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

    Neuronpedia

    Jacobian LensNEW
    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. 17-LLAMASCOPE-RES-131K
    11. 87505
    Prev
    Next
    INDEX
    Explanations

    instances of the word "sh" followed by varying contexts, indicating a focus on shock or surprise expressions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    strup
    -0.09
    stru
    -0.08
    hend
    -0.08
    arend
    -0.07
    mpar
    -0.07
    imbus
    -0.07
    .resp
    -0.07
    wend
    -0.07
    iaux
    -0.07
    ysa
    -0.07
    POSITIVE LOGITS
    es
    0.07
     warm
    0.07
     sh
    0.06
     Bones
    0.06
    sh
    0.06
    allow
    0.06
    ales
    0.06
     Force
    0.06
     hybrid
    0.06
     coun
    0.05
    Activations Density 0.009%

    No Known Activations