© 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. 31-LLAMASCOPE-RES-32K
    11. 2414
    Prev
    Next
    INDEX
    Explanations

    references to controversial social issues, particularly related to sexual conduct and rights

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     ‘
    -0.21
    util
    -0.21
     utilizing
    -0.20
     utilizes
    -0.20
    Util
    -0.19
     advancements
    -0.19
     utilized
    -0.18
     Util
    -0.18
     utilize
    -0.17
     residing
    -0.17
    POSITIVE LOGITS
     *
    0.25
    (*)
    0.25
     lots
    0.21
    [*
    0.21
     sort
    0.19
     explicitly
    0.19
     ~=
    0.18
     stuff
    0.18
     reasonably
    0.18
     pretty
    0.17
    Activations Density 1.310%

    No Known Activations