© 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. 41751
    Prev
    Next
    INDEX
    Explanations

    phrases that describe actions related to surprise or unexpected encounters

    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
    ,[],
    -0.08
     succ
    -0.07
    اÙĪرÛĮ
    -0.07
     Ú©ÛĮÙģ
    -0.07
    isoft
    -0.07
    Pg
    -0.06
    .experimental
    -0.06
    çak
    -0.06
    .utf
    -0.06
    awe
    -0.06
    POSITIVE LOGITS
     ready
    0.10
     doing
    0.08
     under
    0.07
     Ready
    0.07
    Ready
    0.07
     trying
    0.07
    ready
    0.06
    /net
    0.06
     near
    0.06
    ariant
    0.06
    Activations Density 0.035%

    No Known Activations