© 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. 5923
    Prev
    Next
    INDEX
    Explanations

    instances of emphasis and attention-grabbing punctuation or symbols

    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
    iams
    -0.07
    trys
    -0.06
    uir
    -0.06
    ypass
    -0.06
    auge
    -0.06
    uin
    -0.06
    aces
    -0.06
    icken
    -0.05
    descr
    -0.05
    ãĥĥ
    -0.05
    POSITIVE LOGITS
    istrovstvÃŃ
    0.08
    -REAL
    0.07
    åĿĽ
    0.07
    .FlatStyle
    0.07
    aura
    0.07
    ncy
    0.07
     Gardner
    0.06
    UrlParser
    0.06
    etsk
    0.06
    ξε
    0.06
    Activations Density 0.001%

    No Known Activations