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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 57560
    Prev
    Next
    INDEX
    Explanations

    phrases that indicate arguments or cases being made, especially in a comparative context

    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
    éric
    -0.08
     Passage
    -0.07
     propagate
    -0.07
    erece
    -0.07
    InputDialog
    -0.07
     trú
    -0.07
    éf
    -0.07
    udur
    -0.07
    érica
    -0.07
    Cascade
    -0.07
    POSITIVE LOGITS
     argument
    0.21
     arg
    0.19
     argue
    0.19
    arg
    0.18
     Argument
    0.18
     argued
    0.18
     arguments
    0.17
    argument
    0.17
     Arg
    0.16
     argues
    0.16
    Activations Density 0.024%

    No Known Activations