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. OpenMOSS · Llama Scope: SAEs for Llama-3.1-8B
    3. Llama3.1-8B (Base)
    4. Residual Stream
    5. 16-LLAMASCOPE-RES-32K
    6. 1590
    Prev
    Next
    INDEX
    Explanations

    references to locations and events in Belgium

    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
    icut
    -0.16
     Petr
    -0.15
    vell
    -0.15
     Rajasthan
    -0.15
    qi
    -0.14
    ä¼
    -0.14
    builtin
    -0.14
     Gan
    -0.14
    aida
    -0.14
    ufe
    -0.14
    POSITIVE LOGITS
     Belgian
    0.40
     Belgium
    0.36
     Brussels
    0.33
     Belg
    0.32
     Flem
    0.30
     belg
    0.28
     Cincinnati
    0.22
    .be
    0.21
     Roose
    0.21
     Congo
    0.21
    Activations Density 0.071%

    No Known Activations