© 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. 104062
    Prev
    Next
    INDEX
    Explanations

    phrases related to danger or a sense of urgency

    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
    ød
    -0.07
    braco
    -0.06
    herits
    -0.06
    aroo
    -0.06
     eldre
    -0.06
    ãĤĨ
    -0.06
     Mitarbeiter
    -0.06
    νοι
    -0.06
     PROT
    -0.06
    âľĶ
    -0.06
    POSITIVE LOGITS
     too
    0.09
     might
    0.07
    might
    0.07
     become
    0.07
    mare
    0.07
     Too
    0.06
    too
    0.06
     TOO
    0.06
    bec
    0.06
    479
    0.06
    Activations Density 0.039%

    No Known Activations