© 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. 4-LLAMASCOPE-RES-32K
    11. 12853
    Prev
    Next
    INDEX
    Explanations

    phrases relating to helpful or useful items and their attributes

    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
    341
    -0.19
    weets
    -0.16
     trick
    -0.16
    erah
    -0.16
    otypes
    -0.15
    gles
    -0.15
    æĵį
    -0.15
    ัà¸įà¸Ĭ
    -0.15
    adena
    -0.15
     tricks
    -0.15
    POSITIVE LOGITS
    annes
    0.15
    atio
    0.15
    imits
    0.15
    riere
    0.15
    -stack
    0.15
    rière
    0.15
    ATIO
    0.14
    acz
    0.14
    arella
    0.14
    usch
    0.14
    Activations Density 0.537%

    No Known Activations