© 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. 54736
    Prev
    Next
    INDEX
    Explanations

    themes of danger and risk 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
    .basic
    -0.08
    sek
    -0.07
    amaz
    -0.07
    ail
    -0.06
     Demir
    -0.06
    anik
    -0.06
    ammer
    -0.06
    /basic
    -0.06
     Coin
    -0.06
    enville
    -0.06
    POSITIVE LOGITS
     danger
    0.12
     tendency
    0.11
     traps
    0.11
     trap
    0.10
     dangers
    0.10
     temptation
    0.10
     tendencies
    0.10
     Danger
    0.09
     too
    0.09
     pitfalls
    0.09
    Activations Density 0.053%

    No Known Activations