© 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. 84440
    Prev
    Next
    INDEX
    Explanations

    words and phrases related to endings or conclusions 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
    jer
    -0.08
    ptal
    -0.07
    odos
    -0.07
    opers
    -0.06
    ĨĴ
    -0.06
    uta
    -0.06
    yle
    -0.06
    iper
    -0.06
    ource
    -0.06
    otte
    -0.06
    POSITIVE LOGITS
     middle
    0.10
     end
    0.10
     ÑģеÑĢед
    0.09
    middle
    0.08
     finish
    0.08
     final
    0.08
     towards
    0.08
     toward
    0.08
     finale
    0.08
    final
    0.08
    Activations Density 0.027%

    No Known Activations