© 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. 9-LLAMASCOPE-RES-131K
    11. 96748
    Prev
    Next
    INDEX
    Explanations

    phrases related to operational efficiency and reliability in machinery

    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
    ugh
    -0.08
    emie
    -0.08
    emies
    -0.07
    avr
    -0.07
    artner
    -0.07
    ë¸
    -0.07
    emmel
    -0.07
    /Gate
    -0.07
    ropoda
    -0.07
    codegen
    -0.07
    POSITIVE LOGITS
     intervention
    0.06
     selected
    0.06
    oby
    0.06
    κή
    0.06
     Sel
    0.06
    elay
    0.06
    -ing
    0.06
     Hi
    0.06
     sw
    0.06
     event
    0.06
    Activations Density 0.293%

    No Known Activations