© 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. 4-LLAMASCOPE-RES-32K
    11. 14797
    Prev
    Next
    INDEX
    Explanations

    specific words related to surgical procedures and modifications

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.4.hook_resid_post
    Hook Layer
    4
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    .iOS
    -0.14
    intval
    -0.14
     churn
    -0.14
    aly
    -0.13
     Brennan
    -0.13
    ively
    -0.13
     Gl
    -0.13
    ä¿Ĺ
    -0.13
    "]."
    -0.13
    lt
    -0.13
    POSITIVE LOGITS
    abbo
    0.16
    \grid
    0.14
    oker
    0.14
    arus
    0.14
    izik
    0.14
    unkt
    0.14
    atitude
    0.14
    /Gate
    0.14
    ẩn
    0.14
     grayscale
    0.13
    Activations Density 0.045%

    No Known Activations