı
    0.21
    Activations Density 0.285%

    No Known Activations