© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 14126
    Prev
    Next
    INDEX
    Explanations

    punctuation marks, particularly quotation marks and apostrophes

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Negative Logits
    пÑĢимеÑĢ
    -0.07
     Wie
    -0.07
     Heck
    -0.06
     Carp
    -0.06
     ob
    -0.06
     comp
    -0.06
    hq
    -0.06
     comb
    -0.06
     necessarily
    -0.06
    etri
    -0.05
    POSITIVE LOGITS
    urement
    0.07
     wich
    0.07
    'gc
    0.07
    witter
    0.07
    .Metro
    0.07
    à¸ī
    0.07
    agate
    0.07
    which
    0.07
    ernals
    0.07
    reeze
    0.07
    Activations Density 0.043%

    No Known Activations