© 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. 125517
    Prev
    Next
    INDEX
    Explanations

    concessive phrases that introduce contrast or qualify statements

    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
    oÄŁ
    -0.07
    ;
    -0.07
    ensch
    -0.07
    itt
    -0.06
    eward
    -0.06
    incerely
    -0.06
    emme
    -0.06
     åĽ
    -0.06
    endoza
    -0.06
     conf
    -0.06
    POSITIVE LOGITS
    teness
    0.07
    fila
    0.07
    stein
    0.06
    ÑģÑĤин
    0.06
    adÃŃ
    0.06
     occasionally
    0.06
    ãģłãģ£ãģ¦
    0.06
    isku
    0.06
    tid
    0.06
    acak
    0.06
    Activations Density 0.033%

    No Known Activations