© 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. 9-LLAMASCOPE-RES-32K
    11. 3036
    Prev
    Next
    INDEX
    Explanations

    names of individuals and specific entities related to accountability and trust

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.9.hook_resid_post
    Hook Layer
    9
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Jun
    -0.17
     feed
    -0.16
     Astr
    -0.16
    elman
    -0.15
    BJECT
    -0.15
     action
    -0.15
     occasion
    -0.14
     pocket
    -0.14
    rios
    -0.14
     Gibson
    -0.14
    POSITIVE LOGITS
    <source
    0.16
    Bat
    0.14
    itchens
    0.14
    .enum
    0.14
    wear
    0.14
    樹
    0.14
    项
    0.14
    ych
    0.14
    lator
    0.13
     Processes
    0.13
    Activations Density 0.008%

    No Known Activations