© 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. 78109
    Prev
    Next
    INDEX
    Explanations

    words related to halting or pausing actions

    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
    mite
    -0.07
    á»į
    -0.07
    aln
    -0.07
    arel
    -0.07
     âĦĿ
    -0.06
    еÑĢж
    -0.06
    ucch
    -0.06
    anker
    -0.06
    irsch
    -0.06
    ardi
    -0.06
    POSITIVE LOGITS
    AGMA
    0.07
    axon
    0.06
    AllWindows
    0.06
    545
    0.06
    525
    0.06
     ZákladnÃŃ
    0.06
    275
    0.06
     Branch
    0.06
    ä¸Ģä¸ĭ
    0.06
    åŃĿ
    0.06
    Activations Density 0.003%

    No Known Activations