© 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. 54902
    Prev
    Next
    INDEX
    Explanations

    phrases that indicate intent or desire, particularly associated with actions and initiatives

    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
     cellFor
    -0.07
    ÙĪتÛĮ
    -0.07
    ÑħодиÑĤÑĮ
    -0.07
    оже
    -0.07
    .asInstanceOf
    -0.07
    cent
    -0.07
    enance
    -0.07
    uous
    -0.06
    ah
    -0.06
    λÏį
    -0.06
    POSITIVE LOGITS
    omu
    0.06
    ambre
    0.06
    omore
    0.06
    UNG
    0.06
    çĸ
    0.06
    mlink
    0.06
    lector
    0.06
     hearing
    0.06
    ILED
    0.06
    obar
    0.05
    Activations Density 0.031%

    No Known Activations