© 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. 87140
    Prev
    Next
    INDEX
    Explanations

    titles and names associated with various roles or professions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    Person-descriptor titles or roles that typically end in "er", "man", "guy", "ai", "boy", or similar suffixes indicating a person who performs a specific function, job or role.

    eleuther_acts_top20 · claude-3-7-sonnet-20250219Triggered by @lu-christina
    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
    vale
    -0.08
    ildo
    -0.08
    ajs
    -0.08
    ần
    -0.07
    glas
    -0.07
    pear
    -0.07
    bris
    -0.07
    /perl
    -0.07
    ãĥĨãĥ«
    -0.07
    roje
    -0.07
    POSITIVE LOGITS
    hood
    0.10
     himself
    0.09
     बनन
    0.08
    ship
    0.08
     whom
    0.07
     who
    0.07
    ess
    0.07
    thood
    0.07
    esses
    0.07
    身份
    0.07
    Activations Density 0.065%

    No Known Activations