© 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. 24-LLAMASCOPE-RES-32K
    11. 5670
    Prev
    Next
    INDEX
    Explanations

    code-related constructs and elements, particularly related to function definitions and conditional logic

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.24.hook_resid_post
    Hook Layer
    24
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    elocity
    -0.17
    pir
    -0.15
    ounding
    -0.15
    (æĹ¥
    -0.14
    getDoctrine
    -0.14
    ANGLE
    -0.14
    memset
    -0.14
    ovaly
    -0.14
    (""),
    -0.14
    /rs
    -0.13
    POSITIVE LOGITS
     &&
    0.51
     &&↵
    0.39
     AND
    0.33
    &&
    0.33
     ||
    0.32
     &&č↵
    0.25
     âĪ§
    0.24
     and
    0.24
     ||↵
    0.23
    ||
    0.22
    Activations Density 0.176%

    No Known Activations