© 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. 81313
    Prev
    Next
    INDEX
    Explanations

    instances where different conjunctions and connectors are used to link ideas

    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
    olem
    -0.08
    udas
    -0.07
    radient
    -0.07
    avr
    -0.06
    864
    -0.06
    oles
    -0.06
    angu
    -0.06
    ÏĢή
    -0.06
    deen
    -0.06
    neau
    -0.06
    POSITIVE LOGITS
    erson
    0.10
     half
    0.08
    undry
    0.08
    half
    0.07
    reas
    0.07
    apos
    0.07
    hra
    0.07
    -half
    0.07
    alus
    0.07
    inke
    0.07
    Activations Density 0.026%

    No Known Activations