Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. OpenMOSS · Llama Scope: SAEs for Llama-3.1-8B
    3. Llama3.1-8B (Base)
    4. Residual Stream
    5. 24-LLAMASCOPE-RES-32K
    6. 19030
    Prev
    Next
    INDEX
    Explanations

    references to discovering methods or solutions

    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
    mits
    -0.14
    bakan
    -0.13
     jist
    -0.13
    .Foundation
    -0.13
    ethyl
    -0.12
    女åŃIJ
    -0.12
    gid
    -0.12
    jectories
    -0.12
    collections
    -0.12
    avin
    -0.12
    POSITIVE LOGITS
     ways
    0.88
     Ways
    0.66
    ways
    0.60
     way
    0.47
    WAYS
    0.44
    eways
    0.39
     ÑģпоÑģ
    0.37
     methods
    0.36
    åĬŀæ³ķ
    0.36
    .way
    0.35
    Activations Density 0.137%

    No Known Activations