© 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 EvalsExportsinterp-engineNEWAPI 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. 95454
    Prev
    Next
    INDEX
    Explanations

    phrases related to software performance issues and errors

    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
     Pale
    -0.07
     Eff
    -0.07
     Palestin
    -0.07
    èŀ
    -0.06
    .Void
    -0.06
     enthusi
    -0.06
     redund
    -0.06
    ãĤ¹ãĤ¯
    -0.06
     desper
    -0.06
    egin
    -0.06
    POSITIVE LOGITS
    UserCode
    0.10
    afari
    0.07
     guar
    0.07
     occasional
    0.07
    uisse
    0.07
    ouro
    0.07
     occasionally
    0.07
     repro
    0.07
    ragen
    0.06
     reclaim
    0.06
    Activations Density 0.016%

    No Known Activations