© 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. 80676
    Prev
    Next
    INDEX
    Explanations

    infinitive verb phrases, particularly those that suggest intention or purpose

    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
    trak
    -0.08
    ults
    -0.07
    γγ
    -0.07
    trinsic
    -0.07
    ÃŃsto
    -0.07
    anne
    -0.07
    iasi
    -0.07
    \application
    -0.07
    baÅŁ
    -0.07
     Yao
    -0.07
    POSITIVE LOGITS
     adequately
    0.07
     Proper
    0.07
    proper
    0.07
     anymore
    0.07
     yet
    0.07
     notice
    0.06
    æĦıä¹ī
    0.06
     ex
    0.06
     properly
    0.06
    BV
    0.06
    Activations Density 0.014%

    No Known Activations