© 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. 5-LLAMASCOPE-RES-131K
    11. 126009
    Prev
    Next
    INDEX
    Explanations

    terms related to separation or avoidance

    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
     slik
    -0.06
    æ³ī
    -0.06
    ablish
    -0.06
    hack
    -0.06
    ollah
    -0.06
    yled
    -0.06
    exter
    -0.06
    aul
    -0.06
    ollo
    -0.06
    éºĹ
    -0.05
    POSITIVE LOGITS
    erre
    0.07
    tures
    0.07
    938
    0.06
    ائÙĤ
    0.06
     Geh
    0.06
    atism
    0.06
    lain
    0.06
    939
    0.06
    erson
    0.06
    721
    0.06
    Activations Density 0.001%

    No Known Activations