© 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. 5431
    Prev
    Next
    INDEX
    Explanations

    phrases expressing preference or comparison

    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
    ito
    -0.07
    rac
    -0.06
    oha
    -0.06
    ña
    -0.06
     cro
    -0.06
     Ø®ÙĦ
    -0.06
    anes
    -0.06
     Gra
    -0.05
    ras
    -0.05
    yme
    -0.05
    POSITIVE LOGITS
     rather
    0.18
    rather
    0.16
     Rather
    0.15
    Rather
    0.15
     plutôt
    0.13
     пÑĢедпоÑĩ
    0.12
    Better
    0.12
     prefer
    0.12
     preference
    0.12
     preferable
    0.11
    Activations Density 0.089%

    No Known Activations