© 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. 24-LLAMASCOPE-RES-32K
    11. 4272
    Prev
    Next
    INDEX
    Explanations

    references to up-to-date information and resources

    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
    Uvs
    -0.14
    fter
    -0.14
    ίκη
    -0.14
    ela
    -0.14
    atha
    -0.13
    336
    -0.13
    ssp
    -0.13
    hcp
    -0.13
    edx
    -0.13
    achment
    -0.13
    POSITIVE LOGITS
     latest
    0.77
    latest
    0.66
    æľĢæĸ°
    0.65
     up
    0.63
    Latest
    0.60
     Latest
    0.59
     current
    0.56
     newest
    0.54
    /latest
    0.53
    _latest
    0.52
    Activations Density 0.196%

    No Known Activations