© 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. 104524
    Prev
    Next
    INDEX
    Explanations

    phrases that express contrast or negation, often in relation to expectations versus reality

    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
    adar
    -0.07
    .dt
    -0.07
    enko
    -0.06
    جا
    -0.06
     Bind
    -0.06
    hi
    -0.06
    ran
    -0.06
    olle
    -0.06
     cant
    -0.06
    WithError
    -0.06
    POSITIVE LOGITS
    ä¸įæĺ¯
    0.15
     neither
    0.14
     wasn
    0.11
     Neither
    0.11
     NOT
    0.11
     नह
    0.11
    ãģ§ãģ¯ãģªãģı
    0.11
    Neither
    0.11
     nicht
    0.10
     isn
    0.10
    Activations Density 0.046%

    No Known Activations