© 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. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 13608
    Prev
    Next
    INDEX
    Explanations

    trust

    np_max-act · gemini-2.0-flash

    The neuron fires on occurrences of the word “trust” (and its close variants like trustee, settlor, etc.) in legal or tax‐planning contexts.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.11.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     vẽ
    -0.07
    áce
    -0.07
     mannen
    -0.06
     대부분
    -0.06
     ціл
    -0.06
     False
    -0.06
    ContentSize
    -0.06
     equality
    -0.06
     genital
    -0.06
     brigade
    -0.06
    POSITIVE LOGITS
     trust
    0.09
     Trust
    0.09
     trusts
    0.08
    -us
    0.07
    purpose
    0.07
    ブ
    0.07
    Reflect
    0.07
     sentimental
    0.06
    307
    0.06
     تع
    0.06
    Activations Density 0.004%

    No Known Activations