© 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. 31-LLAMASCOPE-RES-32K
    11. 17569
    Prev
    Next
    INDEX
    Explanations

    instances of the word "complicit" or related variations, indicating a focus on complicity in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    _compress
    -0.18
     halde
    -0.17
    833
    -0.15
    anmar
    -0.15
     hâl
    -0.15
    addtogroup
    -0.15
    ippy
    -0.15
     Rooney
    -0.15
    ÑĭÑĪ
    -0.15
     Bare
    -0.15
    POSITIVE LOGITS
    ounding
    0.33
    licit
    0.31
    licity
    0.31
    licate
    0.29
    ulsion
    0.29
    atri
    0.27
    end
    0.25
    action
    0.24
    unction
    0.24
    acts
    0.24
    Activations Density 0.006%

    No Known Activations