© 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. 18-LLAMASCOPE-RES-32K
    11. 5394
    Prev
    Next
    INDEX
    Explanations

    keywords and terms related to specific categories or subjects, often with emphasis on costs, classifications, and items in a list format

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ÃŃas
    -0.15
    áp
    -0.14
    ï¼Ĺ
    -0.13
    ĮĴ
    -0.13
     )↵↵↵↵↵↵↵↵
    -0.13
    oned
    -0.13
    泡
    -0.13
    usercontent
    -0.12
    .writeValue
    -0.12
    à¸Ĺร
    -0.12
    POSITIVE LOGITS
    린ìĿ´
    0.18
    /trunk
    0.16
    umba
    0.14
    rial
    0.14
    vertising
    0.14
     queryInterface
    0.13
    bsolute
    0.13
    sumer
    0.13
    Ïĥμο
    0.13
    echa
    0.13
    Activations Density 0.373%

    No Known Activations