© 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. 10-LLAMASCOPE-RES-32K
    11. 21778
    Prev
    Next
    INDEX
    Explanations

    phrases related to legal acts and regulations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 10-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L10R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    âĢ¦↵
    -0.16
    âĢ¦↵↵
    -0.13
     wireType
    -0.13
     [âĢ¦]↵
    -0.12
     âĢ¦↵
    -0.12
    .fromFunction
    -0.12
     eskort
    -0.12
    swire
    -0.12
    _Renderer
    -0.12
    ...↵
    -0.12
    POSITIVE LOGITS
     chatte
    0.14
    uliar
    0.14
    displayText
    0.13
    apore
    0.13
    isos
    0.13
    /preferences
    0.13
    _unicode
    0.13
    pNet
    0.13
    vrier
    0.13
    skb
    0.12
    Activations Density 4.449%

    No Known Activations