Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 9-LLAMASCOPE-RES-32K
    6. 20889
    Prev
    Next
    INDEX
    Explanations

    references to specific architectural locations or notable structures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.9.hook_resid_post
    Hook Layer
    9
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ecast
    -0.17
    alach
    -0.14
    alous
    -0.14
     conver
    -0.13
     tiers
    -0.13
    ani
    -0.13
    /tutorial
    -0.13
    ÑĢÑĥн
    -0.13
    xico
    -0.13
    xEF
    -0.13
    POSITIVE LOGITS
    esen
    0.15
    bilt
    0.14
    ional
    0.14
    شر
    0.14
    heets
    0.14
    esi
    0.14
    elder
    0.14
    allah
    0.14
    uddy
    0.13
    oses
    0.13
    Activations Density 0.377%

    No Known Activations