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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 16642
    Prev
    Next
    INDEX
    Explanations

    variations of the word "can" and its negations in different 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
    RIX
    -0.07
    zcze
    -0.07
    gence
    -0.07
    olars
    -0.07
    hta
    -0.06
    loth
    -0.06
     gratuita
    -0.06
    веÑģÑĤи
    -0.06
    ATS
    -0.06
    APT
    -0.06
    POSITIVE LOGITS
    رخ
    0.07
    ’t
    0.07
    _cpp
    0.07
    't
    0.07
    º
    0.06
     Lump
    0.06
    rab
    0.06
    ÑĢовод
    0.06
    itational
    0.06
     haz
    0.06
    Activations Density 0.020%

    No Known Activations