© 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. 54342
    Prev
    Next
    INDEX
    Explanations

    various forms of the word "sink" or related concepts of sinking and submerging

    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
     windshield
    -0.06
    icina
    -0.06
    OLF
    -0.06
    tuk
    -0.06
    anmar
    -0.06
    prox
    -0.06
    nant
    -0.06
    ãĥĪãĥ«
    -0.06
    arf
    -0.06
    lifetime
    -0.06
    POSITIVE LOGITS
    holes
    0.08
    hole
    0.08
     sunk
    0.07
    heim
    0.07
    aka
    0.07
     deep
    0.07
    bound
    0.07
    ieder
    0.07
    /dd
    0.07
     deeper
    0.06
    Activations Density 0.006%

    No Known Activations