© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. OpenMOSS · Llama Scope: SAEs for Llama-3.1-8B
    5. Llama3.1-8B (Base)
    7. Residual Stream
    9. 2-LLAMASCOPE-RES-32K
    11. 29866
    Prev
    Next
    INDEX
    Explanations

    common phrases that express visibility or perception, especially in terms of sharing information

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.2.hook_resid_post
    Hook Layer
    2
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    iani
    -0.15
    hift
    -0.14
    ControlItem
    -0.14
    azel
    -0.14
    vpn
    -0.14
    itele
    -0.14
    çļĦè¯Ŀ
    -0.14
     Heritage
    -0.14
    iyan
    -0.13
    legen
    -0.13
    POSITIVE LOGITS
    osaur
    0.17
    éIJĺ
    0.16
    atti
    0.14
     Gil
    0.13
     constexpr
    0.13
    asio
    0.13
    orative
    0.13
    _pointer
    0.13
     fat
    0.13
    obby
    0.13
    Activations Density 0.111%

    No Known Activations