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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 119117
    Prev
    Next
    INDEX
    Explanations

    terms associated with violent incidents and criminal activity

    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
     YYSTYPE
    -0.07
     Unsafe
    -0.07
    .getRaw
    -0.07
    umpt
    -0.07
    fcn
    -0.06
    едж
    -0.06
    irma
    -0.06
     Ago
    -0.06
    orget
    -0.06
    Sleep
    -0.06
    POSITIVE LOGITS
     targeting
    0.07
    uÄį
    0.06
    ello
    0.06
     Pé
    0.06
    ì´Ŀ
    0.06
     Fors
    0.06
     Eigen
    0.06
     Reid
    0.05
     Perez
    0.05
    óm
    0.05
    Activations Density 0.011%

    No Known Activations