© 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. 9634
    Prev
    Next
    INDEX
    Explanations

    references to the "Star Wars" franchise and its associated content

    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
    /XML
    -0.18
    .XML
    -0.17
    çĵ
    -0.16
    rob
    -0.16
    ATRIX
    -0.16
    oulos
    -0.15
    rud
    -0.15
     Nicar
    -0.15
     Colombian
    -0.15
    ignon
    -0.15
    POSITIVE LOGITS
     Rey
    0.39
     Ky
    0.38
     Finn
    0.33
     Sno
    0.32
     Poe
    0.31
    Ky
    0.30
     Ren
    0.29
     BB
    0.28
     Leia
    0.28
     Maz
    0.27
    Activations Density 0.012%

    No Known Activations