© 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. 28195
    Prev
    Next
    INDEX
    Explanations

    instances of high-value or significant elements in various contexts

    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
    istrovstvÃŃ
    -0.08
    ÄĽr
    -0.08
    ì§Ģëħ¸
    -0.08
     bordel
    -0.08
     dilig
    -0.07
     resil
    -0.07
     porr
    -0.07
     smarty
    -0.07
     poil
    -0.07
    alnız
    -0.07
    POSITIVE LOGITS
    âĢ¦↵↵↵
    0.07
    InThe
    0.06
     overposting
    0.06
    /from
    0.06
     âĢ¦
    0.06
    izens
    0.06
    vertise
    0.06
     herself
    0.05
    ropdown
    0.05
    blings
    0.05
    Activations Density 0.354%

    No Known Activations