© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 75435
    Prev
    Next
    INDEX
    Explanations

    instances of the word "that" in various contexts, often emphasizing its significance in different phrases

    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
    imbus
    -0.07
    ypy
    -0.07
    æĢ
    -0.06
    ãĥ¼ãĥ
    -0.06
    rof
    -0.06
    ette
    -0.06
     BILL
    -0.06
     wur
    -0.06
    gne
    -0.06
     Kup
    -0.06
    POSITIVE LOGITS
     much
    0.08
    much
    0.08
     anymore
    0.07
    ulong
    0.07
    away
    0.07
     necessarily
    0.07
    abouts
    0.07
    ley
    0.07
    respect
    0.07
    Much
    0.07
    Activations Density 0.005%

    No Known Activations