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

    Neuronpedia

    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. 4793
    Prev
    Next
    INDEX
    Explanations

    URLs, specifically those with the domain "github.com" and its various subdirectories

    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
    arine
    -0.16
    nearest
    -0.15
    semb
    -0.15
    ighbor
    -0.15
    ;base
    -0.14
    æīĢ
    -0.14
    LOB
    -0.14
    rani
    -0.14
    idious
    -0.14
    udiant
    -0.13
    POSITIVE LOGITS
    ellen
    0.18
    anton
    0.17
    anto
    0.17
    ?q
    0.16
    phil
    0.15
    addy
    0.15
    oss
    0.15
     ÙĨØŃ
    0.14
    antly
    0.14
     mat
    0.14
    Activations Density 0.031%

    No Known Activations