© 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. 40212
    Prev
    Next
    INDEX
    Explanations

    possessive forms of nouns, particularly those indicating ownership or association

    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
    mw
    -0.07
    'icon
    -0.06
    ocy
    -0.06
    ocol
    -0.06
     respectively
    -0.06
    @(
    -0.06
    oc
    -0.06
     jugg
    -0.06
    hread
    -0.06
    ates
    -0.06
    POSITIVE LOGITS
     aim
    0.12
     goal
    0.10
    ureau
    0.09
    oret
    0.08
    缮
    0.08
    goal
    0.08
     mere
    0.07
     focus
    0.07
    ews
    0.07
    ificio
    0.07
    Activations Density 0.068%

    No Known Activations