© 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. 21-LLAMASCOPE-RES-32K
    11. 17046
    Prev
    Next
    INDEX
    Explanations

    phrases indicating deception or manipulated narratives

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    666
    -0.15
     underrated
    -0.15
    rist
    -0.15
    lfw
    -0.15
    uder
    -0.15
     Pist
    -0.14
    308
    -0.14
     selfish
    -0.14
     underst
    -0.14
    oge
    -0.13
    POSITIVE LOGITS
     fabrication
    0.24
     invention
    0.23
     invented
    0.23
    fabric
    0.23
     fiction
    0.22
     inventions
    0.21
     Fabric
    0.21
     fabric
    0.20
     fabricated
    0.20
     invent
    0.19
    Activations Density 0.178%

    No Known Activations